<date> ... ... http://opensolaris.pastebin.ca/702814 gisburn Sep 19 04:51:21 CET 2007 ... http://opensolaris.pastebin.ca/702822 gisburn Sep 19 05:02:52 CET 2007 -- snip -- Finally you can get strings/data/text/etc. from http://opensolaris.pastebin.ca using... -- snip -- $ shnote get http://opensolaris.pastebin.ca/702814 # Record name is '702814' demo #8932001 -- snip -- ...or just short: -- snip -- $ shnote get 702814 # Record name is '702814' demo #8932001 -- snip -- Suggestions/comments/rants/etc. welcome... :-) ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) -------------- next part -------------- #!/bin/ksh93 # # CDDL HEADER START # # The contents of this file are subject to the terms of the # Common Development and Distribution License (the "License"). # You may not use this file except in compliance with the License. # # You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE # or http://www.opensolaris.org/os/licensing. # See the License for the specific language governing permissions # and limitations under the License. # # When distributing Covered Code, include this CDDL HEADER in each # file and include the License file at usr/src/OPENSOLARIS.LICENSE. # If applicable, add the following below this CDDL HEADER, with the # fields enclosed by brackets "[]" replaced with your own identifying # information: Portions Copyright [yyyy] [name of copyright owner] # # CDDL HEADER END # # # Copyright 2007 Sun Microsystems, Inc. All rights reserved. # Use is subject to license terms. # # ident "%Z%%M% %I% %E% SMI" # # Solaris needs /usr/xpg4/bin/ because the tools in /usr/bin are not POSIX-conformant export PATH=/usr/xpg4/bin:/bin:/usr/bin # Make sure all math stuff runs in the "C" locale to avoid problems with alternative # radix point representations (e.g. ',' instead of '.' in de_DE.*-locales). This # needs to be set _before_ any floating-point constants are defined in this script) if [[ "${LC_ALL}" != "" ]] ; then export \ LC_MONETARY="${LC_ALL}" \ LC_MESSAGES="${LC_ALL}" \ LC_COLLATE="${LC_ALL}" \ LC_CTYPE="${LC_ALL}" unset LC_ALL fi export LC_NUMERIC=C function fatal_error { print -u 2 "${progname}: $@" exit 1 } # fixme: use "stat" builtin instead of "ls" function get_filesize { typeset filename="$1" typeset dummy # dummy string integer filesize=-1 if [[ -f "${filename}" ]] ; then ls -lb "${filename}" | read dummy dummy dummy dummy filesize dummy fi print -- ${filesize} return 0 } function encode_multipart_form_data { #set -o xtrace nameref formdata=$1 nameref content="${1}.content" # ToDO: should be |content="formdata.content"| integer numformelements=${#formdata.form[*]} integer i content="" # todo: add support to upload files for (( i=0 ; i < numformelements ; i++ )) ; do nameref element="${1}.form[${i}]" # ToDo: should be |element="formdata.form[${i}]"| content+="--${formdata.boundary}\n" content+="Content-Disposition: form-data; name=\"${element.name}\"\n" content+="\n" content+="${element.data}\n" # fixme: may need encoding for non-ASCII data done formdata.content_length=${#content} # add content tail (which MUST not be added to the content length) content+="--${formdata.boundary}--\n" return 0 } # parse return code, cookies etc. function parse_http_response { nameref response=$1 typeset h c i IFS=$' \t' read -r h c # read HTTP/1.[01] <code> [[ "$h" != ~(Eil)HTTP/.* ]] && { print -u2 "$0: HTTP/ header missing" ; return 1 ; } response.httpcode="$c" # skip remaining headers while IFS='' read -r i ; do [[ "$i" = $'\r' ]] && break done return 0 } function history_write_record { # rec: history record: # rec.title # rec.description # rec.provider # rec.providertoken # rec.url nameref rec="$1" integer histfd mkdir -p "${HOME}/.shnote" { # write a single-line record which can be read # as a compound variable back into the shell printf "title=%q description=%q date=%q provider=%q providertoken=%q url=%q\n" \ "${rec.title}" \ "${rec.description}" \ "$(date)" \ "${rec.provider}" \ "${rec.providertoken}" \ "${rec.url}" } >>"${history_file}" return $? } function print_history { integer histfd # http stream number typeset line # default output format is: # <access url>/<title> <date> <access url> [[ "$1" = "-l" ]] || printf "# %s\t\t\t\t\t%s\t%s\n" "<url>" "<title>" "<date>" # no history file ? if [[ ! -f "${history_file}" ]] ; then return 0 fi # open history file exec {histfd}<>"${history_file}" while read -u${histfd} line ; do typeset rec=() # This is a bit messy - "eval" should be avoided but there is currently # no better way to read a serialised compound variable from a file back # into the shell. eval "rec=( $line )" if [[ "$1" = "-l" ]] ; then print -- "${rec}" else printf "%s\t%s\t%s\n" "${rec.url}" "${rec.title}" "${rec.date}" fi unset rec done # close history file exec {histfd}<&- } function put_note_pastebin_ca { #set -o xtrace # key to autheticate this script against pastebin.ca typeset -r pastebin_ca_key="9CFXFyeNC3iga/vthok75kTBu5kSSLPD" # site setup typeset url_host="opensolaris.pastebin.ca" typeset url_path="/quiet-paste.php?api=${pastebin_ca_key}" typeset url="http://${url_host}${url_path}" integer netfd # http stream number # argument for "encode_multipart_form_data" typeset mimeform=( # input typeset boundary typeset -a form # output typeset content integer content_length ) typeset request="" typeset content="" typeset -r boundary="--------shnote_${RANDOM}_Xfish_${RANDOM}_Yeats_${RANDOM}_Zchicken_${RANDOM}monster_--------" # ToDo: Use "mimeform.form+=(" once ksh93 has been fixed. mimeform.boundary="${boundary}" mimeform.form[0]=( name="name" data="${LOGNAME}" ) mimeform.form[1]=( name="expiry" data="Never" ) mimeform.form[2]=( name="type" data="1" ) mimeform.form[3]=( name="description" data="logname=${LOGNAME};hostname=$(hostname);date=$(date)" ) mimeform.form[4]=( name="content" data="$1" ) encode_multipart_form_data mimeform content="${mimeform.content}" request="POST ${url_path} HTTP/1.1\n" request+="Host: ${url_host}\n" request+="User-Agent: ${http_user_agent}\n" request+="Connection: close\n" request+="Content-Type: multipart/form-data; boundary=${boundary}\n" request+="Content-Length: $(( ${mimeform.content_length} ))\n" exec {netfd}<>"/dev/tcp/${url_host}/80" (( $? != 0 )) && { print -u2 "Couldn't open connection to ${url_host}." ; return 1 ; } # send http post { print -- "${request}" print -- "${content}" } >&${netfd} # process reply parse_http_response httpresponse <&${netfd} response="$(cat <&${netfd})" # close connection exec {netfd}<&- if [[ "${response}" = ~(E).*SUCCESS.* ]] ; then typeset response_token="${response/~(E).*SUCCESS:/}" print "SUCCESS: http://opensolaris.pastebin.ca/${response_token}" # write history entry typeset histrec=( title="${mimeform.form[0].data}" description="${mimeform.form[3].data}" providertoken="${response_token}" provider="opensolaris.pastebin.ca" url="http://opensolaris.pastebin.ca/${response_token}" ) history_write_record histrec else print "ERROR: ${response}" return 1 fi return 0 } function get_note_pastebin_ca { typeset recordname="$1" integer netfd # http stream number case "${recordname}" in ~(Elr)[0-9][0-9]*) # pass-through ;; ~(Elr)http://opensolaris.pastebin.ca/raw/[0-9]*) recordname="${recordname/~(El)http:\/\/opensolaris.pastebin.ca\/raw\//}" ;; ~(Elr)http://opensolaris.pastebin.ca/[0-9]*) recordname="${recordname/~(El)http:\/\/opensolaris.pastebin.ca\//}" ;; *) fatal_error "Unsupported record name ${recordname}." esac print -u2 -f "# Record name is '%s'\n" "${recordname}" typeset url_host="opensolaris.pastebin.ca" typeset url_path="/raw/${recordname}" typeset url="http://${url_host}${url_path}" # I hereby curse Solaris for not having an entry for "http" in /etc/services # open TCP channel exec {netfd}<>"/dev/tcp/${url_host}/80" (( $? != 0 )) && { print -u2 "Couldn't open connection to ${url_host}." ; return 1 ; } # send HTTP request request="GET ${url_path} HTTP/1.1\n" request+="Host: ${url_host}\n" request+="User-Agent: ${http_user_agent}\n" request+="Connection: close\n" print -u${netfd} -- "${request}\n" # collect response and send it to stdout parse_http_response httpresponse <&${netfd} cat <&${netfd} # close connection exec {netfd}<&- print # add newline return 0 } function usage { OPTIND=0 getopts -a "${progname}" "${USAGE}" OPT '-?' exit 2 } # program start builtin basename builtin cat builtin date builtin uname typeset progname="$(basename "${0}")" # HTTP protocol client identifer typeset -r http_user_agent="shnote/ksh93 (2007-09-18; $(uname -s -r -p)" # name of history log (the number after "history" is some kind of version # counter to handle incompatible changes to the history file format) typeset -r history_file="${HOME}/.shnote/history0.txt" USAGE=$' [-?\n@(#)\$Id: shnote (Roland Mainz) 2007-09-18 \$\n] [-author?Roland Mainz <roland.mainz at nrubsig.org] [+NAME?shnote - read/write text data to internet clipboards] [+DESCRIPTION?\bshnote\b is a small utilty which can read and write text data to internet "clipboards" such as opensolaris.pastebin.ca.] [+?The first arg \bmethod\b describes one of the methods, "put" saves a string to the internet clipboard, returning an identifer and the full URL where the data are stored. The method "get" retrives the raw information using the identifer from the previous "put" action.] [+?The second arg \bstring\b contains either the string data which should be stored on the clipboard using the "put" method, the "get" method uses this information as identifer to retrive the raw data from the clipboard.] method [ string ] [+SEE ALSO?\bksh93\b(1), \brssread\b(1), http://opensolaris.pastebin.ca] ' while getopts -a "${progname}" "${USAGE}" OPT ; do # printmsg "## OPT=|${OPT}|, OPTARG=|${OPTARG}|" case ${OPT} in *) usage ;; esac done shift $((OPTIND-1)) typeset method="$1" shift # todo: "history" mode case "${method}" in put) put_note_pastebin_ca "$@" ; exit $? ;; get) get_note_pastebin_ca "$@" ; exit $? ;; hist) print_history "$@" ; exit $? ;; *) usage ;; esac fatal_error "not reached." # EOF. From brandorr at opensolaris.org Tue Sep 18 21:29:10 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 00:29:10 -0400 Subject: [opensolaris-summit] Everyone who proposed a topic for the forum on the wiki needs to update their topic with an explanation of why the topic is a critical issue.. Message-ID: <5b5090780709182129n1b9919f6icbf30d5fc2d40588@mail.gmail.com> The table has been modified to allow for this info. Please update your topics by the end of this week, as we will be asking everyone to select 5 of the most critical topics and submit them to a poll. (Post to this list if for some reason you can not update your topic by the end of the week). (Only 2 1/2 weeks to go). Cheers, Brian -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Tim.Foster at Sun.COM Wed Sep 19 02:16:34 2007 From: Tim.Foster at Sun.COM (Tim Foster) Date: Wed, 19 Sep 2007 10:16:34 +0100 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F065A2.6010006@sun.com> References: <46F065A2.6010006@sun.com> Message-ID: <1190193394.22489.73.camel@haiiro> On Tue, 2007-09-18 at 16:56 -0700, Jesse Silver wrote: > Server availability- > Suggestions: > - Have as many different kinds of machines available as possible, split > between Sparc and x86. > - Have at least one central server, stable and powerful, like an M4000 > with a bunch of storage. Why servers ? I thought this was a collective "bang our heads together" meeting, rather than an actual hackfest ? If if everyone has their noses buried in laptops for the duration of the weekend, then I'm not sure we'll have achieved our aims imho. [ I'm not bringing a laptop (yeah okay, mostly because I don't have one ;-) ] cheers, tim -- Tim Foster, Sun Microsystems Inc, Solaris Engineering Ops http://blogs.sun.com/timf From swalker at opensolaris.org Wed Sep 19 06:30:55 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Wed, 19 Sep 2007 08:30:55 -0500 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <1190193394.22489.73.camel@haiiro> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> Message-ID: <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> On 19/09/2007, Tim Foster <Tim.Foster at sun.com> wrote: > On Tue, 2007-09-18 at 16:56 -0700, Jesse Silver wrote: > > Server availability- > > Suggestions: > > - Have as many different kinds of machines available as possible, split > > between Sparc and x86. > > - Have at least one central server, stable and powerful, like an M4000 > > with a bunch of storage. > > Why servers ? I thought this was a collective "bang our heads together" > meeting, rather than an actual hackfest ? If if everyone has their noses > buried in laptops for the duration of the weekend, then I'm not sure > we'll have achieved our aims imho. > > [ I'm not bringing a laptop (yeah okay, mostly because I don't have > one ;-) ] There may be a few folks in the later hours of the conference that get very little chance for "bang their heads together on a software problem time." Otherwise, I tend to agree that I hope that this is more about problem solving than hacking... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From John.Plocher at Sun.COM Wed Sep 19 09:00:14 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 19 Sep 2007 09:00:14 -0700 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> Message-ID: <46F1478E.1080103@Sun.Com> Shawn Walker wrote: > I tend to agree that I hope that this is > more about problem solving than hacking... I see room for both - especially if we (Sun) can arrange for a reasonable set of high end systems to play with. While some topics (politics/governance/ARCs/...) won't be impacted much by the compute infrastructure at the conference, I can see other tracks taking a more hands-on dig in and play with code attitude. It would be VERY USEFUL for us all to articulate what exactly we wish to get out of the conference - what problems need to be solved that would benefit from a concentrated, in person focus? (See the Wiki at http://www.genunix.org/wiki/index.php/OpenSolaris_Summit for some suggestions) -John From James.Walker at Sun.COM Wed Sep 19 10:38:18 2007 From: James.Walker at Sun.COM (Jim Walker) Date: Wed, 19 Sep 2007 11:38:18 -0600 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F1478E.1080103@Sun.Com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> Message-ID: <46F15E8A.70407@sun.com> If useful, I could use some of the test farm automation to manage reservation time slots on whatever hardware we have at the summit. In addition, if we have good network access we can get to remote test machines. Cheers, Jim John Plocher wrote: > Shawn Walker wrote: >> I tend to agree that I hope that this is >> more about problem solving than hacking... > > > I see room for both - especially if we (Sun) can arrange for > a reasonable set of high end systems to play with. > > While some topics (politics/governance/ARCs/...) won't be impacted > much by the compute infrastructure at the conference, I can see other > tracks taking a more hands-on dig in and play with code attitude. > > It would be VERY USEFUL for us all to articulate what exactly we wish > to get out of the conference - what problems need to be solved that > would benefit from a concentrated, in person focus? (See the Wiki at > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit for some > suggestions) > > -John > > > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit -- Jim Walker, http://blogs.sun.com/jwalker Sun Microsystems, Software, Solaris QE x77744, 500 Eldorado Blvd, Broomfield CO 80021 From alan.duboff at sun.com Wed Sep 19 10:37:41 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 19 Sep 2007 10:37:41 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46ECD251.1090409@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> Message-ID: <Pine.GSO.4.64.0709191031530.1532@jazz> On Sat, 15 Sep 2007, Garrett D'Amore wrote: > Again, my biggest question/concern is, how many people will be in > attendance that the gory details of driver/kernel programming will > appeal to? I know you've voted, so that's one... are there more? I'm sure there are more, but the problem is that most of the people that are interested, are spread out more globally, as a whole, so targeting specific venues will be hit/miss at best in most cases. We need this information online, with the ability to stream in any location, so that the right people will have easy access to it. That's where the efforts should be focused rather than issolated efforts with less impact. My $0.02 (and where I'm trying to put some of my effort) Max Brunning has a lecture that he offers through SunU, and that is another example of a more issolated effort. It's great for folks that can get to SunU, but doesn't help someone in Europe or other countries, and we have many potential driver writers in those locales. -- Alan DuBoff - Solaris x86 IHV/OEM Group From brandorr at opensolaris.org Wed Sep 19 11:36:00 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 14:36:00 -0400 Subject: [opensolaris-summit] Everyone who proposed a topic for the forum on the wiki needs to update their topic with an explanation of why the topic is a critical issue.. In-Reply-To: <5b5090780709182129n1b9919f6icbf30d5fc2d40588@mail.gmail.com> References: <5b5090780709182129n1b9919f6icbf30d5fc2d40588@mail.gmail.com> Message-ID: <5b5090780709191136l39651cf3uc375eb2c11022fbe@mail.gmail.com> http://www.genunix.org/wiki/index.php/OpenSolaris_Summit I forgot to include the link, but I see many of you found your way there anyhow. :) On 9/19/07, Brandorr <brandorr at opensolaris.org> wrote: > The table has been modified to allow for this info. > > Please update your topics by the end of this week, as we will be > asking everyone to select 5 of the most critical topics and submit > them to a poll. (Post to this list if for some reason you can not > update your topic by the end of the week). > > (Only 2 1/2 weeks to go). > > Cheers, > Brian > > -- > - Brian Gupta > > http://opensolaris.org/os/project/nycosug/ > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Darren.Reed at Sun.COM Wed Sep 19 11:46:43 2007 From: Darren.Reed at Sun.COM (Darren.Reed at Sun.COM) Date: Wed, 19 Sep 2007 11:46:43 -0700 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F1478E.1080103@Sun.Com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> Message-ID: <46F16E93.7060801@Sun.COM> John Plocher wrote: >Shawn Walker wrote: > > >>I tend to agree that I hope that this is >>more about problem solving than hacking... >> >> > > >I see room for both - especially if we (Sun) can arrange for >a reasonable set of high end systems to play with. > >While some topics (politics/governance/ARCs/...) won't be impacted >much by the compute infrastructure at the conference, I can see other >tracks taking a more hands-on dig in and play with code attitude. > > As originally advertised, the event was to be about discussion relating to the project itself - it wasn't billed as a "hackathon" - and more specifically, to discuss Indiana. If we're going to be more generic and even think about making it a coding/hacking session then we need to readvertise it as such. If there aren't going to be a lot of these events held then I can definately see there being room for using this (and other) opensolaris summits as the means to get people together and look at problems we've got and how to solve them. However it might be more appropriate for _that_ kind of event to be held during the week, rather than the weekend, so that (some) people don't feel like they're working a 7 day week. I don't know about others, but this was one issue I had to settle with myself doing opensolaris stuff on the weekend, as being a Sun person makes it very easy to look at this weekend as being "work time". Darren From swalker at opensolaris.org Wed Sep 19 11:57:05 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Wed, 19 Sep 2007 13:57:05 -0500 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F16E93.7060801@Sun.COM> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> Message-ID: <b9c544f0709191157r217f13dew2699f6c59abf1bf2@mail.gmail.com> On 19/09/2007, Darren.Reed at sun.com <Darren.Reed at sun.com> wrote: > John Plocher wrote: > > >Shawn Walker wrote: > > > > > >>I tend to agree that I hope that this is > >>more about problem solving than hacking... > >> > >> > > > > > >I see room for both - especially if we (Sun) can arrange for > >a reasonable set of high end systems to play with. > > > >While some topics (politics/governance/ARCs/...) won't be impacted > >much by the compute infrastructure at the conference, I can see other > >tracks taking a more hands-on dig in and play with code attitude. > > > > > > As originally advertised, the event was to be about discussion > relating to the project itself - it wasn't billed as a "hackathon" - > and more specifically, to discuss Indiana. > > If we're going to be more generic and even think about making > it a coding/hacking session then we need to readvertise it as > such. > > If there aren't going to be a lot of these events held then I can > definately see there being room for using this (and other) > opensolaris summits as the means to get people together and > look at problems we've got and how to solve them. However > it might be more appropriate for _that_ kind of event to be > held during the week, rather than the weekend, so that > (some) people don't feel like they're working a 7 day week. > > I don't know about others, but this was one issue I had to settle > with myself doing opensolaris stuff on the weekend, as being > a Sun person makes it very easy to look at this weekend as > being "work time". ...and from those of us who aren't at Sun, we thank you for that sacrifice. I would not have been able to attend if it had been during the week as I'm independent and not doing this on behalf of my employer. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Glynn.Foster at Sun.COM Wed Sep 19 15:19:21 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Thu, 20 Sep 2007 10:19:21 +1200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F16E93.7060801@Sun.COM> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> Message-ID: <46F1A069.3030903@sun.com> Darren.Reed at Sun.COM wrote: > If there aren't going to be a lot of these events held then I can > definately see there being room for using this (and other) > opensolaris summits as the means to get people together and > look at problems we've got and how to solve them. However > it might be more appropriate for _that_ kind of event to be > held during the week, rather than the weekend, so that > (some) people don't feel like they're working a 7 day week. > > I don't know about others, but this was one issue I had to settle > with myself doing opensolaris stuff on the weekend, as being > a Sun person makes it very easy to look at this weekend as > being "work time". Hopefully you could organize with your manager to get some time off in lieu for working the weekend. I'd like to think most managers would happily accommodate that. Yes, it's sucky for those people that work hard during the week and have to come in during the weekend and lose the precious time they spend with their families. It's also sucky for those people who don't work for Sun to take time off during the week to attend a summit that doesn't necessarily have any overlap with their day job. The summit is going to mean different things to a lot of people - our hope is to accommodate people as much as possible. Glynn From brandorr at opensolaris.org Wed Sep 19 16:10:42 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 19:10:42 -0400 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F065A2.6010006@sun.com> References: <46F065A2.6010006@sun.com> Message-ID: <5b5090780709191610w7b7df16asa56f404dcda3e755@mail.gmail.com> > Action Items: > - Update Wiki, adding fields to expand on the topic suggestions. I sent out an email.. looks like 7/18 have expounded on their topics. http://www.genunix.org/wiki/index.php/OpenSolaris_Summit#Topic_Suggestions > - Add voting system to Wiki Ok, I may have misheard, but I could have sworn I heard Sara say that someone she knew was going to use some third party system for voting? Did I hear correctly? I am going to start a discussion thread about the topics. (NO reason we can't start discussing them now. -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Glynn.Foster at Sun.COM Wed Sep 19 16:18:38 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Thu, 20 Sep 2007 11:18:38 +1200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <5b5090780709191610w7b7df16asa56f404dcda3e755@mail.gmail.com> References: <46F065A2.6010006@sun.com> <5b5090780709191610w7b7df16asa56f404dcda3e755@mail.gmail.com> Message-ID: <46F1AE4E.3020209@sun.com> Hey, Brandorr wrote: >> Action Items: >> - Update Wiki, adding fields to expand on the topic suggestions. > > I sent out an email.. looks like 7/18 have expounded on their topics. > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit#Topic_Suggestions > >> - Add voting system to Wiki > > Ok, I may have misheard, but I could have sworn I heard Sara say that > someone she knew was going to use some third party system for voting? > Did I hear correctly? Yeah, we're trying to get a simple poll site together. I've sent the list to Moazam and he said he'd get something together on unixville.com. Glynn From brandorr at opensolaris.org Wed Sep 19 18:08:02 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 21:08:02 -0400 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics Message-ID: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> I figured I'd roll this down the hill and see what others say. (I just realized that biggest thing I want to hear from the Indiana team, is not on the list: What's been done and what needs to be done. The meeting should be about fleshing that out and obtaining buy in to get it done.) (I think) Anyway, on with the list.... --------------------------------------------------------------------------------------------------------------------------------------------------- Darren Reed Community Structure Elaboration needed. --------------------------------------------------------------------------------------------------------------------------------------------------- Glynn Foster How to plan time based releases in a non-time based release world One of the primary social goals of Project Indiana focus around the release roadmap, and in particular getting to a point where we can regularly do 6 month time based releases of our binaries creating predictable cycles for developers who target that distribution, and users downloading, installing and upgrading. Many popular open source communities have found benefits of this time based approach, and others are starting to move towards it as a way of managing their software. While this is nice in theory, it very much relies on a strict development ethic. How can we get to a stage of providing time based releases when the world around us has traditionally been feature based? Is there a best of both worlds approach, and how can our influence change the way distributions are created both internal and external to Sun? Mark Kupfer's blogged about this very issue. I suggest reading it. http://blogs.sun.com/kupfer/entry/what_i_learned_from_ubuntu --------------------------------------------------------------------------------------------------------------------------------------------------- Brian Gupta Developing/supporting a User community, and portal, etc. As Sun is marketing/pushing Indiana, Solaris, and OpenSolaris, we are seeing many people that aren't developers coming to this site. As this is the first thing they see when they get here, I feel that this should be the logical location for a user oriented welcome page. I also propose that this welcome page is much simpler, as the current page is a bit overwhelming for newcomers. (Hard to find certain links) Developers would still have access to a developer homepage that is linked off the new homepage. (We might even setup a redirect: dev.opensolaris.org) http://www.opensolaris.org/jive/thread.jspa?messageID=148391 No comment since this is my suggestion. :) --------------------------------------------------------------------------------------------------------------------------------------------------- John Plocher Architecture Community's relationship to Indiana, PSARC and other distros. I'm looking for direction and consensus on several fundamental issues: Problem: What is needed to do good systems architecture in a world that has multiple systems? i.e., what is good for Solaris Express may not be good for Indiana or Schillix... I think this is an Indiana talk... this may be off topic Problem: What needs to be done to enable effective ARC participation by external-to-Sun members? i.e., web services, tools, materials submission, case creation, voting, etc I think this might be incorporate in a larger question about including the community. Problem: What does it mean to be an external ARC member? Job description, scope of job, time needed, tasks, authority...? I view this as an important question, but might we be able to take this question directly to the arc members on the mailing list? Problem: How are appeals handled? Who makes the unpopular decisions? Where is the "teeth"? As I understand there is a Constitution that handles appeals. Unpopular with whom? Teeth?? --------------------------------------------------------------------------------------------------------------------------------------------------- Stephen Lau Is it hard to contribute to OpenSolaris? How can we make it easier? The "is it hard" is a rhetorical question, really. Its fairly obvious there are some barriers to contribution. This will be a gathering of people to figure out what those barriers are, identify concrete things that suck, and try to figure out how to reduce the suck. It is not a session about distro specific stuff, it is not a session about adding new features or changing things in OpenSolaris other than the contribution barriers. NC --------------------------------------------------------------------------------------------------------------------------------------------------- Ben Rockwood The Future of Patching and Upgrades (OS & Apps) Is there more to this topic than the current SXCE upgrade process? --------------------------------------------------------------------------------------------------------------------------------------------------- John Sonnenschein Increasing community participation while maintaining traditional stability & compatibility, or "avoiding the Linux free-for-all mess" Please elaborate and give examples... (Remember we are talking about the Indiana distro, not everything in th OpenSolaris/Solaris universe.) --------------------------------------------------------------------------------------------------------------------------------------------------- Tim Foster Project Indiana, an exercise in compromise (in the UK/Ireland sense) Because we all have to agree to drop things that may be near/dear to us in order to actually get something out the door. We don't have to get everything right, but we do have to show progress. Endlessly arguing about what's the default shell, or which version of tar we use isn't the way to get from here to there. This get's me plus one, with a slight change. Basically we need to decide what is important for Project Indiana. What do we need to address before our first binary shipment? --------------------------------------------------------------------------------------------------------------------------------------------------- Jim Walker How do we test OpenSolaris? This get's my plus one. As I understand all Solaris Express testing is currently done behind closed walls. If this can't be brought out into the open, what is involved in setting up an equivelant e;nvironment --------------------------------------------------------------------------------------------------------------------------------------------------- Dave Stewart What is Intel doing to help OpenSolaris drivers? Please eleborate how the Summit can address this? --------------------------------------------------------------------------------------------------------------------------------------------------- Ian Murdock ZFS to the max (justification from timf) # fsck /dev/dsk/c0d0s4 ** /dev/rdsk/c0d0s4 CANNOT READ: BLK 62217648 CONTINUE? n This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Ian Murdock Laptop support This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Ian Murdock User experience This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Sara Dornsife Naming This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Brian Gupta Opensolaris.org OSOL Infrastructure Support to the outside community. Currently the Opensolaris.org infrastructure ( www.opensolaris.org and the mailing lists being two example pieces of infrastructure) is completely maintained by Sun employees. This limits available resources. There are many of sys-admin types out in the community that are capable, and most likely willing, to help out. (As well as businesses that might be willing to fund contract employees to aid in this endeavor) Might we explore a path to opening Opensolaris.org's care and feeding to the greater community? No comment, my topic again. :) --------------------------------------------------------------------------------------------------------------------------------------------------- Shawn Walker Creating custom distributions using Indiana and possible legal obligations / requirements. I don't know if there are going to be lawyers there. If not, I suggest we best not speculate. --------------------------------------------------------------------------------------------------------------------------------------------------- Stefan Teleman C++ Standard Compliant: Sun Studio 12, BOOST, and the Apache Standard C++ Library I am not sure how this directly relates to project Indiana. --------------------------------------------------------------------------------------------------------------------------------------------------- Stuart Kreitman Parity: Making Open Solaris equal or better than Linux as a development desktop Please eleborate --------------------------------------------------------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070919/36d888d9/attachment.html> From John.Plocher at Sun.COM Wed Sep 19 21:26:42 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 19 Sep 2007 21:26:42 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> Message-ID: <46F1F682.2030008@Sun.Com> Brandorr wrote: > John Plocher ... > Problem: ... Focusing on Indiana: Indiana is all about building a system (aka distro) out of OS.o parts. The ARC is a systems architecture body. What can the ARC do for Indiana? That is, what does Indiana need from the ARC so Indiana can be successful for its various releases? [The ideas I've heard concern things like refactoring "the WOS" to allow different rates of change in the various sections, keeping the "core" of OpenSolaris completely compatible while allowing incompatible and disruptive innovation to happen closer to the edge. What explicit actions are needed to make that a reality?] What can Indiana do for the ARC? Since it is an "Us-ARCing-Our-Stuff" world, what ARC-related resources are Indiana developers willing or able to provide to help deliver those things? (Or is the expectation really "Us-vs-Them", where someone /else/ is expected to do the ARC work?) In any case, what does the ARC community need to do to ensure that those resources are effectively and productively used? How does the existing community structure need to change? What should happen when the systems architectural needs of Indiana conflict with those of Solaris or one of the other distros? Should there be an "Indiana ARC", a "Solaris ARC", a "Schillix ARC", ...etc? If so, where does an "OpenSolaris ARC" fit? What is in and out of scope for each? The needs of a kernel.org-style community are not the same as the needs of an ubuntu-style one. What needs to be done to move the larger OS.o community from its current kernel.org-ish mindset to more of the uubuntu-style one that Indiana envisions? > Problem: How are appeals handled? Who makes the unpopular > decisions? Where is the "teeth"? You are right - this is bigger than just Indiana. Nevertheless, for context, if nothing else: > As I understand there is a Constitution that handles > appeals. Unpopular with whom? Teeth?? Say the ARC says "do /this/" or even "we don't approve this proposal", and the project team or its sponsoring community disagrees. This is a common situation where the short term needs of a project may conflict with the long term needs of the larger OS.o community. 20 years of Sun experience shows that there needs to be an intermediate step in those "appeals" to ensure that they all don't immediately overload the OGB with hearings or cause expensive OS.o community-wide formal votes. What happens if a project or community chooses to make architectural- level changes to an OS.o component without engaging with the ARC? Who gets to choose whether or not ARCing applies to something? Even if one is in a part of the OS.o community where the ARC is known to apply, what happens if a project [tries to] integrate without doing an ARC review? If they get stopped, who does the stopping? By what authority does the stopper actually obtain the right to stop it? That's teeth. From alan.duboff at sun.com Wed Sep 19 22:25:13 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 19 Sep 2007 22:25:13 -0700 (PDT) Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <46F1F682.2030008@Sun.Com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> Message-ID: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> On Wed, 19 Sep 2007, John Plocher wrote: > Say the ARC says "do /this/" or even "we don't approve this proposal", > and the project team or its sponsoring community disagrees. This is > a common situation where the short term needs of a project may conflict > with the long term needs of the larger OS.o community. 20 years of Sun > experience shows that there needs to be an intermediate step in those > "appeals" to ensure that they all don't immediately overload the OGB > with hearings or cause expensive OS.o community-wide formal votes. This is where an ARC would do the OpenSolaris community an injustice if they impose limitations on the software that can be contained in a repository of user-land, open source software. For a long time I've hoped that we would see open source software install in /usr/bin:/usr/sbin as it should, but it's not clear if Indiana will take things to that extent yet. But I did sit in the session that was given at IDF today in San Francisco, and I got the impression that Ian could be thinking that way (and if so, I would support that thought;-). The current ARCs would blow a fuse at the thought of putting anything in /usr/bin:/usr/sbin I suspect, but am only speculating. However, this is where the software should go in a system, IMO, unless there is conflict (similar functions like SysV tar vs. gnu tar as an example). > What happens if a project or community chooses to make architectural- > level changes to an OS.o component without engaging with the ARC? Who > gets to choose whether or not ARCing applies to something? I don't think that should be an option, because they would need to, or should be required to go through an ARC to get anything into the ON portion of OpenSolaris, and that certainly should not change. > Even if > one is in a part of the OS.o community where the ARC is known to > apply, what happens if a project [tries to] integrate without doing > an ARC review? If they get stopped, who does the stopping? By what > authority does the stopper actually obtain the right to stop it? > That's teeth. The way some have gotten around this happening in open source software today, is to have unstable and stable distributions, where unstable is more tolerant to accepting software. In Sun's case with Solaris, they will certainly need to decide how they take software from the open source community, just as they do today, unless they change their ways possibly. I don't see that for Solaris, but stranger things have happened. A good question is how the OpenSolaris will keep control over what goes into any "general" repository though, I don't think anyone should have the access to do a putback if the code is not packaged properly for instance. At some point I want to see a large repository of software, today we have that scattered all around, blastwave, pkgsrc, polaris, belenix, schillix, Nexenta, etc...I think Nexenta gets more right than any of the others to date, not that there is no flaw in their solution, just that today they seem to have more in place and understand how that software should be placed in the system. Maybe because they follow Debian's lead in how they operate, but I just mean in regards to having a system that can be updated and be inclusive of a large repository. What if there was a large repository of packages that anyone could use on an OpenSolaris system? I think it would be good to elliminate the walls that seperate the above mentioned efforts, if possible, get our community to unify once and for all...and to do that we need to encourage folks to help us package and contribute to some large effort. For better or worse it seems to me that Indiana is as good of a reason to kickoff some type of effort like that, as anything else. -- Alan DuBoff - (wearing his nomex suit, go ahead and flame me cause he's wearing his honest face;-) From alan.duboff at sun.com Wed Sep 19 22:35:00 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 19 Sep 2007 22:35:00 -0700 (PDT) Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F1A069.3030903@sun.com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> <46F1A069.3030903@sun.com> Message-ID: <Pine.GSO.4.64.0709192230380.3804@eagle.softorchestra.com> On Thu, 20 Sep 2007, Glynn Foster wrote: > Hopefully you could organize with your manager to get some time off in > lieu for working the weekend. I'd like to think most managers would > happily accommodate that. I think the community should fall how it falls. IOW, if someone needs to feel they need to be paid to work on OpenSolaris, that's ok, and if they feel they would like to participate for another reason, that's ok also. Ultimately OpenSolaris will be comprised of the folks that do participate in it and are a part of it. I'm not trying to say that Darren should feel as if he should be one way or the other, I'm trying not to critisize him for his own decision just as I wouldn't another. > Yes, it's sucky for those people that work hard during the week and have > to come in during the weekend and lose the precious time they spend with > their families. And how is that any different for the community people that do other things during the week and may spend the time to attend? It is held on the weekend to get as many folks as they can, I think you would be in agreement with that. Can you imagine if it was held during the week? There should be many community people saying, "hey, I have to support my family and can't get off work". > It's also sucky for those people who don't work for Sun > to take time off during the week to attend a summit that doesn't > necessarily have any overlap with their day job. Exactly.;-) > The summit is going to mean different things to a lot of people - our > hope is to accommodate people as much as possible. Amen, let the cards fall how they fall and be glad there cards at all.;-) -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.coopersmith at sun.com Thu Sep 20 08:07:03 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Thu, 20 Sep 2007 08:07:03 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> Message-ID: <46F28C97.6010103@sun.com> Alan DuBoff wrote: > On Wed, 19 Sep 2007, John Plocher wrote: > The current ARCs would blow a fuse at the thought of putting anything in > /usr/bin:/usr/sbin I suspect, but am only speculating. However, this is > where the software should go in a system, IMO, unless there is conflict > (similar functions like SysV tar vs. gnu tar as an example). That would be the ARC's that already decided that everything in /usr/sfw/bin that doesn't conflict with something already in /usr/bin should move to /usr/bin? I think you're a bit out of touch with the ARC's... -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From brandorr at opensolaris.org Thu Sep 20 08:21:47 2007 From: brandorr at opensolaris.org (Brandorr) Date: Thu, 20 Sep 2007 11:21:47 -0400 Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. Message-ID: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> I an catching an early flight to San Jose. (landing at 12:30pm). Does anyone want to try and meet up during the day on Friday? -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From John.Plocher at Sun.COM Thu Sep 20 10:19:57 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 20 Sep 2007 10:19:57 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> Message-ID: <46F2ABBD.2070306@Sun.Com> Alan DuBoff wrote: > This is where an ARC would do the OpenSolaris community an > injustice if they impose limitations on the software that > can be contained in a repository of user-land, open source software. Nobody but you has said anything about general software availability in random repositories. I'm talking about the systems we choose to build out of specific things from specific repositories. It may well be that we get a severe case of stupidity and create several incompatible versions of libc and place them in various repositories. No problem there (other than being stupid, of course). It should be obvious that one can't simply randomly pick one of those libc's and expect the rest of their system to "just work" - there is a dependency chain that links your libc choice to the stuff that will work with it. Same for most everything that is in or near the core of a system. The systems engineering and architecture I'm concerned with focuses on the decisions that a distro producer needs to make to ensure that their distro starts off self consistent - and remains so over time. If you want to go off and do your own thing without any regard to how your behavior impacts projects that have come to depend on you, feel free to do so. Just don't expect others to be very happy with you if you jerk them around by behaving irresponsibly. What do I mean by irresponsible behavior? Effectively, not setting expectations for those who would consume your stuff, or once having set them, breaking them. > The current ARCs would blow a fuse at the thought of putting anything in > /usr/bin:/usr/sbin I suspect, but am only speculating. You /are/ speculating - wildly and without much basis in fact. Most of the ARC's focus in the last couple of years has been on doing exactly that. The whole point of Systems Engineering - as opposed to unconstrained hacking - is to develop a model for how we /desire/ a system to behave and to ensure that what we /do/ matches that model. This includes changing that model as time goes by and expectations change. The Solaris model is one of stability of interfaces over time. The model says that we desire that all code and scripts that were written following the rules will continue to operate correctly on any and all minor and micro releases of Solaris in the future. This obviously is not the same model that is desired for Indiana. Unfortunately, nobody has articulated the specifics of what this new model are, nor have they put any effort in developing transition plans to move between the old and new models. Topping it all off, we (Sun, OpenSolaris,...) only have a "stable tree" to work with. There is no "unstable development branch" where we can do all those chaotic things that you seem to desire. This all makes it hard for developers, architects, distro producers and users to understand what they need to do. -John From Glynn.Foster at Sun.COM Thu Sep 20 12:48:31 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Fri, 21 Sep 2007 07:48:31 +1200 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <46F2ABBD.2070306@Sun.Com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> <46F2ABBD.2070306@Sun.Com> Message-ID: <46F2CE8F.5040703@sun.com> John Plocher wrote: > The Solaris model is one of stability of interfaces over time. The > model says that we desire that all code and scripts that were written > following the rules will continue to operate correctly on any and > all minor and micro releases of Solaris in the future. > > This obviously is not the same model that is desired for Indiana. > > Unfortunately, nobody has articulated the specifics of what this > new model are, nor have they put any effort in developing transition > plans to move between the old and new models. Topping it all off, > we (Sun, OpenSolaris,...) only have a "stable tree" to work with. > There is no "unstable development branch" where we can do all those > chaotic things that you seem to desire. Very much what I'm keen to tackle at the summit. Most of my concerns during the Indiana PSARC meeting a couple of months ago revolved around this. As you point out, it's certainly a pretty hard for anything that isn't Solaris to know what to do best, without losing the unique benefits of those guarantees (and by Solaris I mean 'official release' rather than the express train). Glynn From phil at bolthole.com Thu Sep 20 15:13:19 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 15:13:19 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com>; from alan.duboff@sun.com on Wed, Sep 19, 2007 at 10:25:13PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> Message-ID: <20070920151319.A23828@bolthole.com> On Wed, Sep 19, 2007 at 10:25:13PM -0700, Alan DuBoff wrote: > ... > For a long time I've hoped that we would see open source software install > in /usr/bin:/usr/sbin as it should, but it's not clear if Indiana will > take things to that extent yet. But I did sit in the session that was > given at IDF today in San Francisco, and I got the impression that Ian > could be thinking that way (and if so, I would support that thought;-). My opinion on that, is that it should depend on "who do you call, for an 'indiana' system, if something in /usr/bin breaks?" There is still some fuziness in my mind, as to what indiana is. Even if indiana is supposed to be a more "do it yourself" option, I could still envision some situationally dependant reasons where "everything goes in /usr/bin" may not be the best choice. That is to say, even if Sun is not the force behind "supporting" an indiana based install.. SOMEONE/some group/forum will be. For that entity, there will still be value for knowing something about what to expect from stuff in /usr/bin. Envision this: Someone posts on a debian user mailing list, "Help! my [foo] isnt working right!" people try to help, and after much oddity and headscratching and time spent, someone says, [hypothetical: this doesnt work too well if there are 2,000 files there!] "well show us what ls -l /usr/bin looks like", and it is then determined that; 1. this user just "happened" to have installed some experimental version of grep in /usr/bin, instead of the usual debian package 2. this version of grep has some bugs or altered functionality in edge cases 3. This other version was what caused [foo] to not work right. Tens of hours have been wasted trying to track down a problem "with debian", when it was actually a problem with an experimental user dumping some alternative binary in /bin, because after all, 'everything goes in bin'. If you dont set up some kind of expectation of, "dont muck with bin like that", then messes like that will be bound to happen. In the case of Debian, even though I'm not aware of any explicit rule against that sort of thing... I think the person would be lambasted for wasting peoples time. There is some amount of expectation that "a debian system", would have the debian version of grep installed. In the case of Indiana... I think it would be beneficial to have some kind of explicit statement of what to expect from /usr/bin, and what kinds of things should be installed "elsewhere". (and where "elsewhere" should be!) In addition to avoiding timewasters like the above, i think it would help to build confidence in what to expect from "Indiana". But, all this again depends on what exactly Indiana is supposed to be. If it is just supposed to be "a more fluidly updated version of Solaris", (with a nod to also making additional freeware more easily installable) then I think the rules on what goes in /usr/bin, should be quite strict, similar to regular Solaris. In my opinion, "non-sun-tracked" software should still go somewhere else. If, on the other hand, Indiana is targetted as a more "anything goes" environment, then perhaps not. But if it's an "anything goes" environment, then I almost dont see the point of it. From Darren.Reed at Sun.COM Thu Sep 20 15:34:23 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Thu, 20 Sep 2007 15:34:23 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> Message-ID: <46F2F56F.70706@Sun.COM> Brandorr wrote: > I figured I'd roll this down the hill and see what others say. > > (I just realized that biggest thing I want to hear from the Indiana > team, is not on the list: What's been done and what needs to be done. > The meeting should be about fleshing that out and obtaining buy in to > get it done.) (I think) > > Anyway, on with the list.... > > --------------------------------------------------------------------------------------------------------------------------------------------------- > > Darren Reed Community Structure > > Elaboration needed. See: http://blogs.sun.com/avalon/entry/manufacturing_a_community And to be more specific: - are contributors and core contributors required? - is the voting mechanism correct to bring in new contributors? - if leaders are just those that can edit web pages, is this the right label? - etc. Darren From swalker at opensolaris.org Thu Sep 20 15:55:32 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 20 Sep 2007 17:55:32 -0500 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> Message-ID: <b9c544f0709201555v2e8c8549q32d774070eedad71@mail.gmail.com> On 19/09/2007, Brandorr <brandorr at opensolaris.org> wrote: > Shawn Walker Creating custom distributions using Indiana and possible > legal obligations / requirements. > > I don't know if there are going to be lawyers there. If not, I > suggest we best not speculate. I wasn't suggesting that we speculate. I've updated the entry to clarify: Project Indiana, as far as I know, has a goal of being a base for others to create their own distributions upon. I would like to hear more about how it might approach some of the issues that will face people wanting to create their own distribution, or what the plan to support them is. Some questions might include: * What is the plan to support distribution creators? * Will there be a special mailing list just for folks that are creating their own distribution based off of Indiana? * Will there be special documentation explaining the legal, technical, and other requirements? (i.e. possible redistribution issues, copyright and trademark issues, licensing issues, etc.) * Can I create a distribution that has a Solaris kernel and GNU userland (and the possible legal implications)? * Can I redistribute the nVidia drivers with my distribution? While legal people may not be present, that should not stop getting a list of possible questions for someone qualified to answer at a later date. I would like to see this area addressed within the scope of the OpenSolaris community (not just the Indiana project, although that would be good too). -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From phil at bolthole.com Thu Sep 20 15:59:52 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 15:59:52 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F2F56F.70706@Sun.COM>; from Darren.Reed@Sun.COM on Thu, Sep 20, 2007 at 03:34:23PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> Message-ID: <20070920155952.A17130@bolthole.com> On Thu, Sep 20, 2007 at 03:34:23PM -0700, Darren Reed wrote: > > See: > http://blogs.sun.com/avalon/entry/manufacturing_a_community > interesting blog. I'd say one core problem in there, is due to sun marketspeak misusing the term "community". Someone apparently got all caught up in "Open Source Development is 'community based'! We've Got To Have Comunit(ies) !!!" Seems like the marketdroids have achieved way too much power inside sun lately, as evidenced by stupidity like renaming SUNW to "JAVA". Maybe the entities currently described under http://www.opensolaris.org/os/communities/ should have a name other than "community", and then have a more hierarchical organization to them. There's just too many of them at the top level right now. What might be better,could be an approach like the old USENET style approach, where you first start with the HIGH level comp.os.opsol Split it up into comp.os.opsol.kernel comp.os.opsol.install comp.os.opsol.graphics comp.os.opsol.blah and then split up into finer grained pieces, ONLY WHEN participants of a particular "leaf" node, vote and agree that the leaf should now become a node with child leaves. (and usually, there were rules about that sort of thing, such as, the vote has to continue for X amount of time, so as to avoid making a split about something that is going to die out in 1 month anyway) getting a bit off-topic for this list now, I suppose. From alan.coopersmith at sun.com Thu Sep 20 16:12:31 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Thu, 20 Sep 2007 16:12:31 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070920155952.A17130@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> Message-ID: <46F2FE5F.3070508@sun.com> Philip Brown wrote: > There's just too many of them at the top level right now. We tried to reorganize a few months ago, but most of the communities felt that giving up their top-level status diminished their importance and they didn't want any part of it. (As I understand, the OpenSolaris structure was designed after the Apache Foundation structure, but they're more of a loosely affialiated group of software projects than something trying to produce a coherent OS, so it may not have been the best fit.) -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From phil at bolthole.com Thu Sep 20 16:29:35 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 16:29:35 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F2FE5F.3070508@sun.com>; from alan.coopersmith@sun.com on Thu, Sep 20, 2007 at 04:12:31PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> Message-ID: <20070920162935.B17130@bolthole.com> On Thu, Sep 20, 2007 at 04:12:31PM -0700, Alan Coopersmith wrote: > Philip Brown wrote: > > There's just too many of them at the top level right now. > > We tried to reorganize a few months ago, but most of the communities > felt that giving up their top-level status diminished their importance > and they didn't want any part of it. > Well, then there's a core problem with opensolaris.org right there. Someone needs to have the authority, and B... erm.. "gumption".. to tell them, "Too bad, this is for the better good of OpenSolaris.". Isnt that what the (previously named?) "CAB" is for? You cant make everyone happy, all the time. Obviously, some groups are going to be unhappy to be moved down. Doesnt mean it shouldnt be done. If they were thinking more sensibly, they should be happy about it, because better organization should actually mean they would be easier to find. But whether they come to that realization or not, doesnt change the need for it. From Darren.Reed at Sun.COM Thu Sep 20 16:57:40 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Thu, 20 Sep 2007 16:57:40 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070920162935.B17130@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> Message-ID: <46F308F4.8050101@Sun.COM> Philip Brown wrote: >On Thu, Sep 20, 2007 at 04:12:31PM -0700, Alan Coopersmith wrote: > > >>Philip Brown wrote: >> >> >>>There's just too many of them at the top level right now. >>> >>> >>We tried to reorganize a few months ago, but most of the communities >>felt that giving up their top-level status diminished their importance >>and they didn't want any part of it. >> >> >> > >Well, then there's a core problem with opensolaris.org right there. >Someone needs to have the authority, and B... erm.. "gumption".. to tell >them, "Too bad, this is for the better good of OpenSolaris.". > >Isnt that what the (previously named?) "CAB" is for? > > Possibly. And having put the problem like this, for better or worse, what OpenSolaris needs is a board of directors (or the equivalent): a body that can make decisions; not just have discussions via email and try to resolve "issues". How would that be different to the OGB? No open board-discuss for starters. Have it function like a real board: meet monthly or quarterly, publish the minutes of the meetings, appoint a chairman, give the board power to make decisions that change opensolaris without needing votes, etc... >You cant make everyone happy, all the time. >Obviously, some groups are going to be unhappy to be moved down. Doesnt >mean it shouldnt be done. > > Yes. And so long as we allow even one "I disagree" to hold up change, we'll never really accomplish very much at all in a timely fashion. Darren From phil at bolthole.com Thu Sep 20 17:20:25 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 17:20:25 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F308F4.8050101@Sun.COM>; from Darren.Reed@Sun.COM on Thu, Sep 20, 2007 at 04:57:40PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> Message-ID: <20070920172025.D17130@bolthole.com> On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: > Philip Brown wrote: > >Isnt that what the (previously named?) "CAB" is for? > >... > > How would that be different to the OGB? No open board-discuss > for starters. Have it function like a real board: meet monthly or > quarterly, publish the minutes of the meetings, appoint a chairman, > give the board power to make decisions that change opensolaris > without needing votes, etc... The one trouble with doing stuff politically like that, is that there is then a tendancy for those positions to be filled with politicians. There need to be rules, but not to the point of needing a "full time politician" to handle it, so to speak. > >You cant make everyone happy, all the time. > >Obviously, some groups are going to be unhappy to be moved down. Doesnt > >mean it shouldnt be done. > > Yes. And so long as we allow even one "I disagree" to hold up change, > we'll never really accomplish very much at all in a timely fashion. The one softening I would like to add to my comments, is a clarification of my general principles on that sort of thing. "We shouldnt do it because x,y/z" should be carefully considered. Whines about, "I dont like it/I dont want to change", should be ignored, in my opinion [if the change has long term benefits] A single person dissenting, doesnt mean that single person should be automatically ignored, though. The more important question is whether the dissent is on rational grounds or not. (and it then becomes important for the people making that determination, to be both mentally able to discern that, and also to have the commitment and integrity to do so. Having a "majority vote" is no substitute for technical competance and understanding) From Darren.Reed at Sun.COM Thu Sep 20 17:41:25 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Thu, 20 Sep 2007 17:41:25 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070920172025.D17130@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> Message-ID: <46F31335.6060700@Sun.COM> Philip Brown wrote: >On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: > > >>Philip Brown wrote: >> >> >>>Isnt that what the (previously named?) "CAB" is for? >>> >>> >>... >> >>How would that be different to the OGB? No open board-discuss >>for starters. Have it function like a real board: meet monthly or >>quarterly, publish the minutes of the meetings, appoint a chairman, >>give the board power to make decisions that change opensolaris >>without needing votes, etc... >> >> > >The one trouble with doing stuff politically like that, is that there is >then a tendancy for those positions to be filled with politicians. >There need to be rules, but not to the point of needing a >"full time politician" to handle it, so to speak. > > To which I'd say that the board must be comprised of only those who commit or make changes to the source code of opensolaris, not those that are only involved in the ARC, website, etc. That there are few "code cutters" that are "politicians" tends to solve the "politician" problem (or at least this is my observation.) Open source projects are fundamentally about code, everything else is just window dressing. The right to be taken seriously should be inline with the contribution to the source code base. Darren From stevel at sun.com Thu Sep 20 17:49:52 2007 From: stevel at sun.com (Stephen Lau) Date: Thu, 20 Sep 2007 17:49:52 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F31335.6060700@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> Message-ID: <46F31530.9020102@sun.com> Darren Reed wrote: > Philip Brown wrote: > >> On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: >> >> >>> Philip Brown wrote: >>> >>> >>>> Isnt that what the (previously named?) "CAB" is for? >>>> >>>> >>> ... >>> >>> How would that be different to the OGB? No open board-discuss >>> for starters. Have it function like a real board: meet monthly or >>> quarterly, publish the minutes of the meetings, appoint a chairman, >>> give the board power to make decisions that change opensolaris >>> without needing votes, etc... >>> >>> >> The one trouble with doing stuff politically like that, is that there is >> then a tendancy for those positions to be filled with politicians. >> There need to be rules, but not to the point of needing a >> "full time politician" to handle it, so to speak. >> >> > > To which I'd say that the board must be comprised of only those > who commit or make changes to the source code of opensolaris, > not those that are only involved in the ARC, website, etc. That > there are few "code cutters" that are "politicians" tends to solve > the "politician" problem (or at least this is my observation.) > > Open source projects are fundamentally about code, everything > else is just window dressing. The right to be taken seriously > should be inline with the contribution to the source code base. Which only sets up an even more elitist atmosphere. You would be eliminating the valuable contributions of people who are docs people, translations, artists, evangelists, etc. There is nothing inherent about cutting code that somehow makes someone qualified to run or lead an open source project or community. -steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From Glynn.Foster at Sun.COM Thu Sep 20 18:45:53 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Fri, 21 Sep 2007 13:45:53 +1200 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709191031530.1532@jazz> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> Message-ID: <46F32251.7060102@sun.com> Alan DuBoff wrote: > On Sat, 15 Sep 2007, Garrett D'Amore wrote: > >> Again, my biggest question/concern is, how many people will be in >> attendance that the gory details of driver/kernel programming will >> appeal to? I know you've voted, so that's one... are there more? > > I'm sure there are more, but the problem is that most of the people that > are interested, are spread out more globally, as a whole, so targeting > specific venues will be hit/miss at best in most cases. > > We need this information online, with the ability to stream in any > location, so that the right people will have easy access to it. "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind tour of a week-long class on how to write a driver. Seth touched on the DDI/DKI interface, which provides terrific interface stability from generation to generation, the device tree and tips and tricks. There was a demonstration driver source presented as well, and all of the commands to try it out. I'll have to do that when I get home!" http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ Sounds like the information is there (though possibly not GLDv3?), would probably be good to highlight it more? Glynn From alan.duboff at sun.com Fri Sep 21 01:03:25 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 01:03:25 -0700 (PDT) Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <20070920151319.A23828@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> <20070920151319.A23828@bolthole.com> Message-ID: <Pine.GSO.4.64.0709210027450.5157@eagle.softorchestra.com> On Thu, 20 Sep 2007, Philip Brown wrote: > On Wed, Sep 19, 2007 at 10:25:13PM -0700, Alan DuBoff wrote: >> ... For a long time I've hoped that we would see open source software >> install in /usr/bin:/usr/sbin as it should, but it's not clear if >> Indiana will take things to that extent yet. But I did sit in the >> session that was given at IDF today in San Francisco, and I got the >> impression that Ian could be thinking that way (and if so, I would >> support that thought;-). > > My opinion on that, is that it should depend on "who do you call, for an > 'indiana' system, if something in /usr/bin breaks?" I have a really tweaked vision from the way things are currently done at Sun, and I'm sure many folks won't agree with the way I see things, but from my perspective that is simple. Sun basically sells support, something they're not great at today, but that will need to change. If you don't pay, you go to somewhere like here, or other place on the inet (yahoogroups, usenet, etc...). Pay $x and you can call a live body to talk your problem through with, but it doesn't get any resolution per se, other than the bug can get logged if it does exist. Pay $xx and get the ability to have updates over the wire for a year. Pay $xxx, Sun will not only work the issure with you, they will provide a live body to look into and/or fix the problem. People could get help in broken software, not in the sense that Sun would provide broken software that people need to pay for to get fixed, but more so that the Sun engineer will work with the open source code and/or folks to resolve it. Pay $xxxx, Sun will develop and extend open source software for you. This is very much in the professional services realm. > There is still some fuziness in my mind, as to what indiana is. Don't feel bad, there is in my mind also. Each time I attend one of Ian's presentation, things are becoming more clear, and I thought he gave a pretty good presentation at IDF this week. To me it looks like Sun is going to be working towards providing a binary distribution that will update over the wire, and includes all of the open source software you would expect in today's climate. > Even if indiana is supposed to be a more "do it yourself" option, I > could still envision some situationally dependant reasons where > "everything goes in /usr/bin" may not be the best choice. I personally think Nexenta got a lot of things right. The one thing I don't like is that is has a GNU personality by default. I really need to have SysV compatibility, first and foremost for OpenSolaris to provide backward compatibility to S10 and before. However, Indiana will most likely only install to ZFS, I hope my understanding is correct about that because I think ZFS is one of the best technologies that Sun has at the moment. This evolution in inevitable, IMO. I also don't think that Sun is the only player in this space, and others such as Blastwave could offer their own distribution and provide their own support, just as Sun does, just like Nexenta could, Belenix could, Shillix, Martux, etc...or OTOH, they could just continue to be what they are and not offer support. > That is to say, even if Sun is not the force behind "supporting" an > indiana based install.. SOMEONE/some group/forum will be. For that > entity, there will still be value for knowing something about what to > expect from stuff in /usr/bin. I think Indiana will be a product that Sun will offer support for. I don't forsee Sun offering support for all other OpenSolaris distributions though. > 1. this user just "happened" to have installed some experimental version > of grep in /usr/bin, instead of the usual debian package That's ok with me, pony up to the $xxx plan and you'll get a live body to help you figure that out. > 2. this version of grep has some bugs or altered functionality in edge > cases That's ok also, pony up to the $xxxx program and an engineer will look at it and help assist in getting it to work. At some point, should the use be willing to pay, Sun can work on resolving it on a per hour basis, just as professional services have been done for ages. > 3. This other version was what caused [foo] to not work right. That's ok also, foo was covered in the $xxxx plan <wink>, but in the $xxxxx plan maybe you can get 2 engineers, or the $xxxxxx plan. At some point, just like everything else in life, we find what is feasable to us and/or fits within our budget. For those that can wait and/or find their own workaround, the open source software will work fine for them. Sun can also have releases of Indiana that are considered stable, for instance, and if people need more feature and/or function, they can add updates to it, you can get that in my $xxx plan above!;-) > If you dont set up some kind of expectation of, "dont muck with bin like > that", then messes like that will be bound to happen. I'm fine with that happening, if the user wants Sun to help, cough up for support, or shut up and find a resolution yourself. The sources are there, as they say!;-) > In the case of Debian, even though I'm not aware of any explicit rule > against that sort of thing... I think the person would be lambasted for > wasting peoples time. There is some amount of expectation that "a debian > system", would have the debian version of grep installed. That is bound to happen in the opensolaris community and actually we've seen quite a few flame-ups happen on this very list over controversial issues. > In the case of Indiana... I think it would be beneficial to have some > kind of explicit statement of what to expect from /usr/bin, and what > kinds of things should be installed "elsewhere". (and where "elsewhere" > should be!) In addition to avoiding timewasters like the above, i think > it would help to build confidence in what to expect from "Indiana". Why so? I think that whatever Sun decides, it's their product and ultimately the folks that will want to use it will not care about much of that. They really don't care if it's in /usr/bin, /opt/sfw/, or /usr/local/your/mamma/wears/army/boots...if that makes sense. They just want to use their software to check their mail and surf the web in most cases. > But, all this again depends on what exactly Indiana is supposed to be. I agree, it's been morphing, and I think it's heading in the right direction. Will you be at the summit? A lot has happened over the past 5-6 years, heh? Solaris x86 is more of a contender today than it ever has been. When we get AcroRead, the world should rejoice!<wink> (I did say *when*, because I believe that it will be here at some point) -- Alan DuBoff - Those without dreams will never have dreams come true... From alan.duboff at sun.com Fri Sep 21 01:07:16 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 01:07:16 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F308F4.8050101@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> Message-ID: <Pine.GSO.4.64.0709210106540.5157@eagle.softorchestra.com> On Thu, 20 Sep 2007, Darren Reed wrote: > Possibly. And having put the problem like this, for better or worse, > what OpenSolaris needs is a board of directors (or the equivalent): > a body that can make decisions; not just have discussions via email > and try to resolve "issues". > > How would that be different to the OGB? No open board-discuss > for starters. Have it function like a real board: meet monthly or > quarterly, publish the minutes of the meetings, appoint a chairman, > give the board power to make decisions that change opensolaris > without needing votes, etc... I thought this was what the OGB was doing...silly me... -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Fri Sep 21 01:15:06 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 01:15:06 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F32251.7060102@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> Message-ID: <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> On Fri, 21 Sep 2007, Glynn Foster wrote: > Alan DuBoff wrote: >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: >> >>> Again, my biggest question/concern is, how many people will be in >>> attendance that the gory details of driver/kernel programming will >>> appeal to? I know you've voted, so that's one... are there more? >> >> I'm sure there are more, but the problem is that most of the people that >> are interested, are spread out more globally, as a whole, so targeting >> specific venues will be hit/miss at best in most cases. >> >> We need this information online, with the ability to stream in any >> location, so that the right people will have easy access to it. > > "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind tour of > a week-long class on how to write a driver. Seth touched on the DDI/DKI > interface, which provides terrific interface stability from generation to > generation, the device tree and tips and tricks. There was a demonstration > driver source presented as well, and all of the commands to try it out. I'll > have to do that when I get home!" > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > Sounds like the information is there (though possibly not GLDv3?), would > probably be good to highlight it more? Yes, actually Seth's session was good and pretty well attended. I would like to see Seth do more stuff that is not so crunched in, and to be able to have more shorts that are say, 5 minutes, so that you might have 20 shorts to cover the same thing, or to take 5-10 minutes on each topic. Next week we're going to try to video a session that we're doing in MPK, on using the kernel debugger. I hope this will be the first in a series of content we can offer on YouTube and OpenSolaris. OpenSolaris will have high resolution content, opposed to YouTube which is a great mechanism to reach out (so is facebook), but the quality pales, and having a draw with high resolution video is a good thing, IMO, which will get traffic to OpenSolaris.org. Hey, do I know how I will host it? Nope, don't even care...will figure that out after I have something that is a proof of concept, and make sure that Sun provides the storage and/or bandwidth. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Glynn.Foster at Sun.COM Fri Sep 21 05:17:53 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Sat, 22 Sep 2007 00:17:53 +1200 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> Message-ID: <46F3B671.4090509@sun.com> Hey, Alan DuBoff wrote: > Hey, do I know how I will host it? Nope, don't even care...will figure > that out after I have something that is a proof of concept, and make sure > that Sun provides the storage and/or bandwidth. If you don't manage to make any in-roads on this, please let me know and I'll try from my side. I'm sure it won't be a problem - just a case of finding the right person. I think it would be awesome to have a series of tech talks from various engineers. Glynn From swalker at opensolaris.org Fri Sep 21 08:44:43 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 21 Sep 2007 10:44:43 -0500 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F31335.6060700@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> Message-ID: <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> On 20/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > Philip Brown wrote: > > >On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: > > > > > >>Philip Brown wrote: > >> > >> > >>>Isnt that what the (previously named?) "CAB" is for? > >>> > >>> > >>... > >> > >>How would that be different to the OGB? No open board-discuss > >>for starters. Have it function like a real board: meet monthly or > >>quarterly, publish the minutes of the meetings, appoint a chairman, > >>give the board power to make decisions that change opensolaris > >>without needing votes, etc... > >> > >> > > > >The one trouble with doing stuff politically like that, is that there is > >then a tendancy for those positions to be filled with politicians. > >There need to be rules, but not to the point of needing a > >"full time politician" to handle it, so to speak. > > > > > > To which I'd say that the board must be comprised of only those > who commit or make changes to the source code of opensolaris, > not those that are only involved in the ARC, website, etc. That > there are few "code cutters" that are "politicians" tends to solve > the "politician" problem (or at least this is my observation.) > > Open source projects are fundamentally about code, everything > else is just window dressing. The right to be taken seriously > should be inline with the contribution to the source code base. I don't think you can call the work of writing documentation, providing i18n translations, or running user groups "window dressing." Those are part of the lifeblood of the community. While my contributions are code, and not any of the other things, documentation, etc. are just as important in my mind as code contributions. I'm sure you just accidentally omitted those items... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From phil at bolthole.com Fri Sep 21 08:51:30 2007 From: phil at bolthole.com (Philip Brown) Date: Fri, 21 Sep 2007 08:51:30 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com>; from swalker@opensolaris.org on Fri, Sep 21, 2007 at 10:44:43AM -0500 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> Message-ID: <20070921085130.F7529@bolthole.com> On Fri, Sep 21, 2007 at 10:44:43AM -0500, Shawn Walker wrote: > On 20/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > > Open source projects are fundamentally about code, everything > > else is just window dressing. The right to be taken seriously > > should be inline with the contribution to the source code base. > > I don't think you can call the work of writing documentation, > providing i18n translations, or running user groups "window dressing." > Those are part of the lifeblood of the community. > > While my contributions are code, and not any of the other things, > documentation, etc. are just as important in my mind as code > contributions. Heh. they are "very important", and often make the difference between an "ok" product, and a "great" product. But they are not "just as important as the code". Without the documentation and translations.. the code still results in something useful. Without the code.. the documentation, translations, etc. are useless. That being said: coders, and tech writers, are often separate breeds. That is why for products, there is usually a separate TECH ARC group, vs documentation group. "tech" = "code". Similar authority, but over different areas of the product. Someone who is a great coder, and fully would be qualified for the "tech arc" group, may be completely unfit for the documentation oversight group, and vice versa. From swalker at opensolaris.org Fri Sep 21 09:10:09 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 21 Sep 2007 11:10:09 -0500 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070921085130.F7529@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> <20070921085130.F7529@bolthole.com> Message-ID: <b9c544f0709210910r1dc56d65kef150759e20651a9@mail.gmail.com> On 21/09/2007, Philip Brown <phil at bolthole.com> wrote: > On Fri, Sep 21, 2007 at 10:44:43AM -0500, Shawn Walker wrote: > > On 20/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > > > Open source projects are fundamentally about code, everything > > > else is just window dressing. The right to be taken seriously > > > should be inline with the contribution to the source code base. > > > > I don't think you can call the work of writing documentation, > > providing i18n translations, or running user groups "window dressing." > > Those are part of the lifeblood of the community. > > > > While my contributions are code, and not any of the other things, > > documentation, etc. are just as important in my mind as code > > contributions. > > Heh. > > they are "very important", and often make the difference between an "ok" product, > and a "great" product. > But they are not "just as important as the code". > > Without the documentation and translations.. the code still results in > something useful. > Without the code.. the documentation, translations, etc. are > useless. > > That being said: coders, and tech writers, are often separate breeds. > That is why for products, there is usually a separate TECH ARC group, vs > documentation group. > "tech" = "code". > Similar authority, but over different areas of the product. > > Someone who is a great coder, and fully would be qualified for the > "tech arc" group, may be completely unfit for the documentation oversight > group, and vice versa. But those things are important so that you can have users, and a program without users is a program that probably won't be getting any code... Sort of a chicken-and-egg problem. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Garrett.Damore at Sun.COM Fri Sep 21 09:21:47 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 09:21:47 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F32251.7060102@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> Message-ID: <46F3EF9B.90206@sun.com> Glynn Foster wrote: > Alan DuBoff wrote: > >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: >> >> >>> Again, my biggest question/concern is, how many people will be in >>> attendance that the gory details of driver/kernel programming will >>> appeal to? I know you've voted, so that's one... are there more? >>> >> I'm sure there are more, but the problem is that most of the people that >> are interested, are spread out more globally, as a whole, so targeting >> specific venues will be hit/miss at best in most cases. >> >> We need this information online, with the ability to stream in any >> location, so that the right people will have easy access to it. >> > > "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind tour of > a week-long class on how to write a driver. Seth touched on the DDI/DKI > interface, which provides terrific interface stability from generation to > generation, the device tree and tips and tricks. There was a demonstration > driver source presented as well, and all of the commands to try it out. I'll > have to do that when I get home!" > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > Sounds like the information is there (though possibly not GLDv3?), would > probably be good to highlight it more? > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all (all NIC drivers are STREAMs drivers). What I saw convinced me of the utter foolishness of trying to cram a week's worth of teaching material into one hour. However, I could probably present a simple GLDv2->v3 presentation class. That can be done in an hour if the attendees have at least a tiny inkling of kernel programming. I don't mind covering other topics, and one could probably do a whirlwind tour of kernel/driver APIs in a day, if one had the entire day to do it. -- Garrett > > Glynn > From alan.coopersmith at sun.com Fri Sep 21 09:21:27 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Fri, 21 Sep 2007 09:21:27 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F31335.6060700@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> Message-ID: <46F3EF87.9070705@sun.com> Darren Reed wrote: > To which I'd say that the board must be comprised of only those > who commit or make changes to the source code of opensolaris, > not those that are only involved in the ARC, website, etc. Is there anyone on the ARC who hasn't putback hundreds of changes to some gate? ARC members aren't a separate body of people - they're the senior engineers, who have been in the trenches for years and are sharing their battle scars so others don't have to get hit in the same places they did. In any case, this is way off topic for summit planning, and I await your proposed constitutional amendments on ogb-discuss. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From John.Plocher at Sun.COM Fri Sep 21 09:53:18 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 21 Sep 2007 09:53:18 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070921085130.F7529@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> <20070921085130.F7529@bolthole.com> Message-ID: <46F3F6FE.3000002@Sun.Com> Philip Brown wrote: > But they are not "just as important as the code". In my experience, the act of asking a developer to articulate what exactly their code/program/feature is intended to do (versus what it actually does or how it happens to be implemented today...) invariably causes the developer to produce a better program. Whether this is done at the level of ARC-itectural review or as part of writing the man pages or configuration docs, it /is/ an integral part of software /engineering/. Without this focus on intent, all you have is random hacking, which is far less interesting to this community. -John From brandorr at opensolaris.org Fri Sep 21 10:07:16 2007 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 21 Sep 2007 13:07:16 -0400 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F3EF9B.90206@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> Message-ID: <5b5090780709211007x6f73803bq2c71151e2b291c54@mail.gmail.com> On 9/21/07, Garrett D'Amore <Garrett.Damore at sun.com> wrote: > > Glynn Foster wrote: > > Alan DuBoff wrote: > > > >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: > >> > >> > >>> Again, my biggest question/concern is, how many people will be in > >>> attendance that the gory details of driver/kernel programming will > >>> appeal to? I know you've voted, so that's one... are there more? > >>> > >> I'm sure there are more, but the problem is that most of the people > that > >> are interested, are spread out more globally, as a whole, so targeting > >> specific venues will be hit/miss at best in most cases. > >> > >> We need this information online, with the ability to stream in any > >> location, so that the right people will have easy access to it. > >> > > > > "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind > tour of > > a week-long class on how to write a driver. Seth touched on the DDI/DKI > > interface, which provides terrific interface stability from generation > to > > generation, the device tree and tips and tricks. There was a > demonstration > > driver source presented as well, and all of the commands to try it > out. I'll > > have to do that when I get home!" > > > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > > > Sounds like the information is there (though possibly not GLDv3?), would > > probably be good to highlight it more? > > > > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all > (all NIC drivers are STREAMs drivers). > > What I saw convinced me of the utter foolishness of trying to cram a > week's worth of teaching material into one hour. > > However, I could probably present a simple GLDv2->v3 presentation > class. That can be done in an hour if the attendees have at least a > tiny inkling of kernel programming. > > I don't mind covering other topics, and one could probably do a > whirlwind tour of kernel/driver APIs in a day, if one had the entire day > to do it. Garrett, There will be only be 60-100 attendees to the summit. I think your driver presentations deserve a wider audience. Is there anyway you can work with Alan to make videos and put additional course collaterals online? This way they would be able to help existing driver developers, and would also enable new vendors to contribute Solaris drivers. (Basically I think we should start reaching out to vendors like 3Ware, and offer them unsolicited help porting their drivers... wait a sec,, I just think I thought of a topic for the Summit).. "Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." Do you want to lead up this vendor/community outreach program? If so I could see this as a much more powerful topic/initiative to present at the summit. Cheers, Brian -- Garrett > > > > Glynn > > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070921/d903ecd3/attachment.html> From alan.duboff at sun.com Fri Sep 21 10:46:35 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 10:46:35 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F3B671.4090509@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> <46F3B671.4090509@sun.com> Message-ID: <Pine.GSO.4.64.0709211041180.5157@eagle.softorchestra.com> On Sat, 22 Sep 2007, Glynn Foster wrote: > Hey, > > Alan DuBoff wrote: >> Hey, do I know how I will host it? Nope, don't even care...will figure >> that out after I have something that is a proof of concept, and make >> sure that Sun provides the storage and/or bandwidth. > > If you don't manage to make any in-roads on this, please let me know and > I'll try from my side. I'm sure it won't be a problem - just a case of > finding the right person. I think it would be awesome to have a series > of tech talks from various engineers. I certainly will if I can't make headway with it after I have content. I know that Will Snow was the one responsible for the machines that host opensolaris.org, and he does handle the streaming media for sun.com, AFAIK, or at least used to. I figure however we need to drive that to get content hosted, we will sort that out. The main thing is to get something going where we have actual content and to be able to create shorts that we could use for educational/training/learning. I have a couple engineers in the queue once I get this rolling, and I hope that we can get some useful content to post. My plan is to use YouTube and/or Facebook as an exposure tool, but offer high resolution on opensolaris.org as added value. If folks would like a higher resolution stream, they can get it "from the source" so to speak!;-) -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Fri Sep 21 10:55:22 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 10:55:22 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F3EF9B.90206@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> Message-ID: <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> On Fri, 21 Sep 2007, Garrett D'Amore wrote: > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all (all NIC > drivers are STREAMs drivers). Well, sure his presentation was not intended to do that, he had to dumb it down as it was and was constrained to about 45 minutes. A big round of applause for Seth in being able to work around those contraints and requests from Intel to dumb it down. > What I saw convinced me of the utter foolishness of trying to cram a week's > worth of teaching material into one hour. Absolutely agreed. > However, I could probably present a simple GLDv2->v3 presentation class. > That can be done in an hour if the attendees have at least a tiny inkling of > kernel programming. That would be excellent, where would you do that? Not for the summit, is it? I mean, that really doesn't fit in with the topics at the summit, the summit will most likely be discussing how we could incorporate drivers that are not in OpenSolaris today, not write new ones, but maybe I'm mistaken. > I don't mind covering other topics, and one could probably do a whirlwind > tour of kernel/driver APIs in a day, if one had the entire day to do it. I think preparing a session on that would be good to have for the appropriate venue, certainly. -- Alan DuBoff - Solaris x86 IHV/OEM Group From brandorr at opensolaris.org Fri Sep 21 11:33:12 2007 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 21 Sep 2007 14:33:12 -0400 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." Message-ID: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> I added a new topic to the wiki: http://www.genunix.org/wiki/index.php/OpenSolaris_Summit Blurb: Driver support is an acknowledged area where OpenSolaris has some catching up to do. We need to figure ways to get driver support going. One of these ways could be to setup driver development training materials, as well as by reaching out to vendors to aid in porting.(IE: Lend developers to vendors to assist in writing an OpenSolaris driver). Another way, that I thought of was to setup two teams. Team one would look at a GPL driver and write a document generic describing the hardware. Team two would use that document to write a license compatible OpenSolaris driver. A third way is to setup an opensolaris.org bounty system, where multiple people can join to sponsor a bounty for a particular driver. Heck, Sun could even throw in a little bounty cash for key drivers. :) (To get it started, I am willing to put up $100 for a 3Ware RAID driver.) Anyway I think that we need to put our heads together to figure out how to expand driver support. Cheers, -Brian -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Garrett.Damore at Sun.COM Fri Sep 21 11:34:59 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 11:34:59 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> Message-ID: <46F40ED3.70906@sun.com> Alan DuBoff wrote: > On Fri, 21 Sep 2007, Garrett D'Amore wrote: > >> Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all >> (all NIC drivers are STREAMs drivers). > > Well, sure his presentation was not intended to do that, he had to > dumb it down as it was and was constrained to about 45 minutes. A big > round of applause for Seth in being able to work around those > contraints and requests from Intel to dumb it down. This wasn't a complaint. Just a statement... STREAMs programming is very different from most other kinds of programming, and requires a unique mindset. As it was, I think Seth probably tried to cover too much too quickly. I *know* the material he covered cold, and quite frankly, my eyes were starting to glaze over during the session. Again, this is no fault of Seth's.... its just not practical to cover kernel programming in an hour. I believe a series of lectures, perhaps posted on-line, would be better. If I were designing a kernel class today, I'd break it up into a series of lectures like this: *) Simple kernel mode programming/fundamentals. - compiling a kernel module - memory allocation considerations - simple debug strategies - DDI/DDK - sources of other info - this might include a simple bug fix to an existing module - link flags... kernel linker dependencies (e.g. using -N xxx/xxx) *) Autoconfiguration. - minor numbers, major numbers - the device tree - kinds of kernel modules - clone drivers - pseudo-drivers - device binding - libdevinfo (userland) - detach, attach, getinfo, probe - module _ini, _fini, _info entry points - dynamic reconfiguration *) Character and Block Drivers - entry points to implement - strategy(9e) - memory mapping (mmap, etc.) - copyin, copyout - mapping device registers - interrupt framework - DDI access functions (ddi_get32, etc.) *) Kernel synchronization and threadng - mutexes, condition variables - taskqs - timeout - (maybe?) cyclics - other clock/timing issue (lbolt, etc.) - "contexts" (interrupt, kernel, user) *) 32- versus 64-bit programming - ioctl and pointer passing... type conversion macros - dangers... pointer/integer conversions, etc. - kernel memory maps *) Kernel statistics facility - how kstat works, sample code, etc. *) STREAMs drivers - messages, put, putnext, all the other good stuff. - you cannot be a STREAMs driver and a character/block driver - STREAMs perimeters - other synchronization issues (thou shalt not sleep!) - message types (M_DATA, M_IOCTL, etc.) - copyin/copyout techniques for streams (M_COPYIN, etc.) 5) Network drivers - description of DLPI - GLDv2 - GLDv3 *) SCSI HBA drivers (SCSA) *) USB drivers - kernel apis - libusb - special binding issues. *) Dynamic Reconfiguration - detach, attach issues - rcm - dacf (?) *) Power management - suspend/resume - device power management topics *) Testing and debugging - test tips - kmdb/mdb - DTrace intro Its possible that the break up is not ideal, but this is just a rough draft of what I'm thinking about. I've no idea how such a series of lectures could be developed, however. > >> What I saw convinced me of the utter foolishness of trying to cram a >> week's worth of teaching material into one hour. > > Absolutely agreed. > >> However, I could probably present a simple GLDv2->v3 presentation >> class. That can be done in an hour if the attendees have at least a >> tiny inkling of kernel programming. > > That would be excellent, where would you do that? Not for the summit, > is it? I mean, that really doesn't fit in with the topics at the > summit, the summit will most likely be discussing how we could > incorporate drivers that are not in OpenSolaris today, not write new > ones, but maybe I'm mistaken. I dunno. I was just offering. Its looking like the Summit is going to be about governance and community topics, and have little to do with code... to the point that I'm not sure if any actual developers are going other than myself. That bothers me somewhat. If there is a big disconnect between the engineers doing the work, and the folks who are going to be making policy, we're setting ourselves up for heartache later. It somehow doesn't feel right that a bunch of folks who aren't actually involved intimately with the practice of engineering should be determining exclusively the processes which engineers must follow. Of course, if this is just a pow-wow of marketing type folks who are mostly interested in figuring how to *market* OpenSolaris to the rest of the world, then maybe there's not much point in folks who are mostly interested in development from attending. (And that would, for the most part, include me. Don't get me wrong, I'm as keen as the next person to improve OpenSolaris adoption, but talking about web pages and marketing priorities just doesn't really turn me on.) > >> I don't mind covering other topics, and one could probably do a >> whirlwind tour of kernel/driver APIs in a day, if one had the entire >> day to do it. > > I think preparing a session on that would be good to have for the > appropriate venue, certainly. See above. Probably there are a lot of sessions. I don't think we have any other venue atm, but maybe I'm mistaken? -- Garrett From Joerg.Schilling at fokus.fraunhofer.de Fri Sep 21 11:31:37 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 21 Sep 2007 20:31:37 +0200 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> Message-ID: <46f40e09.1yLPvucU4qmPg+X3%Joerg.Schilling@fokus.fraunhofer.de> Brandorr <brandorr at opensolaris.org> wrote: > Another way, that I thought of was to setup two teams. Team one would > look at a GPL driver and write a document generic describing the > hardware. Team two would use that document to write a license > compatible OpenSolaris driver. If you use Europeans, both teams may be identical. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Garrett.Damore at Sun.COM Fri Sep 21 11:39:42 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 11:39:42 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <5b5090780709211007x6f73803bq2c71151e2b291c54@mail.gmail.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <5b5090780709211007x6f73803bq2c71151e2b291c54@mail.gmail.com> Message-ID: <46F40FEE.7000604@sun.com> I'm not normally in the Bay Area where Alan is, so such a collaborative effort would be 'challenging'. If someone else in southern california wants to try to set something up, I'm willing. The Sun office in San Diego is easily reachable to me, and I can get to the office in LA as well (although that's quite a bit longer drive for me... 3 hours instead of 1 hour.) FWIW, I'm not entirely convinced that HD is necessary or even useful. Although a 640x480 presentation allowing code to be shown on-screen may be useful. -- Garrett Brandorr wrote: > On 9/21/07, *Garrett D'Amore* <Garrett.Damore at sun.com > <mailto:Garrett.Damore at sun.com>> wrote: > > Glynn Foster wrote: > > Alan DuBoff wrote: > > > >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: > >> > >> > >>> Again, my biggest question/concern is, how many people will be in > >>> attendance that the gory details of driver/kernel programming will > >>> appeal to? I know you've voted, so that's one... are there more? > >>> > >> I'm sure there are more, but the problem is that most of the > people that > >> are interested, are spread out more globally, as a whole, so > targeting > >> specific venues will be hit/miss at best in most cases. > >> > >> We need this information online, with the ability to stream in any > >> location, so that the right people will have easy access to it. > >> > > > > "Writing Device Drivers for Solaris - Seth Goldberg provided a > whirlwind tour of > > a week-long class on how to write a driver. Seth touched on the > DDI/DKI > > interface, which provides terrific interface stability from > generation to > > generation, the device tree and tips and tricks. There was a > demonstration > > driver source presented as well, and all of the commands to try > it out. I'll > > have to do that when I get home!" > > > > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > > > > Sounds like the information is there (though possibly not > GLDv3?), would > > probably be good to highlight it more? > > > > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all > (all NIC drivers are STREAMs drivers). > > What I saw convinced me of the utter foolishness of trying to cram a > week's worth of teaching material into one hour. > > However, I could probably present a simple GLDv2->v3 presentation > class. That can be done in an hour if the attendees have at least a > tiny inkling of kernel programming. > > I don't mind covering other topics, and one could probably do a > whirlwind tour of kernel/driver APIs in a day, if one had the > entire day > to do it. > > > Garrett, > > There will be only be 60-100 attendees to the summit. I think your > driver presentations deserve a wider audience. Is there anyway you can > work with Alan to make videos and put additional course collaterals > online? This way they would be able to help existing driver > developers, and would also enable new vendors to contribute Solaris > drivers. (Basically I think we should start reaching out to vendors > like 3Ware, and offer them unsolicited help porting their drivers... > wait a sec,, I just think I thought of a topic for the Summit).. > > "Expanding OpenSolaris hardware support, reaching out to hardware > vendors and helping them port." > > Do you want to lead up this vendor/community outreach program? If so I > could see this as a much more powerful topic/initiative to present at > the summit. > > Cheers, > Brian > > -- Garrett > > > > Glynn > > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > <mailto:opensolaris-summit at opensolaris.org> > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > <http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit> > > > > > -- > - Brian Gupta > > http://opensolaris.org/os/project/nycosug/ From swalker at opensolaris.org Fri Sep 21 11:46:57 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 21 Sep 2007 13:46:57 -0500 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> Message-ID: <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> On 21/09/2007, Brandorr <brandorr at opensolaris.org> wrote: > I added a new topic to the wiki: > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit > > Blurb: > > Driver support is an acknowledged area where OpenSolaris has some > catching up to do. We need to figure ways to get driver support going. > > One of these ways could be to setup driver development training > materials, as well as by reaching out to vendors to aid in > porting.(IE: Lend developers to vendors to assist in writing an > OpenSolaris driver). > > Another way, that I thought of was to setup two teams. Team one would > look at a GPL driver and write a document generic describing the > hardware. Team two would use that document to write a license > compatible OpenSolaris driver. > > A third way is to setup an opensolaris.org bounty system, where > multiple people can join to sponsor a bounty for a particular driver. > Heck, Sun could even throw in a little bounty cash for key drivers. :) > (To get it started, I am willing to put up $100 for a 3Ware RAID > driver.) > > Anyway I think that we need to put our heads together to figure out > how to expand driver support. Before we can get started with adding driver support, we have to know first what is supported. I know there is a project in progress to allow the device detection tool to upload data. OpenSolaris needs its own hardware database to be able to support your proposal before it can move forward. OpenSound (especially now that it is open source) brings a significant amount of hardware support for sound devices to OpenSolaris (yet it may have bugs, but all software arguably does). Finding out what the status of that hardware database is and the device detection tool source code (which was supposed to be posted, haven't checked lately) would be the first step I think towards supporting your proposal. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Joerg.Schilling at fokus.fraunhofer.de Fri Sep 21 11:56:15 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 21 Sep 2007 20:56:15 +0200 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> Message-ID: <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" <swalker at opensolaris.org> wrote: > > Anyway I think that we need to put our heads together to figure out > > how to expand driver support. > > Before we can get started with adding driver support, we have to know > first what is supported. I believe that we should set up a list with anorder of importance and I believe that this should be layered to e.g. driver groups. I would be happy to see e.g Blue tooth support in my notebook and suspend (or better suspend) support. Then it makes sense to discuss specific hardware that is more important than others. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Darren.Reed at Sun.COM Fri Sep 21 12:55:57 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Fri, 21 Sep 2007 12:55:57 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709210106540.5157@eagle.softorchestra.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <Pine.GSO.4.64.0709210106540.5157@eagle.softorchestra.com> Message-ID: <46F421CD.4000102@Sun.COM> Alan DuBoff wrote: > On Thu, 20 Sep 2007, Darren Reed wrote: > >> Possibly. And having put the problem like this, for better or worse, >> what OpenSolaris needs is a board of directors (or the equivalent): >> a body that can make decisions; not just have discussions via email >> and try to resolve "issues". >> >> How would that be different to the OGB? No open board-discuss >> for starters. Have it function like a real board: meet monthly or >> quarterly, publish the minutes of the meetings, appoint a chairman, >> give the board power to make decisions that change opensolaris >> without needing votes, etc... > > > I thought this was what the OGB was doing...silly me... Maybe they are - there are better things to do than waste time trying to follow ogb-discuss and opensolaris-discuss. Darren From alan.duboff at sun.com Fri Sep 21 13:42:33 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 13:42:33 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F40ED3.70906@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> Message-ID: <Pine.GSO.4.64.0709211331260.8861@konocti> On Fri, 21 Sep 2007, Garrett D'Amore wrote: > This wasn't a complaint. Just a statement... STREAMs programming is > very different from most other kinds of programming, and requires a > unique mindset. As it was, I think Seth probably tried to cover too > much too quickly. Sure, but it was more of an overview session, not a hands on training course on writing device drivers. No fault of Seth for that, yes, there was quite a bit of information to absorb. I know we're in agreement on this. > I believe a series of lectures, perhaps posted on-line, would be > better. If I were designing a kernel class today, I'd break it up into > a series of lectures like this: Sure, I think that I stated something to the safe effect. > *) Simple kernel mode programming/fundamentals. > *) Autoconfiguration. > *) Character and Block Drivers > *) Kernel synchronization and threadng > *) 32- versus 64-bit programming > *) Kernel statistics facility > *) STREAMs drivers > 5) Network drivers > *) SCSI HBA drivers (SCSA) > *) USB drivers > *) Dynamic Reconfiguration > *) Power management > *) Testing and debugging These are all good and it would seem we get back to a similar course that Max Brunning currently teaches at SunU. >> That would be excellent, where would you do that? Not for the summit, >> is it? I mean, that really doesn't fit in with the topics at the >> summit, the summit will most likely be discussing how we could >> incorporate drivers that are not in OpenSolaris today, not write new >> ones, but maybe I'm mistaken. > > I dunno. I was just offering. Its looking like the Summit is going to > be about governance and community topics, and have little to do with > code... to the point that I'm not sure if any actual developers are > going other than myself. I think that was always the intention. Again, I could be wrong. > That bothers me somewhat. If there is a big disconnect between the > engineers doing the work, and the folks who are going to be making > policy, we're setting ourselves up for heartache later. It somehow > doesn't feel right that a bunch of folks who aren't actually involved > intimately with the practice of engineering should be determining > exclusively the processes which engineers must follow. You wouldn't be alone in this feeling, I suspect there are quite a few folks in engineering that are still scratching their heads trying to figure out what exactly Indiana is. From the engineering standpoint, Sun hired Ian and they all of the sudden started hearing things that supposedly they were working on. I believe Ian is trying to do the best he can in the situation he's also been put in, but that doesn't help the communication between those folks involved in that and the actual engineers. > Of course, if this is just a pow-wow of marketing type folks who are > mostly interested in figuring how to *market* OpenSolaris to the rest of > the world, then maybe there's not much point in folks who are mostly > interested in development from attending. I don't think it's fair to call it marketing type folks, there will be quite a few engineers there who are involved in pieces that will be used for Indiana, several of which were already under development before anyone heard light of Indiana. > (And that would, for the most > part, include me. Don't get me wrong, I'm as keen as the next person > to improve OpenSolaris adoption, but talking about web pages and > marketing priorities just doesn't really turn me on.) I don't think it's just about Solaris adoption, I think it's about figuring out a way that those marketing types and engineering types such as yourself to find a way to make all of this work, and be able to create a product that can be distributed in a given timeframe. Again, maybe I'm offbase on this concept as well, I'm just grasping what exactly Indiana is myself, and the definition has been a moving target. > See above. Probably there are a lot of sessions. I don't think we have > any other venue atm, but maybe I'm mistaken? We have Tech Days that travels around, and does present similar session, most of similar length however. This is something that might be worth talking to Chris Atwood about, he's in charge of training at Sun, and offering courses for that purpose (i.e., SunU). -- Alan DuBoff - Solaris x86 IHV/OEM Group From Garrett.Damore at Sun.COM Fri Sep 21 14:15:15 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 14:15:15 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709211331260.8861@konocti> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> Message-ID: <46F43463.9080106@sun.com> > > We have Tech Days that travels around, and does present similar > session, most of similar length however. > > This is something that might be worth talking to Chris Atwood about, > he's in charge of training at Sun, and offering courses for that > purpose (i.e., SunU). One question, and I don't know what the answer is, is how SunU fits in with what I was talking about. SunU charges lots of $$$ for their courses. But you get certified instructors, course materials, and hands-on lab time. What I was talking about was something that would be free, might cover topics that SunU doesn't (such as new information relating specifically to Nevada), and would be downloadable. I don't know how we avoid stepping on toes, etc. Anyway, I'm going to back away from this for now, only because what I'm hearing is a lot of reasons why it won't work, and not enough reason/request/desire to see such a thing work out. I have enough stuff on m plate to do that I know people want not to spend time spinning my wheels on stuff other people aren't interested in. -- Garrett From alan.duboff at sun.com Fri Sep 21 14:41:39 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 14:41:39 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F43463.9080106@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> Message-ID: <Pine.GSO.4.64.0709211434150.8861@konocti> On Fri, 21 Sep 2007, Garrett D'Amore wrote: >> We have Tech Days that travels around, and does present similar >> session, most of similar length however. >> >> This is something that might be worth talking to Chris Atwood about, >> he's in charge of training at Sun, and offering courses for that >> purpose (i.e., SunU). > > One question, and I don't know what the answer is, is how SunU fits in > with what I was talking about. I can't speak for those folks, but I have talked to them. I have pointed out to them that we need a more cost effective solution, to educate the community. I would like to see something creative, like using a black box that we drop ship to a specific location to accomodate folks for such training. Or using universities as a venue where we could do that with an army of laptops. There are various ways we could leverage new ideas and concepts to the old fashioned SunU methodology. My $0.02. > SunU charges lots of $$$ for their courses. But you get certified > instructors, course materials, and hands-on lab time. > > What I was talking about was something that would be free, might cover > topics that SunU doesn't (such as new information relating specifically > to Nevada), and would be downloadable. Yes, we're talking about the same thing, but after I contacted and/or talked to them myself, I felt it would be a better investment of time and $$$s to produce our own content and post it ourselves, and be done with it. > I don't know how we avoid stepping on toes, etc. Anyway, I'm going to > back away from this for now, only because what I'm hearing is a lot of > reasons why it won't work, and not enough reason/request/desire to see > such a thing work out. I have enough stuff on m plate to do that I know > people want not to spend time spinning my wheels on stuff other people > aren't interested in. I think there is interest, just that in the global community we're in now, it needs to be leveraged in a way that can encapsulate those folks and be able to reach them in their specific locales. Providing streaming content is one such way, and one of the more likely ones. The folks that are initmately associated with OpenSolaris.org must not have that on the top of their list, which is understandable, but I think it's a very important aspect for our community. Apple would have had QuickTime content riddled throughout the site had they rolled something out, and as such Apple does think differently, at least than Sun. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Moazam.Raja at Sun.COM Fri Sep 21 14:43:57 2007 From: Moazam.Raja at Sun.COM (Moazam Raja) Date: Fri, 21 Sep 2007 14:43:57 -0700 Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F43463.9080106@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> Message-ID: <46F43B1D.4060609@sun.com> > One question, and I don't know what the answer is, is how SunU fits in > with what I was talking about. > > SunU charges lots of $$$ for their courses. But you get certified > instructors, course materials, and hands-on lab time. > > What I was talking about was something that would be free, might cover > topics that SunU doesn't (such as new information relating specifically > to Nevada), and would be downloadable. > > -- Garrett This sounds like something the local OpenSolaris user groups would be good for (providing 'classes' for free). I'm going to be working on some podcast/video-podcasts geared towards beginner SysAdmins, that might be a start for SVOSUG. -Moazam From storycrafter at gmail.com Fri Sep 21 14:45:08 2007 From: storycrafter at gmail.com (Mark Martin) Date: Fri, 21 Sep 2007 16:45:08 -0500 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F43463.9080106@sun.com> References: <46EB4DE2.8080808@Sun.COM> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> Message-ID: <e40c28290709211445p72deec88i8f20547b1f2293cb@mail.gmail.com> On 9/21/07, Garrett D'Amore <Garrett.Damore at sun.com> wrote: > > > I don't know how we avoid stepping on toes, etc. Anyway, I'm going to > back away from this for now, only because what I'm hearing is a lot of > reasons why it won't work, and not enough reason/request/desire to see > such a thing work out. I have enough stuff on m plate to do that I know > people want not to spend time spinning my wheels on stuff other people > aren't interested in. > > FWIW, I hope this doesn't die completely. I have been lurking in interest because: a) I was hoping there would be some way to consume content from the summit without being able to attend b) That some of that content would be useful to a (potential) contributor outside Sundom c) I might learn more about how to navigate the waters into actually contributing, as I believe I'm only about 10% into that process Broad sips would help me more now than deep pulls from the information hose, and any tidbits I can get from a coding and process standpoint would be 100% useful to me, especially in easily digested streaming video form. I'm hoping Alan's vision comes forward more -- low-res content on YouTube and hi-res on opensolaris.org. To that end, I'd even volunteer time for editing and managing video if that's required. Either way, at some point in my personal future timeline I expect to want to have the in-depth stuff -- the 1 and 2 hour or even 8 hour sessions. I realize that may not come from 15 minute session talks from a summit or TechDay or whatever but hopefully I don't have to go too far away (virtually) from opensolaris.org to find that. And honestly, I don't want to have to go to SunU to get it either unless that's a no-cost proposition. If there's a format, curriculum, or content that can be "leveraged" from there, that seems ostensibly like a win, though. Device driver implementation is of concern for me -- even if I don't have qualifications in that particular area to contribute effectively -- because that's a perceived weakness of OpenSolaris as far as I'm concerned. I'd volunteer what meager resources I could to see that any and all information, especially in streaming media format, came to light. Mark -- ------------------------------------------------------ Born to the false world, the wanderer, Storyteller, The Pied Piper On a quest for immortality Gathering a troop to find the fantasy -- Nightwish -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070921/d5a98bf8/attachment.html> From alan.duboff at sun.com Fri Sep 21 14:57:34 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 14:57:34 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F43B1D.4060609@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> Message-ID: <Pine.GSO.4.64.0709211455010.8861@konocti> On Fri, 21 Sep 2007, Moazam Raja wrote: >> One question, and I don't know what the answer is, is how SunU fits in >> with what I was talking about. >> >> SunU charges lots of $$$ for their courses. But you get certified >> instructors, course materials, and hands-on lab time. >> >> What I was talking about was something that would be free, might cover >> topics that SunU doesn't (such as new information relating specifically >> to Nevada), and would be downloadable. > > >> -- Garrett > > This sounds like something the local OpenSolaris user groups would be > good for (providing 'classes' for free). I'm going to be working on some > podcast/video-podcasts geared towards beginner SysAdmins, that might be > a start for SVOSUG. Absolutely! Seems we do have a location to host our videos, I saw Will Snow at lunch, he manages the opensolaris.org servers. He also handles streaming media, and said that we can host it on media.sun.com (I believe that is the site but it doesn't pull anything up). We can host any content that is created though, Moazam, please feel free to move forward with any ideas you have. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.coopersmith at sun.com Fri Sep 21 15:10:23 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Fri, 21 Sep 2007 15:10:23 -0700 Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <Pine.GSO.4.64.0709211455010.8861@konocti> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> <Pine.GSO.4.64.0709211455010.8861@konocti> Message-ID: <46F4414F.8010405@sun.com> Alan DuBoff wrote: > Absolutely! Seems we do have a location to host our videos, I saw Will > Snow at lunch, he manages the opensolaris.org servers. He also handles > streaming media, and said that we can host it on media.sun.com (I believe > that is the site but it doesn't pull anything up). Try http://mediacast.sun.com/ -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From brandorr at opensolaris.org Fri Sep 21 15:19:37 2007 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 21 Sep 2007 18:19:37 -0400 Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F4414F.8010405@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> <Pine.GSO.4.64.0709211455010.8861@konocti> <46F4414F.8010405@sun.com> Message-ID: <5b5090780709211519y1618dd79l83ffed27721b915@mail.gmail.com> On 9/21/07, Alan Coopersmith <alan.coopersmith at sun.com> wrote: > Alan DuBoff wrote: > > Absolutely! Seems we do have a location to host our videos, I saw Will > > Snow at lunch, he manages the opensolaris.org servers. He also handles > > streaming media, and said that we can host it on media.sun.com (I believe > > that is the site but it doesn't pull anything up). > > Try http://mediacast.sun.com/ Also the documentation community has some videos listed on there homepage: http://www.opensolaris.org/os/community/documentation/ It looks like they are being hosted here: http://frsun.downloads.edgesuite.net/sun/07C00892/media/demos/OpenSolarisDualBoot-Step4-OS-Installation.html -Brian > > -- > -Alan Coopersmith- alan.coopersmith at sun.com > Sun Microsystems, Inc. - X Window System Engineering > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From alan.duboff at sun.com Fri Sep 21 15:24:35 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 15:24:35 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F4414F.8010405@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> <Pine.GSO.4.64.0709211455010.8861@konocti> <46F4414F.8010405@sun.com> Message-ID: <Pine.GSO.4.64.0709211517220.8861@konocti> On Fri, 21 Sep 2007, Alan Coopersmith wrote: > Alan DuBoff wrote: >> Absolutely! Seems we do have a location to host our videos, I saw Will >> Snow at lunch, he manages the opensolaris.org servers. He also handles >> streaming media, and said that we can host it on media.sun.com (I >> believe that is the site but it doesn't pull anything up). > > Try http://mediacast.sun.com/ > Yes, that's it Alan, thanks. Will said we can upload stuff there. BTW, a short video on using XRandR would be a good one to have, that's a very handy utility for folks to know about. After we get things going maybe you or Stuart would like to help with that. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Jim.Grisanzio at Sun.COM Fri Sep 21 18:53:20 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Sat, 22 Sep 2007 10:53:20 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <18163.64275.538538.803812@gargle.gargle.HOWL> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> Message-ID: <46F47590.4010309@sun.com> James Carlson wrote: > Stephen Lau writes: >> Darren Reed wrote: >>> To which I'd say that the board must be comprised of only those >>> who commit or make changes to the source code of opensolaris, >>> not those that are only involved in the ARC, website, etc. > [...] >>> Open source projects are fundamentally about code, everything >>> else is just window dressing. The right to be taken seriously >>> should be inline with the contribution to the source code base. >> Which only sets up an even more elitist atmosphere. You would be >> eliminating the valuable contributions of people who are docs people, >> translations, artists, evangelists, etc. > > Indeed. > >> There is nothing inherent about cutting code that somehow makes someone >> qualified to run or lead an open source project or community. > > I agree. I agree as well. Many people have worked hard to build a diverse community open to all. I think we've come a long way, too, given the restrictions and dependencies we've been working with all alone. These things take time, and I think Indiana will help us grow to a new level of diversity. Diversity has been the intention all along from our very earliest planning meetings four years ago. The code is at the center and that first layer of coders is critical as a foundation for the entire community, no question about it. But communities are more than foundations. Also, as we grow in emerging regions around the world, we'll be engaging significantly more non-coders than coders (at least initially). Jim > Perhaps more importantly for this thread, though, I would like some > clarification from Darren on what he thinks are the existing > deficiencies in the OGB. In other words, I can't tell if he's > speaking in generic terms about all open source projects aside from > OpenSolaris, or if he has some clear complaint to be made about the > current membership. The wording seems deliberately inflammatory, > though. > > If he does in fact have a complaint, I'd like to hear it. I *am* an > ARC member, along with one of the other elected OGB members, but that > somehow doesn't keep me (or Alan) from writing code. Am I the > unqualified one due to a lack of political prowess, serving on the > wrong committees, or just for my code? I'd really like to know. > > Who are those drag-inducing people of whom Darren writes? If I'm it, > please make sure I don't end up winning the first round of the vote > next time we do this. ;-} > From Garrett.Damore at Sun.COM Fri Sep 21 19:04:23 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 19:04:23 -0700 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F47827.5000208@sun.com> Joerg Schilling wrote: > "Shawn Walker" <swalker at opensolaris.org> wrote: > > >>> Anyway I think that we need to put our heads together to figure out >>> how to expand driver support. >>> >> Before we can get started with adding driver support, we have to know >> first what is supported. >> > > I believe that we should set up a list with anorder of importance and I believe > that this should be layered to e.g. driver groups. I would be happy to see e.g > Blue tooth support in my notebook and suspend (or better suspend) support. > Suspend is working just fine now. Its *resume* that is the tricky part. ;-) Seriously, though, I have just been asked to start looking at Suspend/Resume for a certain brand of laptop. Its likely that once that work is done, we'll expand the scope to include support for a variety of other models. Suspend in this case means S3 (suspend to ram) initially. S4 (suspend to disk) is something we still need on x86 as well. (We already have it on SPARC.) > Then it makes sense to discuss specific hardware that is more important than > others. > There will always be favorite hardware that is missing from the list. Making such a list is useful, but shouldn't gate the effort of getting information out to folks who want to help. I can think probably of a half-dozen or so simple driver projects for folks to undertake, if they were so inclined. Right now, the number of people with time and inclination, and ability (or ability to learn) is the limiting factor. Anything that addresses *that* gap is likely to be of the most use. -- Garrett > J?rg > > From Darren.Reed at Sun.COM Fri Sep 21 19:54:41 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Fri, 21 Sep 2007 19:54:41 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <18163.64275.538538.803812@gargle.gargle.HOWL> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> Message-ID: <46F483F1.1050805@Sun.COM> James Carlson wrote: >Stephen Lau writes: > >>Darren Reed wrote: >> >>>To which I'd say that the board must be comprised of only those >>>who commit or make changes to the source code of opensolaris, >>>not those that are only involved in the ARC, website, etc. >>> >[...] > >>>Open source projects are fundamentally about code, everything >>>else is just window dressing. The right to be taken seriously >>>should be inline with the contribution to the source code base. >>> >>Which only sets up an even more elitist atmosphere. You would be >>eliminating the valuable contributions of people who are docs people, >>translations, artists, evangelists, etc. >> > >Indeed. > The suggestion was an off-the-cuff remark. A more accurate pool of people from which to make up a board would be the contributors and core contributors combined as this should (I hope) include those who right documentation, etc. I see no reason why there should be any distinction between core and non-core contributors. But to this day, I've not seen any open source project dominated by "politicians" and thus I'd opine that to suggest that this would happen is not entirely accurate. Or to put it differently, if it did happen within OpenSolaris then it would be the first place it had and would make me wonder more deeply about where OpenSolaris is going. On top of that there are ways and means (like 1 or 2 year limits on terms, combined with limits on how frequently people can come back) to limit the attractiveness of using a position for political gain only. My observation of other projects is that people get "tired" of such positions after a period of time so the other issues become less of a concern than one might expect - especially since these offices are all pro-bono (and should remain that way.) >>There is nothing inherent about cutting code that somehow makes someone >>qualified to run or lead an open source project or community. >> > >I agree. > >Perhaps more importantly for this thread, though, I would like some >clarification from Darren on what he thinks are the existing >deficiencies in the OGB. In other words, I can't tell if he's >speaking in generic terms about all open source projects aside from >OpenSolaris, or if he has some clear complaint to be made about the >current membership. The wording seems deliberately inflammatory, >though. > The OGB doesn't appear to me to be actually doing anything. It appears more more like an arbitration council that also sanctions community decisions about which communities to create and how to make a contributor/core contributor. "So what". In this guise it feels more like a rubber stamp operation than anything meaningful. Darren p.s. Meeting minutes should always include the full name of all those present, not just first names. From swalker at opensolaris.org Fri Sep 21 22:56:39 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 22 Sep 2007 00:56:39 -0500 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <46F47827.5000208@sun.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> <46F47827.5000208@sun.com> Message-ID: <b9c544f0709212256r11508ca1w7a848c049856b602@mail.gmail.com> On 21/09/2007, Garrett D'Amore <Garrett.Damore at sun.com> wrote: > Joerg Schilling wrote: > > "Shawn Walker" <swalker at opensolaris.org> wrote: > > > > > >>> Anyway I think that we need to put our heads together to figure out > >>> how to expand driver support. > >>> > >> Before we can get started with adding driver support, we have to know > >> first what is supported. > >> > > > > I believe that we should set up a list with anorder of importance and I believe > > that this should be layered to e.g. driver groups. I would be happy to see e.g > > Blue tooth support in my notebook and suspend (or better suspend) support. > > > > > Suspend is working just fine now. Its *resume* that is the tricky part. ;-) > > Seriously, though, I have just been asked to start looking at > Suspend/Resume for a certain brand of laptop. Its likely that once that > work is done, we'll expand the scope to include support for a variety of > other models. > > Suspend in this case means S3 (suspend to ram) initially. S4 (suspend > to disk) is something we still need on x86 as well. (We already have it > on SPARC.) > > > Then it makes sense to discuss specific hardware that is more important than > > others. > > > > There will always be favorite hardware that is missing from the list. > Making such a list is useful, but shouldn't gate the effort of getting > information out to folks who want to help. > > I can think probably of a half-dozen or so simple driver projects for > folks to undertake, if they were so inclined. Right now, the number of > people with time and inclination, and ability (or ability to learn) is > the limiting factor. Anything that addresses *that* gap is likely to > be of the most use. I tend to agree; which is why I suggested we need a hardware database of some sort so we can track what does work, or what is supposed to work, along with what needs support. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Joerg.Schilling at fokus.fraunhofer.de Sat Sep 22 03:30:19 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Sat, 22 Sep 2007 12:30:19 +0200 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <46F47827.5000208@sun.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> <46F47827.5000208@sun.com> Message-ID: <46f4eebb.GvMw5wwYg+SRiQVQ%Joerg.Schilling@fokus.fraunhofer.de> "Garrett D'Amore" <Garrett.Damore at Sun.COM> wrote: > > I believe that we should set up a list with anorder of importance and I believe > > that this should be layered to e.g. driver groups. I would be happy to see e.g > > Blue tooth support in my notebook and suspend (or better suspend) support. > > > > > Suspend is working just fine now. Its *resume* that is the tricky part. ;-) You don't have a IBM notebook? Suspend _did_ work with Solaris 10 GA, it currently does not work. The resume problem was a frozen clock that could be reanimated with some tricks. > Suspend in this case means S3 (suspend to ram) initially. S4 (suspend > to disk) is something we still need on x86 as well. (We already have it > on SPARC.) Suspend to RAM is the more frequent case that already allows a day or more battery time. > There will always be favorite hardware that is missing from the list. > Making such a list is useful, but shouldn't gate the effort of getting > information out to folks who want to help. We are more than a few people, let us vote.... As long as Sun votes with one Dollar == one vote, some features will never make it into OpenSolaris. Solaris sells to other people than it lives from. > I can think probably of a half-dozen or so simple driver projects for > folks to undertake, if they were so inclined. Right now, the number of > people with time and inclination, and ability (or ability to learn) is > the limiting factor. Anything that addresses *that* gap is likely to > be of the most use. For smaller driver projects, the biggest impact currently is burocracy and the absence of a useful way to integrate code. As we should not impact the quality, I propose a 2 stage system. - A testing source tree allows anyone who has a sufficient credability to add code. If this code causes problems that are not fixed within a specific time, it will beremoved by a gatekeeper. - The primary source is fed by the testing source in case that there was no technical problem for some time and after a legal review was done. - Modifying sources including CLI "enhancements" still needs ARC approvement. - Pure add ons get a reversed ARC system: A sufficient amount of people need to convince the community that the additional code does not fit properly into OpenSolaris. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From shivakumar.gn at gmail.com Sat Sep 22 03:52:42 2007 From: shivakumar.gn at gmail.com (S h i v) Date: Sat, 22 Sep 2007 16:22:42 +0530 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> Message-ID: <319ee2b10709220352h4fb9af78la12855577ba41468@mail.gmail.com> On 9/22/07, Shawn Walker <swalker at opensolaris.org> wrote: > Before we can get started with adding driver support, we have to know > first what is supported. > > I know there is a project in progress to allow the device detection > tool to upload data. > > OpenSolaris needs its own hardware database to be able to support your > proposal before it can move forward. > I see that submission of sddt output results is already there in B72 ISO. Tried submitting, the tool said results successfully submitted to Sun Microsystems Inc. If at all, this needs to be moved to a opensolaris.org database but the feature itself is already there. regards Shiv From Jim.Grisanzio at Sun.COM Sat Sep 22 04:01:00 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Sat, 22 Sep 2007 20:01:00 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F483F1.1050805@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> Message-ID: <46F4F5EC.9090909@sun.com> Darren Reed wrote: > The OGB doesn't appear to me to be actually doing anything. It > appears more more like an arbitration council that also sanctions > community decisions about which communities to create and how to > make a contributor/core contributor. "So what". In this guise > it feels more like a rubber stamp operation than anything meaningful. There was a lot of discussion about this when the initial OGB was writing the Constitution. I think the intent was to distribute leadership and decision making power among the communities and projects and not centralize it in the OGB. I agree with that model because it offers significant incentive and opportunity for participation across the entire community. That one element is what makes the OpenSolaris Constitution so meaningful. The OGB's job was to create and maintain the structure in which a community could grow. They did that. And they did that pretty well, actually. The OGB /represents/ the community, but it doesn't /run/ the community. Operations are distributed by design. The community runs itself via individual community groups and projects. When there are problems or meta issues to consider, the OGB has the authority to step in and mediate and resolve disputes. That seems to be a perfectly reasonable system to me. That's my impression of things, anyway. Jim -- http://blogs.sun.com/jimgris From roland.mainz at nrubsig.org Sun Sep 23 18:03:11 2007 From: roland.mainz at nrubsig.org (Roland Mainz) Date: Mon, 24 Sep 2007 03:03:11 +0200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> Message-ID: <46F70CCF.DD249188@nrubsig.org> Darren.Reed at Sun.COM wrote: > John Plocher wrote: > >Shawn Walker wrote: > >>I tend to agree that I hope that this is > >>more about problem solving than hacking... > > > >I see room for both - especially if we (Sun) can arrange for > >a reasonable set of high end systems to play with. > > > >While some topics (politics/governance/ARCs/...) won't be impacted > >much by the compute infrastructure at the conference, I can see other > >tracks taking a more hands-on dig in and play with code attitude. > > As originally advertised, the event was to be about discussion > relating to the project itself - it wasn't billed as a "hackathon" - > and more specifically, to discuss Indiana. > > If we're going to be more generic and even think about making > it a coding/hacking session then we need to readvertise it as > such. Erm, the idea the behind the requested stuff (short: one server and 2x x86 + 2xSPARC) was only for a few people sticking their heads together on some software problems, including booting custom kernels (e.g. the "server" is userland only (e.g. SunRay, home dirs, editors, compilers, OS/Net build tools etc.) and for jumpstarting the other machines). For a "hackathon" with 100 people we would AFAIK at least need ten times more machines... ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) From roland.mainz at nrubsig.org Sun Sep 23 18:14:08 2007 From: roland.mainz at nrubsig.org (Roland Mainz) Date: Mon, 24 Sep 2007 03:14:08 +0200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F15E8A.70407@sun.com> Message-ID: <46F70F60.662239E7@nrubsig.org> Jim Walker wrote: > John Plocher wrote: > > Shawn Walker wrote: > >> I tend to agree that I hope that this is > >> more about problem solving than hacking... > > > > I see room for both - especially if we (Sun) can arrange for > > a reasonable set of high end systems to play with. > > > > While some topics (politics/governance/ARCs/...) won't be impacted > > much by the compute infrastructure at the conference, I can see other > > tracks taking a more hands-on dig in and play with code attitude. > > > > It would be VERY USEFUL for us all to articulate what exactly we wish > > to get out of the conference - what problems need to be solved that > > would benefit from a concentrated, in person focus? (See the Wiki at > > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit for some > > suggestions) > > If useful, I could use some of the test farm automation > to manage reservation time slots on whatever hardware > we have at the summit. In addition, if we have good > network access we can get to remote test machines. Erm, AFAIK the test farm is non-interactive (e.g. you can't do something like "login and hack the kernel to death"), right ? The idea of the five machines was to have one "central hub" which is user-land only (e.g. nooone except two or three people have "root" access) which handles SunRay services, home dirs (which are exported to the four other machines), jumpstart and has some development tools (compiler, OS/Net build tools etc.) installed and then 2x x86 and 2x SPARC machines to boot custom kernels from the server on demand. This stuff is only thought to give people the opportunity to stick their heads together and try some stuff without having to improvise something with laptops (or worse) ... ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) From roland.mainz at nrubsig.org Sun Sep 23 18:28:00 2007 From: roland.mainz at nrubsig.org (Roland Mainz) Date: Mon, 24 Sep 2007 03:28:00 +0200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> Message-ID: <46F712A0.443DB9B8@nrubsig.org> Tim Foster wrote: > On Tue, 2007-09-18 at 16:56 -0700, Jesse Silver wrote: > > Server availability- > > Suggestions: > > - Have as many different kinds of machines available as possible, split > > between Sparc and x86. > > - Have at least one central server, stable and powerful, like an M4000 > > with a bunch of storage. > > Why servers ? I thought this was a collective "bang our heads together" > meeting, rather than an actual hackfest ? If if everyone has their noses > buried in laptops for the duration of the weekend, then I'm not sure > we'll have achieved our aims imho. Erm, seem other emails for a response. It's only to prevent the case of having many people around from all around thje globe who may want discuss some practical code stuff and then test or demonstrate it (including custom kernel mods) and then have no machine to do it (or rely on machines which need a day or two to crawl over the OS/Net codebase for a complete rebuild from scratch). --- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) From phil at bolthole.com Mon Sep 24 09:53:46 2007 From: phil at bolthole.com (Philip Brown) Date: Mon, 24 Sep 2007 09:53:46 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F483F1.1050805@Sun.COM>; from Darren.Reed@Sun.COM on Fri, Sep 21, 2007 at 07:54:41PM -0700 References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> Message-ID: <20070924095346.F14087@bolthole.com> On Fri, Sep 21, 2007 at 07:54:41PM -0700, Darren Reed wrote: > > But to this day, I've not seen any open source project dominated > by "politicians" and thus I'd opine that to suggest that this would > happen is not entirely accurate. I have. The reason that YOU havent, is that those projects FAILED. :-} hence, my concern. > Or to put it differently, if it did happen within OpenSolaris then > it would be the first place it had and would make me wonder more > deeply about where OpenSolaris is going. indeed. From phil at bolthole.com Mon Sep 24 10:09:26 2007 From: phil at bolthole.com (Philip Brown) Date: Mon, 24 Sep 2007 10:09:26 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F47590.4010309@sun.com>; from Jim.Grisanzio@Sun.COM on Sat, Sep 22, 2007 at 10:53:20AM +0900 References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> Message-ID: <20070924100926.G14087@bolthole.com> On Sat, Sep 22, 2007 at 10:53:20AM +0900, Jim Grisanzio wrote: > ... > I agree as well. Many people have worked hard to build a diverse > community open to all. I think we've come a long way, too, given the > restrictions and dependencies we've been working with all alone. These > things take time, and I think Indiana will help us grow to a new level > of diversity. Diversity has been the intention all along from our very > earliest planning meetings four years ago. oooboy.. there's another 10-point buzzword again. When we successfully pair "diversity" with "community", the resulting fusion of synergized energies is certain to enhance the value-add of our open-source paradigm! :-P Open Solaris is not going to do well because of "diversity". It will succeed if we keep good people, motivated to keep doing what they are good at, within overall processes that keep sane and good engineering design as the underlying framework for what they do. Thats why this sub-topic came up. Because good document writers, etc. arent usually particularly gifted at good engineering principles. Whereas good coders... no, make that /excellent coders/... are. From alan.duboff at sun.com Mon Sep 24 11:29:19 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Mon, 24 Sep 2007 11:29:19 -0700 (PDT) Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. In-Reply-To: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> References: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> Message-ID: <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> On Thu, 20 Sep 2007, Brandorr wrote: > I an catching an early flight to San Jose. (landing at 12:30pm). Does > anyone want to try and meet up during the day on Friday? Are you talking about the Fri. before the summit? I might be, not sure, we're a few weeks away. -- Alan DuBoff - Solaris x86 IHV/OEM Group From brandorr at opensolaris.org Mon Sep 24 11:34:46 2007 From: brandorr at opensolaris.org (Brandorr) Date: Mon, 24 Sep 2007 14:34:46 -0400 Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. In-Reply-To: <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> References: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> Message-ID: <5b5090780709241134p4ed1b898pc163db250f3445d2@mail.gmail.com> On 9/24/07, Alan DuBoff <alan.duboff at sun.com> wrote: > On Thu, 20 Sep 2007, Brandorr wrote: > > > I an catching an early flight to San Jose. (landing at 12:30pm). Does > > anyone want to try and meet up during the day on Friday? > > Are you talking about the Fri. before the summit? > > I might be, not sure, we're a few weeks away. > > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > Yes, this is the Friday before. (Oct 12th).. -Brian -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Sara.Dornsife at Sun.COM Mon Sep 24 12:44:34 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Mon, 24 Sep 2007 14:44:34 -0500 Subject: [opensolaris-summit] Tuesday planning meetings - 8am and 3pm PT 9/25 Message-ID: <46F813A2.8030404@sun.com> Reminder that we will have our next planning meetings tomorrow Tues 9/25 at 8am and 3pm PT. Dial-in information: (866)230-6968 Int'l Access/Caller Paid Dial In Number: (865)544-7856 ACCESS CODE: 2192132 Agenda: Let's get our agenda nailed down Thanks, Sara From Darren.Reed at Sun.COM Mon Sep 24 12:48:37 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Mon, 24 Sep 2007 12:48:37 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F4F5EC.9090909@sun.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> Message-ID: <46F81495.7080704@Sun.COM> Jim Grisanzio wrote: >Darren Reed wrote: > > >>The OGB doesn't appear to me to be actually doing anything. It >>appears more more like an arbitration council that also sanctions >>community decisions about which communities to create and how to >>make a contributor/core contributor. "So what". In this guise >>it feels more like a rubber stamp operation than anything meaningful. >> >> > >There was a lot of discussion about this when the initial OGB was >writing the Constitution. I think the intent was to distribute >leadership and decision making power among the communities and projects >and not centralize it in the OGB. I agree with that model because it >offers significant incentive and opportunity for participation across >the entire community. That one element is what makes the OpenSolaris >Constitution so meaningful. > >The OGB's job was to create and maintain the structure in which a >community could grow. They did that. And they did that pretty well, >actually. The OGB /represents/ the community, but it doesn't /run/ the >community. Operations are distributed by design. The community runs >itself via individual community groups and projects. When there are >problems or meta issues to consider, the OGB has the authority to step >in and mediate and resolve disputes. That seems to be a perfectly >reasonable system to me. That's my impression of things, anyway. > > All of which means that if the way OpenSolaris has been constructed is faulty then there's no way to fix it because all that the communities can do is look after themselves, not OpenSolaris as a whole. http://blogs.sun.com/avalon/entry/manufacturing_a_community Darren From Stuart.Kreitman at Sun.COM Mon Sep 24 12:57:41 2007 From: Stuart.Kreitman at Sun.COM (Stuart Kreitman) Date: Mon, 24 Sep 2007 12:57:41 -0700 Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. In-Reply-To: <5b5090780709241134p4ed1b898pc163db250f3445d2@mail.gmail.com> References: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> <5b5090780709241134p4ed1b898pc163db250f3445d2@mail.gmail.com> Message-ID: <46F816B5.10304@sun.com> Brian: Are you wanting to meet people in San Jose or in Santa Cruz? Since Santa Cruz is only 45 minutes from the Bay Area, I suspect most of us Bay Area (Sun) people will probably drive down in the morning. There's an Open Solaris reception in Santa Cruz Friday night at 7pm. (location?) If there's interest, we could do a parallel happy hour at the same time on this side of the hill. I recommend Fault Line Brewery, in Sunnyvale, just north of the airport. Stuart Brandorr wrote: > On 9/24/07, Alan DuBoff <alan.duboff at sun.com> wrote: > >> On Thu, 20 Sep 2007, Brandorr wrote: >> >> >>> I an catching an early flight to San Jose. (landing at 12:30pm). Does >>> anyone want to try and meet up during the day on Friday? >>> >> Are you talking about the Fri. before the summit? >> >> I might be, not sure, we're a few weeks away. >> >> -- >> >> Alan DuBoff - Solaris x86 IHV/OEM Group >> >> > > Yes, this is the Friday before. (Oct 12th).. > > -Brian > > > From Darren.Reed at Sun.COM Mon Sep 24 16:28:09 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Mon, 24 Sep 2007 16:28:09 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070924200946.GA868815@sun.com> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> Message-ID: <46F84809.500@Sun.COM> Keith M Wesolowski wrote: > ... > >>http://blogs.sun.com/avalon/entry/manufacturing_a_community >> > >I'm going to read this a few more times before responding. You seem >very angry, and some of what you've written appears at first blush to >consist of non sequitors. But I'll keep trying. In the meantime, is >there some concrete problem you're facing that we should be looking >into, or at least know about to have some context. > In short, the OpenSolaris community hasn't built itself or sat down and thought about what structure is best for it. And we;ve now got a structure where various groups are uninterested in change because it might mean that they are "demoted" from being a community because there is a perception that being one is a good thing or means you are important or.... Compare, for example, the decision to have communities for ZFS and SMF with that for security and networking. On the one hand you have two (ongoing) projects within Sun, on the other you have two communities of people with lots of different projects involved. The problem we have is that there's only one level or way to represent a group here: as a community. But who made all of the decisions about this? Sun. Not the OpenSolaris community. Darren From Jim.Grisanzio at Sun.COM Mon Sep 24 16:44:15 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 08:44:15 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070924100926.G14087@bolthole.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> Message-ID: <46F84BCF.1050801@sun.com> Philip Brown wrote: > On Sat, Sep 22, 2007 at 10:53:20AM +0900, Jim Grisanzio wrote: >> ... >> I agree as well. Many people have worked hard to build a diverse >> community open to all. I think we've come a long way, too, given the >> restrictions and dependencies we've been working with all alone. These >> things take time, and I think Indiana will help us grow to a new level >> of diversity. Diversity has been the intention all along from our very >> earliest planning meetings four years ago. > > oooboy.. there's another 10-point buzzword again. You like that, eh? > When we successfully pair "diversity" with "community", the resulting > fusion of synergized energies is certain to enhance the value-add of our > open-source paradigm! > > :-P > > Open Solaris is not going to do well because of "diversity". Really? So we only want coders (as Darren intimated)? And in which geo would those coders live? > It will succeed if we keep good people, motivated to keep doing what they > are good at, within overall processes that keep sane and good engineering > design as the underlying framework for what they do. I basically agree. And I've said that the coders are the core and always will be. But to have a viable global community we also need great diversity in skills, and we need to be able to work across cultures and languages and locations. > Thats why this sub-topic came up. Because good document writers, etc. arent > usually particularly gifted at good engineering principles. Whereas > good coders... no, make that /excellent coders/... are. That's fine. I generally agree. But we need the doc writers too. And the users. And the marketers. And the program managers. And the students. And the professors. And the admins. And the app developers. And the [insert whoever else wants to be a part of the community here]. Jim -- http://blogs.sun.com/jimgris From Jim.Grisanzio at Sun.COM Mon Sep 24 17:23:28 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 09:23:28 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F81495.7080704@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> Message-ID: <46F85500.2070903@sun.com> Darren Reed wrote: > All of which means that if the way OpenSolaris has been > constructed is faulty then there's no way to fix it because > all that the communities can do is look after themselves, > not OpenSolaris as a whole. No, it doesn't mean that at all. How did you arrive at that conclusion? > http://blogs.sun.com/avalon/entry/manufacturing_a_community Thanks for the reminder. I had been planning to respond to this post. Jim -- http://blogs.sun.com/jimgris From swalker at opensolaris.org Mon Sep 24 17:50:49 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Mon, 24 Sep 2007 19:50:49 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F84809.500@Sun.COM> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> Message-ID: <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> On 24/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > Keith M Wesolowski wrote: > > > ... > > > >>http://blogs.sun.com/avalon/entry/manufacturing_a_community > >> > > > >I'm going to read this a few more times before responding. You seem > >very angry, and some of what you've written appears at first blush to > >consist of non sequitors. But I'll keep trying. In the meantime, is > >there some concrete problem you're facing that we should be looking > >into, or at least know about to have some context. > > > > In short, the OpenSolaris community hasn't built itself or sat > down and thought about what structure is best for it. And we;ve > now got a structure where various groups are uninterested in > change because it might mean that they are "demoted" from > being a community because there is a perception that being one > is a good thing or means you are important or.... > > Compare, for example, the decision to have communities for > ZFS and SMF with that for security and networking. On the > one hand you have two (ongoing) projects within Sun, on the > other you have two communities of people with lots of different > projects involved. The problem we have is that there's only > one level or way to represent a group here: as a community. > > But who made all of the decisions about this? Sun. > Not the OpenSolaris community. The community ratified the constitution, they didn't have to. The community elected the ogb, they didn't have to. You could argue that there needs to be more non-Sun folks that have voting privileges, but those Sun folks are still part of the community. Not only that, it was an absolute disappointment that so few of the people that could vote ended up doing so. Sun hasn't made all the decisions. I'm sure many of us would like to see something better. I'm sure many of us would like the community to live on it's own. However, as you well know, the OpenSolaris community didn't grow in a typical open-source "organic" fashion since what it is centered around was a very mature project already before it became an open one. Given the absolute turmoil and cost to Sun on the inside, I think folks like Jim G. and many others at Sun have done a great job in helping to open internal processes and encourage the growth of a community. Nonetheless, I think everyone can agree that things should and can be improved. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From alan.duboff at sun.com Mon Sep 24 17:57:46 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Mon, 24 Sep 2007 17:57:46 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F84809.500@Sun.COM> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> Message-ID: <Pine.GSO.4.64.0709241749350.26931@konocti> On Mon, 24 Sep 2007, Darren Reed wrote: > In short, the OpenSolaris community hasn't built itself or sat > down and thought about what structure is best for it. And we;ve > now got a structure where various groups are uninterested in > change because it might mean that they are "demoted" from > being a community because there is a perception that being one > is a good thing or means you are important or.... > > Compare, for example, the decision to have communities for > ZFS and SMF with that for security and networking. On the > one hand you have two (ongoing) projects within Sun, on the > other you have two communities of people with lots of different > projects involved. The problem we have is that there's only > one level or way to represent a group here: as a community. > > But who made all of the decisions about this? Sun. > Not the OpenSolaris community. I was talking with John Plocher today, and one of the things that we both agreed with is that we need some way for the community members to understand where and how they can participate, so that we can grow the community as a whole. This would make a great topic for the summit, and possible give some of the community people that are attending, the knowledge and/or understanding on how they can contribute and help. As it is, I think a lot of folks (myself included) are trying to find a place to help, but having a hard time. I know there's a lot of other community people that in the same boat, and it would be good if we could come up with some type of solution where it is known, to be of help. I don't think that the above scenario and/or some of your point were intentional from Sun, even though some exist, it's just the nature of evolution and getting from point A to point B. I don't know how else they would have done things differently without causing too much rocking of the boat. Moving forward is going to need more help and participation from the community, outside of what is ON as we know it today. There's a lot of software that we need to port and or bring over to OpenSolaris, and getting the community members to understand how they can help in that would be most useful, IMO. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Mon Sep 24 18:01:59 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Mon, 24 Sep 2007 18:01:59 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F84BCF.1050801@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> Message-ID: <Pine.GSO.4.64.0709241800240.26931@konocti> On Tue, 25 Sep 2007, Jim Grisanzio wrote: > That's fine. I generally agree. But we need the doc writers too. And the > users. And the marketers. And the program managers. And the students. > And the professors. And the admins. And the app developers. And the > [insert whoever else wants to be a part of the community here]. I agree, we need everyone and everyone should be welcome. FWIW, the current systems does not encourage developers to participate, IMO, it encourages the people that have been doing the work to continue doing so. Many of the community members feel alienated. We need to overcome that hurdle. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Darren.Reed at Sun.COM Mon Sep 24 18:21:22 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Mon, 24 Sep 2007 18:21:22 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070925001601.GB868815@sun.com> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> Message-ID: <46F86292.20607@Sun.COM> Keith M Wesolowski wrote: >On Mon, Sep 24, 2007 at 04:28:09PM -0700, Darren Reed wrote: > > >>In short, the OpenSolaris community hasn't built itself or sat >>down and thought about what structure is best for it. And we;ve >>now got a structure where various groups are uninterested in >>change because it might mean that they are "demoted" from >>being a community because there is a perception that being one >>is a good thing or means you are important or.... >> > >... >Not - as you suggest - because people were afraid of being "demoted", >but because many people felt that their approach to a particular >problem space was radically different from others' and that achieving >consensus within such a framework would be too difficult. In other >words, we left Community Groups as relatively fine-grained entities >precisely because we wanted to help people be more productive. > The other way to look at this is the way in which we achieve things needs to change, or the vehicle to use needs to be different, depending on the need. >>Compare, for example, the decision to have communities for >>ZFS and SMF with that for security and networking. On the >>one hand you have two (ongoing) projects within Sun, on the >>other you have two communities of people with lots of different >>projects involved. The problem we have is that there's only >>one level or way to represent a group here: as a community. >> > >The notion of an "ongoing project" is part of the problem, I think. >In principle, there's no such thing - there are active Projects, there >are completed (or abandoned) Projects, and there are Community Groups. >Logically, if a community forms around a completed Project's artifacts >and - this is the key - wishes to sponsor additional active Projects >to improve and extend that work - then by all means it makes sense to >represent that as a Community Group. Not because it's a reward for >success or as a "promotion," but because that's what it *is*. > By presenting projects at the same level as a community, I believe we've done ourselves a disservice and have possibly made it harder to engage the broaders communities. >>But who made all of the decisions about this? Sun. >>Not the OpenSolaris community. >> > >Really? Here's what actually happened: > ... None of which pertains to the architecting and formation of OpenSolaris as a community, only the formation of the OGB and the constitution. Darren From stevel at sun.com Mon Sep 24 18:59:24 2007 From: stevel at sun.com (Stephen Lau) Date: Mon, 24 Sep 2007 18:59:24 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F86292.20607@Sun.COM> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> <46F86292.20607@Sun.COM> Message-ID: <46F86B7C.70604@sun.com> Darren Reed wrote: > By presenting projects at the same level as a community, > I believe we've done ourselves a disservice and have > possibly made it harder to engage the broaders communities. How are they presented at the same level? >>> But who made all of the decisions about this? Sun. >>> Not the OpenSolaris community. >>> >> >> Really? Here's what actually happened: >> > ... > > None of which pertains to the architecting and formation > of OpenSolaris as a community, only the formation of the > OGB and the constitution. My observation is that architecture and formation of the OpenSolaris community arise from that very same constitution. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From Jim.Grisanzio at Sun.COM Mon Sep 24 23:19:43 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 15:19:43 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <1190698157.1139.7.camel@localhost> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <1190698157.1139.7.camel@localhost> Message-ID: <46F8A87F.2030406@sun.com> Damian Wojslaw wrote: > As one no-coder I have to agree. I did most of Polish ZFS Admin Guide > translation and much of Polish OpenSolaris portal translation. I put > first Polish OpenSolaris conference in motion and hopethat it comes > true. I'm thinking of two more things to make here, so that more people > hear of OpenSolaris. And all that I do is translate docs. I care for > what OpenSolaris/Solaris is and where it's going a great deal and I'd > hate to be excluded from community, or whatever it's called these days. > While I agree that OpenSolaris needs to maintain a high quality of > engineering process and of code, I don't believe that the circle should > be closed for nonprogrammers. For once, receiveng core contributor > status for translation made me even more interesteed in doing next > translations in future. You bring up very good points around communications. In some of the fastest growing markets in the world, there are also some big language barriers. We need to be able to engage large numbers of people (who /want/ to get to know us) in their own language. The translation of documents and the website are critical community-building activities. Spending even a few minutes in China brings this point home real quick. The same in slower growing markets like Korea and Japan. There is probably more English in Eastern Europe, but the point is that things can happen much, much faster if we are multi-lingual. Jim -- Jim Grisanzio http://blogs.sun.com/jimgris From Jim.Grisanzio at Sun.COM Mon Sep 24 23:42:47 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 15:42:47 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709241800240.26931@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> Message-ID: <46F8ADE7.2040707@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Jim Grisanzio wrote: > >> That's fine. I generally agree. But we need the doc writers too. And the >> users. And the marketers. And the program managers. And the students. >> And the professors. And the admins. And the app developers. And the >> [insert whoever else wants to be a part of the community here]. > > I agree, we need everyone and everyone should be welcome. > > FWIW, the current systems does not encourage developers to participate, > IMO, it encourages the people that have been doing the work to continue > doing so. Many of the community members feel alienated. We need to > overcome that hurdle. I would certainly agree that we can do more. That's obvious. But I also see a lot of progress around here as well. I'm just trying to remind everyone that people /are/ participating and we /are/ making progress. I also think it's better to build on what has already been done rather than tearing each other down. Let's cut ourselves some slack from time to time. :) I think a lot of these issues will be better resolved at the Summit when people have a change to meet face-to-face. I'm sorry I'm going to miss that. Oh, well, next time ... Jim -- Jim Grisanzio http://blogs.sun.com/jimgris From alan.duboff at sun.com Tue Sep 25 02:24:13 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 02:24:13 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F8ADE7.2040707@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> Message-ID: <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> On Tue, 25 Sep 2007, Jim Grisanzio wrote: > Alan DuBoff wrote: >> On Tue, 25 Sep 2007, Jim Grisanzio wrote: >> >>> That's fine. I generally agree. But we need the doc writers too. And the >>> users. And the marketers. And the program managers. And the students. >>> And the professors. And the admins. And the app developers. And the >>> [insert whoever else wants to be a part of the community here]. >> >> I agree, we need everyone and everyone should be welcome. >> >> FWIW, the current systems does not encourage developers to participate, >> IMO, it encourages the people that have been doing the work to continue >> doing so. Many of the community members feel alienated. We need to >> overcome that hurdle. > > > I would certainly agree that we can do more. That's obvious. But I also > see a lot of progress around here as well. I'm just trying to remind > everyone that people /are/ participating and we /are/ making progress. I > also think it's better to build on what has already been done rather > than tearing each other down. Let's cut ourselves some slack from time > to time. :) I think a lot of these issues will be better resolved at the > Summit when people have a change to meet face-to-face. I'm sorry I'm > going to miss that. Oh, well, next time ... Jim, Sorry that you won't be at the summit, AFAIK, I will be barring any emergencies. People are participating, agreed, but not in any great capacity. Progress is being made, certainly, and Mercurial will help that to change that even more. I am not faulting anyone with my comments, more so would like to see some type of way that would allow folks to participate and contribute packages of software back to the presumed repositories that will be forming. One very confusing piece is WTF OpenSolaris actually is. To me it means a certain things, and to another person it means something else, as it does another. Sun has done a good job at confusing people with their usage of it. I had a conversation with John Plocher about this a few weeks ago, and we seem to agree that it is not clearly defined what it actually is, and unfortunately it does mean different things to different people. However, with all of this in mind, we will soon have the ability to grow a large repository of open source software, and I think it's important to get people involved with OpenSolaris in this capacity, rather than just working on the kernel pieces (ON, or what I consider to be OpenSolaris at this point, what is available). Consider there are 20,000+ packages that will potentially be available on OpenSolaris, it makes sense to try and organize this piece in a coherent way that will attract and encourage folks to participate. This is very different than the way folks participate today, and potentially where we will need the most help. It would be great if many distributions could build upon this repository as a base, in the same way that folks like Ubuntu leverage the base packages from Debian. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 02:49:30 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 11:49:30 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> Message-ID: <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" <swalker at opensolaris.org> wrote: > > But who made all of the decisions about this? Sun. > > Not the OpenSolaris community. > > The community ratified the constitution, they didn't have to. > > The community elected the ogb, they didn't have to. > > You could argue that there needs to be more non-Sun folks that have > voting privileges, but those Sun folks are still part of the > community. After it looked as it Sun did like to dominate OpenSolaris, I did ask to create a constitution an s government board in November 2004. I did this because I wanted to make sure that we will not get a similar situation as with Linux wehre Torvalds dominates all decisions. There was a hard discussion and I was attacked and told that this will never work. Later (in January 2005) people from Sun started to offer the same as their idea...... Now we have it and my impression is that it's main problem is missing visibility. Missing visibility results in missing interest. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 03:02:11 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 12:02:11 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709241749350.26931@konocti> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> Message-ID: <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> Alan DuBoff <alan.duboff at sun.com> wrote: > I was talking with John Plocher today, and one of the things that we both > agreed with is that we need some way for the community members to > understand where and how they can participate, so that we can grow the > community as a whole. I believe there are two problems: missing visibility of the OGB and missing power of the community. Another big problem is burocracy that makes it hard to contribute and that makes progress slow and this cannot be changed by the OGB. As burocracy is a self serving and self fostering system, only someone from outside that has the needed power may change it.The only such person I am currently aware is Ian Murdock. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 03:16:32 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 12:16:32 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F86292.20607@Sun.COM> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> <46F86292.20607@Sun.COM> Message-ID: <46f8e000.zsYm/LjN85c9owNs%Joerg.Schilling@fokus.fraunhofer.de> Darren Reed <Darren.Reed at Sun.COM> wrote: > By presenting projects at the same level as a community, > I believe we've done ourselves a disservice and have > possibly made it harder to engage the broaders communities. I did not like this disposition from the beginning. It prevented e.g. the creation of a general filesystem community and it artificially limits the power of important people who did not become a "community leader" because of this sructure. I believe we should have an interest based structure where people who are interested in the same topic just communicate with each other. One place where this should already happen is ARC cases. Unfortunately, it is possible to foil the ARC by using "fast tracks" for things that need a detailed discussion. The fact that nobody is able to stop this creates a sense of powerlessness. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From dougs at truemail.co.th Tue Sep 25 03:37:39 2007 From: dougs at truemail.co.th (Doug Scott) Date: Tue, 25 Sep 2007 17:37:39 +0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> Message-ID: <46F8E4F3.4050707@truemail.co.th> Alan DuBoff wrote: > However, with all of this in mind, we will soon have the ability to grow a > large repository of open source software, and I think it's important to > get people involved with OpenSolaris in this capacity, rather than just > working on the kernel pieces (ON, or what I consider to be OpenSolaris at > this point, what is available). Consider there are 20,000+ packages that > will potentially be available on OpenSolaris, it makes sense to try and > organize this piece in a coherent way that will attract and encourage > folks to participate. This is very different than the way folks > participate today, and potentially where we will need the most help. It > would be great if many distributions could build upon this repository as a > base, in the same way that folks like Ubuntu leverage the base packages > from Debian Alan, I very much agree. I think while the ON project is the most important project within OpenSolaris, I think too much "contributing" emphasis is placed on it. I keep reading that people think that it is difficult to contribute to OpenSolaris, and if we limit ourselves to thinking that the ON project = OpenSolaris than this is true. It is actually very easy to contribute if you do not try to work on something that somebody else already doing. As you pointed out that there are thousands of packages and applications which can be ported to OpenSolaris. We should be looking at how we can utilise the efforts from blastwave, pkgsrc and spec-files-extra etc to build a large package repository. This is where I think we actually need the masses to contribute to. Doug From Jim.Grisanzio at Sun.COM Tue Sep 25 05:16:53 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 21:16:53 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> Message-ID: <46F8FC35.2000904@sun.com> Alan DuBoff wrote: > Jim, > > Sorry that you won't be at the summit, AFAIK, I will be barring any > emergencies. Do some surfing for me. :) I'll catch you in January. > People are participating, agreed, but not in any great capacity. I'm thinking of almost 200 putbacks, another 100 or so in the queue, a hand full of ARC cases, the first putbacks from the Intel deal, hundreds of bugs submitted, etc. Also, ksh93 was a nice one. I think there was some nice community work on the specification/selection of Mercurial and the whole conversation around SCM generally and SCM migration. Indiana has generated a lot of participation recently. There's a respectable number of user groups. People are presenting all around the world. There are pretty active university programs in China and India and the US. Lots of projects now, and even some using Mercurial. Docs contributions. Portal website contributions/translations. Entire distros. Etc ... So, with some thought, I bet we could triple that rather random list pretty easily. There is no single source for all the contributions to OpenSolaris, but I would argue that if we spent some time collecting the data we'd find out that it's significant given our current constraints. > Progress > is being made, certainly, and Mercurial will help that to change that even > more. I am not faulting anyone with my comments, I know, sorry about that. I didn't mean to imply that you were criticizing. I just think that we are too hard on ourselves generally. I do, however, understand some of the reasons why, and I understand that much better now that I work remotely. > more so would like to see > some type of way that would allow folks to participate and contribute > packages of software back to the presumed repositories that will be > forming. That sounds very cool. > One very confusing piece is WTF OpenSolaris actually is. To me it means a > certain things, and to another person it means something else, as it does > another. Sun has done a good job at confusing people with their usage of > it. Sure. But again, I'd argue that this is part of the normal birth and growth of a very big project that, by necessity, has come into being in stages over a very long period of time. The confusion is natural. Now, that doesn't mean that we ought to just leave it be. Far from it. We continue to work and complete the project. Honestly, I don't see a problem. I see a work in progress. And I also see Indiana as the next logical step in our evolution. > I had a conversation with John Plocher about this a few weeks ago, and > we seem to agree that it is not clearly defined what it actually is, and > unfortunately it does mean different things to different people. > > However, with all of this in mind, we will soon have the ability to grow a > large repository of open source software, and I think it's important to > get people involved with OpenSolaris in this capacity, rather than just > working on the kernel pieces (ON, or what I consider to be OpenSolaris at > this point, what is available). Consider there are 20,000+ packages that > will potentially be available on OpenSolaris, it makes sense to try and > organize this piece in a coherent way that will attract and encourage > folks to participate. This is very different than the way folks > participate today, and potentially where we will need the most help. It > would be great if many distributions could build upon this repository as a > base, in the same way that folks like Ubuntu leverage the base packages > from Debian. And I think that's great. We'll be able to engage many new levels of developers and users as a result, and participation will significantly increase and diversify. This is a huge opportunity for us, no question about it. Jim -- http://blogs.sun.com/jimgris From Garrett.Damore at Sun.COM Tue Sep 25 05:13:29 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 05:13:29 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f8e000.zsYm/LjN85c9owNs%Joerg.Schilling@fokus.fraunhofer.de> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> <46F86292.20607@Sun.COM> <46f8e000.zsYm/LjN85c9owNs%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F8FB69.8060302@sun.com> Joerg Schilling wrote: > Darren Reed <Darren.Reed at Sun.COM> wrote: > > >> By presenting projects at the same level as a community, >> I believe we've done ourselves a disservice and have >> possibly made it harder to engage the broaders communities. >> > > I did not like this disposition from the beginning. > > It prevented e.g. the creation of a general filesystem community and it > artificially limits the power of important people who did not become > a "community leader" because of this sructure. > > I believe we should have an interest based structure where people who are > interested in the same topic just communicate with each other. > > One place where this should already happen is ARC cases. Unfortunately, > it is possible to foil the ARC by using "fast tracks" for things that need > a detailed discussion. The fact that nobody is able to stop this creates a > sense of powerlessness. > Actually, that is totally untrue. ARC members can derail any case that they feel needs a full review. They can (and often do) ask for more time, as well, when ongoing discussions need more time to converge. In the past, the ARC members have done this frequently. However, in the one particular case I know about, where you were pretty much the only dissenter, and where everyone else had converged, the membership voted to go ahead and approve the case. Note that in that particular case it was an active decision to approve the fasttrack... there was enough discussion between the membership to ensure that the ARC was pretty much unanimous in the belief that a) the case didn't warrant derailing, b) letting the case run for more time would not have resulted in any further convergence, and c) the case would have been approved if it were a full case. Furthermore, you were not present at the ARC meeting where this was decided, despite the fact that you could have dialed in and that the agenda was public. Penalizing the project team and the ARC because one person in the broader community objects doesn't seem fair. (The cost of effort to go from fasttrack to full case review is fairly high for both the project team and the ARC.) Nobody has absolute veto power. However, the ARC membership (which is only a small handful of technical leaders) can derail the case, if it is warranted. It doesn't happen *often*, but it does happen occasionally. If you're not happy with the way ARC runs its affairs, then I'd recommend that you volunteer as a PSARC intern. Or at least attend a few meetings to see how the process goes, before complaining. -- Garrett > > J?rg > > From Jim.Grisanzio at Sun.COM Tue Sep 25 05:53:48 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 21:53:48 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F904DC.8010305@sun.com> Joerg Schilling wrote: > After it looked as it Sun did like to dominate OpenSolaris, I did > ask to create a constitution an s government board in November 2004. > I did this because I wanted to make sure that we will not get a similar > situation as with Linux wehre Torvalds dominates all decisions. > > There was a hard discussion and I was attacked and told that this will never > work. > > Later (in January 2005) people from Sun started to offer the same as their > idea...... There were many people who had ideas about boards -- corporate boards, partner/customer boards, community boards, elections, non-elected appointees, no board at all. But you are correct in that we really didn't focus on this to any great extent in our external communications or even with the pilot project until Jan 05 when we released DTrace and announced the CDDL. The intention to form the CAB was part of that 1/05 announcement, and I think we didn't communicate it well at all and we didn't involve the pilot early enough, either. Point taken. I remember that time and I was pretty pissed about it. But I don't think anyone intended to take your ideas. :) > Now we have it and my impression is that it's main problem is missing > visibility. Missing visibility results in missing interest. I think Keith outlined beautifully the CAB-OGB sequence -- all of which was pretty darn visible. I can also remember pinging opensolaris-discuss on various CAB issues and getting very little response. Interest in governance was always rather low. But if that's increasing now than that's cool. Remember, the CAB/OGB was never intended to be a high profile, operational organization for OpenSolaris engineering and community development. It was designed as a bootstrap to get us going with a foundational structure. And it did just that. Future elections will determine how/if its current role changes. Jim -- http://blogs.sun.com/jimgris From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 06:23:13 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 15:23:13 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F904DC.8010305@sun.com> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> <46F904DC.8010305@sun.com> Message-ID: <46f90bc1.ymFwPOaU2f0l+kaH%Joerg.Schilling@fokus.fraunhofer.de> Jim Grisanzio <Jim.Grisanzio at Sun.COM> wrote: > Joerg Schilling wrote: > > > After it looked as it Sun did like to dominate OpenSolaris, I did > > ask to create a constitution an s government board in November 2004. > > I did this because I wanted to make sure that we will not get a similar > > situation as with Linux wehre Torvalds dominates all decisions. > > > > There was a hard discussion and I was attacked and told that this will never > > work. > > > > Later (in January 2005) people from Sun started to offer the same as their > > idea...... > > > There were many people who had ideas about boards -- corporate boards, > partner/customer boards, community boards, elections, non-elected > appointees, no board at all. But you are correct in that we really If this was true, why then was the discussion all against my proposal? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From stevel at sun.com Tue Sep 25 08:25:19 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 08:25:19 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f90bc1.ymFwPOaU2f0l+kaH%Joerg.Schilling@fokus.fraunhofer.de> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> <46F904DC.8010305@sun.com> <46f90bc1.ymFwPOaU2f0l+kaH%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F9285F.7090407@sun.com> Joerg Schilling wrote: > Jim Grisanzio <Jim.Grisanzio at Sun.COM> wrote: > >> Joerg Schilling wrote: >> >>> After it looked as it Sun did like to dominate OpenSolaris, I did >>> ask to create a constitution an s government board in November 2004. >>> I did this because I wanted to make sure that we will not get a similar >>> situation as with Linux wehre Torvalds dominates all decisions. >>> >>> There was a hard discussion and I was attacked and told that this will never >>> work. >>> >>> Later (in January 2005) people from Sun started to offer the same as their >>> idea...... >> >> There were many people who had ideas about boards -- corporate boards, >> partner/customer boards, community boards, elections, non-elected >> appointees, no board at all. But you are correct in that we really > > If this was true, why then was the discussion all against my proposal? I wasn't around for that discussion, so I have no history or context about it.... but more importantly, I don't see what good or use will come out of dredging it up out of the past again. Whoever originated the proposal: kudos to you. We've made progress, and we have a board - let's move forward rather than looking back. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From stevel at sun.com Tue Sep 25 08:32:28 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 08:32:28 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F92A0C.9090706@sun.com> Joerg Schilling wrote: > Alan DuBoff <alan.duboff at sun.com> wrote: > >> I was talking with John Plocher today, and one of the things that we both >> agreed with is that we need some way for the community members to >> understand where and how they can participate, so that we can grow the >> community as a whole. > > I believe there are two problems: missing visibility of the OGB and missing > power of the community. Why do you perceive the former to be a problem? Visibility of the OGB is a double-edged sword. Yes, we want people to know there is an OGB so that they know there is a path of appeals and such; but we don't want the OGB to be so visible to the point that it is "in-your-face"[1]. The OGB has taken the view that the community is to be self-governing. > Another big problem is burocracy that makes it hard to contribute and that > makes progress slow and this cannot be changed by the OGB. Perhaps. Perhaps not. The topic I suggested for the summit is to identify specific barriers to contribution, so I'd like to hear more details (either now by email, or in a couple of weeks in person) about what specific barriers you see so we can figure out how to break them down. > As burocracy is a self serving and self fostering system, only someone from > outside that has the needed power may change it.The only such person I am > currently aware is Ian Murdock. What is the criteria for being considered "outside"? While Ian clearly has exhibited past leadership, it's not clear to me that he is "outside" (I'm making the assumption that "outside" means external to Sun, apologies if I'm wrong...) or that he has the "needed power." I would argue that people like yourself who have been in the community longer would have just as much credibility and/or ability. cheers, steve [1] http://www.thefreedictionary.com/in-your-face for non-English speakers who may be unfamiliar with the phrase -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From John.Plocher at Sun.COM Tue Sep 25 09:01:40 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 25 Sep 2007 09:01:40 -0700 Subject: [opensolaris-summit] Extended participation... In-Reply-To: <Pine.GSO.4.64.0709241800240.26931@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> Message-ID: <46F930E4.8090708@Sun.Com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Jim Grisanzio wrote: > >> That's fine. I generally agree. But we need the doc writers too. And the >> users. And the marketers. And the program managers. And the students. >> And the professors. And the admins. And the app developers. And the >> [insert whoever else wants to be a part of the community here]. > > I agree, we need everyone and everyone should be welcome. I've been trolling around looking at other FOSS communities, and have noticed that many of them make a distinction between the sites devoted to developers -vs- others. On my mind when I was poking at these sites was AlanB's comment on website-discuss that it was extremely important for all the parts of the OS.o extended site to share the exact same look and feel; however, I didn't see this behavior on these sites. They certainly shared some of the same branding (icons, color schemes...), so they were easily identifiable as being related, but (probably because changeable skins and themes were one of the project's features) there wasn't a slavish obsession with consistency. And it didn't seem to matter. The Gallery photo archive, the E107 CMS, Mozilla/Firefox, Wordpress, etc, each have several "related" web sites, each with a slightly different focus. While some (like e107) seem to have taken this to extremes, the idea that one site does not have to fit all seems to be a good one. (e.g., e107.org, e107coders.org, plugins.e107.org, e107themes.com, e107tutorials.com, e107faq.org ...) All these projects have a similar organization: Their main site is entirely devoted to facilitating downloads, installs and upgrades. (e.g., http://wordpress.org/ ) Once you have installed their particular thing, you get exposed to the next level - themes and skins, plugins and add-on customizations and the like. There were *many* competing sites for these additions, and, for the most part, they all seem to work well together. (e.g., http://www.google.com/search?q=wordpress+themes&aq=t ) You also start getting exposed to the beginnings of "here is how it works under the hood", so you can customize things yourself. Of course this all leads to the developer site(s), where you find the people who are developing the project, skins, themes and plugins... (e.g., http://codex.gallery2.org/Gallery2:Themes ) My point is simply that we need to figure out how to make Alan's website efforts a success in the short term so that we all can move forward with the goal of creating opportunities for others to contribute. Not just in putbacks to ON, but with wiki content, webmastering, repository creation and hosting, doc writing and editing, ARC, Design and Code reviewing and the like. -John From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 09:42:48 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 18:42:48 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070925164008.GA277983@sun.com> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> Message-ID: <46f93a88.q8BIaA3GH5/e0PGr%Joerg.Schilling@fokus.fraunhofer.de> Keith M Wesolowski <Keith.Wesolowski at sun.com> wrote: > > I believe there are two problems: missing visibility of the OGB and missing > > power of the community. > > Are you concerned about a lack of *transparency* or a lack of > *visibility*? I don't believe the OGB needs to be more visible; if Ok, it is missing visibility. > > Another big problem is burocracy that makes it hard to contribute and that > > makes progress slow and this cannot be changed by the OGB. > > It's not bureaucracy (why does everyone think there's so much of it? > Do you know what the word actually means?) but a lack of adequate As I know what it means, I used it. > > As burocracy is a self serving and self fostering system, only someone from > > outside that has the needed power may change it.The only such person I am > > currently aware is Ian Murdock. > > You're joking, right? No, but it seems that you did not understand me. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From James.Walker at Sun.COM Tue Sep 25 10:35:21 2007 From: James.Walker at Sun.COM (Jim Walker) Date: Tue, 25 Sep 2007 11:35:21 -0600 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F70F60.662239E7@nrubsig.org> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F15E8A.70407@sun.com> <46F70F60.662239E7@nrubsig.org> Message-ID: <46F946D9.7060704@sun.com> Roland Mainz wrote: > Erm, AFAIK the test farm is non-interactive (e.g. you can't do something > like "login and hack the kernel to death"), right ? If you mean the test farm we are setting up for OpenSolaris, then it is fully "interactive". It allows you to get access to a nice new test machine that is bare metal with an os build, that you can use to test your opensolaris software. You get root access, so you can hack away. If you want early access to it, signup to be a beta tester: http://opensolaris.org/os/community/testing/testfarm/ > The idea of the five machines was to have one "central hub" which is > user-land only (e.g. nooone except two or three people have "root" > access) which handles SunRay services, home dirs (which are exported to > the four other machines), jumpstart and has some development tools > (compiler, OS/Net build tools etc.) installed and then 2x x86 and 2x > SPARC machines to boot custom kernels from the server on demand. This > stuff is only thought to give people the opportunity to stick their > heads together and try some stuff without having to improvise something > with laptops (or worse) ... Since it looks like the network will be good, I plan on having some test farm machines available remotely for the summit, in addition to whatever we have locally. That said, I agree with many that at this stage in organization development, the summit should focus on discussions that inform the community on where we are at, and where we would like to go, which have the effect of fostering contributions to the overall strategy of OpenSolaris. But, we should have time during mornings, evenings and breaks to work specific hardware and software issues and do installs. And, what you map out above looks good. I don't know who is providing this and what the limits are though. Some virtualization on the machines may make them go farther. Cheers, Jim -- Jim Walker, http://blogs.sun.com/jwalker Sun Microsystems, Software, Solaris QE x77744, 500 Eldorado Blvd, Broomfield CO 80021 From Darren.Reed at Sun.COM Tue Sep 25 12:14:28 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 12:14:28 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070925164008.GA277983@sun.com> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> Message-ID: <46F95E14.4000008@Sun.COM> Keith M Wesolowski wrote: >... >It's not bureaucracy (why does everyone think there's so much of it? >Do you know what the word actually means?) but a lack of adequate >infrastructure that makes contributing such a pain. I blame Sun for >this; its insistence on maintaining control of the infrastructure and >its tortuous legal and policy constraints are a main reason more >progress has not been made. That said, anyone could put together a >proposal to move all this outside Sun's control and accelerate the >process, yet no one has. It's unclear whether that's because of >laziness, lack of interest, or lack of means. > > There are a number of things that need to happen here and they've been raised before (at least by myself): 1) move the opensolaris machinery (web server, repository, etc) away from being owned by Sun; 2) to achieve (1), create an opensolaris entity that can own pieces of hardware, etc; 3) in order to fund both (1) and (2), make an opensolaris entity that is a non profit organisation. If we were a true bona-fide opensource project and with a bit of luck, we can get a server into a back-bone ISP's colocation facility using 1RU or less of space. To what extent Sun is required to cooperate here, I don't know. But, now that OpenSolaris is open, if there were enough interested parties we could build a new distribution that we all worked on (with a new name) with the model *we* want and just leave OpenSolaris in Sun's hands. We can still pull in code from OS.o, after it gets mirrored there from nevada. Heck, maybe we participants should just do that anyway, rather than wait for Sun to work out what it wants to do with OS.o? Darren From brandorr at opensolaris.org Tue Sep 25 13:44:47 2007 From: brandorr at opensolaris.org (Brandorr) Date: Tue, 25 Sep 2007 16:44:47 -0400 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F95E14.4000008@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> Message-ID: <5b5090780709251344u5b7221efo1d3b68965a97f53b@mail.gmail.com> On 9/25/07, Darren Reed <Darren.Reed at sun.com> wrote: > Heck, maybe we participants should just do that anyway, rather > than wait for Sun to work out what it wants to do with OS.o? While I understand your frustration at a low level of outside contribution, I don't think Sun is to be blamed. I think it stems from the fact that Solaris is a mature, formerly proprietary, product. Many users of Solaris today don't give a care about any notions of community of contribution. (So long as the get what their support contracts cover) We here at OpenSolaris.org are currently in the process of building a community. This community currently exists mostly of Sun employees, and a few brave new worlders, and a few excited Solaris customers. We are in a bit of a chicken and egg situation. What we need right now is more outside manpower, to build the infrastructure to make it easier for outside contributers to contribute. Right now, one way we can address this, is to do outreach. Currently this is on the agenda for our next user group meeting: Get involved: ------------------ - Mentoring - We are looking at starting a new mentoring program. We are looking for Seniors that would be interested in mentoring Juniors. (Developers and sysadmins). Please email Brian Gupta (brandorr at opensolaris.org) if you are interested. - The OpenSolaris.org Systems administration Community Group is looking to expand membership. syadmin-discuss-subscribe at opensolaris.org - Volunteering to help build a new OpenSolaris user portal, as well as an NYOSUG.org website. (email Brian Gupta (brandorr at opensolaris.org) if you are interested in either.) - We will be launching/announcing a virtual development lab next month, that will allow anyone who is interested in developing OpenSolaris or porting open source apps to OpenSolaris, to gain free access to VMWare ESX Virtual Machines. Let Brian Gupta (brandorr at opensolaris.org) know if you are interested in either helping administer ESX, or if you want access to a VM. (It's not setup yet). - We can use help planning these UG meetings. Mail Isaac or Brian Gupta "Isaac Rozenfeld - NEW YORK" <Isaac.Rozenfeld at sun.com>, Brian Gupta (brandorr at opensolaris.org) - Brian Gupta (brandorr at opensolaris.org) is going to the first OpenSolaris Developers Summit, and will have more to report next month. (http://www.genunix.org/wiki/index.php?title=OpenSolaris_Summit.) - Brian Gupta (brandorr at opensolaris.org) assembled the following new user FAQ: assembling: http://www.genunix.org/wiki/index.php?title=FAQ_Prototype What I would ask each contributing member of OpenSolaris.org, is to find at least one other person that has the ability to contribute, and walk them through their first contribution. (We'd double the number of contributers over night. Also, UserGroups should be actively advertising outside of their own users. e.g. - Reach out to Linux and BSD user groups, as well as local UNIGROUP, Lisa and SAGE chapters. The goal being to increase the number of attendees to your meetings. Basically, I feel that by working around the barriers for now, we will be able to grow the community enough to remove them down the road. (With the added manpower). > Darren > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From stevel at sun.com Tue Sep 25 14:48:49 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 14:48:49 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F95E14.4000008@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> Message-ID: <46F98241.8030207@sun.com> Darren Reed wrote: > Keith M Wesolowski wrote: > >> ... >> It's not bureaucracy (why does everyone think there's so much of it? >> Do you know what the word actually means?) but a lack of adequate >> infrastructure that makes contributing such a pain. I blame Sun for >> this; its insistence on maintaining control of the infrastructure and >> its tortuous legal and policy constraints are a main reason more >> progress has not been made. That said, anyone could put together a >> proposal to move all this outside Sun's control and accelerate the >> process, yet no one has. It's unclear whether that's because of >> laziness, lack of interest, or lack of means. >> >> > > There are a number of things that need to happen here and > they've been raised before (at least by myself): > > 1) move the opensolaris machinery (web server, repository, > etc) away from being owned by Sun; While this is a noteworthy long-term goal, I fail to see what problems this solves in the short-term. Who will own them? Who will maintain them? Who will be paid to maintain them? > 2) to achieve (1), create an opensolaris entity that can own > pieces of hardware, etc; > 3) in order to fund both (1) and (2), make an opensolaris > entity that is a non profit organisation. okay, that answers my questions from #1... but... > If we were a true bona-fide opensource project and with a bit > of luck, we can get a server into a back-bone ISP's colocation > facility using 1RU or less of space. There are plenty of open source projects who succeed massively without the need for non-profit foundations. > To what extent Sun is required to cooperate here, I don't know. > > But, now that OpenSolaris is open, if there were enough interested > parties we could build a new distribution that we all worked on > (with a new name) with the model *we* want and just leave > OpenSolaris in Sun's hands. We can still pull in code from OS.o, > after it gets mirrored there from nevada. > > Heck, maybe we participants should just do that anyway, rather > than wait for Sun to work out what it wants to do with OS.o? That's certainly a prerogative that anyone has... after all, the code is open. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From stevel at sun.com Tue Sep 25 15:04:24 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 15:04:24 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F984CE.3000508@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <46F98241.8030207@sun.com> <46F984CE.3000508@Sun.COM> Message-ID: <46F985E8.7020100@sun.com> James C. McPherson wrote: > Stephen Lau wrote: >> Darren Reed wrote: >>> Keith M Wesolowski wrote: >>> >>>> ... >>>> It's not bureaucracy (why does everyone think there's so much of it? >>>> Do you know what the word actually means?) but a lack of adequate >>>> infrastructure that makes contributing such a pain. I blame Sun for >>>> this; its insistence on maintaining control of the infrastructure and >>>> its tortuous legal and policy constraints are a main reason more >>>> progress has not been made. That said, anyone could put together a >>>> proposal to move all this outside Sun's control and accelerate the >>>> process, yet no one has. It's unclear whether that's because of >>>> laziness, lack of interest, or lack of means. >>>> >>>> >>> There are a number of things that need to happen here and >>> they've been raised before (at least by myself): >>> >>> 1) move the opensolaris machinery (web server, repository, >>> etc) away from being owned by Sun; >> >> While this is a noteworthy long-term goal, I fail to see what problems >> this solves in the short-term. >> >> Who will own them? Who will maintain them? Who will be paid to >> maintain them? > > > Who owns, runs and maintains kernel.org? kernel.org isn't much more than a download site. > Dare I suggest slashdot, sourceforge and even OSNews as > examples here? I don't believe /. & osnews establish communities. Sourceforge is probably a more relevant example, but Sourceforge creates micro-communities with hardly any attempt to have cross-project communication. opensolaris.org is attempting to be everything (or at least all-things OpenSolaris related) to everyone, and I would argue that that is probably its fallacy. There is no *single* Linux community site, why should there need to be a single opensolaris.org site? Frankly, I think it'd be good for the health of our community if people did take stuff elsewhere and do interesting things on their own. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From John.Plocher at Sun.COM Tue Sep 25 15:09:55 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 25 Sep 2007 15:09:55 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F983E4.4080804@Sun.COM> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> Message-ID: <46F98733.9020701@Sun.Com> Darren Reed wrote: > ie. Instead of trying to get part of bugster out of Sun, create > a new bug reporting database The dealbreaker for me is that there is no easy way to initially populate this new database with the hundreds of thousands of open bugs/rfes already in bugster, which effectively means that nobody without bugster access will be able to fix/manage bugs, and if they have bugster access, they have no need to use something else... Catch-22. See Machiavelli for more insight. -John From Darren.Reed at Sun.COM Tue Sep 25 15:12:01 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 15:12:01 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F98241.8030207@sun.com> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <46F98241.8030207@sun.com> Message-ID: <46F987B1.4050601@Sun.COM> Stephen Lau wrote: >Darren Reed wrote: > > > ... > >>If we were a true bona-fide opensource project and with a bit >>of luck, we can get a server into a back-bone ISP's colocation >>facility using 1RU or less of space. >> >> > >There are plenty of open source projects who succeed massively without >the need for non-profit foundations. > > True. Outside of the BSDs and a few large proejcts, I'm not sure of any that have pursued this path (strangely, it doesn't come for free.) I see OpenSolaris as being big enough that it is worth considering. It depends on whether the goal of OpenSolaris is to become "business" in the sense that RedHat and others have or the foundations for others, such as BSD does. In this vein, Indiana is a different approach again. It is shaping up to be the first distribution to also come from the base platform vendor. Given that, it is unclear whether becoming non-profit is in the best interests of OpenSolaris. But I will add this: if it does become non-profit then it becomes more attractive for 3rd party people who want to use OpenSolaris to contribute financially to its survival. That should be a carrot hanging from a stick for Sun :) Darren From John.Plocher at Sun.COM Tue Sep 25 13:44:24 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 25 Sep 2007 13:44:24 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> Message-ID: <46F97328.90502@Sun.Com> Peter Tribble wrote: > So how do we encourage communities and individuals to take up the > reins of governance? Not to be flippant, but why would they want to? That is, what problems are they currently having that could be solved by their spending energy on self-governance? My perspective is that the answer is that there are lots of serious problems, none of which can effectively be solved thru self-governance alone. Problem 1: The structure of the meta-community echoes the confusion in people's minds of what exactly OpenSolaris is and should be. Because people are not in agreement with what it is, there is no consensus on how it should be structured. Problem 2: The observable artifacts of the community are still locked away from direct community manipulation: The source tree, the bug database, the web site, the mailing list infrastructure, the ON C-Team and even the ARCs are /ALL/ solely maintainable by a set of under-resourced, under-appreciated and constantly dwindling group of Sun employees. Solve these problems and the community of developers will grow. Ignore them, and it will stagnate and die. -John From Darren.Reed at Sun.COM Tue Sep 25 14:55:48 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 14:55:48 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F97328.90502@Sun.Com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> Message-ID: <46F983E4.4080804@Sun.COM> John Plocher wrote: > Peter Tribble wrote: > >> So how do we encourage communities and individuals to take up the >> reins of governance? > > > Not to be flippant, but why would they want to? > > That is, what problems are they currently having that could > be solved by their spending energy on self-governance? > > My perspective is that the answer is that there are lots of > serious problems, none of which can effectively be solved thru > self-governance alone. > > Problem 1: The structure of the meta-community echoes the > confusion in people's minds of what exactly OpenSolaris is > and should be. Because people are not in agreement with > what it is, there is no consensus on how it should be > structured. > > Problem 2: The observable artifacts of the community are > still locked away from direct community manipulation: The > source tree, the bug database, the web site, the mailing > list infrastructure, the ON C-Team and even the ARCs are > /ALL/ solely maintainable by a set of under-resourced, > under-appreciated and constantly dwindling group of Sun > employees. > > Solve these problems and the community of developers will > grow. Ignore them, and it will stagnate and die. So a question that needs to be asked is, considering where we want to go, is it easier to just build something "new" and let Sun work out how it wants to join up with that, rather than to try and morph what we have into something else? ie. Instead of trying to get part of bugster out of Sun, create a new bug reporting database for OpenSolaris, independant of Sun's internal database, and so on. Darren From alan.coopersmith at sun.com Tue Sep 25 15:05:07 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Tue, 25 Sep 2007 15:05:07 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F983E4.4080804@Sun.COM> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> Message-ID: <46F98613.1060108@sun.com> Darren Reed wrote: > ie. Instead of trying to get part of bugster out of Sun, create > a new bug reporting database for OpenSolaris, independant > of Sun's internal database, That's already the plan, hardly anything new there - the problem isn't deciding to break new ground, it's finding people who want to do more than just complain about the broken bug problem and are willing to put in time to figure out what new system should be used and get it going. If you have the time, please go join the Defect Tracking System (DTS) discussions on tools-discuss. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From stevel at sun.com Tue Sep 25 15:15:23 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 15:15:23 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F983E4.4080804@Sun.COM> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> Message-ID: <46F9887B.5010604@sun.com> Darren Reed wrote: > John Plocher wrote: > >> Peter Tribble wrote: >> >>> So how do we encourage communities and individuals to take up the >>> reins of governance? >> >> >> Not to be flippant, but why would they want to? >> >> That is, what problems are they currently having that could >> be solved by their spending energy on self-governance? >> >> My perspective is that the answer is that there are lots of >> serious problems, none of which can effectively be solved thru >> self-governance alone. >> >> Problem 1: The structure of the meta-community echoes the >> confusion in people's minds of what exactly OpenSolaris is >> and should be. Because people are not in agreement with >> what it is, there is no consensus on how it should be >> structured. >> >> Problem 2: The observable artifacts of the community are >> still locked away from direct community manipulation: The >> source tree, the bug database, the web site, the mailing >> list infrastructure, the ON C-Team and even the ARCs are >> /ALL/ solely maintainable by a set of under-resourced, >> under-appreciated and constantly dwindling group of Sun >> employees. >> >> Solve these problems and the community of developers will >> grow. Ignore them, and it will stagnate and die. > > > So a question that needs to be asked is, considering where > we want to go, is it easier to just build something "new" and > let Sun work out how it wants to join up with that, rather than > to try and morph what we have into something else? > > ie. Instead of trying to get part of bugster out of Sun, create > a new bug reporting database for OpenSolaris, independant > of Sun's internal database, and so on. We've got an existing DTS evaluation going on - it's floundering not because Sun is holding anything back, but because nobody is making the effort to evaluate any of the DTS (or at least they haven't mentioned it publicly). -steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From alan.duboff at sun.com Tue Sep 25 16:08:37 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 16:08:37 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7myvapsae.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Richard Lowe wrote: >> I'm just trying to remind everyone that people /are/ participating >> and we /are/ making progress. > > Slowly, when it comes to development. Very, very slowly. We're > making progress as far as mktg (I guess), and various other things. > Open development? Not even close. > >> I also think it's better to build on what has already been done >> rather than tearing each other down. > > Sure, but pretending the problems don't exist gets us nowhere. It > seems nearly everything I read about opensolaris tells me how peachy > things are, which is, of course, largely crap. Rich, I don't think it's completly fair to say it is not open, not do I think we can just say it hasn't evolved. I do appreciate you giving your honest view though. Please see my blog entry: http://blogs.sun.com/aland/entry/one_package_per_community_member -- Alan DuBoff - Solaris x86 IHV/OEM Group From stevel at sun.com Tue Sep 25 16:28:33 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 16:28:33 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> Message-ID: <46F999A1.7090900@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Richard Lowe wrote: > >>> I'm just trying to remind everyone that people /are/ participating >>> and we /are/ making progress. >> >> Slowly, when it comes to development. Very, very slowly. We're >> making progress as far as mktg (I guess), and various other things. >> Open development? Not even close. >> >>> I also think it's better to build on what has already been done >>> rather than tearing each other down. >> >> Sure, but pretending the problems don't exist gets us nowhere. It >> seems nearly everything I read about opensolaris tells me how peachy >> things are, which is, of course, largely crap. > > Rich, > > I don't think it's completly fair to say it is not open, not do I think > we can just say it hasn't evolved. I do appreciate you giving your > honest view though. > > Please see my blog entry: > > http://blogs.sun.com/aland/entry/one_package_per_community_member One major problem that I read is: > If we look at the kernel components, do we even want to make it easier for members to contribute? I don't think so, in fact I believe it should be more difficult. To date it has been mostly Sun folks that have contributed the bulk of code in that space, and that should be expected. The engineers that have managed the code over the years are still managing it to date, and this is a good thing from my perspective. They know the code better and have worked on it more than others, it absolutely makes sense for them to do as they have done, wether it's open or not, it's a must to keep those sources well maintained. While I agree with the idea that it should be well maintained, I think perhaps your wording was poor. It came across (to me anyway) that you don't believe non-Sun members should be allow to contribute kernel code? cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From swalker at opensolaris.org Tue Sep 25 16:40:04 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:40:04 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F95E14.4000008@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> Message-ID: <b9c544f0709251640n63a77ca6rae158d89d9acffc9@mail.gmail.com> On 25/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > Keith M Wesolowski wrote: > > >... > >It's not bureaucracy (why does everyone think there's so much of it? > >Do you know what the word actually means?) but a lack of adequate > >infrastructure that makes contributing such a pain. I blame Sun for > >this; its insistence on maintaining control of the infrastructure and > >its tortuous legal and policy constraints are a main reason more > >progress has not been made. That said, anyone could put together a > >proposal to move all this outside Sun's control and accelerate the > >process, yet no one has. It's unclear whether that's because of > >laziness, lack of interest, or lack of means. > > > > > > There are a number of things that need to happen here and > they've been raised before (at least by myself): > > 1) move the opensolaris machinery (web server, repository, > etc) away from being owned by Sun; Sounds great, but who pays for it? What benefit would it have for us? What's stopping someone from starting up a fork and doing this themselves right now? > 2) to achieve (1), create an opensolaris entity that can own > pieces of hardware, etc; > > 3) in order to fund both (1) and (2), make an opensolaris > entity that is a non profit organisation. You talk about project success all the time, but the things you talk about are not the things I think it needs to succeed right now. From my perspective as a indepdenent, non-Sun affiiliated person, your proposals would just burn energy on things now that is better directed at more important issues. I don't think the issues that are facing us right now would be solved by what you are proposing. However, they are all great goals to have for a future date, despite the challenges in achieving them. > If we were a true bona-fide opensource project and with a bit > of luck, we can get a server into a back-bone ISP's colocation > facility using 1RU or less of space. "bona-fide"...interesting terminology. What exactly is the "accepted definition" of "bona-fide opensource project". > But, now that OpenSolaris is open, if there were enough interested > parties we could build a new distribution that we all worked on > (with a new name) with the model *we* want and just leave > OpenSolaris in Sun's hands. We can still pull in code from OS.o, > after it gets mirrored there from nevada. > > Heck, maybe we participants should just do that anyway, rather > than wait for Sun to work out what it wants to do with OS.o? What's stopping someone from doing this now? Personally, I don't see a community division as the "Right Thing To Do" at the moment. I think divided efforts would hurt us even more. I'm fairly frustrated with the progress in certain areas, but I'm not ready to "throw the baby out with the bath water" just yet... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From swalker at opensolaris.org Tue Sep 25 16:49:58 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:49:58 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F985E8.7020100@sun.com> References: <46F31335.6060700@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <46F98241.8030207@sun.com> <46F984CE.3000508@Sun.COM> <46F985E8.7020100@sun.com> Message-ID: <b9c544f0709251649n58c65762rbd8704bcd2e6263c@mail.gmail.com> On 25/09/2007, Stephen Lau <stevel at sun.com> wrote: > James C. McPherson wrote: > > Stephen Lau wrote: > >> Darren Reed wrote: > >>> Keith M Wesolowski wrote: > >>> > >>>> ... > >>>> It's not bureaucracy (why does everyone think there's so much of it? > >>>> Do you know what the word actually means?) but a lack of adequate > >>>> infrastructure that makes contributing such a pain. I blame Sun for > >>>> this; its insistence on maintaining control of the infrastructure and > >>>> its tortuous legal and policy constraints are a main reason more > >>>> progress has not been made. That said, anyone could put together a > >>>> proposal to move all this outside Sun's control and accelerate the > >>>> process, yet no one has. It's unclear whether that's because of > >>>> laziness, lack of interest, or lack of means. > >>>> > >>>> > >>> There are a number of things that need to happen here and > >>> they've been raised before (at least by myself): > >>> > >>> 1) move the opensolaris machinery (web server, repository, > >>> etc) away from being owned by Sun; > >> > >> While this is a noteworthy long-term goal, I fail to see what problems > >> this solves in the short-term. > >> > >> Who will own them? Who will maintain them? Who will be paid to > >> maintain them? > > > > > > Who owns, runs and maintains kernel.org? > > kernel.org isn't much more than a download site. > > > Dare I suggest slashdot, sourceforge and even OSNews as > > examples here? > > I don't believe /. & osnews establish communities. > Sourceforge is probably a more relevant example, but Sourceforge creates > micro-communities with hardly any attempt to have cross-project > communication. > > opensolaris.org is attempting to be everything (or at least all-things > OpenSolaris related) to everyone, and I would argue that that is > probably its fallacy. > > There is no *single* Linux community site, why should there need to be a > single opensolaris.org site? Frankly, I think it'd be good for the > health of our community if people did take stuff elsewhere and do > interesting things on their own. Isn't that part of the "Linux problem" though? Hasn't the "Linux world" (which doesn't have a "unified" place the community can call home) fractured into a "bajillion" communities, websites, and distributions leading to uncoordinated, wasted efforts? I think Ubuntu proved how valuable having a centralized user community, bug tracking, repository, etc. can be. It's the same model that many other projects have as well (FreeBSD, etc.). -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From swalker at opensolaris.org Tue Sep 25 16:51:22 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:51:22 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F98733.9020701@Sun.Com> References: <46F2FE5F.3070508@sun.com> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F98733.9020701@Sun.Com> Message-ID: <b9c544f0709251651t321edf8ct604fd6719adc2659@mail.gmail.com> On 25/09/2007, John Plocher <John.Plocher at sun.com> wrote: > Darren Reed wrote: > > ie. Instead of trying to get part of bugster out of Sun, create > > a new bug reporting database > > The dealbreaker for me is that there is no easy way to initially > populate this new database with the hundreds of thousands of open > bugs/rfes already in bugster, which effectively means that nobody > without bugster access will be able to fix/manage bugs, and if they > have bugster access, they have no need to use something else... Right, but we're always going to have that problem. Sun has proprietary, confidential data in there, and there's no way practical way to scrub those millions of bits of data properly. We may just have to pay this price and move on...the important thing will be whether we can deal with new bugs in both systems going forward. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Darren.Reed at Sun.COM Tue Sep 25 17:07:17 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 17:07:17 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F98733.9020701@Sun.Com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F98733.9020701@Sun.Com> Message-ID: <46F9A2B5.5000509@Sun.COM> John Plocher wrote: > Darren Reed wrote: > >> ie. Instead of trying to get part of bugster out of Sun, create >> a new bug reporting database > > > The dealbreaker for me is that there is no easy way to initially > populate this new database with the hundreds of thousands of open > bugs/rfes already in bugster, which effectively means that nobody > without bugster access will be able to fix/manage bugs, and if they > have bugster access, they have no need to use something else... So OpenSolaris starts with a clean slate. What's wrong with that? :) This can only work in OpenSolaris' favour: it'll have less bugs than Solaris ;-) And if there is a desire to import bugster information later, just reserve some number space for that and start at some other place for OpenSolaris. Darren From alan.duboff at sun.com Tue Sep 25 18:07:35 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 18:07:35 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F999A1.7090900@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> Message-ID: <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Stephen Lau wrote: > One major problem that I read is: >> If we look at the kernel components, do we even want to make it easier for >> members to contribute? I don't think so, in fact I believe it should be >> more difficult. To date it has been mostly Sun folks that have contributed >> the bulk of code in that space, and that should be expected. The engineers >> that have managed the code over the years are still managing it to date, >> and this is a good thing from my perspective. They know the code better >> and have worked on it more than others, it absolutely makes sense for them >> to do as they have done, wether it's open or not, it's a must to keep those >> sources well maintained. > > While I agree with the idea that it should be well maintained, I think > perhaps your wording was poor. It came across (to me anyway) that you don't > believe non-Sun members should be allow to contribute kernel code? Steve, I will edit my blog to reflect what I really meant, that it really is not community or not, it's that the people that understand and know the code should be the ones that maintain it. This is so today with the kernel engineers that are doing just that. In the future, when community people come up to speed, there is no reason they would not fall into the same category. More what I meant was that we just do not see the skills and knowledge in the community to work with and maintain the code in the same capacity. I'm sure that will change over time. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Garrett.Damore at Sun.COM Tue Sep 25 20:45:35 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 20:45:35 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> Message-ID: <46F9D5DF.7030502@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Stephen Lau wrote: > > >> One major problem that I read is: >> >>> If we look at the kernel components, do we even want to make it easier for >>> members to contribute? I don't think so, in fact I believe it should be >>> more difficult. To date it has been mostly Sun folks that have contributed >>> the bulk of code in that space, and that should be expected. The engineers >>> that have managed the code over the years are still managing it to date, >>> and this is a good thing from my perspective. They know the code better >>> and have worked on it more than others, it absolutely makes sense for them >>> to do as they have done, wether it's open or not, it's a must to keep those >>> sources well maintained. >>> >> While I agree with the idea that it should be well maintained, I think >> perhaps your wording was poor. It came across (to me anyway) that you don't >> believe non-Sun members should be allow to contribute kernel code? >> > > Steve, > > I will edit my blog to reflect what I really meant, that it really is not > community or not, it's that the people that understand and know the code > should be the ones that maintain it. This is so today with the kernel > engineers that are doing just that. > > In the future, when community people come up to speed, there is no reason > they would not fall into the same category. More what I meant was that we > just do not see the skills and knowledge in the community to work with and > maintain the code in the same capacity. I'm sure that will change over > time. > Actually, I take pretty strong exception to that fact. Until recently I was such a community member. Other community members are themselves former Sun employees... And still others work at places like Tadpole, where a lot of real kernel work does in fact take place. (To the point that our current cardbus stack is largely derived from original Tadpole work!) Andrew Gallatin, at Myricom for example, has had a lot of positive impact and feedback into what we're doing in the kernel layer. There are undoubtedly many others. To suggest that the community lacks the skill set suggests either a) a level of elitism that I hope we don't exude to the rest of the world, or b) ignorance about what other folks are doing out there. Now, having said all that, we do indeed have a number of processes in place which I don't want want to go away... codereview, test validation, and RTI advocacy are all steps along the way to ensure that the overall quality of what is in Solaris (Open or otherwise) meets some stringent quality considerations. In fact, even the userland code we have in Solaris should meet these guidelines too, its just that (IMO) its easier to find less thorough advocates and code reviewers for userland bits of code. :-) -- Garrett > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From alan.duboff at sun.com Tue Sep 25 21:17:33 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 21:17:33 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F9D5DF.7030502@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> Message-ID: <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Garrett D'Amore wrote: > Actually, I take pretty strong exception to that fact. > > Until recently I was such a community member. And before that you worked at Sun. > Other community members are themselves former Sun employees... Yes, but do we see these folks flocking to contribute code back? Those people, as yourself, know the process and how it works. They have the knowledge and understanding, those are not the people I refer to. Those people as yourself are in a minority and I stated there are a few that do in fact have the skills. Again, you folks are in the minority. > To suggest that the community lacks the skill set suggests either a) a > level of elitism that I hope we don't exude to the rest of the world, or > b) ignorance about what other folks are doing out there. Whatever...split hairs if you like, fine. > Now, having said all that, we do indeed have a number of processes in > place which I don't want want to go away... codereview, test validation, > and RTI advocacy are all steps along the way to ensure that the overall > quality of what is in Solaris (Open or otherwise) meets some stringent > quality considerations. Do you think that folks should be allowed a putback if they don't completely meet the test validation process? -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Tue Sep 25 21:20:17 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 21:20:17 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x74phikrek.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <x74phikrek.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709252117420.7188@eagle.softorchestra.com> On Wed, 26 Sep 2007, Richard Lowe wrote: > For what it's worth, the update pissed me off more than the original > text, for pretty much the reasons you outline below. And why you can have your own blog to say what you like,;-) How would you change the process to work better today? > I most definitely don't want that to go away either, as you know. > Making external development easier is not about removing valuable > process, it's about getting that process to the point it works for > *everyone*. That's a large part of what I think we're making very > little progress on. Sun has provided the information on their process. Have you been able to understand it? How could they make better progress on that? -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Tue Sep 25 23:09:17 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 23:09:17 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7zlzaj9lv.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> On Wed, 26 Sep 2007, Richard Lowe wrote: >>> Now, having said all that, we do indeed have a number of processes in >>> place which I don't want want to go away... codereview, test validation, >>> and RTI advocacy are all steps along the way to ensure that the overall >>> quality of what is in Solaris (Open or otherwise) meets some stringent >>> quality considerations. >> >> Do you think that folks should be allowed a putback if they don't >> completely meet the test validation process? >> > > Garrett totally didn't say that, he in fact suggested quite the > opposite, very strongly. Rich, Yeah, I know what he said, but sometimes people say something different than they actually do. Since I had been working with him recentely, I just wanted to get clarification from him on that point. I agree with you that he didn't say that, and suggested quite the opposite. I think I stand strongly on the current process and the test validation that has been put in place. Although there is some case to be argued that the test validation shouldn't be so rigorous, the current system in place is what it is, and has been put in place for a reason. When Sun changes that policy, it could change the putback process (at least for some components like device drivers). I was talking with Randy Fishel recently about this, and it's possible in the future that Sun could lighten up on the test validation for components that do not fall into their top tier, but that hasn't happened yet and the process for doing a putback requires quite a rigorous test validation. So far I haven't seen any suggestions from you, just criticism on the process Sun has put in place. Can you elaborate on what exactly Sun could do better? From my viewpoint Sun pushed their process outside the firewall. While it's probably not the easiest process to follow, I'm not sure how else they could have transitioned the code out to the community and still ensure that the process retains the quality they put in place. What can you suggest that would help that better? How would you retain the quality of Solaris as it has been, yet allow the community to participate in the development more so? Your previous comments about the development process not being open must relate to the ON sources that are currently out in the community, after all, there really is nothing more to OpenSolaris at this point aside from some packages on opensolaris.org. But those that have not been putback are not a part of any development process. Can you elaborate on what you would do different in regard to being more open in the development of the current sources? -- Alan DuBoff - Solaris x86 IHV/OEM Group From Jim.Grisanzio at Sun.COM Tue Sep 25 23:30:33 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Wed, 26 Sep 2007 15:30:33 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> Message-ID: <46F9FC89.2030306@sun.com> Peter Tribble wrote: > But does the community know that it's supposed to be governing itself? Many people do, but I'd say most do not. I'd also say that's normal. It's our job to educate people within each Community Group as we go along. Over time if we develop active leaders across the CGs the situation will improve. But as new people become involved, the process of education needs to be constant. Jim -- http://blogs.sun.com/jimgris From Garrett.Damore at Sun.COM Tue Sep 25 23:46:41 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 23:46:41 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> Message-ID: <46FA0051.2010800@sun.com> Alan DuBoff wrote: > On Wed, 26 Sep 2007, Richard Lowe wrote: > > >>>> Now, having said all that, we do indeed have a number of processes in >>>> place which I don't want want to go away... codereview, test validation, >>>> and RTI advocacy are all steps along the way to ensure that the overall >>>> quality of what is in Solaris (Open or otherwise) meets some stringent >>>> quality considerations. >>>> >>> Do you think that folks should be allowed a putback if they don't >>> completely meet the test validation process? >>> >>> >> Garrett totally didn't say that, he in fact suggested quite the >> opposite, very strongly. >> > > Rich, > > Yeah, I know what he said, but sometimes people say something different > than they actually do. Since I had been working with him recentely, I just > wanted to get clarification from him on that point. I agree with you that > For the background here, Alan and I had a disagreement about what the test matrix required when I putback my drivers (afe and mxfe) without performing a certain group of stress tests on all possible platform configurations, as I believed (and still do) that they were redundant when taken against the rest of the tests that had already been performed, and not strictly required for final putback. Anyone else who wants the particulars of the that particular event is welcome to contact me offline. Anyway, that small matter aside, I do believe that a reasonably complete amount of testing is required, and indeed, we have a number of test suites in place to help with that (though they are far from exhaustive.) -- Garrett From Garrett.Damore at Sun.COM Tue Sep 25 23:55:11 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 23:55:11 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F9887B.5010604@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> Message-ID: <46FA024F.2020002@sun.com> Stephen Lau wrote: > Darren Reed wrote: > >> John Plocher wrote: >> >> >>> Peter Tribble wrote: >>> >>> >>>> So how do we encourage communities and individuals to take up the >>>> reins of governance? >>>> >>> Not to be flippant, but why would they want to? >>> >>> That is, what problems are they currently having that could >>> be solved by their spending energy on self-governance? >>> >>> My perspective is that the answer is that there are lots of >>> serious problems, none of which can effectively be solved thru >>> self-governance alone. >>> >>> Problem 1: The structure of the meta-community echoes the >>> confusion in people's minds of what exactly OpenSolaris is >>> and should be. Because people are not in agreement with >>> what it is, there is no consensus on how it should be >>> structured. >>> >>> Problem 2: The observable artifacts of the community are >>> still locked away from direct community manipulation: The >>> source tree, the bug database, the web site, the mailing >>> list infrastructure, the ON C-Team and even the ARCs are >>> /ALL/ solely maintainable by a set of under-resourced, >>> under-appreciated and constantly dwindling group of Sun >>> employees. >>> >>> Solve these problems and the community of developers will >>> grow. Ignore them, and it will stagnate and die. >>> >> So a question that needs to be asked is, considering where >> we want to go, is it easier to just build something "new" and >> let Sun work out how it wants to join up with that, rather than >> to try and morph what we have into something else? >> >> ie. Instead of trying to get part of bugster out of Sun, create >> a new bug reporting database for OpenSolaris, independant >> of Sun's internal database, and so on. >> > > We've got an existing DTS evaluation going on - it's floundering not > because Sun is holding anything back, but because nobody is making the > effort to evaluate any of the DTS (or at least they haven't mentioned it > publicly). > There is another real problem here. I think the only people who are qualified to "fully" evaluate the bug database are those who are also Sun employees. Why? Because only they really know what the current bugster contains, and what features are needed, and what aren't. Of course, all the Sun employees I know are up to their neck in other work. And it isn't entirely clear that the Sun folks who should be paying for an evaluation of a new DTS have any motivation to do so; after all, bugster works just fine for *their* purposes. Speaking for myself, I don't have the time to do a full evaluation of a DTS. And, I'm not really qualified as a DTS administrator. *HOWEVER*, I know that I would be willing to play in a sandbox that someone else built. And even use it to track real project bugs. I *suspect* that the same is true of other people. So, if someone is energetic enough (a big if, I know) to set up the demo sandbox, then maybe an announcement in a broader context inviting others to participate in the test (as developers, not DTS administrators!) would be helpful. -- Garrett From Jim.Grisanzio at Sun.COM Wed Sep 26 00:06:18 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Wed, 26 Sep 2007 16:06:18 +0900 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7myvapsae.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> Message-ID: <46FA04EA.7060503@sun.com> Richard Lowe wrote: > Sure, but pretending the problems don't exist gets us nowhere. It > seems nearly everything I read about opensolaris tells me how peachy > things are, which is, of course, largely crap. Who's saying nearly everything is peachy? The OpenSolaris lists are filled with strong arguments about all the problems we have to deal with. I don't get the impression things are peachy from a development perspective at all. Things are also not largely crap. I don't support either extreme. Jim -- Jim Grisanzio http://blogs.sun.com/jimgris From alan.duboff at sun.com Wed Sep 26 00:37:45 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 00:37:45 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA0051.2010800@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> Message-ID: <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Garrett D'Amore wrote: > For the background here Wasn't trying to split hairs with you, but you were pretty strong in your statement...<LOL!> > Anyway, that small matter aside, I do believe that a reasonably complete > amount of testing is required, and indeed, we have a number of test suites in > place to help with that (though they are far from exhaustive.) I know you do, and I believe your intentions are quite good. As a case in point, I know you try to follow the DDI strictly. Sun internal drivers seem to have been some of the worst offenders of sticking to the DDI. FWIW, I think it would be worth considering getting drivers put back that didn't meet the rigorous testing, at least in some form of more open system like Indiana is rumored to be. It is still not clear to me what exactly Indiana will include. In regards to skills and/or the offense you took with my comments, which I removed from my blog and made a slight reference change to the previous, let me say this. I have seen several putbacks with massive amounts of files in the short time I've been at Sun. While not common, it's not a surprise to see putbacks that total thousands of files. To work with a set of source and manage such a putback is not for the average developer. It should also be of no surprise that most all developers who have in depth knowledge of Solaris have some ties back to Sun. As a case in point, there are people that could manage putbacks with thousands of files to OpenSolaris that work at VMWare, Google, or Netapp for instance...And in most all cases I bet you can tie them back to Sun. Nothing wrong with that, they are the community now. The point is that there are folks around that have done and could do it again through the community...as long as they followed the process to do so... Those people are in the minority when we look at the community as a whole. One of the things we can fault Sun with is time, it is slow to evolve. I mean that in the sense of getting out of the "Darwin Syndrome" that Andy Tucker used to talk about (coincidentally at VMWare I believe). Seems that Andy projected a 6 month to 1 year duration of the "Darwin Syndrome", and it still feels like we're in it today...:-/ For those that have never heard this term, it was based on Apple's work with the community on Darwin, which was later taken back inside Apple as it failed in the sense of an open source project. Sun has opened some of the processes, in the sense that community folks can attend ARC meetings, for instance...but the bulk of the process still resides inside the SWAN, hence my reference to the "Darwin Syndrome". -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Wed Sep 26 00:47:04 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 00:47:04 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA024F.2020002@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> Message-ID: <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Garrett D'Amore wrote: > There is another real problem here. I think the only people who are > qualified to "fully" evaluate the bug database are those who are also Sun > employees. Why? Because only they really know what the current bugster > contains, and what features are needed, and what aren't. I just chopped out a bunch of folks on the cc list. But I wanted to ask, what's the difference anyway? I have seen people complain that they can't read the entire bugs on the OpenSolaris bug database. Ok, even if you could read them internally, there are cases where you still need to look at the actual files to fully understand what is being changed. Hypothetically, let's say you created a process that made it so difficult for engineers to do their putback, that they started to slip changes into another change request to get their changes back? This would never happen at Sun, right? My point is that even if you could read the bug database from outside SWAN, in it's whole, it still wouldn't ensure that you could know exactly what a putback did unless you really looked at every single file and evaluated how they relate to the change request anyway. My $0.02. For the record, I'm not claiming this goes on willy nilly, just that in any large set of sources where you have massive amounts of changes happening on a daily basis, there is a higher chance that engineers will integrate more changes to a CR than logical. Maybe it's a part of common habit. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Wed Sep 26 00:55:42 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 00:55:42 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7odfpkhh8.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> On Wed, 26 Sep 2007, Richard Lowe wrote: > We don't lighten the process, we certainly don't drop any quality > control aspects thereof, we just make it work without the need for > someone on SWAN to hand-hold, go try and find where a bug ended up, > etc, etc. I'm not saying that this shouldn't be the way *everyone* > (to include new hires at Sun) works with our gates at first. I'm > saying there needs to be an alternative. The sponsor process doesn't > scale, and was never intended to. For better or worse, Sun is moving to get outside the SWAN, it's just slow, but as I said in a previous message, I can only fault them on the time it's taking. There is no having your cake and eating it too, if you want to keep the processes in place and not lighten up on anything, it will continue to require a sponsor. The process is really too complicated for the average person to follow without help, IMO. However, testing is the most difficult piece with device drivers for instance. Ok, you don't want to lighten the process, it will continue to take a long time to putback each driver and work through the issues and fix them to be able to meet the rigorous testing that is currently in place. I feel the testing needs to be evaluated in some cases. For instance, testing a NIC driver to be loaded and unloaded thousands of times will probably just tell you if there is a memory leak. There could be better ways to test that, IMO. Loading and unloading a driver is not a common practice on most systems. >> Can you elaborate on what you would do different in regard >> to being more open in the development of the current sources? > > We get to the point where external developers have equal access to the > tools and processes, and useful things can be done without latent > intermediaries. I really don't care what developers inside Sun (or > outside Sun) choose to do. Only that they have the ability to > actually do it themselves, and that the decisions made about changes > made to opensolaris gates are taken by opensolaris, in places > opensolaris can see, rather than by Sun alone. Right, when you can do a putback to OpenSolaris, and Sun can determine if they want to accept that into Solaris, then we will have reached that level. FWIW, we are moving there, it's just moving there slowly. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Sara.Dornsife at Sun.COM Wed Sep 26 06:29:27 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Wed, 26 Sep 2007 08:29:27 -0500 Subject: [opensolaris-summit] Tuesday planning meetings - 8am and 3pm PT 9/25 In-Reply-To: <46F813A2.8030404@sun.com> References: <46F813A2.8030404@sun.com> Message-ID: <46FA5EB7.1030208@sun.com> Notes from today's meeting(s): Attendees: 8:00am Ian Murdock Stuart Kreitman Stefan Teleman Jim Walker Sara Dornsife 3:00pm Ian Murdock Stuart Kreitman Glynn Foster Patrick Finch Shawn Walker John Plocher Jesse Silver Sara Dornsife We decided that based on the topics proposed, we should dedicate all of Saturday to 3 topics. - Packaging System - Installer (including the Distro Constructor) - Modernization Each of these 3 topics will have a 30 minute presentation slot at the beginning of the day to give a high-level overview of the topic and what the topic owner hopes to accomplish during the Summit. My AI on that was to contact the owners and ask them what sub-topics they would be discussing for the rest of the times alloted. Email sent. Stuart asked for projectors for IRC chats to be projected onto a screen. He's going to bring an extra laptop and keep an IRC channel open. We spent some time discussing what our top 3 goals for this Summit were. - Level-setting - what is OpenSolaris? Get everyone in agreement - Packaging system - everyone should leave this Summit with a full understanding of what it is, how to use it, how it is changing the game - Decision-making - we should all have a clear understanding of the decision-making process for OpenSolaris Feel free to insert your own goals here. Glynn is going to formally invite the rest of the OGB. We currently have 3 of the 7 members attending. Glynn has created a schedule and posted it http://www.opensolaris.org/os/project/indiana/summit/ Please take a look and comment. I think it's coming together nicely. For next week, we need owners and secretaries for each section. We also need to make sure that everyone has a ride in from the airport. Please update the wiki if you can pick people up or if you need a ride. http://www.genunix.org/wiki/index.php/OpenSolaris_Summit. As of this morning (9/26) we're at 77 people. Sara Sara Dornsife wrote: > Reminder that we will have our next planning meetings tomorrow Tues 9/25 > at 8am and 3pm PT. > > Dial-in information: > (866)230-6968 > Int'l Access/Caller Paid Dial In Number: (865)544-7856 > ACCESS CODE: 2192132 > > Agenda: > Let's get our agenda nailed down > > Thanks, > Sara > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From alan.coopersmith at sun.com Wed Sep 26 07:52:08 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 07:52:08 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> Message-ID: <46FA7218.50301@sun.com> Alan DuBoff wrote: > What can you suggest that would help that better? How would you retain the > quality of Solaris as it has been, yet allow the community to participate > in the development more so? Your previous comments about the development > process not being open must relate to the ON sources that are currently > out in the community, after all, there really is nothing more to > OpenSolaris at this point aside from some packages on opensolaris.org. You really are hopeless, you know that? I know you'll never believe it for some insane reason, but just in case anyone else reading this thread believes you, there is far more than just ON on OpenSolaris, as is obvious to anyone who opens their eyes and looks. JDS has beat everyone else in putting their master source tree on OpenSolaris.org, where anyone can pull and community members can commit, and other consolidations like X, SFW, Admin, and NWS have their buildable sources out there as well. Is there any point in you continuing to belittle the members of the community and the people at Sun who have worked their asses off making this happen or are you just trying to insult as many people as possible and tear apart the community? -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From Garrett.Damore at Sun.COM Wed Sep 26 08:08:33 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 08:08:33 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> Message-ID: <46FA75F1.9010302@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Garrett D'Amore wrote: > >> There is another real problem here. I think the only people who are >> qualified to "fully" evaluate the bug database are those who are also >> Sun employees. Why? Because only they really know what the current >> bugster contains, and what features are needed, and what aren't. > > I just chopped out a bunch of folks on the cc list. > > But I wanted to ask, what's the difference anyway? I have seen people > complain that they can't read the entire bugs on the OpenSolaris bug > database. Ok, even if you could read them internally, there are cases > where you still need to look at the actual files to fully understand > what is being changed. > > Hypothetically, let's say you created a process that made it so > difficult for engineers to do their putback, that they started to slip > changes into another change request to get their changes back? That would be an error. IF the process were so heavy weight, then something is wrong. I've slipped many changes into a single putback ... *but* I try to make sure I have only one CR. (The exception being bugs against a new driver or other software, which I haven't integrated. In that case there is only one CR, which is the integration of the driver in the first place... in that case it makes sense to try to make the first integration as bug-free as you can.) And, FWIW, I think a lot of the RTI advocates would catch this. Most of the advocates I've used in the past have acted as another layer of code review... even though that isn't strictly part of their RTI duties. > > This would never happen at Sun, right? > > My point is that even if you could read the bug database from outside > SWAN, in it's whole, it still wouldn't ensure that you could know > exactly what a putback did unless you really looked at every single > file and evaluated how they relate to the change request anyway. My > $0.02. You're misunderstanding the problem. People use the CR database for a lot more than knowing what was in a putback. It contains customer call records, sample code, attachments, etc. A lot of those use cases happen internally to Sun. I'm not sure I even know what they all are. Someone should involve, for example, RPE and CTE in such an evaluation... they are amongst the heaviest bugster users. > > For the record, I'm not claiming this goes on willy nilly, just that > in any large set of sources where you have massive amounts of changes > happening on a daily basis, there is a higher chance that engineers > will integrate more changes to a CR than logical. Maybe it's a part of > common habit. I don't think so. Certainly I don't see that in the kernel. Codereview and RTI is supposed to protect against it. -- Garrett > > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group From Garrett.Damore at Sun.COM Wed Sep 26 08:18:21 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 08:18:21 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.16644.873233.737411@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> Message-ID: <46FA783D.7050803@sun.com> James Carlson wrote: > Alan DuBoff writes: > >> I know you do, and I believe your intentions are quite good. As a case in >> point, I know you try to follow the DDI strictly. Sun internal drivers >> seem to have been some of the worst offenders of sticking to the DDI. >> > > You seem to say that as though it were a bad thing. > > For drivers that deliver via ON, there's no reason that they need to > stick to the DDI at all. All projects integrating in ON (including > drivers) are able to use ON consolidation private interfaces without > additional review. It's one of the perks of delivering synchronously > with ON. > > The only considerations that matter here are the usual ones, such as > design correctness and code clarity. Adherence to rules that apply > for non-ON code isn't part of the bargain. > > I agree that when driver designs are required to reach outside of the > DDI on a regular basis, then that signals a problem. It most often > signals a problem with the DDI: it means that the functions that the > designers need are not present. It rarely (though perhaps > occasionally) means that the driver writer is doing the wrong thing. > > So, case in point, the GLDv3/Nemo interfaces are not part of the DDI. > Many of our drivers use them anyway. If someone proposed to add a new > Ethernet driver to ON, and that driver used only DDI interfaces (thus > ignoring Nemo), I would certainly consider the driver to be defective. > > Thus, DDI isn't a mandate. > Historically, believe I had the only DDI complaint ethernet drivers that were to be found... either inside ON or outside of ON. I believe that this signaled both deficiencies in ON, and in the various unbundled drivers. I believe that sticking to the DDI, while not a hard requirement for drivers in ON, is a worthwhile cause even for the drivers in ON. 1) It reduces the pain and suffering when non-DDI interfaces change 2) By sticking to documented APIs (or APIs you expect to be documented) you're unlikely to be bitten later by changes in undocumented behavior. 3) By using the DDI, you know that you're using the same APIs that we recommend to external developers. That means you help to uncover any limitations in the DDI. I think if the drivers in ON "had" to use the DDI or go thru the process of getting a waiver/contract, then its likely the DDI would be a lot richer than it is today. (For one, we'd probably have a public API for nexus drivers!) Note that for APIs like GLDv3, kEF, etc. I think use is appropriate... largely because we are using those APIs as we expect an external consumer would, and we're in the process of trying to make them public. This is quite different than drivers that start up their own kernel threads, etc. -- Garrett From Garrett.Damore at Sun.COM Wed Sep 26 08:28:02 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 08:28:02 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA75F1.9010302@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> <46FA75F1.9010302@sun.com> Message-ID: <46FA7A82.7070406@sun.com> Garrett D'Amore wrote: > Alan DuBoff wrote: > >> On Tue, 25 Sep 2007, Garrett D'Amore wrote: >> >> >>> There is another real problem here. I think the only people who are >>> qualified to "fully" evaluate the bug database are those who are also >>> Sun employees. Why? Because only they really know what the current >>> bugster contains, and what features are needed, and what aren't. >>> >> I just chopped out a bunch of folks on the cc list. >> >> But I wanted to ask, what's the difference anyway? I have seen people >> complain that they can't read the entire bugs on the OpenSolaris bug >> database. Ok, even if you could read them internally, there are cases >> where you still need to look at the actual files to fully understand >> what is being changed. >> >> Hypothetically, let's say you created a process that made it so >> difficult for engineers to do their putback, that they started to slip >> changes into another change request to get their changes back? >> > > That would be an error. IF the process were so heavy weight, then > something is wrong. > > I've slipped many changes into a single putback ... *but* I try to make > sure I have only one CR. This didn't come out right. I meant to say, "one CR per bug". So one putback, many bugs, and hence many CRs. Sorry for the confusion. -- Garrett From stevel at sun.com Wed Sep 26 08:42:28 2007 From: stevel at sun.com (Stephen Lau) Date: Wed, 26 Sep 2007 08:42:28 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA024F.2020002@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> Message-ID: <46FA7DE4.7010208@sun.com> Garrett D'Amore wrote: > Stephen Lau wrote: >> We've got an existing DTS evaluation going on - it's floundering not >> because Sun is holding anything back, but because nobody is making the >> effort to evaluate any of the DTS (or at least they haven't mentioned >> it publicly). >> > > There is another real problem here. I think the only people who are > qualified to "fully" evaluate the bug database are those who are also > Sun employees. Why? Because only they really know what the current > bugster contains, and what features are needed, and what aren't. > > Of course, all the Sun employees I know are up to their neck in other > work. And it isn't entirely clear that the Sun folks who should be > paying for an evaluation of a new DTS have any motivation to do so; > after all, bugster works just fine for *their* purposes. > > Speaking for myself, I don't have the time to do a full evaluation of a > DTS. And, I'm not really qualified as a DTS administrator. *HOWEVER*, > I know that I would be willing to play in a sandbox that someone else > built. And even use it to track real project bugs. I *suspect* that > the same is true of other people. > > So, if someone is energetic enough (a big if, I know) to set up the demo > sandbox, then maybe an announcement in a broader context inviting others > to participate in the test (as developers, not DTS administrators!) > would be helpful. Good points Garrett, but I do think there are still some pieces where non-Sun people should certainly be able to help out. We've outlined a specific list of requirements: http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt It should certainly be at least possible for people to help eliminate potential candidates based on whether or not they meet the requirements or not. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From John.Plocher at Sun.COM Wed Sep 26 08:43:41 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 26 Sep 2007 08:43:41 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> Message-ID: <46FA7E2D.3050508@Sun.Com> Alan DuBoff wrote: > Do you think that folks should be allowed a putback if they don't > completely meet the test validation process? I think this is an example of a myopic viewpoint, and that there are places where this behavior might be perfectly acceptable. Think website wikis, package repository experimentation, prototyping, ... OpenSolaris != the ON Consolidation. The level of detail you (and Garrett) are looking at is really only valid for some segments of the OpenSolaris community. Darren also seems stuck there. There are other segments of the larger OpenSolaris Community where there are different (or even non-existent) test validation processes, places where the bar to participation is intentionally lower. Painting all of OS.o with the "ON brush" does us all a disservice, as does blindly assuming that only ON matters. -John (As I said to Alan in a side conversation, we really should start using the word "OpenSolaris" as an adjective and not a noun. That is, Alan and Garrett are concerned with the OpenSolaris ON Community and associated policies, while others (JimG...) are focused on the OpenSolaris User Groups and Portals; Ian is focused on an OpenSolaris Distro, etc.) From sch at sun.com Wed Sep 26 09:34:16 2007 From: sch at sun.com (Stephen Hahn) Date: Wed, 26 Sep 2007 09:34:16 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> References: <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> Message-ID: <20070926163416.GC6132@eng.sun.com> * Alan DuBoff <alan.duboff at sun.com> [2007-09-26 06:11]: > So far I haven't seen any suggestions from you, just criticism on the > process Sun has put in place. Can you elaborate on what exactly Sun could > do better? I would suggest, Alan, that you haven't looked widely enough. Rich has provided fixes and constructive commentary on tools, processes, and product code, in such a fashion to get Contributor or Core Contributor status in four community groups. He's even given feedback during the review of the DTS requirements... - Stephen -- sch at sun.com http://blogs.sun.com/sch/ From alan.duboff at sun.com Wed Sep 26 10:29:56 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 10:29:56 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.16644.873233.737411@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> Message-ID: <Pine.GSO.4.64.0709261018130.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, James Carlson wrote: > Alan DuBoff writes: >> I know you do, and I believe your intentions are quite good. As a case in >> point, I know you try to follow the DDI strictly. Sun internal drivers >> seem to have been some of the worst offenders of sticking to the DDI. > > You seem to say that as though it were a bad thing. Not a bad thing directly, but if Sun creates and advocates that driver writers adhere to the DDI, but doesn't do it them self, it's a kind of double speak. > So, case in point, the GLDv3/Nemo interfaces are not part of the DDI. > Many of our drivers use them anyway. If someone proposed to add a new > Ethernet driver to ON, and that driver used only DDI interfaces (thus > ignoring Nemo), I would certainly consider the driver to be defective. > > Thus, DDI isn't a mandate. GLDv3 is an edge case right now as the interfaces are going to be public soon. Currently I am working with driver that was written to them, and it requires you copy/use mac.h and mac_ether.h from uts/common/sys/. This is only due to evolution of the development process. That was not what I was referring to in my post per se. This is a valid case for using private interfaces, IMO. It's not that using private interfaces is a bad thing, it's just that the driver stands the chance of breaking if those interfaces change. Any public interfaces are guaranteed to stay stable, I believe. -- Alan DuBoff - Solaris x86 IHV/OEM Group From james.liu at sun.com Wed Sep 26 10:36:10 2007 From: james.liu at sun.com (James C. Liu) Date: Wed, 26 Sep 2007 10:36:10 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> Message-ID: <46FA988A.5020608@sun.com> Alan DuBoff wrote: > There is no having your cake and eating it too, if you want to keep the > processes in place and not lighten up on anything, it will continue to > require a sponsor. The process is really too complicated for the average > person to follow without help, IMO. > Why can't we have our cake and eat it too? We don't need to dump the rigorousness of the existing process. We can still require some sponsor. It may not scale over time and then we'll have to rethink perhaps. But we've had this hallway discussion before, and the solution then was to promote the idea of an "experimental" section for drivers and apps that haven't gone through full testing. Some stuff (like kernel modules) would be available on the install media with the right disclaimers and caveats, and other packages would be available elsewhere. Over time, the idea would be to do the testing and certify code as official and move it over. But that would be transparent to endusers who would already have had access to the pkgs if they needed it. -James Liu - Solaris x86 IHV/OEM Engineering From alan.duboff at sun.com Wed Sep 26 10:51:37 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 10:51:37 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA7A82.7070406@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> <46FA75F1.9010302@sun.com> <46FA7A82.7070406@sun.com> Message-ID: <Pine.GSO.4.64.0709261039310.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, Garrett D'Amore wrote: >> That would be an error. IF the process were so heavy weight, then >> something is wrong. >> >> I've slipped many changes into a single putback ... *but* I try to make >> sure I have only one CR. > > This didn't come out right. I meant to say, "one CR per bug". So one > putback, many bugs, and hence many CRs. Sorry for the confusion. Garrett, I guess my point is that I have heard engineers say they have done that because the process is so difficult. I won't name who, but I think that to some extent this does go on some, today. I was more using this as an example of how the engineers have learned to work around some of the limitation. I don't know that this is a bad thing, just that it can and does happen. It makes the engineer more aware so in the case where something does break, they know they're on the line for it. I wasn't pointing any finger at you or anyone else specific, just pointing out that there are many cases where files are going back on a daily basis. You really need to look at the changes to fully understand what is going back. I agree with what you said about workarounds, examples, contact info, etc...Sun is moving towards getting the bug database out to the community though, time being the only real process holding it up. I was also more using that as a case where the community claims foul, that Sun has not opened up the bug database. I was only pointing out that it wouldn't magically help people understand every CR that gets putback, they would still need to look at and evaluate the files and fixes to fully understand each case. Again, I don't know how you could easily transition such a large code base to a community and not cause some issues as the bug database, unless it just all magically happened together at the same time (source code management, bug database, sources, ARCs, etc...). In the case of Solaris, as it was, there was just too much process for that to happen cleanly. As it is they did a good job, IMO, and our community is moving towards a pure, open model. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Wed Sep 26 11:02:28 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 11:02:28 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA7E2D.3050508@Sun.Com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> Message-ID: <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, John Plocher wrote: > Alan DuBoff wrote: >> Do you think that folks should be allowed a putback if they don't >> completely meet the test validation process? > > > I think this is an example of a myopic viewpoint, and that there > are places where this behavior might be perfectly acceptable. Think > website wikis, package repository experimentation, prototyping, ... > > OpenSolaris != the ON Consolidation. Let me clear something up, before AlanC's jugular explodes.;-) I realize that OpenSolaris is not just ON, however today it pretty much is from a sources standpoint. When someone refers to OpenSolaris today, that is what is there. > The level of detail you (and Garrett) are looking at is really only > valid for some segments of the OpenSolaris community. Darren also > seems stuck there. There are other segments of the larger OpenSolaris > Community where there are different (or even non-existent) test > validation processes, places where the bar to participation is > intentionally lower. Agreed. However, today what we have are ON sources that primarily comprise what is OpenSolaris. This is not to say we are limited to this or that other pieces will not be added, I am just pointing out where we are today. > (As I said to Alan in a side conversation, we really should start > using the word "OpenSolaris" as an adjective and not a noun. That > is, Alan and Garrett are concerned with the OpenSolaris ON Community > and associated policies, while others (JimG...) are focused on the > OpenSolaris User Groups and Portals; Ian is focused on an OpenSolaris > Distro, etc.) And add AlanC to the mix, he has his view also, as others do. My point was that it's been confusing at best to figure out what OpenSolaris really is for some, like me for instance. In speaking for myself, and only myself, I can say that I have found it confusing what OpenSolaris really is. To further complicate that, marketing has also been unclear on how they intend to use the OpenSolaris name them self. It has been unclear to me anyway. -- Alan DuBoff - Solaris x86 IHV/OEM Group From randy.fishel at sun.com Wed Sep 26 11:21:41 2007 From: randy.fishel at sun.com (Randy Fishel) Date: Wed, 26 Sep 2007 11:21:41 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FA783D.7050803@sun.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> Message-ID: <Pine.GSO.4.64.0709261050440.2186@grimmy> On Wed, 26 Sep 2007, Garrett D'Amore wrote: > I believe that sticking to the DDI, while not a hard requirement for > drivers in ON, is a worthwhile cause even for the drivers in ON. I would argue that it must be a hard requirement, and for the reasons you state below. > > 1) It reduces the pain and suffering when non-DDI interfaces change > 2) By sticking to documented APIs (or APIs you expect to be documented) > you're unlikely to be bitten later by changes in undocumented behavior. > 3) By using the DDI, you know that you're using the same APIs that we > recommend to external developers. That means you help to uncover any > limitations in the DDI. > > I think if the drivers in ON "had" to use the DDI or go thru the process > of getting a waiver/contract, then its likely the DDI would be a lot > richer than it is today. (For one, we'd probably have a public API for > nexus drivers!) > > Note that for APIs like GLDv3, kEF, etc. I think use is appropriate... > largely because we are using those APIs as we expect an external > consumer would, and we're in the process of trying to make them public. > This is quite different than drivers that start up their own kernel > threads, etc. > > -- Garrett We do make the requirement of drivers to specify all the interfaces they expect to import and export. We don't make them list the DDI/DKI, as that is a set of implied interfaces. They must, however, specify if they intend to use GLDvX, kEF, and even FMA. If a driver is using any other interface, it should have listed it, and would be in violation of its ARC case if it didn't (IMHO, they should even list consolidation private interfaces). The problem, with this "requirement", is that we have no way to validate or enforce that only the ARC'd API's have been used. We leave it to the developer (in theory, checked by the RTI advocate) to appropriately use the API's. But if you know of a driver that uses a non-public API, and it doesn't have an ARC case stating this use (so all the GLDv3 drivers should be OK), then you can file a bug on the driver. Just maybe one of the things we need to think about is some set of driver validation tools. ---- Randy From alan.duboff at sun.com Wed Sep 26 11:25:26 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 11:25:26 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA988A.5020608@sun.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> <46FA988A.5020608@sun.com> Message-ID: <Pine.GSO.4.64.0709261118000.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, James C. Liu wrote: > > Alan DuBoff wrote: > >> There is no having your cake and eating it too, if you want to keep the >> processes in place and not lighten up on anything, it will continue to >> require a sponsor. The process is really too complicated for the average >> person to follow without help, IMO. >> > > Why can't we have our cake and eat it too? We don't need to dump > the rigorousness of the existing process. We can still require some > sponsor. It may not scale over time and then we'll have to rethink > perhaps. But we've had this hallway discussion before, and the > solution then was to promote the idea of an "experimental" section > for drivers and apps that haven't gone through full testing. Some > stuff (like kernel modules) would be available on the install media > with the right disclaimers and caveats, and other packages would be > available elsewhere. Over time, the idea would be to do the testing > and certify code as official and move it over. But that would > be transparent to endusers who would already have had access to the > pkgs if they needed it. > > -James Liu - Solaris x86 IHV/OEM Engineering > James, Yes, this would allow for drivers to be included that haven't passed the rigorous testing. However, the OP stated that they didn't want to lighten up on testing, or any current requirements. I was only pointing out that while it doesn't need to be mandatory for them to have a sponsor (today they are required to have one), it's all in their best interest. If there was no sponsor requirement today, it would be more difficult for folks to grasp the process. I think the sponsor idea is to assist more than anything else, at least from my view. If all community members understood the putback process well enough there would be no need for a sponsor, they would be able to do everything. Oh, and if all process was outside the SWAN, that would also help and where I believe we're headed. -- Alan DuBoff - Solaris x86 IHV/OEM Group From John.Plocher at Sun.COM Wed Sep 26 11:32:50 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 26 Sep 2007 11:32:50 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> Message-ID: <46FAA5D2.20300@Sun.Com> Alan DuBoff wrote: >> OpenSolaris != the ON Consolidation. > > Let me clear something up, before AlanC's jugular explodes.;-) > > I realize that OpenSolaris is not just ON, however today it pretty much > is from a sources standpoint. When someone refers to OpenSolaris today, > that is what is there. You seem to be showing either an amazing degree of either ignorance or intentional disrespect for those who have built out the non-ON parts of the community. It may be true that when /YOU/ say "OpenSolaris", /YOU/ mean ON, but please don't generalize your feelings without some serious data to back up your claims. Some conversations don't work well in email; this seems to have become one of them. Why don't we let this topic die? -John From Garrett.Damore at Sun.COM Wed Sep 26 11:30:03 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 11:30:03 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <Pine.GSO.4.64.0709261050440.2186@grimmy> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <Pine.GSO.4.64.0709261050440.2186@grimmy> Message-ID: <46FAA52B.6090403@sun.com> Randy Fishel wrote: > On Wed, 26 Sep 2007, Garrett D'Amore wrote: > > >> I believe that sticking to the DDI, while not a hard requirement for >> drivers in ON, is a worthwhile cause even for the drivers in ON. >> > > I would argue that it must be a hard requirement, and for the > reasons you state below. > > >> 1) It reduces the pain and suffering when non-DDI interfaces change >> 2) By sticking to documented APIs (or APIs you expect to be documented) >> you're unlikely to be bitten later by changes in undocumented behavior. >> 3) By using the DDI, you know that you're using the same APIs that we >> recommend to external developers. That means you help to uncover any >> limitations in the DDI. >> >> I think if the drivers in ON "had" to use the DDI or go thru the process >> of getting a waiver/contract, then its likely the DDI would be a lot >> richer than it is today. (For one, we'd probably have a public API for >> nexus drivers!) >> >> Note that for APIs like GLDv3, kEF, etc. I think use is appropriate... >> largely because we are using those APIs as we expect an external >> consumer would, and we're in the process of trying to make them public. >> This is quite different than drivers that start up their own kernel >> threads, etc. >> >> -- Garrett >> > > > We do make the requirement of drivers to specify all the interfaces > they expect to import and export. We don't make them list the > DDI/DKI, as that is a set of implied interfaces. They must, however, > specify if they intend to use GLDvX, kEF, and even FMA. If a driver > is using any other interface, it should have listed it, and would be > in violation of its ARC case if it didn't (IMHO, they should even list > consolidation private interfaces). > > The problem, with this "requirement", is that we have no way to > validate or enforce that only the ARC'd API's have been used. We > leave it to the developer (in theory, checked by the RTI advocate) to > appropriately use the API's. But if you know of a driver that uses a > non-public API, and it doesn't have an ARC case stating this use (so > all the GLDv3 drivers should be OK), then you can file a bug on the > driver. > > > Just maybe one of the things we need to think about is some set of > driver validation tools. > There used to be one.. .it was called "ddict". I sometimes believe I'm the only person on the planet who ever actually used "ddict". -- Garrett > > > ---- Randy > From Garrett.Damore at Sun.COM Wed Sep 26 11:44:01 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 11:44:01 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261118000.7356@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> <46FA988A.5020608@sun.com> <Pine.GSO.4.64.0709261118000.7356@eagle.softorchestra.com> Message-ID: <46FAA871.1070707@sun.com> Alan DuBoff wrote: > On Wed, 26 Sep 2007, James C. Liu wrote: > > >> Alan DuBoff wrote: >> >> >>> There is no having your cake and eating it too, if you want to keep the >>> processes in place and not lighten up on anything, it will continue to >>> require a sponsor. The process is really too complicated for the average >>> person to follow without help, IMO. >>> >>> >> Why can't we have our cake and eat it too? We don't need to dump >> the rigorousness of the existing process. We can still require some >> sponsor. It may not scale over time and then we'll have to rethink >> perhaps. But we've had this hallway discussion before, and the >> solution then was to promote the idea of an "experimental" section >> for drivers and apps that haven't gone through full testing. Some >> stuff (like kernel modules) would be available on the install media >> with the right disclaimers and caveats, and other packages would be >> available elsewhere. Over time, the idea would be to do the testing >> and certify code as official and move it over. But that would >> be transparent to endusers who would already have had access to the >> pkgs if they needed it. >> >> -James Liu - Solaris x86 IHV/OEM Engineering >> >> > > James, > > Yes, this would allow for drivers to be included that haven't passed the > rigorous testing. However, the OP stated that they didn't want to lighten > up on testing, or any current requirements. I was only pointing out that > while it doesn't need to be mandatory for them to have a sponsor (today > they are required to have one), it's all in their best interest. > > If there was no sponsor requirement today, it would be more difficult for > folks to grasp the process. I think the sponsor idea is to assist more > than anything else, at least from my view. If all community members > understood the putback process well enough there would be no need for a > sponsor, they would be able to do everything. Oh, and if all process was > outside the SWAN, that would also help and where I believe we're headed. > Let me state this without reservation: * I believe that all device drivers, and any other kernel software, should pass a fairly standardized set of test requirements. * I do not believe that "waiving" test failures is a reasonable thing to do, *unless* it can be shown that the failures meet one of a *very* small set of criteria: 1) The failure is due to a bug or bad assumption in the test suite. (Yes, I've had a few of these.) And the driver is performing properly. (Demonstrated by hand outside of the test suite, if necessary.) 2) The failure is benign and only shows up under unrealistic test scenarios. For example, a network driver that fails due to performance limitations when pushing out 1-byte UDP packets at 200Kpps/sec might be reasonable... as long as once the test concluded the NIC resumed behaving normally without administrator intervention. In case 1, a bug is warranted. In case 2, it may or may not be warranted, depending on the actual severity and likelihood of anyone ever hitting it. (Some of the test scenarios are very, very demanding, and very, very strict... e.g. dropping even a single ethernet packet can cause some of the NICDRV tests to fail!) There may be other cases than the three above, but I definitely think they should be examined closely before any waiver would ever be granted. -- Garrett > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From phil at bolthole.com Wed Sep 26 11:51:01 2007 From: phil at bolthole.com (Philip Brown) Date: Wed, 26 Sep 2007 11:51:01 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com>; from alan.duboff@sun.com on Wed, Sep 26, 2007 at 12:55:42AM -0700 References: <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> Message-ID: <20070926115101.A91989@bolthole.com> Putting my dusty "Driver developer" hat on for a second.. .:-) On Wed, Sep 26, 2007 at 12:55:42AM -0700, Alan DuBoff wrote: > ... > I feel the testing needs to be evaluated in some cases. For instance, > testing a NIC driver to be loaded and unloaded thousands of times will > probably just tell you if there is a memory leak. There could be better > ways to test that, IMO. Loading and unloading a driver is not a common > practice on most systems. Actually, that sounds like a GOOD TEST to me, that should be kept in! I've occasionally had bugs in drivers of mine, that really only manifested themselves in a comprehensible way, at driver load time. Very rare ones, to be sure. But then, that's a core component of driver testing: working out all the "very rare" cases, and making them bulletproof. From Garrett.Damore at Sun.COM Wed Sep 26 11:46:57 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 11:46:57 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.42089.235484.84538@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <Pine.GSO.4.64.0709261050440.2186@grimmy> <18170.42089.235484.84538@gargle.gargle.HOWL> Message-ID: <46FAA921.5070107@sun.com> James Carlson wrote: > Randy Fishel writes: > >> We do make the requirement of drivers to specify all the interfaces >> they expect to import and export. We don't make them list the >> DDI/DKI, as that is a set of implied interfaces. They must, however, >> specify if they intend to use GLDvX, kEF, and even FMA. If a driver >> is using any other interface, it should have listed it, and would be >> in violation of its ARC case if it didn't (IMHO, they should even list >> consolidation private interfaces). >> > > You're welcome to that opinion, and you can certainly demand it in any > reviews you do yourself, but it's not actually an architectural > requirement. > > It'd be possible to set some sort of OpenSolaris-wide policy here. We > could have an ARC case that specifies what things must be "declared" > and how they're declared. Such a thing (other than the existing > interface taxonomy) does not currently exist for the special case of > drivers. A fast-track specifying this (and the detailed rationale > behind it) might be helpful. > +1. I certainly never heard, or believed, that imported ON consolidation private interfaces required to be called out in ARC reviews. I always figured at that point, they sort of feel underneath ARC's radar. (Now importing project-private interfaces, or using ON private APIs from unbundled drivers, is a totally different story.) -- Garrett From phil at bolthole.com Wed Sep 26 12:03:06 2007 From: phil at bolthole.com (Philip Brown) Date: Wed, 26 Sep 2007 12:03:06 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FA783D.7050803@sun.com>; from Garrett.Damore@Sun.COM on Wed, Sep 26, 2007 at 08:18:21AM -0700 References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> Message-ID: <20070926120306.B91989@bolthole.com> On Wed, Sep 26, 2007 at 08:18:21AM -0700, Garrett D'Amore wrote: > James Carlson wrote: > >.... > > Thus, DDI isn't a mandate. > > > > Historically, believe I had the only DDI complaint ethernet drivers that > were to be found... either inside ON or outside of ON. I believe that > this signaled both deficiencies in ON, and in the various unbundled drivers. > > I believe that sticking to the DDI, while not a hard requirement for > drivers in ON, is a worthwhile cause even for the drivers in ON. Gaaaah... DDI *should be* a mandate, even for drivers in ON. Why? because it ensures that the DDI level stuff, stays actually USEFUL. Multiple times, I think, I was trying to write a driver, and looking through the DDI for nice clean ways to do it.. and, there just wasnt a way to do it! There were existing sun drivers that did similar things, and they got the job done just fine... by going around DDI. But for external people, "you cant get there from here". So, the answer to the long-standing question/gripe of, "why are there so few drivers by non-sun people out there?" would be at least in part, "Well, gee, becuase sun didnt provide non-src-privileged people a means of writing them!" So, I doubly underscore the later paragraph from Garrett: > 3) By using the DDI, you know that you're using the same APIs that we > recommend to external developers. That means you help to uncover any > limitations in the DDI. Even in this modern day and age, where people now have "access to the source"... there is something very positive to be said about having an official DDI for things, that must be adhered to.. because then, (ideally), the interfaces would actually get DOCUMENTED...? Believe it or not, open source developers DO actually appreciate documentation (even if they rarely write any themselves ;-) From alan.coopersmith at sun.com Wed Sep 26 12:07:05 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 12:07:05 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> Message-ID: <46FAADD9.4090509@sun.com> Alan DuBoff wrote: > I realize that OpenSolaris is not just ON, however today it pretty much is > from a sources standpoint. Except that JDS's sources are about twice the size of ON's last I heard (19 million lines for JDS vs. 10 million for ON), and there's a huge amount of source from all the other consolidations. For regular Solaris, ON is only around 20% of the software - for OpenSolaris, it's probably closer to 30-40%, since we have some large chunks of Solaris like CDE that aren't released. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From swalker at opensolaris.org Tue Sep 25 16:43:12 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:43:12 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> Message-ID: <b9c544f0709251643o15ad5a8fraeaa174bf4ce0d32@mail.gmail.com> On 25/09/2007, Peter Tribble <peter.tribble at gmail.com> wrote: > On 9/25/07, Stephen Lau <stevel at sun.com> wrote: > > Joerg Schilling wrote: > > > > > > I believe there are two problems: missing visibility of the OGB and missing > > > power of the community. > > > > Why do you perceive the former to be a problem? Visibility of the OGB > > is a double-edged sword. Yes, we want people to know there is an OGB so > > that they know there is a path of appeals and such; but we don't want > > the OGB to be so visible to the point that it is "in-your-face"[1]. The > > OGB has taken the view that the community is to be self-governing. > > But does the community know that it's supposed to be governing itself? Given our voter turnout, and the problems we had in getting leaders to communicate about their community, one could conclude they are strangely not aware (or apathetic). > I suspect that most members of the community simply don't understand > what powers, rights, and responsibilities they have, nor how to exercise > them. I think part of the problem is that most people's eyes are going to glaze over once you start talking about those things. Unless you can sum them all up in a few bullet lists, most folks are going to ignore them... > So how do we encourage communities and individuals to take up the > reins of governance? A better user-based web portal with tools to ease that process with short-and-to-the-point summaries would go a long way... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From swalker at opensolaris.org Tue Sep 25 16:45:04 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:45:04 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F97328.90502@Sun.Com> References: <46F2FE5F.3070508@sun.com> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> Message-ID: <b9c544f0709251645h65020ed2rcb7d9cdfe1e1d736@mail.gmail.com> On 25/09/2007, John Plocher <John.Plocher at sun.com> wrote: > Peter Tribble wrote: > > So how do we encourage communities and individuals to take up the > > reins of governance? > > Not to be flippant, but why would they want to? > > That is, what problems are they currently having that could > be solved by their spending energy on self-governance? > > My perspective is that the answer is that there are lots of > serious problems, none of which can effectively be solved thru > self-governance alone. > > Problem 1: The structure of the meta-community echoes the > confusion in people's minds of what exactly OpenSolaris is > and should be. Because people are not in agreement with > what it is, there is no consensus on how it should be > structured. +1 "Where there is no vision the people perish..." or something like that... > Problem 2: The observable artifacts of the community are > still locked away from direct community manipulation: The > source tree, the bug database, the web site, the mailing > list infrastructure, the ON C-Team and even the ARCs are > /ALL/ solely maintainable by a set of under-resourced, > under-appreciated and constantly dwindling group of Sun > employees. +1 Power to the people > Solve these problems and the community of developers will > grow. Ignore them, and it will stagnate and die. Agreed. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Garrett.Damore at Sun.COM Wed Sep 26 12:12:45 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 12:12:45 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <20070926120306.B91989@bolthole.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> Message-ID: <46FAAF2D.4030901@sun.com> So, now my question for PSARC is, if someone (say me, or Philip) wanted to start up a case to make this a new requirement going forward (and presumably we could allow for drivers to waive the requirement, but they should need to be justified) wanted to sponsor a case, does this fall under the realm of what would be called a fasttrack? Such a proposal would itself be very light weight, but I'm not at all convinced it would be non-controversial enough to qualify for less than a full case. If nothing else, it gives PSARC a chance to express a stronger opinion. I'm willing to author the original document (fasttrack or otherwise), although if its a full case then I need a sponsor on PSARC, I believe. -- Garrett Philip Brown wrote: > On Wed, Sep 26, 2007 at 08:18:21AM -0700, Garrett D'Amore wrote: > >> James Carlson wrote: >> >>> .... >>> Thus, DDI isn't a mandate. >>> >>> >> Historically, believe I had the only DDI complaint ethernet drivers that >> were to be found... either inside ON or outside of ON. I believe that >> this signaled both deficiencies in ON, and in the various unbundled drivers. >> >> I believe that sticking to the DDI, while not a hard requirement for >> drivers in ON, is a worthwhile cause even for the drivers in ON. >> > > Gaaaah... > > DDI *should be* a mandate, even for drivers in ON. > > Why? because it ensures that the DDI level stuff, stays actually USEFUL. > > Multiple times, I think, I was trying to write a driver, and looking > through the DDI for nice clean ways to do it.. > and, there just wasnt a way to do it! > There were existing sun drivers that did similar things, and they got the > job done just fine... by going around DDI. > But for external people, "you cant get there from here". > > > So, the answer to the long-standing question/gripe of, > "why are there so few drivers by non-sun people out there?" > would be at least in part, > > "Well, gee, becuase sun didnt provide non-src-privileged people a means of > writing them!" > > > So, I doubly underscore the later paragraph from Garrett: > > > >> 3) By using the DDI, you know that you're using the same APIs that we >> recommend to external developers. That means you help to uncover any >> limitations in the DDI. >> > > > > Even in this modern day and age, where people now have "access to the > source"... there is something very positive to be said about having an > official DDI for things, that must be adhered to.. because then, > (ideally), the interfaces would actually get DOCUMENTED...? > > Believe it or not, open source developers DO actually appreciate > documentation (even if they rarely write any themselves ;-) > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From phil at bolthole.com Wed Sep 26 12:37:23 2007 From: phil at bolthole.com (Philip Brown) Date: Wed, 26 Sep 2007 12:37:23 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAAF2D.4030901@sun.com>; from Garrett.Damore@Sun.COM on Wed, Sep 26, 2007 at 12:12:45PM -0700 References: <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> Message-ID: <20070926123723.D5803@bolthole.com> On Wed, Sep 26, 2007 at 12:12:45PM -0700, Garrett D'Amore wrote: > So, now my question for PSARC is, if someone (say me, or Philip) wanted > to start up a case to make this a new requirement going forward (and > presumably we could allow for drivers to waive the requirement, but they > should need to be justified) wanted to sponsor a case, does this fall > under the realm of what would be called a fasttrack? > > Such a proposal would itself be very light weight, but I'm not at all > convinced it would be non-controversial enough to qualify for less than > a full case. If nothing else, it gives PSARC a chance to express a > stronger opinion. for the record.. I believe I actually raised this proposal in front of a few high level people, a few years back (and I think Stephen Hahn was present at that meeting, too ;-) [*waves to Stephen*] I even had the stipulation of, "requirement could be waived, but had to justify" then too. but it still got shot down, in the meeting. From randy.fishel at sun.com Wed Sep 26 12:39:02 2007 From: randy.fishel at sun.com (Randy Fishel) Date: Wed, 26 Sep 2007 12:39:02 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.42089.235484.84538@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <Pine.GSO.4.64.0709261050440.2186@grimmy> <18170.42089.235484.84538@gargle.gargle.HOWL> Message-ID: <Pine.GSO.4.64.0709261237320.8725@grimmy> (resend) > Randy Fishel writes: > > We do make the requirement of drivers to specify all the interfaces > > they expect to import and export. We don't make them list the > > DDI/DKI, as that is a set of implied interfaces. They must, however, > > specify if they intend to use GLDvX, kEF, and even FMA. If a driver > > is using any other interface, it should have listed it, and would be > > in violation of its ARC case if it didn't (IMHO, they should even list > > consolidation private interfaces). > > You're welcome to that opinion, and you can certainly demand it in any > reviews you do yourself, but it's not actually an architectural > requirement. Agreed, it is my opinion, and I try to work that into my reviews of drivers. I would, however, like it to be more an architectual requirement, and would be willing to make some arguments or proposals in the near future (it would be real nice to get my current project into Solaris). > > It'd be possible to set some sort of OpenSolaris-wide policy here. We > could have an ARC case that specifies what things must be "declared" > and how they're declared. Such a thing (other than the existing > interface taxonomy) does not currently exist for the special case of > drivers. A fast-track specifying this (and the detailed rationale > behind it) might be helpful. > As Garrett, I am in agreement, and am more than willing to work with and/or sponsor such an effort (though again, after I integrate my current set of bits). Garrett- you in for a collaboration here? ---- Randy From alan.duboff at sun.com Wed Sep 26 15:06:47 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 15:06:47 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FAADD9.4090509@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> <46FAADD9.4090509@sun.com> Message-ID: <Pine.GSO.4.64.0709261458090.17209@konocti> On Wed, 26 Sep 2007, Alan Coopersmith wrote: > Alan DuBoff wrote: >> I realize that OpenSolaris is not just ON, however today it pretty much >> is from a sources standpoint. > > Except that JDS's sources are about twice the size of ON's last I heard > (19 million lines for JDS vs. 10 million for ON), and there's a huge > amount of source from all the other consolidations. > > For regular Solaris, ON is only around 20% of the software - for > OpenSolaris, it's probably closer to 30-40%, since we have some large > chunks of Solaris like CDE that aren't released. Alan, Not wanting to debate what OpenSolaris actually is, but it's interesting that you can take a set of sources that were already open sourced by other folks, include them on the OpenSolaris website and they become one and the same. Maybe this is one large distinction between (Open)Solaris and Linux, because Linux is really just a kernel and the other software is the system, although the majority of folks refer to the entire system when they say "Linux". I read a day or two ago (from you I believe) that Sun will not be calling the desktop JDS anymore and that it would be calling it GNOME. I remember GNOME as an open source project before OpenSolaris was created. Sun didn't do anything different with GNOME to include it with OpenSolaris, yet somehow it has become a part of it all. If Sun would clearly define how OpenSolaris is determined, there wouldn't be various views of it, including me, you, John, Jim, or others... >From the above I suspect we're gonna be using IBM's ancient definition of MLOC to determine what is a part of OpenSolaris.:-/ -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.coopersmith at sun.com Wed Sep 26 16:08:14 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 16:08:14 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAE504.6040108@Sun.COM> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> Message-ID: <46FAE65E.1040003@sun.com> Joseph Kowalski wrote: > Where "lsarc" could be what I believe are the interested parties. > Somethings may be interesting to {psarc, lsarc, wsarc}, but not (for > example) fwarc. Are drivers ever reviewed by any arc other than PSARC? I'm not sure LSARC, WSARC or FWARC need to be involved with this. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From David.Clack at Sun.COM Wed Sep 26 17:13:11 2007 From: David.Clack at Sun.COM (David Clack) Date: Wed, 26 Sep 2007 17:13:11 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261458090.17209@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> <46FAADD9.4090509@sun.com> <Pine.GSO.4.64.0709261458090.17209@konocti> Message-ID: <1190851991.27503.70.camel@localhost> On Wed, 2007-09-26 at 15:06 -0700, Alan DuBoff wrote: > On Wed, 26 Sep 2007, Alan Coopersmith wrote: > > > Alan DuBoff wrote: > >> I realize that OpenSolaris is not just ON, however today it pretty much > >> is from a sources standpoint. > > > > Except that JDS's sources are about twice the size of ON's last I heard > > (19 million lines for JDS vs. 10 million for ON), and there's a huge > > amount of source from all the other consolidations. > > > > For regular Solaris, ON is only around 20% of the software - for > > OpenSolaris, it's probably closer to 30-40%, since we have some large > > chunks of Solaris like CDE that aren't released. > > Alan, > > Not wanting to debate what OpenSolaris actually is, but it's interesting > that you can take a set of sources that were already open sourced by other > folks, include them on the OpenSolaris website and they become one and the > same. Maybe this is one large distinction between (Open)Solaris and Linux, > because Linux is really just a kernel and the other software is the > system, although the majority of folks refer to the entire system when > they say "Linux". > > I read a day or two ago (from you I believe) that Sun will not be calling > the desktop JDS anymore and that it would be calling it GNOME. I remember > GNOME as an open source project before OpenSolaris was created. Sun didn't > do anything different with GNOME to include it with OpenSolaris, yet > somehow it has become a part of it all. Not true, take a look at svn co svn +ssh://anon at svn.opensolaris.org/svn/jds/spec-files/trunk We patch it to make it run on our kernel, fix some bugs, add trusted Solaris extensions Lots of the good original work for parts of Gnome are done by Sun JDS people. I hate it when people put down the hard work of our people. > If Sun would clearly define how > OpenSolaris is determined, there wouldn't be various views of it, > including me, you, John, Jim, or others... > > >From the above I suspect we're gonna be using IBM's ancient definition of > MLOC to determine what is a part of OpenSolaris.:-/ > > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit David Clack Solaris X86 Evangelist Senior Systems Engineer OEM Software Sales Sun Microsystems 642, Chinook Ave SE, Ocean Shores, WA, USA, 98569 Phone +1-360-289-2158 Fax +1-360-289-2091 Mobile +1-206-265-1904 Email David.Clack at Sun.COM -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070926/cc5dc1a9/attachment.html> From alan.coopersmith at sun.com Wed Sep 26 17:21:16 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 17:21:16 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261458090.17209@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> <46FAADD9.4090509@sun.com> <Pine.GSO.4.64.0709261458090.17209@konocti> Message-ID: <46FAF77C.10502@sun.com> Alan DuBoff wrote: > Not wanting to debate what OpenSolaris actually is, but it's interesting > that you can take a set of sources that were already open sourced by > other folks, include them on the OpenSolaris website and they become one > and the same. You prefer 100% original work then? Look no further than the Storage community, with their non-ON releases such as QFS & SAM - they even have a simple diagram, easily colored to show how much of their OpenSolaris work is outside of ON: http://www.opensolaris.org/os/community/storage/ But then, ON doesn't count if you're looking for 100% original - how much of it came from BSD's open source releases? -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From alan.duboff at sun.com Wed Sep 26 17:31:12 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 17:31:12 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAE65E.1040003@sun.com> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> <46FAE65E.1040003@sun.com> Message-ID: <Pine.GSO.4.64.0709261729580.17209@konocti> On Wed, 26 Sep 2007, Alan Coopersmith wrote: > Are drivers ever reviewed by any arc other than PSARC? I'm not sure > LSARC, WSARC or FWARC need to be involved with this. I don't think so, but they are supposed to be by c-team. PSARC seems more interested in the interfaces and/or wether new ones are required or if the driver uses existing interfaces. -- Alan DuBoff - Solaris x86 IHV/OEM Group From John.Plocher at Sun.COM Wed Sep 26 17:43:19 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 26 Sep 2007 17:43:19 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAE504.6040108@Sun.COM> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> Message-ID: <46FAFCA7.2090701@Sun.Com> Joseph Kowalski wrote: > I'd assume that: > To: psarc-ext > CC: lsarc > CC: wsarc > ... > CC: <submitter> > > Should be OK. (John?) The problem is that external people can't send mail to WSARC and LSARC aliases directly, so those ARC members would be cut off from all external replies. The "best" answer right now is to create the case as a PSARC case, put the wsarc and lsarc aliases in the IAM:Interest list and send the mail to PSARC-ext only, letting the mail subsystem handle the notification and inclusion of the other ARCs. -John From Garrett.Damore at Sun.COM Wed Sep 26 21:44:40 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 21:44:40 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAFCA7.2090701@Sun.Com> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> <46FAFCA7.2090701@Sun.Com> Message-ID: <46FB3538.4000302@sun.com> John Plocher wrote: > Joseph Kowalski wrote: > >> I'd assume that: >> To: psarc-ext >> CC: lsarc >> CC: wsarc >> ... >> CC: <submitter> >> >> Should be OK. (John?) > > The problem is that external people can't send mail to WSARC and LSARC > aliases directly, so those ARC members would be cut off from all external > replies. > > The "best" answer right now is to create the case as a PSARC case, > put the wsarc and lsarc aliases in the IAM:Interest list and send > the mail to PSARC-ext only, letting the mail subsystem handle the > notification and inclusion of the other ARCs. I really believe that this is only a PSARC issue. I doubt very much that other ARCs look at software that imports both the DDI and non-public kernel APIs. And note that I'm really only talking about *device drivers*.... not other kernel software. -- Garrett > > -John > > > From Garrett.Damore at Sun.COM Wed Sep 26 21:46:25 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 21:46:25 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAFE9F.50809@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> Message-ID: <46FB35A1.2030101@sun.com> Artem Kachitchkine wrote: > > Why is this an ARC issue? Drivers don't usually go through ARC. it > should probably be on the ON C-Team Driver Checklist. Add DDICT to the > ON tool set and integrate it with wx. What I'm talking about is an ARC requirement to *disclose* non-DDI interface consumption by drivers, and to strongly discourage the practice of going outside of the DDI for device drivers. I think that *is* an issue for PSARC at least. (We're talking about interfaces here, not implementation.) -- Garrett > > -Artem From Garrett.Damore at Sun.COM Wed Sep 26 22:06:56 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 22:06:56 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB38F2.5090604@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> Message-ID: <46FB3A70.2020906@sun.com> Artem Kachitchkine wrote: > >> What I'm talking about is an ARC requirement to *disclose* non-DDI >> interface consumption by drivers, and to strongly discourage the >> practice of going outside of the DDI for device drivers. I think >> that *is* an issue for PSARC at least. > > How do you envision to enforce this in practice? Should C-team > communicate back to PSARC? It used to be mostly one-way, making it > bounce back and forth is thickening the process, while we're trying to > thin it, to encourage external contributors. > > > (We're talking about interfaces here, not implementation.) > > In this particular case, the line is unclear. A minor bugfix can turn > a DDI compliant driver into a non-compliant one. In my mind it's more > implementation (think "lint clean") than architecture. > > -Artem As a policy, I'm not sure how *hard and fast* the rule needs to be. I certainly think that when a case for a new driver goes before ARC, it should list the APIs it is using. This may mean that driver projects which were able to skip ARC altogether (although a poor choice, IMO, self-review should be used instead in such cases, IMO) now need to expose themselves to ARC. In and of itself, that's not too bad. And some APIs, like the GLDv3, should probably not cause a case to do more than self-review, so the cost of effort is minimal in such cases. Bug fixes that import APIs that the driver didn't have before.... I'm not so sure about. I'd hope that a fasttrack is not too burdensome. And again, maybe even some would qualify for self-review. Certainly the transition in going from DDI compliant to non-DDI-compliant is not one that I would want to just sweep under the rug. Better to announce that transition up front, IMO. (And if going non-DDI compliant is required to fix a bug, my first question would be "why?") -- Garrett From Artem.Kachitchkin at Sun.COM Wed Sep 26 22:00:34 2007 From: Artem.Kachitchkin at Sun.COM (Artem Kachitchkine) Date: Wed, 26 Sep 2007 22:00:34 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB35A1.2030101@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> Message-ID: <46FB38F2.5090604@sun.com> > What I'm talking about is an ARC requirement to *disclose* non-DDI > interface consumption by drivers, and to strongly discourage the > practice of going outside of the DDI for device drivers. I think that > *is* an issue for PSARC at least. How do you envision to enforce this in practice? Should C-team communicate back to PSARC? It used to be mostly one-way, making it bounce back and forth is thickening the process, while we're trying to thin it, to encourage external contributors. > (We're talking about interfaces here, not implementation.) In this particular case, the line is unclear. A minor bugfix can turn a DDI compliant driver into a non-compliant one. In my mind it's more implementation (think "lint clean") than architecture. -Artem From Artem.Kachitchkin at Sun.COM Wed Sep 26 22:17:07 2007 From: Artem.Kachitchkin at Sun.COM (Artem Kachitchkine) Date: Wed, 26 Sep 2007 22:17:07 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3A70.2020906@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> Message-ID: <46FB3CD3.5000909@sun.com> > Certainly the transition in going from DDI compliant to > non-DDI-compliant is not one that I would want to just sweep under the > rug. Better to announce that transition up front, IMO. (And if going > non-DDI compliant is required to fix a bug, my first question would be > "why?") I'm not disagreeing this that. I'm just wondering why ARC should be the one asking that question and not C-team/CRT. -Artem From Jim.Grisanzio at Sun.COM Thu Sep 27 06:56:57 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Thu, 27 Sep 2007 22:56:57 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <5b5090780709251344u5b7221efo1d3b68965a97f53b@mail.gmail.com> References: <46F31335.6060700@Sun.COM> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <5b5090780709251344u5b7221efo1d3b68965a97f53b@mail.gmail.com> Message-ID: <46FBB6A9.8050307@sun.com> Brandorr wrote: > On 9/25/07, Darren Reed <Darren.Reed at sun.com> wrote: >> Heck, maybe we participants should just do that anyway, rather >> than wait for Sun to work out what it wants to do with OS.o? > > While I understand your frustration at a low level of outside > contribution, I don't think Sun is to be blamed. I think it stems from > the fact that Solaris is a mature, formerly proprietary, product. Many > users of Solaris today don't give a care about any notions of > community of contribution. (So long as the get what their support > contracts cover) > > We here at OpenSolaris.org are currently in the process of building a > community. This community currently exists mostly of Sun employees, > and a few brave new worlders, and a few excited Solaris customers. > > We are in a bit of a chicken and egg situation. What we need right now > is more outside manpower, to build the infrastructure to make it > easier for outside contributers to contribute. > > Right now, one way we can address this, is to do outreach. Currently > this is on the agenda for our next user group meeting: This is a very nice list below. Perhaps you can post these in Advocacy? I'm sure there are many contribution docs within various community groups and projects, but here are two general contribution docs: http://opensolaris.org/os/community/advocacy/help/ http://opensolaris.org/os/communities/participation/ Jim > Get involved: > ------------------ > - Mentoring - We are looking at starting a new mentoring program. We > are looking for Seniors that would be interested in mentoring Juniors. > (Developers and sysadmins). Please email Brian Gupta > (brandorr at opensolaris.org) if you are interested. > - The OpenSolaris.org Systems administration Community Group is > looking to expand membership. > syadmin-discuss-subscribe at opensolaris.org > - Volunteering to help build a new OpenSolaris user portal, as well as > an NYOSUG.org website. (email Brian Gupta (brandorr at opensolaris.org) > if you are interested in either.) > - We will be launching/announcing a virtual development lab next > month, that will allow anyone who is interested in developing > OpenSolaris or porting open source apps to OpenSolaris, to gain free > access to VMWare ESX Virtual Machines. Let Brian Gupta > (brandorr at opensolaris.org) know if you are interested > in either helping administer ESX, or if you want access to a VM. (It's > not setup yet). > - We can use help planning these UG meetings. Mail Isaac or Brian Gupta > "Isaac Rozenfeld - NEW YORK" <Isaac.Rozenfeld at sun.com>, Brian Gupta > (brandorr at opensolaris.org) > - Brian Gupta (brandorr at opensolaris.org) is going to the first > OpenSolaris Developers Summit, and will have more to report next > month. > (http://www.genunix.org/wiki/index.php?title=OpenSolaris_Summit.) > - Brian Gupta (brandorr at opensolaris.org) assembled the following new > user FAQ: assembling: > http://www.genunix.org/wiki/index.php?title=FAQ_Prototype > > What I would ask each contributing member of OpenSolaris.org, is to > find at least one other person that has the ability to contribute, > and walk them through their first contribution. (We'd double the > number of contributers over night. > > Also, UserGroups should be actively advertising outside of their own > users. e.g. - Reach out to Linux and BSD user groups, as well as local > UNIGROUP, Lisa and SAGE chapters. The goal being to increase the > number of attendees to your meetings. > > Basically, I feel that by working around the barriers for now, we will > be able to grow the community enough to remove them down the road. > (With the added manpower). From John.Plocher at Sun.COM Thu Sep 27 09:08:29 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 27 Sep 2007 09:08:29 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB35A1.2030101@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> Message-ID: <46FBD57D.1020100@Sun.Com> Garrett D'Amore wrote: > Artem Kachitchkine wrote: >> Why is this an ARC issue? > I think that > *is* an issue for PSARC at least. This issue is on the boundry between the "systems" architecture of OpenSolaris and the design/implementation of the ON subsystem. IFF device drivers were an implementation detail of the OpenSolaris ON consolidation alone, then the ARCs would not really care about it, since they could safely delegate the whole shebang to the people managing ON. If ON wished to do a global refactoring or flag day or whatever, they /could/ also go thru and update every single driver to match, and all such changes would be invisible to everyone outside of ON. Of course, the above assumption is false. Drivers are written outside of the ON consolidation; there is even early pie-in-the-sky talk about refactoring the existing ON to potentially separate out things like drivers into their own consolidation(s). This makes the driver interface to ON part of the systems integration conversation - it no longer is simply an implementation detail of ON. -John From John.Plocher at Sun.COM Thu Sep 27 09:14:53 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 27 Sep 2007 09:14:53 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3CD3.5000909@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> <46FB3CD3.5000909@sun.com> Message-ID: <46FBD6FD.60003@Sun.Com> Artem Kachitchkine wrote: > I'm not disagreeing this that. I'm just wondering why ARC should be the > one asking that question and not C-team/CRT. One reason may be that the ON C-Team right now is a Sun-only closed entity, this is the /Open/Solaris community, and the ARC has more visibility and/or transparency in the community than does the C-Team. Another is that not all drivers from the community are going to be of interest to the Sun ON C-Team... -John From james.d.carlson at Sun.COM Thu Sep 27 05:33:00 2007 From: james.d.carlson at Sun.COM (James Carlson) Date: Thu, 27 Sep 2007 08:33:00 -0400 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3CD3.5000909@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> <46FB3CD3.5000909@sun.com> Message-ID: <18171.41724.616477.425290@gargle.gargle.HOWL> Artem Kachitchkine writes: > > Certainly the transition in going from DDI compliant to > > non-DDI-compliant is not one that I would want to just sweep under the > > rug. Better to announce that transition up front, IMO. (And if going > > non-DDI compliant is required to fix a bug, my first question would be > > "why?") > > I'm not disagreeing this that. I'm just wondering why ARC should be the > one asking that question and not C-team/CRT. I think you might be confusing policy with enforcement. The requirement itself (if we were to approve one) is a system architectural issue. We would be saying that, despite the clear indications to the contrary in the interface taxonomy and existing ARC practice, projects of a certain type may not actually use Consolidation Private interfaces from their own consolidation without additional review. That's the ARC issue. _Enforcement_ of these requirements is strictly in the hands of the applicable C-teams. Just as with our other rules, it's up to the C-team to make sure that the project team has properly complied with all applicable review requirements, architectural restrictions, and the like. Yes, if I were on the C-team, I'd be advocating a new checklist item and probably some new local policies to make sure that we catch any mistakes that might be made. Those things aren't architectural, though. Only the policy (disallowing unreviewed use of private interfaces, and thus mandating architectural review) itself is. -- James Carlson, Solaris Networking <james.d.carlson at sun.com> Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677 From sommerfeld at sun.com Thu Sep 27 05:30:36 2007 From: sommerfeld at sun.com (Bill Sommerfeld) Date: Thu, 27 Sep 2007 08:30:36 -0400 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3CD3.5000909@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> <46FB3CD3.5000909@sun.com> Message-ID: <1190896236.5428.26.camel@localhost> On Wed, 2007-09-26 at 22:17 -0700, Artem Kachitchkine wrote: > I'm not disagreeing this that. I'm just wondering why ARC should be the > one asking that question and not C-team/CRT. If our goal is to have a DDI that is sufficient for all types of drivers then the ARC needs to know about the holes in it that are found in practice. The c-team/crt is the last line of defense as always and would in fact be the ones asking the question if the project team hadn't figured it out for themselves. When the answer is "no, we couldn't get it done within the DDI", it's a deficiency in the DDI and (in a world where the DDI is supposed to be truly complete) the ARC really needs to know about this. - Bill From james.d.carlson at Sun.COM Thu Sep 27 05:34:53 2007 From: james.d.carlson at Sun.COM (James Carlson) Date: Thu, 27 Sep 2007 08:34:53 -0400 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB38F2.5090604@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> Message-ID: <18171.41837.171584.86169@gargle.gargle.HOWL> Artem Kachitchkine writes: > > (We're talking about interfaces here, not implementation.) > > In this particular case, the line is unclear. A minor bugfix can turn a > DDI compliant driver into a non-compliant one. In my mind it's more > implementation (think "lint clean") than architecture. That has always been true. Even a one-line bug fix may well have important architectural implications, just as a multi-kiloline change might not have any architectural issues worth reviewing. You can't determine ARC issues by running "wc -l". -- James Carlson, Solaris Networking <james.d.carlson at sun.com> Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677 From jek3 at sun.com Thu Sep 27 10:53:25 2007 From: jek3 at sun.com (Joseph Kowalski) Date: Thu, 27 Sep 2007 07:53:25 -1000 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB35A1.2030101@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> Message-ID: <46FBEE15.5010101@sun.com> Garrett D'Amore wrote: > Artem Kachitchkine wrote: >> >> Why is this an ARC issue? Drivers don't usually go through ARC. it >> should probably be on the ON C-Team Driver Checklist. Add DDICT to >> the ON tool set and integrate it with wx. > > What I'm talking about is an ARC requirement to *disclose* non-DDI > interface consumption by drivers, and to strongly discourage the > practice of going outside of the DDI for device drivers. I think that > *is* an issue for PSARC at least. (We're talking about interfaces > here, not implementation.) Sorry, I was suggesting methods for the general case. Yea, this is pretty much a "PSARC only" issue. Right now, drivers often don't need ARC involvement (beyond the "register the name" paperwork - closed approved automatic). I think the "closed approved" path is only appropriate (already) for DDI-compliant drivers... if <DDI-compliant> -> closed approved automatic else -> fast-track (or greater) I think the only thing we need to do is to advertise and enforce the above, existing rule. The question is "how to enforce". I really don't have a suggestion for that. (I suspect somebody will!) - jek3 From jek3 at sun.com Thu Sep 27 11:07:58 2007 From: jek3 at sun.com (Joseph Kowalski) Date: Thu, 27 Sep 2007 08:07:58 -1000 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18171.41837.171584.86169@gargle.gargle.HOWL> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <18171.41837.171584.86169@gargle.gargle.HOWL> Message-ID: <46FBF17E.1060400@sun.com> James Carlson wrote: > You can't determine ARC issues by running "wc -l". > Poster boy: The DVD worth of materials for SUS6 was a fast-track. - jek3 From Jesse.Silver at Sun.COM Sat Sep 29 13:32:46 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Sat, 29 Sep 2007 13:32:46 -0700 Subject: [opensolaris-summit] Hotel Reservations Message-ID: <46FEB66E.30109@sun.com> To those in need of hotel accommodations: The Summit is closing in fast. If you need a hotel room and have yet to reserve one, please call the Coast Santa Cruz Hotel immediately at 831-426-4330. Reference "Sun Microsystems" to get our special rate of $189/night. The rooms WILL BE FILLED by the end of this week, so please call now to secure one for yourself. If you have any trouble with the reservations people, or are told that the room block is full, please call me, Jesse Silver, at 877-868-6937. Thanks, and we're looking forward to a great summit, Jesse

From Jesse.Silver at Sun.COM Fri Sep 7 15:11:49 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Fri, 07 Sep 2007 15:11:49 -0700 Subject: [opensolaris-summit] ping Message-ID: <46E1CCA5.9000604@sun.com> From Jesse.Silver at Sun.COM Fri Sep 7 15:16:42 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Fri, 07 Sep 2007 15:16:42 -0700 Subject: [opensolaris-summit] REMINDER Message-ID: <46E1CDCA.2000503@sun.com> Reminder: The weekend of October 13^th is little more than a month away and the OpenSolaris Developer Summit is fast approaching. Registration: Thanks to those who have already registered. If you are planning to attend but haven't registered yet, please add your name to our Summit Wiki as soon as possible. We are trying to get a good count for planning purposes. Also be sure to join our mailing list, opensolaris-summit at opensolaris.org . Agenda: We've had the following topics suggested: * Packaging * Drivers * Desktop * Patching * Installation * Community growth and participation * Whatever else comes up Travel Sponsorships: Travel sponsorships covering air and hotel accommodations are still available! To request travel sponsorship or to make any summit related inquiries, please e-mail jesse.silver at sun.com . Deadline is September 10th for requests for sponsorship. Planning: Get involved in the planning. We'll start on Tuesday, September 11th at 8am PST and 3pm PST to accommodate multiple timezones. Anyone who would like to is encouraged to participate. Toll Free Dial In Number: (866)230-6968 Int'l Access/Caller Paid Dial In Number: (865)544-7856 ACCESS CODE: 2192132 See you at the Summit! -------------- next part -------------- An HTML attachment was scrubbed... URL: From brandorr at opensolaris.org Fri Sep 7 15:22:00 2007 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 7 Sep 2007 18:22:00 -0400 Subject: [opensolaris-summit] ping In-Reply-To: <46E1CCA5.9000604@sun.com> References: <46E1CCA5.9000604@sun.com> Message-ID: <5b5090780709071522k14f0ba10yf16b075b4aa08c7a@mail.gmail.com> pong On 9/7/07, Jesse Silver wrote: > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From shivakumar.gn at gmail.com Fri Sep 7 19:37:39 2007 From: shivakumar.gn at gmail.com (S h i v) Date: Sat, 8 Sep 2007 08:07:39 +0530 Subject: [opensolaris-summit] [osol-discuss] REMINDER: OpenSolaris Developer Summit In-Reply-To: <46E1CC25.4040603@sun.com> References: <46E1CC25.4040603@sun.com> Message-ID: <319ee2b10709071937q44159199k5786de5a286efe05@mail.gmail.com> On 9/8/07, Sara Dornsife wrote: > > Agenda: > We've had the following topics suggested: > > Packaging > Drivers > Desktop > Patching > Installation > Community growth and participation Am not a participant. I topic I would have suggested if I were one is already there on the wiki but not in the above list, "Is it hard to contribute to OpenSolaris? How can we make it easier?" suggested by Stephen Lau. Note that "Community growth and participation" is likely to lead into different (outward looking - advocacy related) discussions than Stephen Lau's that might result in inward looking(tools, processes) discussions. My .02c regards Shiv From Jesse.Silver at Sun.COM Sat Sep 8 01:38:14 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Sat, 08 Sep 2007 01:38:14 -0700 Subject: [opensolaris-summit] Quick note about Hotels Message-ID: <46E25F76.6010105@sun.com> A quick note for non-Sun employees who have not asked for sponsored travel and for Sun employees whose departments are paying for their travel: So as to keep everyone centralized and minimize the complexities of transporting participants to and from UCSC and our Saturday evening event, please make your best effort to book your room at the Coast Santa Cruz Hotel. If for any reason you do not stay at the Coast Santa Cruz Hotel and will not have a car, please try to book a room close by (the Santa Cruz Beach Boardwalk vicinity). Thanks for your cooperation! From Sara.Dornsife at Sun.COM Tue Sep 11 11:37:02 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Tue, 11 Sep 2007 13:37:02 -0500 Subject: [opensolaris-summit] REMINDER: Planning meeting today 3p PT Message-ID: <46E6E04E.8000308@sun.com> Agenda: - Do we have everyone who needs to be there? - Sponsorships are still available.... - Topics/Tracks/Sessions - Round Table (866)230-6968 Int'l Access/Caller Paid Dial In Number: (865)544-7856 ACCESS CODE: 2192132 Talk to y'all at 3p PT Sara From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 11 11:44:30 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 11 Sep 2007 20:44:30 +0200 Subject: [opensolaris-summit] REMINDER: Planning meeting today 3p PT In-Reply-To: <46E6E04E.8000308@sun.com> References: <46E6E04E.8000308@sun.com> Message-ID: <46e6e20e.g3HrmUJsj2tbaWWj%Joerg.Schilling@fokus.fraunhofer.de> Sara Dornsife wrote: > Agenda: > - Do we have everyone who needs to be there? > - Sponsorships are still available.... > - Topics/Tracks/Sessions > - Round Table > > (866)230-6968 > Int'l Access/Caller Paid Dial In Number: (865)544-7856 > ACCESS CODE: 2192132 > > Talk to y'all at 3p PT Who should dial in? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Sara.Dornsife at Sun.COM Tue Sep 11 12:48:40 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Tue, 11 Sep 2007 14:48:40 -0500 Subject: [opensolaris-summit] REMINDER: Planning meeting today 3p PT In-Reply-To: <46e6e20e.g3HrmUJsj2tbaWWj%Joerg.Schilling@fokus.fraunhofer.de> References: <46E6E04E.8000308@sun.com> <46e6e20e.g3HrmUJsj2tbaWWj%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46E6F118.1090008@sun.com> Anyone interested in helping with the planning or if you have any questions or comments.... Sara Joerg Schilling wrote: > Sara Dornsife wrote: > > >> Agenda: >> - Do we have everyone who needs to be there? >> - Sponsorships are still available.... >> - Topics/Tracks/Sessions >> - Round Table >> >> (866)230-6968 >> Int'l Access/Caller Paid Dial In Number: (865)544-7856 >> ACCESS CODE: 2192132 >> >> Talk to y'all at 3p PT >> > > Who should dial in? > > J?rg > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 11 15:01:04 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Wed, 12 Sep 2007 00:01:04 +0200 Subject: [opensolaris-summit] REMINDER: Planning meeting today 3p PT In-Reply-To: <46E6E04E.8000308@sun.com> References: <46E6E04E.8000308@sun.com> Message-ID: <46e71020.67gOQ3+SXZvtRCDU%Joerg.Schilling@fokus.fraunhofer.de> Sara Dornsife wrote: > Agenda: > - Do we have everyone who needs to be there? > - Sponsorships are still available.... > - Topics/Tracks/Sessions > - Round Table > > (866)230-6968 > Int'l Access/Caller Paid Dial In Number: (865)544-7856 ^^^^^^^^^^^^ This number does not seem to work. I get a repeated ring tone..... J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Jesse.Silver at Sun.COM Tue Sep 11 15:07:07 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Tue, 11 Sep 2007 15:07:07 -0700 Subject: [opensolaris-summit] REMINDER: Planning meeting today 3p PT In-Reply-To: <46e71020.67gOQ3+SXZvtRCDU%Joerg.Schilling@fokus.fraunhofer.de> References: <46E6E04E.8000308@sun.com> <46e71020.67gOQ3+SXZvtRCDU%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46E7118B.9060307@sun.com> Don't know. We're on the phone. Try Skype? Joerg Schilling wrote: > Sara Dornsife wrote: > > >> Agenda: >> - Do we have everyone who needs to be there? >> - Sponsorships are still available.... >> - Topics/Tracks/Sessions >> - Round Table >> >> (866)230-6968 >> Int'l Access/Caller Paid Dial In Number: (865)544-7856 >> > > ^^^^^^^^^^^^ > This number does not seem to > work. > > I get a repeated ring tone..... > > J?rg > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From brandorr at opensolaris.org Thu Sep 13 08:28:49 2007 From: brandorr at opensolaris.org (Brandorr) Date: Thu, 13 Sep 2007 11:28:49 -0400 Subject: [opensolaris-summit] It seems many of us are on Twitter already. Message-ID: <5b5090780709130828r5a56b2c4t6573e721863c6830@mail.gmail.com> I thought it might be a useful method of keeping in touch while we are at the summit, so I added a column to the table here: http://www.genunix.org/wiki/index.php/OpenSolaris_Summit Please feel free to add your twitter ID. (Mine is htttp://twitter.com/HelixOne) -- - Brian Gupta P.S. - Any minutes from yesterday's meeting? http://opensolaris.org/os/project/nycosug/ From Sara.Dornsife at Sun.COM Thu Sep 13 15:53:18 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Thu, 13 Sep 2007 17:53:18 -0500 Subject: [opensolaris-summit] Notes from 9/11 OpenSolaris Developer Summit planning meeting Message-ID: <46E9BF5E.8000401@sun.com> We held our first OpenSolaris Developer Summit meetings. The first one at 8am PT and the send at 3p PT on Tuesday 9/11/07. The first call was poorly attended having just me, Jesse Silver, and Ian Murdock in attendance (plus a mysterious and cheeky silent guest who never identified themselves). The second call had a few more people.... Eric Boutilier Jesse Silver Dave Stewart Shawn Walker Ian Murdock Joerg Schilling Chris Baker Glynn Foster Sara Dornsife The purpose of our first call was to try to nail down the agenda and make sure that we had the right people in attendance. We created two lists. One of the people who should be there, but hadn't signed up. And the second of topics, based on what had already been submitted. Most of the people needed work for Sun and we split up the efforts to contact them directly and ask if they attend. Please let me know what I missed or got wrong. Next week we will focus on narrowing and slotting these into the schedule. And we will make sure that we have leaders for each session. Sara Topics: - Installer - Snap upgrade - Core - Distribution Consctructor - Packaging System - build process - repositories (how/where) - patching and upgrades - ZFS to the Max (based on long email conversation on the list) - Laptop Support and Drivers - Intel Drivers - Desktop User Experience - Intel on the desktop - GUI - Governance of user community - Ways to collaborate - Decision-making processees - binary compatibity - contributions by project - guidelines - Web site arch for user communities - Test suites and farm - Sun Processees - ARC - PSARC - PMT - PAC - Release engineering - Mercurial - source code evolution over time - Redistribution - compiling for OpenSolaris on OpenSolaris - Emancipation - Virtualization - Naming (I just added this one) From Glynn.Foster at Sun.COM Thu Sep 13 19:49:14 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Fri, 14 Sep 2007 14:49:14 +1200 Subject: [opensolaris-summit] Figuring out the schedule Message-ID: <46E9F6AA.1010506@sun.com> Hey, As Sara posted earlier, a few of us met to start figuring out some of the schedule for the summit in October, identifying various high level themes for the 2 days. One of the things proposed was the thought of doing a set of 10 minute lightning (ok, not so lightning) talks to set the scene for the day, and then break off into a number of topics that would be more interactive. Sara, Jesse and I took a stab at coming up with this, so we could start some discussion around it - http://www.opensolaris.org/os/project/indiana/summit/ which has assigned 2 blocks of lightning talks - Image Package System - Stephen Hahn New Solaris Installer - David Miner Laptop and Desktop Experience - Dave Stewart ZFS - Eric Kustarz/Tim Foster Sun Developer Processes - John Plocher Virtualization - Joost Pronk Governance - Steve Lau Naming - Sara Dornsife OpenSolaris Test Farm - Jim Walker Website - Derek Cicero Emancipation - Bonnie Corwin Modernization - David Comay Before anyone dives off at the deep-end about this, I'll re-emphasize this is only an initial collection of thoughts, and none of the speakers or topics should consider themselves confirmed. What do you think about the general idea? We're still very much accepting topic ideas on the wiki - http://www.genunix.org/wiki/index.php/OpenSolaris_Summit and you can consider that everything suggested thus far will have time during the 2 days. What do people think? Any suggestions for how better to manage this? Glynn From Sara.Dornsife at Sun.COM Fri Sep 14 07:38:26 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Fri, 14 Sep 2007 09:38:26 -0500 Subject: [opensolaris-summit] It seems many of us are on Twitter already. In-Reply-To: <5b5090780709130828r5a56b2c4t6573e721863c6830@mail.gmail.com> References: <5b5090780709130828r5a56b2c4t6573e721863c6830@mail.gmail.com> Message-ID: <46EA9CE2.20906@sun.com> Glynn went ahead and added the planning meeting details and a link to the minutes to the Summit page - http://www.opensolaris.org/os/project/indiana/summit/ Brandorr wrote: > I thought it might be a useful method of keeping in touch while we are > at the summit, so I added a column to the table here: > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit > > Please feel free to add your twitter ID. (Mine is htttp://twitter.com/HelixOne) > > From Stefan.Teleman at Sun.COM Fri Sep 14 20:13:38 2007 From: Stefan.Teleman at Sun.COM (Stefan Teleman) Date: Fri, 14 Sep 2007 23:13:38 -0400 Subject: [opensolaris-summit] OS Summit topic proposal Message-ID: <46EB4DE2.8080808@Sun.COM> Hi. I would like to propose a topic for the OS Summit in October: C++ Standard Compliant: Sun Studio 12, BOOST and the Apache/RogueWave Standard C++ Library This will be a presentation and discussion about Sun Studio 12, C++ Standards Compliance [http://www.open-std.org/jtc1/sc22/wg21/], BOOST [http://www.boost.org/] and the Apache/RogueWave Standard C++ Library [http://incubator.apache.org/stdcxx/]. --Stefan -- Stefan Teleman Sun Microsystems, Inc. Stefan.Teleman at Sun.COM From Garrett.Damore at Sun.COM Fri Sep 14 22:03:43 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 14 Sep 2007 22:03:43 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46EB4DE2.8080808@Sun.COM> References: <46EB4DE2.8080808@Sun.COM> Message-ID: <46EB67AF.2040806@sun.com> Stefan Teleman wrote: > Hi. > > I would like to propose a topic for the OS Summit in October: > > C++ Standard Compliant: Sun Studio 12, BOOST and the Apache/RogueWave Standard > C++ Library > > This will be a presentation and discussion about Sun Studio 12, C++ Standards > Compliance [http://www.open-std.org/jtc1/sc22/wg21/], BOOST > [http://www.boost.org/] and the Apache/RogueWave Standard C++ Library > [http://incubator.apache.org/stdcxx/]. > > > --Stefan > > Btw, if other folks have any interest, I'd be happy to chat about device drivers and OpenSolaris. My personal area of expertise is mainly networking (and lately GLDv3), but I have expertise elsewhere with drivers as well. I don't mind doing a GLDv3 tutorial (either porting from GLDv2 or from pure DLPI) if there is interest.... although I sort of suspect there isn't enough interest. (I can also talk sensibly about power management, dynamic reconfiguration, etc.) I'm not *asking* for a topic, or *asking* to give a talk, but merely suggesting, that if there is sufficient demand from developers or developer-would-bes, then I'm happy to help. I believe that developer adoption is critical for OpenSolaris' long term viability. One of the areas that OpenSolaris as a group needs to address, IMO, is how we handle supporting various bits of hardware, and 3rd party drivers, even ones that Sun may not necessarily have an interest in. (For example, how does Sun deal with device drivers that are "competitive" to their own business... such as 10GbE adapters?) So far its been handled pretty well, but the issues of device qualification, and sustaining support are of significant concern. Another issue that I'd like to see dealt with is defect/CR tracking. Right now, it is impossible to get a bugster category, even as a Sun employee, for a product that does not have a Sun manager. (For example, I was unable to get a category for issues relating to Tadpole platform support... because no manager at Sun owns this, apparently. Now, my current manager has agreed to sign up as the owner for such a category, but that doesn't scale well... longer term we need a solution which accommodates 3rd party hardware better.) -- Garrett From sch at sun.com Sat Sep 15 00:05:03 2007 From: sch at sun.com (Stephen Hahn) Date: Sat, 15 Sep 2007 00:05:03 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46EB67AF.2040806@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> Message-ID: <20070915070502.GB18777@eng.sun.com> * Garrett D'Amore [2007-09-14 22:09]: > Another issue that I'd like to see dealt with is defect/CR tracking. > Right now, it is impossible to get a bugster category, even as a Sun > employee, for a product that does not have a Sun manager. (For example, > I was unable to get a category for issues relating to Tadpole platform > support... because no manager at Sun owns this, apparently. Now, my > current manager has agreed to sign up as the owner for such a category, > but that doesn't scale well... longer term we need a solution which > accommodates 3rd party hardware better.) Yes, the absence of an opensolaris DTS is a significant hindrance (hitting on other aspects as well as this one). We can talk about a proper plan at the summit, but basically we need some help eliminating candidates according to http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt (I think E2 might eliminate almost every choice but Bugzilla 3.x, if anyone's motivated to read some docs and run a sample instance of a few candidates.) - Stephen -- sch at sun.com http://blogs.sun.com/sch/ From dclarke at blastwave.org Sat Sep 15 09:24:05 2007 From: dclarke at blastwave.org (Dennis Clarke) Date: Sat, 15 Sep 2007 12:24:05 -0400 (EDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46EB67AF.2040806@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> Message-ID: <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> > Stefan Teleman wrote: >> Hi. >> >> I would like to propose a topic for the OS Summit in October: >> >> C++ Standard Compliant: Sun Studio 12, BOOST and the Apache/RogueWave >> Standard >> C++ Library >> >> This will be a presentation and discussion about Sun Studio 12, C++ >> Standards >> Compliance [http://www.open-std.org/jtc1/sc22/wg21/], BOOST >> [http://www.boost.org/] and the Apache/RogueWave Standard C++ Library >> [http://incubator.apache.org/stdcxx/]. >> >> >> --Stefan >> >> > Firstly, I want to thank you for your great work on drivers for Solaris and certainly for your recent QFE work : http://gdamore.blogspot.com/2007/08/qfe-gldv3.html The QFE is almost the swiss army knife of ethernet in the Sun hardware world and you have brought this forwards to present day capabilities with GLDv3. Amazing work Garrett. Thank you for your great work. > Btw, if other folks have any interest, I'd be happy to chat about device > drivers and OpenSolaris. My personal area of expertise is mainly > networking (and lately GLDv3), but I have expertise elsewhere with > drivers as well. I don't mind doing a GLDv3 tutorial (either porting > from GLDv2 or from pure DLPI) if there is interest.... although I sort > of suspect there isn't enough interest. (I can also talk sensibly about > power management, dynamic reconfiguration, etc.) There is interest and I wish that you would and could do a total brain dump of everything you know into tons of web based pages of docs for those of us in the mindless masses out here! :-) The more drivers the better. Certainly if they are good ones. > I'm not *asking* for a topic, or *asking* to give a talk, but merely > suggesting, that if there is sufficient demand from developers or > developer-would-bes, then I'm happy to help. I believe that developer > adoption is critical for OpenSolaris' long term viability. I wish that you could do a full day lecture with hands on work. What would that cost? Can we spec out a lecture and then see if we can put together a cost sheet? It would certainly be an easy business case for me. > One of the areas that OpenSolaris as a group needs to address, IMO, is > how we handle supporting various bits of hardware, and 3rd party > drivers, even ones that Sun may not necessarily have an interest in. The OpenSolaris community will have an interest. Someone out there always seems to want a driver for their USB port powered coffee heater. > (For example, how does Sun deal with device drivers that are > "competitive" to their own business... such as 10GbE adapters?) So far > its been handled pretty well, but the issues of device qualification, > and sustaining support are of significant concern. Please explain why that would be *our* problem? (1) Pick a device, any device, (2) craft a driver for it (3) .. release the code. no problem. > Another issue that I'd like to see dealt with is defect/CR tracking. > Right now, it is impossible to get a bugster category, even as a Sun > employee, for a product that does not have a Sun manager. (For example, > I was unable to get a category for issues relating to Tadpole platform > support... because no manager at Sun owns this, apparently. Now, my > current manager has agreed to sign up as the owner for such a category, > but that doesn't scale well... longer term we need a solution which > accommodates 3rd party hardware better.) That is easy to fix also. We simply create a bug tracking site. We can install it into a zone at genunix.org and then go forwards. Dennis From dclarke at blastwave.org Sat Sep 15 09:28:26 2007 From: dclarke at blastwave.org (Dennis Clarke) Date: Sat, 15 Sep 2007 12:28:26 -0400 (EDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <20070915070502.GB18777@eng.sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <20070915070502.GB18777@eng.sun.com> Message-ID: <38187.72.39.216.186.1189873706.squirrel@mail.blastwave.org> > Yes, the absence of an opensolaris DTS is a significant hindrance > (hitting on other aspects as well as this one). We can talk about a > proper plan at the summit, but basically we need some help eliminating > candidates according to > http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt > > (I think E2 might eliminate almost every choice but Bugzilla 3.x, if > anyone's motivated to read some docs and run a sample instance of a > few candidates.) A sample instance can happen without too much debate. Dennis From binarycrusader at gmail.com Sat Sep 15 09:39:53 2007 From: binarycrusader at gmail.com (Shawn Walker) Date: Sat, 15 Sep 2007 11:39:53 -0500 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <20070915070502.GB18777@eng.sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <20070915070502.GB18777@eng.sun.com> Message-ID: On 15/09/2007, Stephen Hahn wrote: > * Garrett D'Amore [2007-09-14 22:09]: > > Another issue that I'd like to see dealt with is defect/CR tracking. > > Right now, it is impossible to get a bugster category, even as a Sun > > employee, for a product that does not have a Sun manager. (For example, > > I was unable to get a category for issues relating to Tadpole platform > > support... because no manager at Sun owns this, apparently. Now, my > > current manager has agreed to sign up as the owner for such a category, > > but that doesn't scale well... longer term we need a solution which > > accommodates 3rd party hardware better.) > > Yes, the absence of an opensolaris DTS is a significant hindrance > (hitting on other aspects as well as this one). We can talk about a > proper plan at the summit, but basically we need some help eliminating > candidates according to > > http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt > > (I think E2 might eliminate almost every choice but Bugzilla 3.x, if > anyone's motivated to read some docs and run a sample instance of a > few candidates.) In case others were wondering, E2 would appear to be this: E2. Selective differentiated access A mechanism must exist to indicate that a defect is private to a set of Participants, and no part of it may be made accessible to Participants not in that set, even if the defect's subcategory would otherwise cause it to be accessible. (DTS Administrators will have access to all defects.) This requirement is to allow for security coordination, and similar efforts. (Requests for differentiated access are expected to be managed by the Tools Group, the Board, or a designee. Distributions are expected to keep customer confidential data external to the Community DTS.) In which case, I agree with Stephen. However, it seems to me that there are some specific requirements in other points that are going to require some customisation to any solution we choose. BugZilla definitely seems the closest of all the candidates I've seen so far. Especially, since BugZilla 3.x appears to have a very good focus on having excellent security. The main piece of functionality that seems to be missing from BugZilla 3.x, at the moment, is the ability to have confidential users. In other words, users that will be "masked" to people outside of a certain group and that will only be available for selection, assignment, etc. to people in that certain group. However, I may have missed this functionality inadvertently. It looks as though the requested evaluation is already happening on the tools-discuss list. Notably, a poster there named "timeless" has been rather tireless in posting evaluations. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From binarycrusader at gmail.com Sat Sep 15 09:40:28 2007 From: binarycrusader at gmail.com (Shawn Walker) Date: Sat, 15 Sep 2007 11:40:28 -0500 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <38187.72.39.216.186.1189873706.squirrel@mail.blastwave.org> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <20070915070502.GB18777@eng.sun.com> <38187.72.39.216.186.1189873706.squirrel@mail.blastwave.org> Message-ID: On 15/09/2007, Dennis Clarke wrote: > > > Yes, the absence of an opensolaris DTS is a significant hindrance > > (hitting on other aspects as well as this one). We can talk about a > > proper plan at the summit, but basically we need some help eliminating > > candidates according to > > > http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt > > > > (I think E2 might eliminate almost every choice but Bugzilla 3.x, if > > anyone's motivated to read some docs and run a sample instance of a > > few candidates.) > > A sample instance can happen without too much debate. I believe that could be quite helpful. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From sch at sun.com Sat Sep 15 10:31:19 2007 From: sch at sun.com (Stephen Hahn) Date: Sat, 15 Sep 2007 10:31:19 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <38187.72.39.216.186.1189873706.squirrel@mail.blastwave.org> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <20070915070502.GB18777@eng.sun.com> <38187.72.39.216.186.1189873706.squirrel@mail.blastwave.org> Message-ID: <20070915173118.GA19122@eng.sun.com> * Dennis Clarke [2007-09-15 09:28]: > > > Yes, the absence of an opensolaris DTS is a significant hindrance > > (hitting on other aspects as well as this one). We can talk about a > > proper plan at the summit, but basically we need some help eliminating > > candidates according to > > > http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt > > > > (I think E2 might eliminate almost every choice but Bugzilla 3.x, if > > anyone's motivated to read some docs and run a sample instance of a > > few candidates.) > > A sample instance can happen without too much debate. Steve's been running a 2.x instance at grommit.com for some time. But the idea is to evaluate against the requirements, not spawn many samples. Come on over to tools-discuss, pick a candidate, work through the list with us. - Stephen -- sch at sun.com http://blogs.sun.com/sch/ From dclarke at blastwave.org Sat Sep 15 10:52:11 2007 From: dclarke at blastwave.org (Dennis Clarke) Date: Sat, 15 Sep 2007 13:52:11 -0400 (EDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <20070915173118.GA19122@eng.sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <20070915070502.GB18777@eng.sun.com> <38187.72.39.216.186.1189873706.squirrel@mail.blastwave.org> <20070915173118.GA19122@eng.sun.com> Message-ID: <51360.72.39.216.186.1189878731.squirrel@mail.blastwave.org> > * Dennis Clarke [2007-09-15 09:28]: >> >> > Yes, the absence of an opensolaris DTS is a significant hindrance >> > (hitting on other aspects as well as this one). We can talk about a >> > proper plan at the summit, but basically we need some help eliminating >> > candidates according to >> >> > http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt >> > >> > (I think E2 might eliminate almost every choice but Bugzilla 3.x, if >> > anyone's motivated to read some docs and run a sample instance of a >> > few candidates.) >> >> A sample instance can happen without too much debate. > > Steve's been running a 2.x instance at grommit.com for some time. But > the idea is to evaluate against the requirements, not spawn many > samples. > > Come on over to tools-discuss, pick a candidate, work through the list > with us. > I was looking in the tools-discuss maillist, coffee cup in hand and I slowed down considerably. Then, with enthusiasm in check, I thought that I had better coordinate as opposed to running around willy nilly. Not that I run willy nilly but .. you know what I mean. Dennis From benr at cuddletech.com Sat Sep 15 18:50:40 2007 From: benr at cuddletech.com (Ben Rockwood) Date: Sat, 15 Sep 2007 18:50:40 -0700 Subject: [opensolaris-summit] Event Format Message-ID: <46EC8BF0.9050906@cuddletech.com> What will the format of the summit be? One big room? Or will we have breakout sessions in various rooms? Further, has anyone thought about making preparations to record the event for distribution for those who can not attend? If we go the "big room" route this is pretty easy to accommodate. benr. From Garrett.Damore at Sun.COM Sat Sep 15 23:50:57 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Sat, 15 Sep 2007 23:50:57 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> Message-ID: <46ECD251.1090409@sun.com> Dennis Clarke wrote: >> Stefan Teleman wrote: >> >>> Hi. >>> >>> I would like to propose a topic for the OS Summit in October: >>> >>> C++ Standard Compliant: Sun Studio 12, BOOST and the Apache/RogueWave >>> Standard >>> C++ Library >>> >>> This will be a presentation and discussion about Sun Studio 12, C++ >>> Standards >>> Compliance [http://www.open-std.org/jtc1/sc22/wg21/], BOOST >>> [http://www.boost.org/] and the Apache/RogueWave Standard C++ Library >>> [http://incubator.apache.org/stdcxx/]. >>> >>> >>> --Stefan >>> >>> >>> > > Firstly, I want to thank you for your great work on drivers for Solaris > and certainly for your recent QFE work : > > http://gdamore.blogspot.com/2007/08/qfe-gldv3.html > > The QFE is almost the swiss army knife of ethernet in the Sun hardware > world and you have brought this forwards to present day capabilities > with GLDv3. Amazing work Garrett. > > Thank you for your great work. > > >> Btw, if other folks have any interest, I'd be happy to chat about device >> drivers and OpenSolaris. My personal area of expertise is mainly >> networking (and lately GLDv3), but I have expertise elsewhere with >> drivers as well. I don't mind doing a GLDv3 tutorial (either porting >> from GLDv2 or from pure DLPI) if there is interest.... although I sort >> of suspect there isn't enough interest. (I can also talk sensibly about >> power management, dynamic reconfiguration, etc.) >> > > There is interest and I wish that you would and could do a total brain > dump of everything you know into tons of web based pages of docs for those > of us in the mindless masses out here! :-) > Hmm... a lot of what I know is already documented in various places, actually. But putting it all together is a bit of the trick. (There are a few things that aren't documented, though....) > The more drivers the better. Certainly if they are good ones. > > >> I'm not *asking* for a topic, or *asking* to give a talk, but merely >> suggesting, that if there is sufficient demand from developers or >> developer-would-bes, then I'm happy to help. I believe that developer >> adoption is critical for OpenSolaris' long term viability. >> > > I wish that you could do a full day lecture with hands on work. > > What would that cost? Can we spec out a lecture and then see if we can put > together a cost sheet? It would certainly be an easy business case for > me. > > As far as I'm concerned, I think I can trivially justify Sun putting this effort in (i.e. paying for my time) to offer such a lecture/lab session either at the Summit, or perhaps in a different venue. Again, my biggest question/concern is, how many people will be in attendance that the gory details of driver/kernel programming will appeal to? I know you've voted, so that's one... are there more? (What is the threshold where it becomes worthwhile? Especially given the limited time available in the Summit, and competition for other topics.) One thought, is that I could ask for a follow on day on Monday, for those interested.... and maybe it would be approved by Sun. :-) That would avoid scheduling conflicts with other Summit-related issues.) The other concern is whether I am even remotely qualified to teach/lecture. There's an adage that those who can, "do", and those who can't "teach". While the adage may or may not be true, it is definitely not necessarily true that I will be any good as a teacher just because I am competent actually doing the work. But maybe other Sun folks might want to help out here? I'm far from the only talented driver engineer within Sun! (And farther still when the broader OpenSolaris community is taken into account.) I could probably even come up with some lecture notes and lab material, but before I spend any time, I'd want to ensure there was demand. Note that the upcoming Intel Developer Forum (IDF) at Moscone next week includes a lecture/lab titled "Writing Solaris Device Drivers" or somesuch... so that may be very much worth attending (though GLDv3 topics are likely to be little covered.) >> One of the areas that OpenSolaris as a group needs to address, IMO, is >> how we handle supporting various bits of hardware, and 3rd party >> drivers, even ones that Sun may not necessarily have an interest in. >> > > The OpenSolaris community will have an interest. Someone out there always > seems to want a driver for their USB port powered coffee heater. > Yes, but that doesn't man that there are enough people interested in *writing* such a driver. > >> (For example, how does Sun deal with device drivers that are >> "competitive" to their own business... such as 10GbE adapters?) So far >> its been handled pretty well, but the issues of device qualification, >> and sustaining support are of significant concern. >> > > Please explain why that would be *our* problem? > > (1) Pick a device, any device, > (2) craft a driver for it > (3) .. release the code. > > no problem. > The problem is that the SCM (and a few other bits) are still controlled at Sun. Unfortunately, we have not managed to divorce OpenSolaris from Sun internal business practices yet. (See, for example, C-Team review, and realize that all members of C-Team are Sun employees, and may even have been directed to consider Sun's business interests *first*.) > >> Another issue that I'd like to see dealt with is defect/CR tracking. >> Right now, it is impossible to get a bugster category, even as a Sun >> employee, for a product that does not have a Sun manager. (For example, >> I was unable to get a category for issues relating to Tadpole platform >> support... because no manager at Sun owns this, apparently. Now, my >> current manager has agreed to sign up as the owner for such a category, >> but that doesn't scale well... longer term we need a solution which >> accommodates 3rd party hardware better.) >> > > That is easy to fix also. We simply create a bug tracking site. We can > install it into a zone at genunix.org and then go forwards. > I think this was covered elsewhere. I hate that Bugzilla looks like the only viable competitor here. But I love that Bugzilla exists and can be a viable competitor. (I.e. I'm grateful that a defect tracking system exists.) I hope someone will start posting about a sandbox... I'd not mind giving it a whirl, or even using it for some of the bugtracking that I'm not able to do properly at Sun right now because the software doesn't have an allocated Sun responsible manager (such as the CRs related to Tadpole hardware.) -- Garrett > Dennis > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From Tim.Foster at Sun.COM Sun Sep 16 06:24:05 2007 From: Tim.Foster at Sun.COM (Tim Foster) Date: Sun, 16 Sep 2007 14:24:05 +0100 Subject: [opensolaris-summit] Event Format In-Reply-To: <46EC8BF0.9050906@cuddletech.com> References: <46EC8BF0.9050906@cuddletech.com> Message-ID: <46ED2E75.5010606@sun.com> hey Ben, Ben Rockwood wrote: > What will the format of the summit be? One big room? Or will we have > breakout sessions in various rooms? Looks like there's 4 rooms, according to the summit web page, with selected stuff being held in one big room - no idea of the other details though. > Further, has anyone thought about making preparations to record the > event for distribution for those who can not attend? If we go the "big > room" route this is pretty easy to accommodate. I'm hoping to record whatever sessions I attend - though I might need to borrow someone's laptop to save the mp3s off to another usb mass-storage device (my recorder can only do about 4hrs at a time to it's builtin flash, and I've no laptop at the moment) 96kps mp3 should be okay, right? With a few more volunteers with recorders, we might have it covered - anyone else? I was planning to add edited highlights to the ie-osug podcast feed anyway, and definitely see the value of recording as much as we can. cheers, tim -- Tim Foster, Sun Microsystems Inc, Solaris Engineering Ops http://blogs.sun.com/timf From John.Plocher at Sun.COM Sun Sep 16 14:11:41 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Sun, 16 Sep 2007 14:11:41 -0700 Subject: [opensolaris-summit] Event Format In-Reply-To: <46ED2E75.5010606@sun.com> References: <46EC8BF0.9050906@cuddletech.com> <46ED2E75.5010606@sun.com> Message-ID: <46ED9C0D.2040509@Sun.Com> Tim Foster wrote: > 96kps mp3 should > be okay, right? With a simpleminded compression/normalization (audio, not bits), 12K to 16K works fine for spoken audio, and takes up much less room. -John From benr at cuddletech.com Sun Sep 16 15:04:39 2007 From: benr at cuddletech.com (Ben Rockwood) Date: Sun, 16 Sep 2007 15:04:39 -0700 Subject: [opensolaris-summit] Event Format In-Reply-To: <46ED9C0D.2040509@Sun.Com> References: <46EC8BF0.9050906@cuddletech.com> <46ED2E75.5010606@sun.com> <46ED9C0D.2040509@Sun.Com> Message-ID: <46EDA877.6010200@cuddletech.com> John Plocher wrote: > Tim Foster wrote: >> 96kps mp3 should >> be okay, right? > > > With a simpleminded compression/normalization (audio, not bits), > 12K to 16K works fine for spoken audio, and takes up much less room. All hail SPEEX. ;) Given the length of most of the presentations hand-held audio recorders are probly the best choice. I've got one that does up to 16 hours of audio and can be used like a coordless mic to move around during questions, etc. They are pretty cheap too. Does anyone know what the AV setup will be? If there is an audio system with mics that will be in use hopefully we can just plug into the mixer direct. benr. From Glynn.Foster at Sun.COM Sun Sep 16 16:47:27 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Mon, 17 Sep 2007 11:47:27 +1200 Subject: [opensolaris-summit] Event Format In-Reply-To: <46ED2E75.5010606@sun.com> References: <46EC8BF0.9050906@cuddletech.com> <46ED2E75.5010606@sun.com> Message-ID: <46EDC08F.8030107@sun.com> Hey, Tim Foster wrote: > hey Ben, > > Ben Rockwood wrote: >> What will the format of the summit be? One big room? Or will we have >> breakout sessions in various rooms? > > Looks like there's 4 rooms, according to the summit web page, with > selected stuff being held in one big room - no idea of the other > details though. We have I think up to 5 rooms, and we're pretty flexible with the schedule. The big issue is trying to figure out how best to use our time. With relatively few people, it doesn't make sense to split up into too many parallel tracks and there's likely to be knowledgeable people missing out on important conversations. If you have suggestions for how best to schedule this, let us know. I guess the trick is to someone schedule discrete topics that don't have much crossover at any given time. >> Further, has anyone thought about making preparations to record the >> event for distribution for those who can not attend? If we go the "big >> room" route this is pretty easy to accommodate. > > I'm hoping to record whatever sessions I attend - though I might need > to borrow someone's laptop to save the mp3s off to another usb > mass-storage device (my recorder can only do about 4hrs at a time to > it's builtin flash, and I've no laptop at the moment) 96kps mp3 should > be okay, right? > > With a few more volunteers with recorders, we might have it > covered - anyone else? Sure, I'll bring a videocam, or buy an MP3/OGG recorder on the way over. > I was planning to add edited highlights to the ie-osug podcast feed > anyway, and definitely see the value of recording as much as we can. Absolutely. I'm very much hoping that people within each of the sessions will nominate someone to take notes and do a write up of the results of the discussion. Glynn From Jesse.Silver at Sun.COM Mon Sep 17 17:50:08 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Mon, 17 Sep 2007 17:50:08 -0700 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST Message-ID: <46EF20C0.9040002@sun.com> Agenda: - Session topic priorities - Attendance - On-site server and workstation availability - Round Table (866)230-6968 Int'l Access/Caller Paid Dial In Number: (865)544-7856 ACCESS CODE: 2192132 Talk to you at 8am and 3pm PST. -Jesse From brandorr at opensolaris.org Mon Sep 17 18:13:33 2007 From: brandorr at opensolaris.org (Brandorr) Date: Mon, 17 Sep 2007 21:13:33 -0400 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST In-Reply-To: <46EF20C0.9040002@sun.com> References: <46EF20C0.9040002@sun.com> Message-ID: <5b5090780709171813g5f4968fen8e16352a8abe676@mail.gmail.com> I can't make the first one as it's during work. Should I call into the evening call? On 9/17/07, Jesse Silver wrote: > Agenda: > - Session topic priorities > - Attendance > - On-site server and workstation availability > - Round Table > > (866)230-6968 > Int'l Access/Caller Paid Dial In Number: (865)544-7856 > ACCESS CODE: 2192132 > > Talk to you at 8am and 3pm PST. > > -Jesse > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Jesse.Silver at Sun.COM Mon Sep 17 18:16:19 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Mon, 17 Sep 2007 18:16:19 -0700 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST In-Reply-To: <5b5090780709171813g5f4968fen8e16352a8abe676@mail.gmail.com> References: <46EF20C0.9040002@sun.com> <5b5090780709171813g5f4968fen8e16352a8abe676@mail.gmail.com> Message-ID: <46EF26E3.2040304@sun.com> Yes please. I did not mean to imply that anyone should attend both meetings, though they are welcome to. I'll fill in the 3pm meeting on the events of the 8am meeting. Then the meeting notes will be sent to the summit list. Brandorr wrote: > I can't make the first one as it's during work. Should I call into the > evening call? > > On 9/17/07, Jesse Silver wrote: > >> Agenda: >> - Session topic priorities >> - Attendance >> - On-site server and workstation availability >> - Round Table >> >> (866)230-6968 >> Int'l Access/Caller Paid Dial In Number: (865)544-7856 >> ACCESS CODE: 2192132 >> >> Talk to you at 8am and 3pm PST. >> >> -Jesse >> >> >> _______________________________________________ >> opensolaris-summit mailing list >> opensolaris-summit at opensolaris.org >> http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit >> >> > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From binarycrusader at gmail.com Mon Sep 17 18:43:59 2007 From: binarycrusader at gmail.com (Shawn Walker) Date: Mon, 17 Sep 2007 20:43:59 -0500 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST In-Reply-To: <46EF20C0.9040002@sun.com> References: <46EF20C0.9040002@sun.com> Message-ID: On 17/09/2007, Jesse Silver wrote: > Agenda: > - Session topic priorities > - Attendance > - On-site server and workstation availability > - Round Table > > (866)230-6968 > Int'l Access/Caller Paid Dial In Number: (865)544-7856 > ACCESS CODE: 2192132 > > Talk to you at 8am and 3pm PST. I'll be in the 2nd call. Thanks, -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Garrett.Damore at Sun.COM Mon Sep 17 19:03:00 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Mon, 17 Sep 2007 19:03:00 -0700 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST In-Reply-To: References: <46EF20C0.9040002@sun.com> Message-ID: <46EF31D4.9090406@sun.com> Shawn Walker wrote: > On 17/09/2007, Jesse Silver wrote: > >> Agenda: >> - Session topic priorities >> - Attendance >> - On-site server and workstation availability >> - Round Table >> >> (866)230-6968 >> Int'l Access/Caller Paid Dial In Number: (865)544-7856 >> ACCESS CODE: 2192132 >> >> Talk to you at 8am and 3pm PST. >> > > I'll be in the 2nd call. > > Thanks > This week is terrible for me, I can make neither. But I'll follow up in e-mail. In particular, I do want to know whether I need to do anything as far as presenting... if dclarke is the only one interested then I won't bother, but if there is demand, then I'm happy to do a public "brain dump". :-) -- Garrett From Jesse.Silver at Sun.COM Tue Sep 18 08:06:18 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Tue, 18 Sep 2007 08:06:18 -0700 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST Message-ID: <0JOK00805KM4BJ70@fe-sfbay-10.sun.com> Sorry to anyone trying to call into the 8am meeting, I will be delayed by 30 minutes from opening up the call. Again sorry, and thanks to those who stick around. -Jesse -----Original Message----- From: "Jesse Silver" To: opensolaris-summit at opensolaris.org Sent: 9/17/07 5:50 PM Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST Agenda: - Session topic priorities - Attendance - On-site server and workstation availability - Round Table (866)230-6968 Int'l Access/Caller Paid Dial In Number: (865)544-7856 ACCESS CODE: 2192132 Talk to you at 8am and 3pm PST. -Jesse _______________________________________________ opensolaris-summit mailing list opensolaris-summit at opensolaris.org http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit From Stuart.Kreitman at Sun.COM Tue Sep 18 08:07:54 2007 From: Stuart.Kreitman at Sun.COM (Stuart Kreitman) Date: Tue, 18 Sep 2007 08:07:54 -0700 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST In-Reply-To: <46EF20C0.9040002@sun.com> References: <46EF20C0.9040002@sun.com> Message-ID: <46EFE9CA.8050606@sun.com> I'm on hold for this call. Is anyone else trying to get in? Stuart Jesse Silver wrote: > Agenda: > - Session topic priorities > - Attendance > - On-site server and workstation availability > - Round Table > > (866)230-6968 > Int'l Access/Caller Paid Dial In Number: (865)544-7856 > ACCESS CODE: 2192132 > > Talk to you at 8am and 3pm PST. > > -Jesse > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From Stuart.Kreitman at Sun.COM Tue Sep 18 08:10:25 2007 From: Stuart.Kreitman at Sun.COM (Stuart Kreitman) Date: Tue, 18 Sep 2007 08:10:25 -0700 Subject: [opensolaris-summit] Event Format In-Reply-To: <46EDC08F.8030107@sun.com> References: <46EC8BF0.9050906@cuddletech.com> <46ED2E75.5010606@sun.com> <46EDC08F.8030107@sun.com> Message-ID: <46EFEA61.2080904@sun.com> Glynn Foster wrote: > Hey, > Sure, I'll bring a videocam, or buy an MP3/OGG recorder on the way over. > > >> I was planning to add edited highlights to the ie-osug podcast feed >> anyway, and definitely see the value of recording as much as we can. >> > > Absolutely. I'm very much hoping that people within each of the sessions will > nominate someone to take notes and do a write up of the results of the discussion. > > I have a DV videocam and tripod, so count me in. Might I suggest asking the presenters if they are willing to be video'd, that will spur them to plan better-organized presentation. From Jesse.Silver at Sun.COM Tue Sep 18 08:22:13 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Tue, 18 Sep 2007 08:22:13 -0700 Subject: [opensolaris-summit] REMINDER: Planning meetings tomorrow at 8am and 3pm PST In-Reply-To: <46EFE9CA.8050606@sun.com> References: <46EF20C0.9040002@sun.com> <46EFE9CA.8050606@sun.com> Message-ID: <46EFED25.4050802@sun.com> Conference is now open. Stuart Kreitman wrote: > I'm on hold for this call. Is anyone else trying to get in? > > Stuart > > Jesse Silver wrote: > >> Agenda: >> - Session topic priorities >> - Attendance >> - On-site server and workstation availability >> - Round Table >> >> (866)230-6968 >> Int'l Access/Caller Paid Dial In Number: (865)544-7856 >> ACCESS CODE: 2192132 >> >> Talk to you at 8am and 3pm PST. >> >> -Jesse >> >> >> _______________________________________________ >> opensolaris-summit mailing list >> opensolaris-summit at opensolaris.org >> http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit >> >> > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Glynn.Foster at Sun.COM Tue Sep 18 14:53:30 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Wed, 19 Sep 2007 09:53:30 +1200 Subject: [opensolaris-summit] Event Format In-Reply-To: <46EFEA61.2080904@sun.com> References: <46EC8BF0.9050906@cuddletech.com> <46ED2E75.5010606@sun.com> <46EDC08F.8030107@sun.com> <46EFEA61.2080904@sun.com> Message-ID: <46F048DA.8010608@sun.com> Hey, Stuart Kreitman wrote: > I have a DV videocam and tripod, so count me in. Might I suggest asking > the presenters if they are willing > to be video'd, that will spur them to plan better-organized presentation. Great, thanks! I don't think we necessarily have to record every session, though certainly it would be good to record bits during the 2 days. Most importantly, having a bunch of people write up the summaries and takeaways from the sessions and post them out on opensolaris.org would be really awesome. Glynn From John.Plocher at Sun.COM Tue Sep 18 14:54:11 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 18 Sep 2007 14:54:11 -0700 Subject: [opensolaris-summit] Event Format In-Reply-To: <46F048DA.8010608@sun.com> References: <46EC8BF0.9050906@cuddletech.com> <46ED2E75.5010606@sun.com> <46EDC08F.8030107@sun.com> <46EFEA61.2080904@sun.com> <46F048DA.8010608@sun.com> Message-ID: <46F04903.5090209@Sun.Com> I'll bring my laptop & the olympus recorder we use for ARC meetings. -John Glynn Foster wrote: > Hey, > > Stuart Kreitman wrote: >> I have a DV videocam and tripod, so count me in. Might I suggest asking >> the presenters if they are willing >> to be video'd, that will spur them to plan better-organized presentation. > > Great, thanks! I don't think we necessarily have to record every session, though > certainly it would be good to record bits during the 2 days. Most importantly, > having a bunch of people write up the summaries and takeaways from the sessions > and post them out on opensolaris.org would be really awesome. > > > Glynn > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit From Jesse.Silver at Sun.COM Tue Sep 18 16:56:18 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Tue, 18 Sep 2007 16:56:18 -0700 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings Message-ID: <46F065A2.6010006@sun.com> Today we held our second round of weekly 8am PST and 3pm PST planning meetings. Thanks to those who participated. Though there are still many unanswered questions, we're making great headway. Hope to see more of you next week. _____________ Despite a late start, our 8am PST meeting turned out to be lively and well attended. Thanks to the following people for participating (please correct the record if I've missed anyone): Darren Reed Stuart Kreitman Tim Foster Dennis Clarke Stefan Teleman John Plocher Jesse Silver Discussion on Format of Sessions: -We are looking for session leaders & secretaries for each topic. -We should set up a mechanism for decisions to be reflected back to the community and voted on -Suggestion: ask session leaders to make proposition cards (deliverables) at the end of each session, then put deliverables up to a community vote Session Topics & Priorities: - Availability, accessibility of up-to-date open source software for OpenSolaris - How to increase relevance of OpenSolaris - How to get young (Linux) developers involved in OpenSolaris community, and developing software on OpenSolaris IT Services Update: - We have a UCSC engineer on-site all day both days. - We have wireless and wired internet access throughout the meeting facilities. - Stewart Kreitman will talk with the UCSC IT Dept. to make sure our bandwidth, security, redundancy and other requirements are met. Media Services Update: - We need people to volunteer to lead the video/audio streaming effort. - There are 5 rooms: 1x 100 people, 2x 59 people, 2x 35 people. Should all these rooms be streamed live? Group is tending toward yes. - We are looking at the possibility of setting up 2 projectors in each room so that an IRC chat window can be projected, providing remote interactivity. Attendance: -Should Solaris/OpenSolaris vendors be invited? ___________________ Glynn Foster Al Hopper Roland Mainz Shawn Walker Jesse Silver Brian Gupta John Plocher Sara Dornsife Most people think we should avoid a nailed down schedule. We will flesh out the topics already proposed on the Wiki, sort them into tracks (process vs. technical or something similar), and create a Wiki page where people can vote on topics/sessions they'd like to participate in. Some new topics: - Driver development tutorial - Roland Mainz doing a ?developing on OpenSolaris? talk, ksh93 - Do we want to talk about naming? Key Proposals: - Maybe we should not preschedule anything but lightning talks. - Make priority list for topics based on how close the topic gets us to the next release of Indiana. - Split tracks by user oriented, political oriented, developer oriented etc... - How about splitting tracks by process/infrastructure (governance, user etc... site) side, and technical (packaging, installing etc...) side. - Maybe 5 minute lightning talks instead of 10? - Set up web based system for voting on topics. - Let's expand the topic suggestions from a few words to a paragraph. Action Items: - Update Wiki, adding fields to expand on the topic suggestions. - Add voting system to Wiki Server availability- Suggestions: - Have as many different kinds of machines available as possible, split between Sparc and x86. - Have at least one central server, stable and powerful, like an M4000 with a bunch of storage. - Get a team together to set up the servers on Friday, coordinate with the UCSC engineering staff to make sure nothing goes wrong, etc... From roland.mainz at nrubsig.org Tue Sep 18 20:26:01 2007 From: roland.mainz at nrubsig.org (Roland Mainz) Date: Wed, 19 Sep 2007 05:26:01 +0200 Subject: [opensolaris-summit] Internet clipboard "opensolaris.pastebin.ca" set up... Message-ID: <46F096C9.33636794@nrubsig.org> Hi! ---- [The users in irc://irc.freenode.org/#opensolaris already know it but AFAIK it was never announced to the general public] I've set-up an OpenSolaris-specific version of the pastebin.ca internet clipboard, e.g. a public place to paste larger texts via browser (or the command-line tool below) which couldn't be transfered easily via other methods (except email) - this is AFAIK quite usefull during IRC conversations, phone meetings and maybe things like the upcoming opensolaris summit (e.g. genereal-purpose scripts/drafts/etc. dumping ground). For those who don't have a browser around to paste things like logs and other stuff to opensolaris.pastebin.ca I wrote a small script (attached as "shnote.ksh.txt" ; note that this script needs Solaris 11/Nevada B72 or higher) which does this without requiring a browser, e.g. install the script with... -- snip -- # cp shnote.ksh.txt /usr/bin/shnote # chmod a+rx,a-w /usr/bin/shnote -- snip -- ...and then use it like this: -- snip -- $ shnote put "demo #8932001" SUCCESS: http://opensolaris.pastebin.ca/702814 -- snip -- ... will write the string "demo #8932001" to http://opensolaris.pastebin.ca and returns the URL where the text can be accessed from a browser. Sending command output works similary, e.g. -- snip -- $ shnote put "$(ls -l)" SUCCESS: http://opensolaris.pastebin.ca/702822 -- snip -- ... will send the output of "ls -l" to http://opensolaris.pastebin.ca and return the access URL. Additionally the script maintains a small history containing the access URL, title (usually the users name) and the date, e.g. -- snip -- $ shnote hist # <date> ... ... http://opensolaris.pastebin.ca/702814 gisburn Sep 19 04:51:21 CET 2007 ... http://opensolaris.pastebin.ca/702822 gisburn Sep 19 05:02:52 CET 2007 -- snip -- Finally you can get strings/data/text/etc. from http://opensolaris.pastebin.ca using... -- snip -- $ shnote get http://opensolaris.pastebin.ca/702814 # Record name is '702814' demo #8932001 -- snip -- ...or just short: -- snip -- $ shnote get 702814 # Record name is '702814' demo #8932001 -- snip -- Suggestions/comments/rants/etc. welcome... :-) ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) -------------- next part -------------- #!/bin/ksh93 # # CDDL HEADER START # # The contents of this file are subject to the terms of the # Common Development and Distribution License (the "License"). # You may not use this file except in compliance with the License. # # You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE # or http://www.opensolaris.org/os/licensing. # See the License for the specific language governing permissions # and limitations under the License. # # When distributing Covered Code, include this CDDL HEADER in each # file and include the License file at usr/src/OPENSOLARIS.LICENSE. # If applicable, add the following below this CDDL HEADER, with the # fields enclosed by brackets "[]" replaced with your own identifying # information: Portions Copyright [yyyy] [name of copyright owner] # # CDDL HEADER END # # # Copyright 2007 Sun Microsystems, Inc. All rights reserved. # Use is subject to license terms. # # ident "%Z%%M% %I% %E% SMI" # # Solaris needs /usr/xpg4/bin/ because the tools in /usr/bin are not POSIX-conformant export PATH=/usr/xpg4/bin:/bin:/usr/bin # Make sure all math stuff runs in the "C" locale to avoid problems with alternative # radix point representations (e.g. ',' instead of '.' in de_DE.*-locales). This # needs to be set _before_ any floating-point constants are defined in this script) if [[ "${LC_ALL}" != "" ]] ; then export \ LC_MONETARY="${LC_ALL}" \ LC_MESSAGES="${LC_ALL}" \ LC_COLLATE="${LC_ALL}" \ LC_CTYPE="${LC_ALL}" unset LC_ALL fi export LC_NUMERIC=C function fatal_error { print -u 2 "${progname}: $@" exit 1 } # fixme: use "stat" builtin instead of "ls" function get_filesize { typeset filename="$1" typeset dummy # dummy string integer filesize=-1 if [[ -f "${filename}" ]] ; then ls -lb "${filename}" | read dummy dummy dummy dummy filesize dummy fi print -- ${filesize} return 0 } function encode_multipart_form_data { #set -o xtrace nameref formdata=$1 nameref content="${1}.content" # ToDO: should be |content="formdata.content"| integer numformelements=${#formdata.form[*]} integer i content="" # todo: add support to upload files for (( i=0 ; i < numformelements ; i++ )) ; do nameref element="${1}.form[${i}]" # ToDo: should be |element="formdata.form[${i}]"| content+="--${formdata.boundary}\n" content+="Content-Disposition: form-data; name=\"${element.name}\"\n" content+="\n" content+="${element.data}\n" # fixme: may need encoding for non-ASCII data done formdata.content_length=${#content} # add content tail (which MUST not be added to the content length) content+="--${formdata.boundary}--\n" return 0 } # parse return code, cookies etc. function parse_http_response { nameref response=$1 typeset h c i IFS=$' \t' read -r h c # read HTTP/1.[01] <code> [[ "$h" != ~(Eil)HTTP/.* ]] && { print -u2 "$0: HTTP/ header missing" ; return 1 ; } response.httpcode="$c" # skip remaining headers while IFS='' read -r i ; do [[ "$i" = $'\r' ]] && break done return 0 } function history_write_record { # rec: history record: # rec.title # rec.description # rec.provider # rec.providertoken # rec.url nameref rec="$1" integer histfd mkdir -p "${HOME}/.shnote" { # write a single-line record which can be read # as a compound variable back into the shell printf "title=%q description=%q date=%q provider=%q providertoken=%q url=%q\n" \ "${rec.title}" \ "${rec.description}" \ "$(date)" \ "${rec.provider}" \ "${rec.providertoken}" \ "${rec.url}" } >>"${history_file}" return $? } function print_history { integer histfd # http stream number typeset line # default output format is: # <access url>/<title> <date> <access url> [[ "$1" = "-l" ]] || printf "# %s\t\t\t\t\t%s\t%s\n" "<url>" "<title>" "<date>" # no history file ? if [[ ! -f "${history_file}" ]] ; then return 0 fi # open history file exec {histfd}<>"${history_file}" while read -u${histfd} line ; do typeset rec=() # This is a bit messy - "eval" should be avoided but there is currently # no better way to read a serialised compound variable from a file back # into the shell. eval "rec=( $line )" if [[ "$1" = "-l" ]] ; then print -- "${rec}" else printf "%s\t%s\t%s\n" "${rec.url}" "${rec.title}" "${rec.date}" fi unset rec done # close history file exec {histfd}<&- } function put_note_pastebin_ca { #set -o xtrace # key to autheticate this script against pastebin.ca typeset -r pastebin_ca_key="9CFXFyeNC3iga/vthok75kTBu5kSSLPD" # site setup typeset url_host="opensolaris.pastebin.ca" typeset url_path="/quiet-paste.php?api=${pastebin_ca_key}" typeset url="http://${url_host}${url_path}" integer netfd # http stream number # argument for "encode_multipart_form_data" typeset mimeform=( # input typeset boundary typeset -a form # output typeset content integer content_length ) typeset request="" typeset content="" typeset -r boundary="--------shnote_${RANDOM}_Xfish_${RANDOM}_Yeats_${RANDOM}_Zchicken_${RANDOM}monster_--------" # ToDo: Use "mimeform.form+=(" once ksh93 has been fixed. mimeform.boundary="${boundary}" mimeform.form[0]=( name="name" data="${LOGNAME}" ) mimeform.form[1]=( name="expiry" data="Never" ) mimeform.form[2]=( name="type" data="1" ) mimeform.form[3]=( name="description" data="logname=${LOGNAME};hostname=$(hostname);date=$(date)" ) mimeform.form[4]=( name="content" data="$1" ) encode_multipart_form_data mimeform content="${mimeform.content}" request="POST ${url_path} HTTP/1.1\n" request+="Host: ${url_host}\n" request+="User-Agent: ${http_user_agent}\n" request+="Connection: close\n" request+="Content-Type: multipart/form-data; boundary=${boundary}\n" request+="Content-Length: $(( ${mimeform.content_length} ))\n" exec {netfd}<>"/dev/tcp/${url_host}/80" (( $? != 0 )) && { print -u2 "Couldn't open connection to ${url_host}." ; return 1 ; } # send http post { print -- "${request}" print -- "${content}" } >&${netfd} # process reply parse_http_response httpresponse <&${netfd} response="$(cat <&${netfd})" # close connection exec {netfd}<&- if [[ "${response}" = ~(E).*SUCCESS.* ]] ; then typeset response_token="${response/~(E).*SUCCESS:/}" print "SUCCESS: http://opensolaris.pastebin.ca/${response_token}" # write history entry typeset histrec=( title="${mimeform.form[0].data}" description="${mimeform.form[3].data}" providertoken="${response_token}" provider="opensolaris.pastebin.ca" url="http://opensolaris.pastebin.ca/${response_token}" ) history_write_record histrec else print "ERROR: ${response}" return 1 fi return 0 } function get_note_pastebin_ca { typeset recordname="$1" integer netfd # http stream number case "${recordname}" in ~(Elr)[0-9][0-9]*) # pass-through ;; ~(Elr)http://opensolaris.pastebin.ca/raw/[0-9]*) recordname="${recordname/~(El)http:\/\/opensolaris.pastebin.ca\/raw\//}" ;; ~(Elr)http://opensolaris.pastebin.ca/[0-9]*) recordname="${recordname/~(El)http:\/\/opensolaris.pastebin.ca\//}" ;; *) fatal_error "Unsupported record name ${recordname}." esac print -u2 -f "# Record name is '%s'\n" "${recordname}" typeset url_host="opensolaris.pastebin.ca" typeset url_path="/raw/${recordname}" typeset url="http://${url_host}${url_path}" # I hereby curse Solaris for not having an entry for "http" in /etc/services # open TCP channel exec {netfd}<>"/dev/tcp/${url_host}/80" (( $? != 0 )) && { print -u2 "Couldn't open connection to ${url_host}." ; return 1 ; } # send HTTP request request="GET ${url_path} HTTP/1.1\n" request+="Host: ${url_host}\n" request+="User-Agent: ${http_user_agent}\n" request+="Connection: close\n" print -u${netfd} -- "${request}\n" # collect response and send it to stdout parse_http_response httpresponse <&${netfd} cat <&${netfd} # close connection exec {netfd}<&- print # add newline return 0 } function usage { OPTIND=0 getopts -a "${progname}" "${USAGE}" OPT '-?' exit 2 } # program start builtin basename builtin cat builtin date builtin uname typeset progname="$(basename "${0}")" # HTTP protocol client identifer typeset -r http_user_agent="shnote/ksh93 (2007-09-18; $(uname -s -r -p)" # name of history log (the number after "history" is some kind of version # counter to handle incompatible changes to the history file format) typeset -r history_file="${HOME}/.shnote/history0.txt" USAGE=$' [-?\n@(#)\$Id: shnote (Roland Mainz) 2007-09-18 \$\n] [-author?Roland Mainz <roland.mainz at nrubsig.org] [+NAME?shnote - read/write text data to internet clipboards] [+DESCRIPTION?\bshnote\b is a small utilty which can read and write text data to internet "clipboards" such as opensolaris.pastebin.ca.] [+?The first arg \bmethod\b describes one of the methods, "put" saves a string to the internet clipboard, returning an identifer and the full URL where the data are stored. The method "get" retrives the raw information using the identifer from the previous "put" action.] [+?The second arg \bstring\b contains either the string data which should be stored on the clipboard using the "put" method, the "get" method uses this information as identifer to retrive the raw data from the clipboard.] method [ string ] [+SEE ALSO?\bksh93\b(1), \brssread\b(1), http://opensolaris.pastebin.ca] ' while getopts -a "${progname}" "${USAGE}" OPT ; do # printmsg "## OPT=|${OPT}|, OPTARG=|${OPTARG}|" case ${OPT} in *) usage ;; esac done shift $((OPTIND-1)) typeset method="$1" shift # todo: "history" mode case "${method}" in put) put_note_pastebin_ca "$@" ; exit $? ;; get) get_note_pastebin_ca "$@" ; exit $? ;; hist) print_history "$@" ; exit $? ;; *) usage ;; esac fatal_error "not reached." # EOF. From brandorr at opensolaris.org Tue Sep 18 21:29:10 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 00:29:10 -0400 Subject: [opensolaris-summit] Everyone who proposed a topic for the forum on the wiki needs to update their topic with an explanation of why the topic is a critical issue.. Message-ID: <5b5090780709182129n1b9919f6icbf30d5fc2d40588@mail.gmail.com> The table has been modified to allow for this info. Please update your topics by the end of this week, as we will be asking everyone to select 5 of the most critical topics and submit them to a poll. (Post to this list if for some reason you can not update your topic by the end of the week). (Only 2 1/2 weeks to go). Cheers, Brian -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Tim.Foster at Sun.COM Wed Sep 19 02:16:34 2007 From: Tim.Foster at Sun.COM (Tim Foster) Date: Wed, 19 Sep 2007 10:16:34 +0100 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F065A2.6010006@sun.com> References: <46F065A2.6010006@sun.com> Message-ID: <1190193394.22489.73.camel@haiiro> On Tue, 2007-09-18 at 16:56 -0700, Jesse Silver wrote: > Server availability- > Suggestions: > - Have as many different kinds of machines available as possible, split > between Sparc and x86. > - Have at least one central server, stable and powerful, like an M4000 > with a bunch of storage. Why servers ? I thought this was a collective "bang our heads together" meeting, rather than an actual hackfest ? If if everyone has their noses buried in laptops for the duration of the weekend, then I'm not sure we'll have achieved our aims imho. [ I'm not bringing a laptop (yeah okay, mostly because I don't have one ;-) ] cheers, tim -- Tim Foster, Sun Microsystems Inc, Solaris Engineering Ops http://blogs.sun.com/timf From swalker at opensolaris.org Wed Sep 19 06:30:55 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Wed, 19 Sep 2007 08:30:55 -0500 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <1190193394.22489.73.camel@haiiro> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> Message-ID: <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> On 19/09/2007, Tim Foster <Tim.Foster at sun.com> wrote: > On Tue, 2007-09-18 at 16:56 -0700, Jesse Silver wrote: > > Server availability- > > Suggestions: > > - Have as many different kinds of machines available as possible, split > > between Sparc and x86. > > - Have at least one central server, stable and powerful, like an M4000 > > with a bunch of storage. > > Why servers ? I thought this was a collective "bang our heads together" > meeting, rather than an actual hackfest ? If if everyone has their noses > buried in laptops for the duration of the weekend, then I'm not sure > we'll have achieved our aims imho. > > [ I'm not bringing a laptop (yeah okay, mostly because I don't have > one ;-) ] There may be a few folks in the later hours of the conference that get very little chance for "bang their heads together on a software problem time." Otherwise, I tend to agree that I hope that this is more about problem solving than hacking... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From John.Plocher at Sun.COM Wed Sep 19 09:00:14 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 19 Sep 2007 09:00:14 -0700 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> Message-ID: <46F1478E.1080103@Sun.Com> Shawn Walker wrote: > I tend to agree that I hope that this is > more about problem solving than hacking... I see room for both - especially if we (Sun) can arrange for a reasonable set of high end systems to play with. While some topics (politics/governance/ARCs/...) won't be impacted much by the compute infrastructure at the conference, I can see other tracks taking a more hands-on dig in and play with code attitude. It would be VERY USEFUL for us all to articulate what exactly we wish to get out of the conference - what problems need to be solved that would benefit from a concentrated, in person focus? (See the Wiki at http://www.genunix.org/wiki/index.php/OpenSolaris_Summit for some suggestions) -John From James.Walker at Sun.COM Wed Sep 19 10:38:18 2007 From: James.Walker at Sun.COM (Jim Walker) Date: Wed, 19 Sep 2007 11:38:18 -0600 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F1478E.1080103@Sun.Com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> Message-ID: <46F15E8A.70407@sun.com> If useful, I could use some of the test farm automation to manage reservation time slots on whatever hardware we have at the summit. In addition, if we have good network access we can get to remote test machines. Cheers, Jim John Plocher wrote: > Shawn Walker wrote: >> I tend to agree that I hope that this is >> more about problem solving than hacking... > > > I see room for both - especially if we (Sun) can arrange for > a reasonable set of high end systems to play with. > > While some topics (politics/governance/ARCs/...) won't be impacted > much by the compute infrastructure at the conference, I can see other > tracks taking a more hands-on dig in and play with code attitude. > > It would be VERY USEFUL for us all to articulate what exactly we wish > to get out of the conference - what problems need to be solved that > would benefit from a concentrated, in person focus? (See the Wiki at > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit for some > suggestions) > > -John > > > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit -- Jim Walker, http://blogs.sun.com/jwalker Sun Microsystems, Software, Solaris QE x77744, 500 Eldorado Blvd, Broomfield CO 80021 From alan.duboff at sun.com Wed Sep 19 10:37:41 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 19 Sep 2007 10:37:41 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46ECD251.1090409@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> Message-ID: <Pine.GSO.4.64.0709191031530.1532@jazz> On Sat, 15 Sep 2007, Garrett D'Amore wrote: > Again, my biggest question/concern is, how many people will be in > attendance that the gory details of driver/kernel programming will > appeal to? I know you've voted, so that's one... are there more? I'm sure there are more, but the problem is that most of the people that are interested, are spread out more globally, as a whole, so targeting specific venues will be hit/miss at best in most cases. We need this information online, with the ability to stream in any location, so that the right people will have easy access to it. That's where the efforts should be focused rather than issolated efforts with less impact. My $0.02 (and where I'm trying to put some of my effort) Max Brunning has a lecture that he offers through SunU, and that is another example of a more issolated effort. It's great for folks that can get to SunU, but doesn't help someone in Europe or other countries, and we have many potential driver writers in those locales. -- Alan DuBoff - Solaris x86 IHV/OEM Group From brandorr at opensolaris.org Wed Sep 19 11:36:00 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 14:36:00 -0400 Subject: [opensolaris-summit] Everyone who proposed a topic for the forum on the wiki needs to update their topic with an explanation of why the topic is a critical issue.. In-Reply-To: <5b5090780709182129n1b9919f6icbf30d5fc2d40588@mail.gmail.com> References: <5b5090780709182129n1b9919f6icbf30d5fc2d40588@mail.gmail.com> Message-ID: <5b5090780709191136l39651cf3uc375eb2c11022fbe@mail.gmail.com> http://www.genunix.org/wiki/index.php/OpenSolaris_Summit I forgot to include the link, but I see many of you found your way there anyhow. :) On 9/19/07, Brandorr <brandorr at opensolaris.org> wrote: > The table has been modified to allow for this info. > > Please update your topics by the end of this week, as we will be > asking everyone to select 5 of the most critical topics and submit > them to a poll. (Post to this list if for some reason you can not > update your topic by the end of the week). > > (Only 2 1/2 weeks to go). > > Cheers, > Brian > > -- > - Brian Gupta > > http://opensolaris.org/os/project/nycosug/ > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Darren.Reed at Sun.COM Wed Sep 19 11:46:43 2007 From: Darren.Reed at Sun.COM (Darren.Reed at Sun.COM) Date: Wed, 19 Sep 2007 11:46:43 -0700 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F1478E.1080103@Sun.Com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> Message-ID: <46F16E93.7060801@Sun.COM> John Plocher wrote: >Shawn Walker wrote: > > >>I tend to agree that I hope that this is >>more about problem solving than hacking... >> >> > > >I see room for both - especially if we (Sun) can arrange for >a reasonable set of high end systems to play with. > >While some topics (politics/governance/ARCs/...) won't be impacted >much by the compute infrastructure at the conference, I can see other >tracks taking a more hands-on dig in and play with code attitude. > > As originally advertised, the event was to be about discussion relating to the project itself - it wasn't billed as a "hackathon" - and more specifically, to discuss Indiana. If we're going to be more generic and even think about making it a coding/hacking session then we need to readvertise it as such. If there aren't going to be a lot of these events held then I can definately see there being room for using this (and other) opensolaris summits as the means to get people together and look at problems we've got and how to solve them. However it might be more appropriate for _that_ kind of event to be held during the week, rather than the weekend, so that (some) people don't feel like they're working a 7 day week. I don't know about others, but this was one issue I had to settle with myself doing opensolaris stuff on the weekend, as being a Sun person makes it very easy to look at this weekend as being "work time". Darren From swalker at opensolaris.org Wed Sep 19 11:57:05 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Wed, 19 Sep 2007 13:57:05 -0500 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F16E93.7060801@Sun.COM> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> Message-ID: <b9c544f0709191157r217f13dew2699f6c59abf1bf2@mail.gmail.com> On 19/09/2007, Darren.Reed at sun.com <Darren.Reed at sun.com> wrote: > John Plocher wrote: > > >Shawn Walker wrote: > > > > > >>I tend to agree that I hope that this is > >>more about problem solving than hacking... > >> > >> > > > > > >I see room for both - especially if we (Sun) can arrange for > >a reasonable set of high end systems to play with. > > > >While some topics (politics/governance/ARCs/...) won't be impacted > >much by the compute infrastructure at the conference, I can see other > >tracks taking a more hands-on dig in and play with code attitude. > > > > > > As originally advertised, the event was to be about discussion > relating to the project itself - it wasn't billed as a "hackathon" - > and more specifically, to discuss Indiana. > > If we're going to be more generic and even think about making > it a coding/hacking session then we need to readvertise it as > such. > > If there aren't going to be a lot of these events held then I can > definately see there being room for using this (and other) > opensolaris summits as the means to get people together and > look at problems we've got and how to solve them. However > it might be more appropriate for _that_ kind of event to be > held during the week, rather than the weekend, so that > (some) people don't feel like they're working a 7 day week. > > I don't know about others, but this was one issue I had to settle > with myself doing opensolaris stuff on the weekend, as being > a Sun person makes it very easy to look at this weekend as > being "work time". ...and from those of us who aren't at Sun, we thank you for that sacrifice. I would not have been able to attend if it had been during the week as I'm independent and not doing this on behalf of my employer. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Glynn.Foster at Sun.COM Wed Sep 19 15:19:21 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Thu, 20 Sep 2007 10:19:21 +1200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F16E93.7060801@Sun.COM> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> Message-ID: <46F1A069.3030903@sun.com> Darren.Reed at Sun.COM wrote: > If there aren't going to be a lot of these events held then I can > definately see there being room for using this (and other) > opensolaris summits as the means to get people together and > look at problems we've got and how to solve them. However > it might be more appropriate for _that_ kind of event to be > held during the week, rather than the weekend, so that > (some) people don't feel like they're working a 7 day week. > > I don't know about others, but this was one issue I had to settle > with myself doing opensolaris stuff on the weekend, as being > a Sun person makes it very easy to look at this weekend as > being "work time". Hopefully you could organize with your manager to get some time off in lieu for working the weekend. I'd like to think most managers would happily accommodate that. Yes, it's sucky for those people that work hard during the week and have to come in during the weekend and lose the precious time they spend with their families. It's also sucky for those people who don't work for Sun to take time off during the week to attend a summit that doesn't necessarily have any overlap with their day job. The summit is going to mean different things to a lot of people - our hope is to accommodate people as much as possible. Glynn From brandorr at opensolaris.org Wed Sep 19 16:10:42 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 19:10:42 -0400 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F065A2.6010006@sun.com> References: <46F065A2.6010006@sun.com> Message-ID: <5b5090780709191610w7b7df16asa56f404dcda3e755@mail.gmail.com> > Action Items: > - Update Wiki, adding fields to expand on the topic suggestions. I sent out an email.. looks like 7/18 have expounded on their topics. http://www.genunix.org/wiki/index.php/OpenSolaris_Summit#Topic_Suggestions > - Add voting system to Wiki Ok, I may have misheard, but I could have sworn I heard Sara say that someone she knew was going to use some third party system for voting? Did I hear correctly? I am going to start a discussion thread about the topics. (NO reason we can't start discussing them now. -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Glynn.Foster at Sun.COM Wed Sep 19 16:18:38 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Thu, 20 Sep 2007 11:18:38 +1200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <5b5090780709191610w7b7df16asa56f404dcda3e755@mail.gmail.com> References: <46F065A2.6010006@sun.com> <5b5090780709191610w7b7df16asa56f404dcda3e755@mail.gmail.com> Message-ID: <46F1AE4E.3020209@sun.com> Hey, Brandorr wrote: >> Action Items: >> - Update Wiki, adding fields to expand on the topic suggestions. > > I sent out an email.. looks like 7/18 have expounded on their topics. > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit#Topic_Suggestions > >> - Add voting system to Wiki > > Ok, I may have misheard, but I could have sworn I heard Sara say that > someone she knew was going to use some third party system for voting? > Did I hear correctly? Yeah, we're trying to get a simple poll site together. I've sent the list to Moazam and he said he'd get something together on unixville.com. Glynn From brandorr at opensolaris.org Wed Sep 19 18:08:02 2007 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 19 Sep 2007 21:08:02 -0400 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics Message-ID: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> I figured I'd roll this down the hill and see what others say. (I just realized that biggest thing I want to hear from the Indiana team, is not on the list: What's been done and what needs to be done. The meeting should be about fleshing that out and obtaining buy in to get it done.) (I think) Anyway, on with the list.... --------------------------------------------------------------------------------------------------------------------------------------------------- Darren Reed Community Structure Elaboration needed. --------------------------------------------------------------------------------------------------------------------------------------------------- Glynn Foster How to plan time based releases in a non-time based release world One of the primary social goals of Project Indiana focus around the release roadmap, and in particular getting to a point where we can regularly do 6 month time based releases of our binaries creating predictable cycles for developers who target that distribution, and users downloading, installing and upgrading. Many popular open source communities have found benefits of this time based approach, and others are starting to move towards it as a way of managing their software. While this is nice in theory, it very much relies on a strict development ethic. How can we get to a stage of providing time based releases when the world around us has traditionally been feature based? Is there a best of both worlds approach, and how can our influence change the way distributions are created both internal and external to Sun? Mark Kupfer's blogged about this very issue. I suggest reading it. http://blogs.sun.com/kupfer/entry/what_i_learned_from_ubuntu --------------------------------------------------------------------------------------------------------------------------------------------------- Brian Gupta Developing/supporting a User community, and portal, etc. As Sun is marketing/pushing Indiana, Solaris, and OpenSolaris, we are seeing many people that aren't developers coming to this site. As this is the first thing they see when they get here, I feel that this should be the logical location for a user oriented welcome page. I also propose that this welcome page is much simpler, as the current page is a bit overwhelming for newcomers. (Hard to find certain links) Developers would still have access to a developer homepage that is linked off the new homepage. (We might even setup a redirect: dev.opensolaris.org) http://www.opensolaris.org/jive/thread.jspa?messageID=148391 No comment since this is my suggestion. :) --------------------------------------------------------------------------------------------------------------------------------------------------- John Plocher Architecture Community's relationship to Indiana, PSARC and other distros. I'm looking for direction and consensus on several fundamental issues: Problem: What is needed to do good systems architecture in a world that has multiple systems? i.e., what is good for Solaris Express may not be good for Indiana or Schillix... I think this is an Indiana talk... this may be off topic Problem: What needs to be done to enable effective ARC participation by external-to-Sun members? i.e., web services, tools, materials submission, case creation, voting, etc I think this might be incorporate in a larger question about including the community. Problem: What does it mean to be an external ARC member? Job description, scope of job, time needed, tasks, authority...? I view this as an important question, but might we be able to take this question directly to the arc members on the mailing list? Problem: How are appeals handled? Who makes the unpopular decisions? Where is the "teeth"? As I understand there is a Constitution that handles appeals. Unpopular with whom? Teeth?? --------------------------------------------------------------------------------------------------------------------------------------------------- Stephen Lau Is it hard to contribute to OpenSolaris? How can we make it easier? The "is it hard" is a rhetorical question, really. Its fairly obvious there are some barriers to contribution. This will be a gathering of people to figure out what those barriers are, identify concrete things that suck, and try to figure out how to reduce the suck. It is not a session about distro specific stuff, it is not a session about adding new features or changing things in OpenSolaris other than the contribution barriers. NC --------------------------------------------------------------------------------------------------------------------------------------------------- Ben Rockwood The Future of Patching and Upgrades (OS & Apps) Is there more to this topic than the current SXCE upgrade process? --------------------------------------------------------------------------------------------------------------------------------------------------- John Sonnenschein Increasing community participation while maintaining traditional stability & compatibility, or "avoiding the Linux free-for-all mess" Please elaborate and give examples... (Remember we are talking about the Indiana distro, not everything in th OpenSolaris/Solaris universe.) --------------------------------------------------------------------------------------------------------------------------------------------------- Tim Foster Project Indiana, an exercise in compromise (in the UK/Ireland sense) Because we all have to agree to drop things that may be near/dear to us in order to actually get something out the door. We don't have to get everything right, but we do have to show progress. Endlessly arguing about what's the default shell, or which version of tar we use isn't the way to get from here to there. This get's me plus one, with a slight change. Basically we need to decide what is important for Project Indiana. What do we need to address before our first binary shipment? --------------------------------------------------------------------------------------------------------------------------------------------------- Jim Walker How do we test OpenSolaris? This get's my plus one. As I understand all Solaris Express testing is currently done behind closed walls. If this can't be brought out into the open, what is involved in setting up an equivelant e;nvironment --------------------------------------------------------------------------------------------------------------------------------------------------- Dave Stewart What is Intel doing to help OpenSolaris drivers? Please eleborate how the Summit can address this? --------------------------------------------------------------------------------------------------------------------------------------------------- Ian Murdock ZFS to the max (justification from timf) # fsck /dev/dsk/c0d0s4 ** /dev/rdsk/c0d0s4 CANNOT READ: BLK 62217648 CONTINUE? n This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Ian Murdock Laptop support This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Ian Murdock User experience This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Sara Dornsife Naming This needs more elaboration --------------------------------------------------------------------------------------------------------------------------------------------------- Brian Gupta Opensolaris.org OSOL Infrastructure Support to the outside community. Currently the Opensolaris.org infrastructure ( www.opensolaris.org and the mailing lists being two example pieces of infrastructure) is completely maintained by Sun employees. This limits available resources. There are many of sys-admin types out in the community that are capable, and most likely willing, to help out. (As well as businesses that might be willing to fund contract employees to aid in this endeavor) Might we explore a path to opening Opensolaris.org's care and feeding to the greater community? No comment, my topic again. :) --------------------------------------------------------------------------------------------------------------------------------------------------- Shawn Walker Creating custom distributions using Indiana and possible legal obligations / requirements. I don't know if there are going to be lawyers there. If not, I suggest we best not speculate. --------------------------------------------------------------------------------------------------------------------------------------------------- Stefan Teleman C++ Standard Compliant: Sun Studio 12, BOOST, and the Apache Standard C++ Library I am not sure how this directly relates to project Indiana. --------------------------------------------------------------------------------------------------------------------------------------------------- Stuart Kreitman Parity: Making Open Solaris equal or better than Linux as a development desktop Please eleborate --------------------------------------------------------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070919/36d888d9/attachment.html> From John.Plocher at Sun.COM Wed Sep 19 21:26:42 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 19 Sep 2007 21:26:42 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> Message-ID: <46F1F682.2030008@Sun.Com> Brandorr wrote: > John Plocher ... > Problem: ... Focusing on Indiana: Indiana is all about building a system (aka distro) out of OS.o parts. The ARC is a systems architecture body. What can the ARC do for Indiana? That is, what does Indiana need from the ARC so Indiana can be successful for its various releases? [The ideas I've heard concern things like refactoring "the WOS" to allow different rates of change in the various sections, keeping the "core" of OpenSolaris completely compatible while allowing incompatible and disruptive innovation to happen closer to the edge. What explicit actions are needed to make that a reality?] What can Indiana do for the ARC? Since it is an "Us-ARCing-Our-Stuff" world, what ARC-related resources are Indiana developers willing or able to provide to help deliver those things? (Or is the expectation really "Us-vs-Them", where someone /else/ is expected to do the ARC work?) In any case, what does the ARC community need to do to ensure that those resources are effectively and productively used? How does the existing community structure need to change? What should happen when the systems architectural needs of Indiana conflict with those of Solaris or one of the other distros? Should there be an "Indiana ARC", a "Solaris ARC", a "Schillix ARC", ...etc? If so, where does an "OpenSolaris ARC" fit? What is in and out of scope for each? The needs of a kernel.org-style community are not the same as the needs of an ubuntu-style one. What needs to be done to move the larger OS.o community from its current kernel.org-ish mindset to more of the uubuntu-style one that Indiana envisions? > Problem: How are appeals handled? Who makes the unpopular > decisions? Where is the "teeth"? You are right - this is bigger than just Indiana. Nevertheless, for context, if nothing else: > As I understand there is a Constitution that handles > appeals. Unpopular with whom? Teeth?? Say the ARC says "do /this/" or even "we don't approve this proposal", and the project team or its sponsoring community disagrees. This is a common situation where the short term needs of a project may conflict with the long term needs of the larger OS.o community. 20 years of Sun experience shows that there needs to be an intermediate step in those "appeals" to ensure that they all don't immediately overload the OGB with hearings or cause expensive OS.o community-wide formal votes. What happens if a project or community chooses to make architectural- level changes to an OS.o component without engaging with the ARC? Who gets to choose whether or not ARCing applies to something? Even if one is in a part of the OS.o community where the ARC is known to apply, what happens if a project [tries to] integrate without doing an ARC review? If they get stopped, who does the stopping? By what authority does the stopper actually obtain the right to stop it? That's teeth. From alan.duboff at sun.com Wed Sep 19 22:25:13 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 19 Sep 2007 22:25:13 -0700 (PDT) Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <46F1F682.2030008@Sun.Com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> Message-ID: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> On Wed, 19 Sep 2007, John Plocher wrote: > Say the ARC says "do /this/" or even "we don't approve this proposal", > and the project team or its sponsoring community disagrees. This is > a common situation where the short term needs of a project may conflict > with the long term needs of the larger OS.o community. 20 years of Sun > experience shows that there needs to be an intermediate step in those > "appeals" to ensure that they all don't immediately overload the OGB > with hearings or cause expensive OS.o community-wide formal votes. This is where an ARC would do the OpenSolaris community an injustice if they impose limitations on the software that can be contained in a repository of user-land, open source software. For a long time I've hoped that we would see open source software install in /usr/bin:/usr/sbin as it should, but it's not clear if Indiana will take things to that extent yet. But I did sit in the session that was given at IDF today in San Francisco, and I got the impression that Ian could be thinking that way (and if so, I would support that thought;-). The current ARCs would blow a fuse at the thought of putting anything in /usr/bin:/usr/sbin I suspect, but am only speculating. However, this is where the software should go in a system, IMO, unless there is conflict (similar functions like SysV tar vs. gnu tar as an example). > What happens if a project or community chooses to make architectural- > level changes to an OS.o component without engaging with the ARC? Who > gets to choose whether or not ARCing applies to something? I don't think that should be an option, because they would need to, or should be required to go through an ARC to get anything into the ON portion of OpenSolaris, and that certainly should not change. > Even if > one is in a part of the OS.o community where the ARC is known to > apply, what happens if a project [tries to] integrate without doing > an ARC review? If they get stopped, who does the stopping? By what > authority does the stopper actually obtain the right to stop it? > That's teeth. The way some have gotten around this happening in open source software today, is to have unstable and stable distributions, where unstable is more tolerant to accepting software. In Sun's case with Solaris, they will certainly need to decide how they take software from the open source community, just as they do today, unless they change their ways possibly. I don't see that for Solaris, but stranger things have happened. A good question is how the OpenSolaris will keep control over what goes into any "general" repository though, I don't think anyone should have the access to do a putback if the code is not packaged properly for instance. At some point I want to see a large repository of software, today we have that scattered all around, blastwave, pkgsrc, polaris, belenix, schillix, Nexenta, etc...I think Nexenta gets more right than any of the others to date, not that there is no flaw in their solution, just that today they seem to have more in place and understand how that software should be placed in the system. Maybe because they follow Debian's lead in how they operate, but I just mean in regards to having a system that can be updated and be inclusive of a large repository. What if there was a large repository of packages that anyone could use on an OpenSolaris system? I think it would be good to elliminate the walls that seperate the above mentioned efforts, if possible, get our community to unify once and for all...and to do that we need to encourage folks to help us package and contribute to some large effort. For better or worse it seems to me that Indiana is as good of a reason to kickoff some type of effort like that, as anything else. -- Alan DuBoff - (wearing his nomex suit, go ahead and flame me cause he's wearing his honest face;-) From alan.duboff at sun.com Wed Sep 19 22:35:00 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 19 Sep 2007 22:35:00 -0700 (PDT) Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F1A069.3030903@sun.com> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> <46F1A069.3030903@sun.com> Message-ID: <Pine.GSO.4.64.0709192230380.3804@eagle.softorchestra.com> On Thu, 20 Sep 2007, Glynn Foster wrote: > Hopefully you could organize with your manager to get some time off in > lieu for working the weekend. I'd like to think most managers would > happily accommodate that. I think the community should fall how it falls. IOW, if someone needs to feel they need to be paid to work on OpenSolaris, that's ok, and if they feel they would like to participate for another reason, that's ok also. Ultimately OpenSolaris will be comprised of the folks that do participate in it and are a part of it. I'm not trying to say that Darren should feel as if he should be one way or the other, I'm trying not to critisize him for his own decision just as I wouldn't another. > Yes, it's sucky for those people that work hard during the week and have > to come in during the weekend and lose the precious time they spend with > their families. And how is that any different for the community people that do other things during the week and may spend the time to attend? It is held on the weekend to get as many folks as they can, I think you would be in agreement with that. Can you imagine if it was held during the week? There should be many community people saying, "hey, I have to support my family and can't get off work". > It's also sucky for those people who don't work for Sun > to take time off during the week to attend a summit that doesn't > necessarily have any overlap with their day job. Exactly.;-) > The summit is going to mean different things to a lot of people - our > hope is to accommodate people as much as possible. Amen, let the cards fall how they fall and be glad there cards at all.;-) -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.coopersmith at sun.com Thu Sep 20 08:07:03 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Thu, 20 Sep 2007 08:07:03 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> Message-ID: <46F28C97.6010103@sun.com> Alan DuBoff wrote: > On Wed, 19 Sep 2007, John Plocher wrote: > The current ARCs would blow a fuse at the thought of putting anything in > /usr/bin:/usr/sbin I suspect, but am only speculating. However, this is > where the software should go in a system, IMO, unless there is conflict > (similar functions like SysV tar vs. gnu tar as an example). That would be the ARC's that already decided that everything in /usr/sfw/bin that doesn't conflict with something already in /usr/bin should move to /usr/bin? I think you're a bit out of touch with the ARC's... -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From brandorr at opensolaris.org Thu Sep 20 08:21:47 2007 From: brandorr at opensolaris.org (Brandorr) Date: Thu, 20 Sep 2007 11:21:47 -0400 Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. Message-ID: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> I an catching an early flight to San Jose. (landing at 12:30pm). Does anyone want to try and meet up during the day on Friday? -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From John.Plocher at Sun.COM Thu Sep 20 10:19:57 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 20 Sep 2007 10:19:57 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> Message-ID: <46F2ABBD.2070306@Sun.Com> Alan DuBoff wrote: > This is where an ARC would do the OpenSolaris community an > injustice if they impose limitations on the software that > can be contained in a repository of user-land, open source software. Nobody but you has said anything about general software availability in random repositories. I'm talking about the systems we choose to build out of specific things from specific repositories. It may well be that we get a severe case of stupidity and create several incompatible versions of libc and place them in various repositories. No problem there (other than being stupid, of course). It should be obvious that one can't simply randomly pick one of those libc's and expect the rest of their system to "just work" - there is a dependency chain that links your libc choice to the stuff that will work with it. Same for most everything that is in or near the core of a system. The systems engineering and architecture I'm concerned with focuses on the decisions that a distro producer needs to make to ensure that their distro starts off self consistent - and remains so over time. If you want to go off and do your own thing without any regard to how your behavior impacts projects that have come to depend on you, feel free to do so. Just don't expect others to be very happy with you if you jerk them around by behaving irresponsibly. What do I mean by irresponsible behavior? Effectively, not setting expectations for those who would consume your stuff, or once having set them, breaking them. > The current ARCs would blow a fuse at the thought of putting anything in > /usr/bin:/usr/sbin I suspect, but am only speculating. You /are/ speculating - wildly and without much basis in fact. Most of the ARC's focus in the last couple of years has been on doing exactly that. The whole point of Systems Engineering - as opposed to unconstrained hacking - is to develop a model for how we /desire/ a system to behave and to ensure that what we /do/ matches that model. This includes changing that model as time goes by and expectations change. The Solaris model is one of stability of interfaces over time. The model says that we desire that all code and scripts that were written following the rules will continue to operate correctly on any and all minor and micro releases of Solaris in the future. This obviously is not the same model that is desired for Indiana. Unfortunately, nobody has articulated the specifics of what this new model are, nor have they put any effort in developing transition plans to move between the old and new models. Topping it all off, we (Sun, OpenSolaris,...) only have a "stable tree" to work with. There is no "unstable development branch" where we can do all those chaotic things that you seem to desire. This all makes it hard for developers, architects, distro producers and users to understand what they need to do. -John From Glynn.Foster at Sun.COM Thu Sep 20 12:48:31 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Fri, 21 Sep 2007 07:48:31 +1200 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <46F2ABBD.2070306@Sun.Com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> <46F2ABBD.2070306@Sun.Com> Message-ID: <46F2CE8F.5040703@sun.com> John Plocher wrote: > The Solaris model is one of stability of interfaces over time. The > model says that we desire that all code and scripts that were written > following the rules will continue to operate correctly on any and > all minor and micro releases of Solaris in the future. > > This obviously is not the same model that is desired for Indiana. > > Unfortunately, nobody has articulated the specifics of what this > new model are, nor have they put any effort in developing transition > plans to move between the old and new models. Topping it all off, > we (Sun, OpenSolaris,...) only have a "stable tree" to work with. > There is no "unstable development branch" where we can do all those > chaotic things that you seem to desire. Very much what I'm keen to tackle at the summit. Most of my concerns during the Indiana PSARC meeting a couple of months ago revolved around this. As you point out, it's certainly a pretty hard for anything that isn't Solaris to know what to do best, without losing the unique benefits of those guarantees (and by Solaris I mean 'official release' rather than the express train). Glynn From phil at bolthole.com Thu Sep 20 15:13:19 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 15:13:19 -0700 Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com>; from alan.duboff@sun.com on Wed, Sep 19, 2007 at 10:25:13PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> Message-ID: <20070920151319.A23828@bolthole.com> On Wed, Sep 19, 2007 at 10:25:13PM -0700, Alan DuBoff wrote: > ... > For a long time I've hoped that we would see open source software install > in /usr/bin:/usr/sbin as it should, but it's not clear if Indiana will > take things to that extent yet. But I did sit in the session that was > given at IDF today in San Francisco, and I got the impression that Ian > could be thinking that way (and if so, I would support that thought;-). My opinion on that, is that it should depend on "who do you call, for an 'indiana' system, if something in /usr/bin breaks?" There is still some fuziness in my mind, as to what indiana is. Even if indiana is supposed to be a more "do it yourself" option, I could still envision some situationally dependant reasons where "everything goes in /usr/bin" may not be the best choice. That is to say, even if Sun is not the force behind "supporting" an indiana based install.. SOMEONE/some group/forum will be. For that entity, there will still be value for knowing something about what to expect from stuff in /usr/bin. Envision this: Someone posts on a debian user mailing list, "Help! my [foo] isnt working right!" people try to help, and after much oddity and headscratching and time spent, someone says, [hypothetical: this doesnt work too well if there are 2,000 files there!] "well show us what ls -l /usr/bin looks like", and it is then determined that; 1. this user just "happened" to have installed some experimental version of grep in /usr/bin, instead of the usual debian package 2. this version of grep has some bugs or altered functionality in edge cases 3. This other version was what caused [foo] to not work right. Tens of hours have been wasted trying to track down a problem "with debian", when it was actually a problem with an experimental user dumping some alternative binary in /bin, because after all, 'everything goes in bin'. If you dont set up some kind of expectation of, "dont muck with bin like that", then messes like that will be bound to happen. In the case of Debian, even though I'm not aware of any explicit rule against that sort of thing... I think the person would be lambasted for wasting peoples time. There is some amount of expectation that "a debian system", would have the debian version of grep installed. In the case of Indiana... I think it would be beneficial to have some kind of explicit statement of what to expect from /usr/bin, and what kinds of things should be installed "elsewhere". (and where "elsewhere" should be!) In addition to avoiding timewasters like the above, i think it would help to build confidence in what to expect from "Indiana". But, all this again depends on what exactly Indiana is supposed to be. If it is just supposed to be "a more fluidly updated version of Solaris", (with a nod to also making additional freeware more easily installable) then I think the rules on what goes in /usr/bin, should be quite strict, similar to regular Solaris. In my opinion, "non-sun-tracked" software should still go somewhere else. If, on the other hand, Indiana is targetted as a more "anything goes" environment, then perhaps not. But if it's an "anything goes" environment, then I almost dont see the point of it. From Darren.Reed at Sun.COM Thu Sep 20 15:34:23 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Thu, 20 Sep 2007 15:34:23 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> Message-ID: <46F2F56F.70706@Sun.COM> Brandorr wrote: > I figured I'd roll this down the hill and see what others say. > > (I just realized that biggest thing I want to hear from the Indiana > team, is not on the list: What's been done and what needs to be done. > The meeting should be about fleshing that out and obtaining buy in to > get it done.) (I think) > > Anyway, on with the list.... > > --------------------------------------------------------------------------------------------------------------------------------------------------- > > Darren Reed Community Structure > > Elaboration needed. See: http://blogs.sun.com/avalon/entry/manufacturing_a_community And to be more specific: - are contributors and core contributors required? - is the voting mechanism correct to bring in new contributors? - if leaders are just those that can edit web pages, is this the right label? - etc. Darren From swalker at opensolaris.org Thu Sep 20 15:55:32 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 20 Sep 2007 17:55:32 -0500 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> Message-ID: <b9c544f0709201555v2e8c8549q32d774070eedad71@mail.gmail.com> On 19/09/2007, Brandorr <brandorr at opensolaris.org> wrote: > Shawn Walker Creating custom distributions using Indiana and possible > legal obligations / requirements. > > I don't know if there are going to be lawyers there. If not, I > suggest we best not speculate. I wasn't suggesting that we speculate. I've updated the entry to clarify: Project Indiana, as far as I know, has a goal of being a base for others to create their own distributions upon. I would like to hear more about how it might approach some of the issues that will face people wanting to create their own distribution, or what the plan to support them is. Some questions might include: * What is the plan to support distribution creators? * Will there be a special mailing list just for folks that are creating their own distribution based off of Indiana? * Will there be special documentation explaining the legal, technical, and other requirements? (i.e. possible redistribution issues, copyright and trademark issues, licensing issues, etc.) * Can I create a distribution that has a Solaris kernel and GNU userland (and the possible legal implications)? * Can I redistribute the nVidia drivers with my distribution? While legal people may not be present, that should not stop getting a list of possible questions for someone qualified to answer at a later date. I would like to see this area addressed within the scope of the OpenSolaris community (not just the Indiana project, although that would be good too). -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From phil at bolthole.com Thu Sep 20 15:59:52 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 15:59:52 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F2F56F.70706@Sun.COM>; from Darren.Reed@Sun.COM on Thu, Sep 20, 2007 at 03:34:23PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> Message-ID: <20070920155952.A17130@bolthole.com> On Thu, Sep 20, 2007 at 03:34:23PM -0700, Darren Reed wrote: > > See: > http://blogs.sun.com/avalon/entry/manufacturing_a_community > interesting blog. I'd say one core problem in there, is due to sun marketspeak misusing the term "community". Someone apparently got all caught up in "Open Source Development is 'community based'! We've Got To Have Comunit(ies) !!!" Seems like the marketdroids have achieved way too much power inside sun lately, as evidenced by stupidity like renaming SUNW to "JAVA". Maybe the entities currently described under http://www.opensolaris.org/os/communities/ should have a name other than "community", and then have a more hierarchical organization to them. There's just too many of them at the top level right now. What might be better,could be an approach like the old USENET style approach, where you first start with the HIGH level comp.os.opsol Split it up into comp.os.opsol.kernel comp.os.opsol.install comp.os.opsol.graphics comp.os.opsol.blah and then split up into finer grained pieces, ONLY WHEN participants of a particular "leaf" node, vote and agree that the leaf should now become a node with child leaves. (and usually, there were rules about that sort of thing, such as, the vote has to continue for X amount of time, so as to avoid making a split about something that is going to die out in 1 month anyway) getting a bit off-topic for this list now, I suppose. From alan.coopersmith at sun.com Thu Sep 20 16:12:31 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Thu, 20 Sep 2007 16:12:31 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070920155952.A17130@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> Message-ID: <46F2FE5F.3070508@sun.com> Philip Brown wrote: > There's just too many of them at the top level right now. We tried to reorganize a few months ago, but most of the communities felt that giving up their top-level status diminished their importance and they didn't want any part of it. (As I understand, the OpenSolaris structure was designed after the Apache Foundation structure, but they're more of a loosely affialiated group of software projects than something trying to produce a coherent OS, so it may not have been the best fit.) -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From phil at bolthole.com Thu Sep 20 16:29:35 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 16:29:35 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F2FE5F.3070508@sun.com>; from alan.coopersmith@sun.com on Thu, Sep 20, 2007 at 04:12:31PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> Message-ID: <20070920162935.B17130@bolthole.com> On Thu, Sep 20, 2007 at 04:12:31PM -0700, Alan Coopersmith wrote: > Philip Brown wrote: > > There's just too many of them at the top level right now. > > We tried to reorganize a few months ago, but most of the communities > felt that giving up their top-level status diminished their importance > and they didn't want any part of it. > Well, then there's a core problem with opensolaris.org right there. Someone needs to have the authority, and B... erm.. "gumption".. to tell them, "Too bad, this is for the better good of OpenSolaris.". Isnt that what the (previously named?) "CAB" is for? You cant make everyone happy, all the time. Obviously, some groups are going to be unhappy to be moved down. Doesnt mean it shouldnt be done. If they were thinking more sensibly, they should be happy about it, because better organization should actually mean they would be easier to find. But whether they come to that realization or not, doesnt change the need for it. From Darren.Reed at Sun.COM Thu Sep 20 16:57:40 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Thu, 20 Sep 2007 16:57:40 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070920162935.B17130@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> Message-ID: <46F308F4.8050101@Sun.COM> Philip Brown wrote: >On Thu, Sep 20, 2007 at 04:12:31PM -0700, Alan Coopersmith wrote: > > >>Philip Brown wrote: >> >> >>>There's just too many of them at the top level right now. >>> >>> >>We tried to reorganize a few months ago, but most of the communities >>felt that giving up their top-level status diminished their importance >>and they didn't want any part of it. >> >> >> > >Well, then there's a core problem with opensolaris.org right there. >Someone needs to have the authority, and B... erm.. "gumption".. to tell >them, "Too bad, this is for the better good of OpenSolaris.". > >Isnt that what the (previously named?) "CAB" is for? > > Possibly. And having put the problem like this, for better or worse, what OpenSolaris needs is a board of directors (or the equivalent): a body that can make decisions; not just have discussions via email and try to resolve "issues". How would that be different to the OGB? No open board-discuss for starters. Have it function like a real board: meet monthly or quarterly, publish the minutes of the meetings, appoint a chairman, give the board power to make decisions that change opensolaris without needing votes, etc... >You cant make everyone happy, all the time. >Obviously, some groups are going to be unhappy to be moved down. Doesnt >mean it shouldnt be done. > > Yes. And so long as we allow even one "I disagree" to hold up change, we'll never really accomplish very much at all in a timely fashion. Darren From phil at bolthole.com Thu Sep 20 17:20:25 2007 From: phil at bolthole.com (Philip Brown) Date: Thu, 20 Sep 2007 17:20:25 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F308F4.8050101@Sun.COM>; from Darren.Reed@Sun.COM on Thu, Sep 20, 2007 at 04:57:40PM -0700 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> Message-ID: <20070920172025.D17130@bolthole.com> On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: > Philip Brown wrote: > >Isnt that what the (previously named?) "CAB" is for? > >... > > How would that be different to the OGB? No open board-discuss > for starters. Have it function like a real board: meet monthly or > quarterly, publish the minutes of the meetings, appoint a chairman, > give the board power to make decisions that change opensolaris > without needing votes, etc... The one trouble with doing stuff politically like that, is that there is then a tendancy for those positions to be filled with politicians. There need to be rules, but not to the point of needing a "full time politician" to handle it, so to speak. > >You cant make everyone happy, all the time. > >Obviously, some groups are going to be unhappy to be moved down. Doesnt > >mean it shouldnt be done. > > Yes. And so long as we allow even one "I disagree" to hold up change, > we'll never really accomplish very much at all in a timely fashion. The one softening I would like to add to my comments, is a clarification of my general principles on that sort of thing. "We shouldnt do it because x,y/z" should be carefully considered. Whines about, "I dont like it/I dont want to change", should be ignored, in my opinion [if the change has long term benefits] A single person dissenting, doesnt mean that single person should be automatically ignored, though. The more important question is whether the dissent is on rational grounds or not. (and it then becomes important for the people making that determination, to be both mentally able to discern that, and also to have the commitment and integrity to do so. Having a "majority vote" is no substitute for technical competance and understanding) From Darren.Reed at Sun.COM Thu Sep 20 17:41:25 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Thu, 20 Sep 2007 17:41:25 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070920172025.D17130@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> Message-ID: <46F31335.6060700@Sun.COM> Philip Brown wrote: >On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: > > >>Philip Brown wrote: >> >> >>>Isnt that what the (previously named?) "CAB" is for? >>> >>> >>... >> >>How would that be different to the OGB? No open board-discuss >>for starters. Have it function like a real board: meet monthly or >>quarterly, publish the minutes of the meetings, appoint a chairman, >>give the board power to make decisions that change opensolaris >>without needing votes, etc... >> >> > >The one trouble with doing stuff politically like that, is that there is >then a tendancy for those positions to be filled with politicians. >There need to be rules, but not to the point of needing a >"full time politician" to handle it, so to speak. > > To which I'd say that the board must be comprised of only those who commit or make changes to the source code of opensolaris, not those that are only involved in the ARC, website, etc. That there are few "code cutters" that are "politicians" tends to solve the "politician" problem (or at least this is my observation.) Open source projects are fundamentally about code, everything else is just window dressing. The right to be taken seriously should be inline with the contribution to the source code base. Darren From stevel at sun.com Thu Sep 20 17:49:52 2007 From: stevel at sun.com (Stephen Lau) Date: Thu, 20 Sep 2007 17:49:52 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F31335.6060700@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> Message-ID: <46F31530.9020102@sun.com> Darren Reed wrote: > Philip Brown wrote: > >> On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: >> >> >>> Philip Brown wrote: >>> >>> >>>> Isnt that what the (previously named?) "CAB" is for? >>>> >>>> >>> ... >>> >>> How would that be different to the OGB? No open board-discuss >>> for starters. Have it function like a real board: meet monthly or >>> quarterly, publish the minutes of the meetings, appoint a chairman, >>> give the board power to make decisions that change opensolaris >>> without needing votes, etc... >>> >>> >> The one trouble with doing stuff politically like that, is that there is >> then a tendancy for those positions to be filled with politicians. >> There need to be rules, but not to the point of needing a >> "full time politician" to handle it, so to speak. >> >> > > To which I'd say that the board must be comprised of only those > who commit or make changes to the source code of opensolaris, > not those that are only involved in the ARC, website, etc. That > there are few "code cutters" that are "politicians" tends to solve > the "politician" problem (or at least this is my observation.) > > Open source projects are fundamentally about code, everything > else is just window dressing. The right to be taken seriously > should be inline with the contribution to the source code base. Which only sets up an even more elitist atmosphere. You would be eliminating the valuable contributions of people who are docs people, translations, artists, evangelists, etc. There is nothing inherent about cutting code that somehow makes someone qualified to run or lead an open source project or community. -steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From Glynn.Foster at Sun.COM Thu Sep 20 18:45:53 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Fri, 21 Sep 2007 13:45:53 +1200 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709191031530.1532@jazz> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> Message-ID: <46F32251.7060102@sun.com> Alan DuBoff wrote: > On Sat, 15 Sep 2007, Garrett D'Amore wrote: > >> Again, my biggest question/concern is, how many people will be in >> attendance that the gory details of driver/kernel programming will >> appeal to? I know you've voted, so that's one... are there more? > > I'm sure there are more, but the problem is that most of the people that > are interested, are spread out more globally, as a whole, so targeting > specific venues will be hit/miss at best in most cases. > > We need this information online, with the ability to stream in any > location, so that the right people will have easy access to it. "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind tour of a week-long class on how to write a driver. Seth touched on the DDI/DKI interface, which provides terrific interface stability from generation to generation, the device tree and tips and tricks. There was a demonstration driver source presented as well, and all of the commands to try it out. I'll have to do that when I get home!" http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ Sounds like the information is there (though possibly not GLDv3?), would probably be good to highlight it more? Glynn From alan.duboff at sun.com Fri Sep 21 01:03:25 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 01:03:25 -0700 (PDT) Subject: [opensolaris-summit] Brandorr's comments on proposed Summit topics In-Reply-To: <20070920151319.A23828@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F1F682.2030008@Sun.Com> <Pine.GSO.4.64.0709192156520.3804@eagle.softorchestra.com> <20070920151319.A23828@bolthole.com> Message-ID: <Pine.GSO.4.64.0709210027450.5157@eagle.softorchestra.com> On Thu, 20 Sep 2007, Philip Brown wrote: > On Wed, Sep 19, 2007 at 10:25:13PM -0700, Alan DuBoff wrote: >> ... For a long time I've hoped that we would see open source software >> install in /usr/bin:/usr/sbin as it should, but it's not clear if >> Indiana will take things to that extent yet. But I did sit in the >> session that was given at IDF today in San Francisco, and I got the >> impression that Ian could be thinking that way (and if so, I would >> support that thought;-). > > My opinion on that, is that it should depend on "who do you call, for an > 'indiana' system, if something in /usr/bin breaks?" I have a really tweaked vision from the way things are currently done at Sun, and I'm sure many folks won't agree with the way I see things, but from my perspective that is simple. Sun basically sells support, something they're not great at today, but that will need to change. If you don't pay, you go to somewhere like here, or other place on the inet (yahoogroups, usenet, etc...). Pay $x and you can call a live body to talk your problem through with, but it doesn't get any resolution per se, other than the bug can get logged if it does exist. Pay $xx and get the ability to have updates over the wire for a year. Pay $xxx, Sun will not only work the issure with you, they will provide a live body to look into and/or fix the problem. People could get help in broken software, not in the sense that Sun would provide broken software that people need to pay for to get fixed, but more so that the Sun engineer will work with the open source code and/or folks to resolve it. Pay $xxxx, Sun will develop and extend open source software for you. This is very much in the professional services realm. > There is still some fuziness in my mind, as to what indiana is. Don't feel bad, there is in my mind also. Each time I attend one of Ian's presentation, things are becoming more clear, and I thought he gave a pretty good presentation at IDF this week. To me it looks like Sun is going to be working towards providing a binary distribution that will update over the wire, and includes all of the open source software you would expect in today's climate. > Even if indiana is supposed to be a more "do it yourself" option, I > could still envision some situationally dependant reasons where > "everything goes in /usr/bin" may not be the best choice. I personally think Nexenta got a lot of things right. The one thing I don't like is that is has a GNU personality by default. I really need to have SysV compatibility, first and foremost for OpenSolaris to provide backward compatibility to S10 and before. However, Indiana will most likely only install to ZFS, I hope my understanding is correct about that because I think ZFS is one of the best technologies that Sun has at the moment. This evolution in inevitable, IMO. I also don't think that Sun is the only player in this space, and others such as Blastwave could offer their own distribution and provide their own support, just as Sun does, just like Nexenta could, Belenix could, Shillix, Martux, etc...or OTOH, they could just continue to be what they are and not offer support. > That is to say, even if Sun is not the force behind "supporting" an > indiana based install.. SOMEONE/some group/forum will be. For that > entity, there will still be value for knowing something about what to > expect from stuff in /usr/bin. I think Indiana will be a product that Sun will offer support for. I don't forsee Sun offering support for all other OpenSolaris distributions though. > 1. this user just "happened" to have installed some experimental version > of grep in /usr/bin, instead of the usual debian package That's ok with me, pony up to the $xxx plan and you'll get a live body to help you figure that out. > 2. this version of grep has some bugs or altered functionality in edge > cases That's ok also, pony up to the $xxxx program and an engineer will look at it and help assist in getting it to work. At some point, should the use be willing to pay, Sun can work on resolving it on a per hour basis, just as professional services have been done for ages. > 3. This other version was what caused [foo] to not work right. That's ok also, foo was covered in the $xxxx plan <wink>, but in the $xxxxx plan maybe you can get 2 engineers, or the $xxxxxx plan. At some point, just like everything else in life, we find what is feasable to us and/or fits within our budget. For those that can wait and/or find their own workaround, the open source software will work fine for them. Sun can also have releases of Indiana that are considered stable, for instance, and if people need more feature and/or function, they can add updates to it, you can get that in my $xxx plan above!;-) > If you dont set up some kind of expectation of, "dont muck with bin like > that", then messes like that will be bound to happen. I'm fine with that happening, if the user wants Sun to help, cough up for support, or shut up and find a resolution yourself. The sources are there, as they say!;-) > In the case of Debian, even though I'm not aware of any explicit rule > against that sort of thing... I think the person would be lambasted for > wasting peoples time. There is some amount of expectation that "a debian > system", would have the debian version of grep installed. That is bound to happen in the opensolaris community and actually we've seen quite a few flame-ups happen on this very list over controversial issues. > In the case of Indiana... I think it would be beneficial to have some > kind of explicit statement of what to expect from /usr/bin, and what > kinds of things should be installed "elsewhere". (and where "elsewhere" > should be!) In addition to avoiding timewasters like the above, i think > it would help to build confidence in what to expect from "Indiana". Why so? I think that whatever Sun decides, it's their product and ultimately the folks that will want to use it will not care about much of that. They really don't care if it's in /usr/bin, /opt/sfw/, or /usr/local/your/mamma/wears/army/boots...if that makes sense. They just want to use their software to check their mail and surf the web in most cases. > But, all this again depends on what exactly Indiana is supposed to be. I agree, it's been morphing, and I think it's heading in the right direction. Will you be at the summit? A lot has happened over the past 5-6 years, heh? Solaris x86 is more of a contender today than it ever has been. When we get AcroRead, the world should rejoice!<wink> (I did say *when*, because I believe that it will be here at some point) -- Alan DuBoff - Those without dreams will never have dreams come true... From alan.duboff at sun.com Fri Sep 21 01:07:16 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 01:07:16 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F308F4.8050101@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> Message-ID: <Pine.GSO.4.64.0709210106540.5157@eagle.softorchestra.com> On Thu, 20 Sep 2007, Darren Reed wrote: > Possibly. And having put the problem like this, for better or worse, > what OpenSolaris needs is a board of directors (or the equivalent): > a body that can make decisions; not just have discussions via email > and try to resolve "issues". > > How would that be different to the OGB? No open board-discuss > for starters. Have it function like a real board: meet monthly or > quarterly, publish the minutes of the meetings, appoint a chairman, > give the board power to make decisions that change opensolaris > without needing votes, etc... I thought this was what the OGB was doing...silly me... -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Fri Sep 21 01:15:06 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 01:15:06 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F32251.7060102@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> Message-ID: <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> On Fri, 21 Sep 2007, Glynn Foster wrote: > Alan DuBoff wrote: >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: >> >>> Again, my biggest question/concern is, how many people will be in >>> attendance that the gory details of driver/kernel programming will >>> appeal to? I know you've voted, so that's one... are there more? >> >> I'm sure there are more, but the problem is that most of the people that >> are interested, are spread out more globally, as a whole, so targeting >> specific venues will be hit/miss at best in most cases. >> >> We need this information online, with the ability to stream in any >> location, so that the right people will have easy access to it. > > "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind tour of > a week-long class on how to write a driver. Seth touched on the DDI/DKI > interface, which provides terrific interface stability from generation to > generation, the device tree and tips and tricks. There was a demonstration > driver source presented as well, and all of the commands to try it out. I'll > have to do that when I get home!" > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > Sounds like the information is there (though possibly not GLDv3?), would > probably be good to highlight it more? Yes, actually Seth's session was good and pretty well attended. I would like to see Seth do more stuff that is not so crunched in, and to be able to have more shorts that are say, 5 minutes, so that you might have 20 shorts to cover the same thing, or to take 5-10 minutes on each topic. Next week we're going to try to video a session that we're doing in MPK, on using the kernel debugger. I hope this will be the first in a series of content we can offer on YouTube and OpenSolaris. OpenSolaris will have high resolution content, opposed to YouTube which is a great mechanism to reach out (so is facebook), but the quality pales, and having a draw with high resolution video is a good thing, IMO, which will get traffic to OpenSolaris.org. Hey, do I know how I will host it? Nope, don't even care...will figure that out after I have something that is a proof of concept, and make sure that Sun provides the storage and/or bandwidth. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Glynn.Foster at Sun.COM Fri Sep 21 05:17:53 2007 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Sat, 22 Sep 2007 00:17:53 +1200 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> Message-ID: <46F3B671.4090509@sun.com> Hey, Alan DuBoff wrote: > Hey, do I know how I will host it? Nope, don't even care...will figure > that out after I have something that is a proof of concept, and make sure > that Sun provides the storage and/or bandwidth. If you don't manage to make any in-roads on this, please let me know and I'll try from my side. I'm sure it won't be a problem - just a case of finding the right person. I think it would be awesome to have a series of tech talks from various engineers. Glynn From swalker at opensolaris.org Fri Sep 21 08:44:43 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 21 Sep 2007 10:44:43 -0500 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F31335.6060700@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> Message-ID: <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> On 20/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > Philip Brown wrote: > > >On Thu, Sep 20, 2007 at 04:57:40PM -0700, Darren Reed wrote: > > > > > >>Philip Brown wrote: > >> > >> > >>>Isnt that what the (previously named?) "CAB" is for? > >>> > >>> > >>... > >> > >>How would that be different to the OGB? No open board-discuss > >>for starters. Have it function like a real board: meet monthly or > >>quarterly, publish the minutes of the meetings, appoint a chairman, > >>give the board power to make decisions that change opensolaris > >>without needing votes, etc... > >> > >> > > > >The one trouble with doing stuff politically like that, is that there is > >then a tendancy for those positions to be filled with politicians. > >There need to be rules, but not to the point of needing a > >"full time politician" to handle it, so to speak. > > > > > > To which I'd say that the board must be comprised of only those > who commit or make changes to the source code of opensolaris, > not those that are only involved in the ARC, website, etc. That > there are few "code cutters" that are "politicians" tends to solve > the "politician" problem (or at least this is my observation.) > > Open source projects are fundamentally about code, everything > else is just window dressing. The right to be taken seriously > should be inline with the contribution to the source code base. I don't think you can call the work of writing documentation, providing i18n translations, or running user groups "window dressing." Those are part of the lifeblood of the community. While my contributions are code, and not any of the other things, documentation, etc. are just as important in my mind as code contributions. I'm sure you just accidentally omitted those items... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From phil at bolthole.com Fri Sep 21 08:51:30 2007 From: phil at bolthole.com (Philip Brown) Date: Fri, 21 Sep 2007 08:51:30 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com>; from swalker@opensolaris.org on Fri, Sep 21, 2007 at 10:44:43AM -0500 References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> Message-ID: <20070921085130.F7529@bolthole.com> On Fri, Sep 21, 2007 at 10:44:43AM -0500, Shawn Walker wrote: > On 20/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > > Open source projects are fundamentally about code, everything > > else is just window dressing. The right to be taken seriously > > should be inline with the contribution to the source code base. > > I don't think you can call the work of writing documentation, > providing i18n translations, or running user groups "window dressing." > Those are part of the lifeblood of the community. > > While my contributions are code, and not any of the other things, > documentation, etc. are just as important in my mind as code > contributions. Heh. they are "very important", and often make the difference between an "ok" product, and a "great" product. But they are not "just as important as the code". Without the documentation and translations.. the code still results in something useful. Without the code.. the documentation, translations, etc. are useless. That being said: coders, and tech writers, are often separate breeds. That is why for products, there is usually a separate TECH ARC group, vs documentation group. "tech" = "code". Similar authority, but over different areas of the product. Someone who is a great coder, and fully would be qualified for the "tech arc" group, may be completely unfit for the documentation oversight group, and vice versa. From swalker at opensolaris.org Fri Sep 21 09:10:09 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 21 Sep 2007 11:10:09 -0500 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070921085130.F7529@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> <20070921085130.F7529@bolthole.com> Message-ID: <b9c544f0709210910r1dc56d65kef150759e20651a9@mail.gmail.com> On 21/09/2007, Philip Brown <phil at bolthole.com> wrote: > On Fri, Sep 21, 2007 at 10:44:43AM -0500, Shawn Walker wrote: > > On 20/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > > > Open source projects are fundamentally about code, everything > > > else is just window dressing. The right to be taken seriously > > > should be inline with the contribution to the source code base. > > > > I don't think you can call the work of writing documentation, > > providing i18n translations, or running user groups "window dressing." > > Those are part of the lifeblood of the community. > > > > While my contributions are code, and not any of the other things, > > documentation, etc. are just as important in my mind as code > > contributions. > > Heh. > > they are "very important", and often make the difference between an "ok" product, > and a "great" product. > But they are not "just as important as the code". > > Without the documentation and translations.. the code still results in > something useful. > Without the code.. the documentation, translations, etc. are > useless. > > That being said: coders, and tech writers, are often separate breeds. > That is why for products, there is usually a separate TECH ARC group, vs > documentation group. > "tech" = "code". > Similar authority, but over different areas of the product. > > Someone who is a great coder, and fully would be qualified for the > "tech arc" group, may be completely unfit for the documentation oversight > group, and vice versa. But those things are important so that you can have users, and a program without users is a program that probably won't be getting any code... Sort of a chicken-and-egg problem. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Garrett.Damore at Sun.COM Fri Sep 21 09:21:47 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 09:21:47 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F32251.7060102@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> Message-ID: <46F3EF9B.90206@sun.com> Glynn Foster wrote: > Alan DuBoff wrote: > >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: >> >> >>> Again, my biggest question/concern is, how many people will be in >>> attendance that the gory details of driver/kernel programming will >>> appeal to? I know you've voted, so that's one... are there more? >>> >> I'm sure there are more, but the problem is that most of the people that >> are interested, are spread out more globally, as a whole, so targeting >> specific venues will be hit/miss at best in most cases. >> >> We need this information online, with the ability to stream in any >> location, so that the right people will have easy access to it. >> > > "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind tour of > a week-long class on how to write a driver. Seth touched on the DDI/DKI > interface, which provides terrific interface stability from generation to > generation, the device tree and tips and tricks. There was a demonstration > driver source presented as well, and all of the commands to try it out. I'll > have to do that when I get home!" > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > Sounds like the information is there (though possibly not GLDv3?), would > probably be good to highlight it more? > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all (all NIC drivers are STREAMs drivers). What I saw convinced me of the utter foolishness of trying to cram a week's worth of teaching material into one hour. However, I could probably present a simple GLDv2->v3 presentation class. That can be done in an hour if the attendees have at least a tiny inkling of kernel programming. I don't mind covering other topics, and one could probably do a whirlwind tour of kernel/driver APIs in a day, if one had the entire day to do it. -- Garrett > > Glynn > From alan.coopersmith at sun.com Fri Sep 21 09:21:27 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Fri, 21 Sep 2007 09:21:27 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F31335.6060700@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> Message-ID: <46F3EF87.9070705@sun.com> Darren Reed wrote: > To which I'd say that the board must be comprised of only those > who commit or make changes to the source code of opensolaris, > not those that are only involved in the ARC, website, etc. Is there anyone on the ARC who hasn't putback hundreds of changes to some gate? ARC members aren't a separate body of people - they're the senior engineers, who have been in the trenches for years and are sharing their battle scars so others don't have to get hit in the same places they did. In any case, this is way off topic for summit planning, and I await your proposed constitutional amendments on ogb-discuss. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From John.Plocher at Sun.COM Fri Sep 21 09:53:18 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 21 Sep 2007 09:53:18 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070921085130.F7529@bolthole.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <b9c544f0709210844p418c87c4ld1859fe70ac72120@mail.gmail.com> <20070921085130.F7529@bolthole.com> Message-ID: <46F3F6FE.3000002@Sun.Com> Philip Brown wrote: > But they are not "just as important as the code". In my experience, the act of asking a developer to articulate what exactly their code/program/feature is intended to do (versus what it actually does or how it happens to be implemented today...) invariably causes the developer to produce a better program. Whether this is done at the level of ARC-itectural review or as part of writing the man pages or configuration docs, it /is/ an integral part of software /engineering/. Without this focus on intent, all you have is random hacking, which is far less interesting to this community. -John From brandorr at opensolaris.org Fri Sep 21 10:07:16 2007 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 21 Sep 2007 13:07:16 -0400 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F3EF9B.90206@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> Message-ID: <5b5090780709211007x6f73803bq2c71151e2b291c54@mail.gmail.com> On 9/21/07, Garrett D'Amore <Garrett.Damore at sun.com> wrote: > > Glynn Foster wrote: > > Alan DuBoff wrote: > > > >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: > >> > >> > >>> Again, my biggest question/concern is, how many people will be in > >>> attendance that the gory details of driver/kernel programming will > >>> appeal to? I know you've voted, so that's one... are there more? > >>> > >> I'm sure there are more, but the problem is that most of the people > that > >> are interested, are spread out more globally, as a whole, so targeting > >> specific venues will be hit/miss at best in most cases. > >> > >> We need this information online, with the ability to stream in any > >> location, so that the right people will have easy access to it. > >> > > > > "Writing Device Drivers for Solaris - Seth Goldberg provided a whirlwind > tour of > > a week-long class on how to write a driver. Seth touched on the DDI/DKI > > interface, which provides terrific interface stability from generation > to > > generation, the device tree and tips and tricks. There was a > demonstration > > driver source presented as well, and all of the commands to try it > out. I'll > > have to do that when I get home!" > > > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > > > Sounds like the information is there (though possibly not GLDv3?), would > > probably be good to highlight it more? > > > > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all > (all NIC drivers are STREAMs drivers). > > What I saw convinced me of the utter foolishness of trying to cram a > week's worth of teaching material into one hour. > > However, I could probably present a simple GLDv2->v3 presentation > class. That can be done in an hour if the attendees have at least a > tiny inkling of kernel programming. > > I don't mind covering other topics, and one could probably do a > whirlwind tour of kernel/driver APIs in a day, if one had the entire day > to do it. Garrett, There will be only be 60-100 attendees to the summit. I think your driver presentations deserve a wider audience. Is there anyway you can work with Alan to make videos and put additional course collaterals online? This way they would be able to help existing driver developers, and would also enable new vendors to contribute Solaris drivers. (Basically I think we should start reaching out to vendors like 3Ware, and offer them unsolicited help porting their drivers... wait a sec,, I just think I thought of a topic for the Summit).. "Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." Do you want to lead up this vendor/community outreach program? If so I could see this as a much more powerful topic/initiative to present at the summit. Cheers, Brian -- Garrett > > > > Glynn > > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070921/d903ecd3/attachment.html> From alan.duboff at sun.com Fri Sep 21 10:46:35 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 10:46:35 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F3B671.4090509@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <Pine.GSO.4.64.0709210109120.5157@eagle.softorchestra.com> <46F3B671.4090509@sun.com> Message-ID: <Pine.GSO.4.64.0709211041180.5157@eagle.softorchestra.com> On Sat, 22 Sep 2007, Glynn Foster wrote: > Hey, > > Alan DuBoff wrote: >> Hey, do I know how I will host it? Nope, don't even care...will figure >> that out after I have something that is a proof of concept, and make >> sure that Sun provides the storage and/or bandwidth. > > If you don't manage to make any in-roads on this, please let me know and > I'll try from my side. I'm sure it won't be a problem - just a case of > finding the right person. I think it would be awesome to have a series > of tech talks from various engineers. I certainly will if I can't make headway with it after I have content. I know that Will Snow was the one responsible for the machines that host opensolaris.org, and he does handle the streaming media for sun.com, AFAIK, or at least used to. I figure however we need to drive that to get content hosted, we will sort that out. The main thing is to get something going where we have actual content and to be able to create shorts that we could use for educational/training/learning. I have a couple engineers in the queue once I get this rolling, and I hope that we can get some useful content to post. My plan is to use YouTube and/or Facebook as an exposure tool, but offer high resolution on opensolaris.org as added value. If folks would like a higher resolution stream, they can get it "from the source" so to speak!;-) -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Fri Sep 21 10:55:22 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 10:55:22 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F3EF9B.90206@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> Message-ID: <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> On Fri, 21 Sep 2007, Garrett D'Amore wrote: > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all (all NIC > drivers are STREAMs drivers). Well, sure his presentation was not intended to do that, he had to dumb it down as it was and was constrained to about 45 minutes. A big round of applause for Seth in being able to work around those contraints and requests from Intel to dumb it down. > What I saw convinced me of the utter foolishness of trying to cram a week's > worth of teaching material into one hour. Absolutely agreed. > However, I could probably present a simple GLDv2->v3 presentation class. > That can be done in an hour if the attendees have at least a tiny inkling of > kernel programming. That would be excellent, where would you do that? Not for the summit, is it? I mean, that really doesn't fit in with the topics at the summit, the summit will most likely be discussing how we could incorporate drivers that are not in OpenSolaris today, not write new ones, but maybe I'm mistaken. > I don't mind covering other topics, and one could probably do a whirlwind > tour of kernel/driver APIs in a day, if one had the entire day to do it. I think preparing a session on that would be good to have for the appropriate venue, certainly. -- Alan DuBoff - Solaris x86 IHV/OEM Group From brandorr at opensolaris.org Fri Sep 21 11:33:12 2007 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 21 Sep 2007 14:33:12 -0400 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." Message-ID: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> I added a new topic to the wiki: http://www.genunix.org/wiki/index.php/OpenSolaris_Summit Blurb: Driver support is an acknowledged area where OpenSolaris has some catching up to do. We need to figure ways to get driver support going. One of these ways could be to setup driver development training materials, as well as by reaching out to vendors to aid in porting.(IE: Lend developers to vendors to assist in writing an OpenSolaris driver). Another way, that I thought of was to setup two teams. Team one would look at a GPL driver and write a document generic describing the hardware. Team two would use that document to write a license compatible OpenSolaris driver. A third way is to setup an opensolaris.org bounty system, where multiple people can join to sponsor a bounty for a particular driver. Heck, Sun could even throw in a little bounty cash for key drivers. :) (To get it started, I am willing to put up $100 for a 3Ware RAID driver.) Anyway I think that we need to put our heads together to figure out how to expand driver support. Cheers, -Brian -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Garrett.Damore at Sun.COM Fri Sep 21 11:34:59 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 11:34:59 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> Message-ID: <46F40ED3.70906@sun.com> Alan DuBoff wrote: > On Fri, 21 Sep 2007, Garrett D'Amore wrote: > >> Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all >> (all NIC drivers are STREAMs drivers). > > Well, sure his presentation was not intended to do that, he had to > dumb it down as it was and was constrained to about 45 minutes. A big > round of applause for Seth in being able to work around those > contraints and requests from Intel to dumb it down. This wasn't a complaint. Just a statement... STREAMs programming is very different from most other kinds of programming, and requires a unique mindset. As it was, I think Seth probably tried to cover too much too quickly. I *know* the material he covered cold, and quite frankly, my eyes were starting to glaze over during the session. Again, this is no fault of Seth's.... its just not practical to cover kernel programming in an hour. I believe a series of lectures, perhaps posted on-line, would be better. If I were designing a kernel class today, I'd break it up into a series of lectures like this: *) Simple kernel mode programming/fundamentals. - compiling a kernel module - memory allocation considerations - simple debug strategies - DDI/DDK - sources of other info - this might include a simple bug fix to an existing module - link flags... kernel linker dependencies (e.g. using -N xxx/xxx) *) Autoconfiguration. - minor numbers, major numbers - the device tree - kinds of kernel modules - clone drivers - pseudo-drivers - device binding - libdevinfo (userland) - detach, attach, getinfo, probe - module _ini, _fini, _info entry points - dynamic reconfiguration *) Character and Block Drivers - entry points to implement - strategy(9e) - memory mapping (mmap, etc.) - copyin, copyout - mapping device registers - interrupt framework - DDI access functions (ddi_get32, etc.) *) Kernel synchronization and threadng - mutexes, condition variables - taskqs - timeout - (maybe?) cyclics - other clock/timing issue (lbolt, etc.) - "contexts" (interrupt, kernel, user) *) 32- versus 64-bit programming - ioctl and pointer passing... type conversion macros - dangers... pointer/integer conversions, etc. - kernel memory maps *) Kernel statistics facility - how kstat works, sample code, etc. *) STREAMs drivers - messages, put, putnext, all the other good stuff. - you cannot be a STREAMs driver and a character/block driver - STREAMs perimeters - other synchronization issues (thou shalt not sleep!) - message types (M_DATA, M_IOCTL, etc.) - copyin/copyout techniques for streams (M_COPYIN, etc.) 5) Network drivers - description of DLPI - GLDv2 - GLDv3 *) SCSI HBA drivers (SCSA) *) USB drivers - kernel apis - libusb - special binding issues. *) Dynamic Reconfiguration - detach, attach issues - rcm - dacf (?) *) Power management - suspend/resume - device power management topics *) Testing and debugging - test tips - kmdb/mdb - DTrace intro Its possible that the break up is not ideal, but this is just a rough draft of what I'm thinking about. I've no idea how such a series of lectures could be developed, however. > >> What I saw convinced me of the utter foolishness of trying to cram a >> week's worth of teaching material into one hour. > > Absolutely agreed. > >> However, I could probably present a simple GLDv2->v3 presentation >> class. That can be done in an hour if the attendees have at least a >> tiny inkling of kernel programming. > > That would be excellent, where would you do that? Not for the summit, > is it? I mean, that really doesn't fit in with the topics at the > summit, the summit will most likely be discussing how we could > incorporate drivers that are not in OpenSolaris today, not write new > ones, but maybe I'm mistaken. I dunno. I was just offering. Its looking like the Summit is going to be about governance and community topics, and have little to do with code... to the point that I'm not sure if any actual developers are going other than myself. That bothers me somewhat. If there is a big disconnect between the engineers doing the work, and the folks who are going to be making policy, we're setting ourselves up for heartache later. It somehow doesn't feel right that a bunch of folks who aren't actually involved intimately with the practice of engineering should be determining exclusively the processes which engineers must follow. Of course, if this is just a pow-wow of marketing type folks who are mostly interested in figuring how to *market* OpenSolaris to the rest of the world, then maybe there's not much point in folks who are mostly interested in development from attending. (And that would, for the most part, include me. Don't get me wrong, I'm as keen as the next person to improve OpenSolaris adoption, but talking about web pages and marketing priorities just doesn't really turn me on.) > >> I don't mind covering other topics, and one could probably do a >> whirlwind tour of kernel/driver APIs in a day, if one had the entire >> day to do it. > > I think preparing a session on that would be good to have for the > appropriate venue, certainly. See above. Probably there are a lot of sessions. I don't think we have any other venue atm, but maybe I'm mistaken? -- Garrett From Joerg.Schilling at fokus.fraunhofer.de Fri Sep 21 11:31:37 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 21 Sep 2007 20:31:37 +0200 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> Message-ID: <46f40e09.1yLPvucU4qmPg+X3%Joerg.Schilling@fokus.fraunhofer.de> Brandorr <brandorr at opensolaris.org> wrote: > Another way, that I thought of was to setup two teams. Team one would > look at a GPL driver and write a document generic describing the > hardware. Team two would use that document to write a license > compatible OpenSolaris driver. If you use Europeans, both teams may be identical. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Garrett.Damore at Sun.COM Fri Sep 21 11:39:42 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 11:39:42 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <5b5090780709211007x6f73803bq2c71151e2b291c54@mail.gmail.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <5b5090780709211007x6f73803bq2c71151e2b291c54@mail.gmail.com> Message-ID: <46F40FEE.7000604@sun.com> I'm not normally in the Bay Area where Alan is, so such a collaborative effort would be 'challenging'. If someone else in southern california wants to try to set something up, I'm willing. The Sun office in San Diego is easily reachable to me, and I can get to the office in LA as well (although that's quite a bit longer drive for me... 3 hours instead of 1 hour.) FWIW, I'm not entirely convinced that HD is necessary or even useful. Although a 640x480 presentation allowing code to be shown on-screen may be useful. -- Garrett Brandorr wrote: > On 9/21/07, *Garrett D'Amore* <Garrett.Damore at sun.com > <mailto:Garrett.Damore at sun.com>> wrote: > > Glynn Foster wrote: > > Alan DuBoff wrote: > > > >> On Sat, 15 Sep 2007, Garrett D'Amore wrote: > >> > >> > >>> Again, my biggest question/concern is, how many people will be in > >>> attendance that the gory details of driver/kernel programming will > >>> appeal to? I know you've voted, so that's one... are there more? > >>> > >> I'm sure there are more, but the problem is that most of the > people that > >> are interested, are spread out more globally, as a whole, so > targeting > >> specific venues will be hit/miss at best in most cases. > >> > >> We need this information online, with the ability to stream in any > >> location, so that the right people will have easy access to it. > >> > > > > "Writing Device Drivers for Solaris - Seth Goldberg provided a > whirlwind tour of > > a week-long class on how to write a driver. Seth touched on the > DDI/DKI > > interface, which provides terrific interface stability from > generation to > > generation, the device tree and tips and tricks. There was a > demonstration > > driver source presented as well, and all of the commands to try > it out. I'll > > have to do that when I get home!" > > > > > http://softwareblogs.intel.com/2007/09/19/opensolaris-at-fall-idf-2007/ > > > > > Sounds like the information is there (though possibly not > GLDv3?), would > > probably be good to highlight it more? > > > > Seth didn't cover GLD (neither v2 nor v3) nor STREAMs drivers at all > (all NIC drivers are STREAMs drivers). > > What I saw convinced me of the utter foolishness of trying to cram a > week's worth of teaching material into one hour. > > However, I could probably present a simple GLDv2->v3 presentation > class. That can be done in an hour if the attendees have at least a > tiny inkling of kernel programming. > > I don't mind covering other topics, and one could probably do a > whirlwind tour of kernel/driver APIs in a day, if one had the > entire day > to do it. > > > Garrett, > > There will be only be 60-100 attendees to the summit. I think your > driver presentations deserve a wider audience. Is there anyway you can > work with Alan to make videos and put additional course collaterals > online? This way they would be able to help existing driver > developers, and would also enable new vendors to contribute Solaris > drivers. (Basically I think we should start reaching out to vendors > like 3Ware, and offer them unsolicited help porting their drivers... > wait a sec,, I just think I thought of a topic for the Summit).. > > "Expanding OpenSolaris hardware support, reaching out to hardware > vendors and helping them port." > > Do you want to lead up this vendor/community outreach program? If so I > could see this as a much more powerful topic/initiative to present at > the summit. > > Cheers, > Brian > > -- Garrett > > > > Glynn > > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > <mailto:opensolaris-summit at opensolaris.org> > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > <http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit> > > > > > -- > - Brian Gupta > > http://opensolaris.org/os/project/nycosug/ From swalker at opensolaris.org Fri Sep 21 11:46:57 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 21 Sep 2007 13:46:57 -0500 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> Message-ID: <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> On 21/09/2007, Brandorr <brandorr at opensolaris.org> wrote: > I added a new topic to the wiki: > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit > > Blurb: > > Driver support is an acknowledged area where OpenSolaris has some > catching up to do. We need to figure ways to get driver support going. > > One of these ways could be to setup driver development training > materials, as well as by reaching out to vendors to aid in > porting.(IE: Lend developers to vendors to assist in writing an > OpenSolaris driver). > > Another way, that I thought of was to setup two teams. Team one would > look at a GPL driver and write a document generic describing the > hardware. Team two would use that document to write a license > compatible OpenSolaris driver. > > A third way is to setup an opensolaris.org bounty system, where > multiple people can join to sponsor a bounty for a particular driver. > Heck, Sun could even throw in a little bounty cash for key drivers. :) > (To get it started, I am willing to put up $100 for a 3Ware RAID > driver.) > > Anyway I think that we need to put our heads together to figure out > how to expand driver support. Before we can get started with adding driver support, we have to know first what is supported. I know there is a project in progress to allow the device detection tool to upload data. OpenSolaris needs its own hardware database to be able to support your proposal before it can move forward. OpenSound (especially now that it is open source) brings a significant amount of hardware support for sound devices to OpenSolaris (yet it may have bugs, but all software arguably does). Finding out what the status of that hardware database is and the device detection tool source code (which was supposed to be posted, haven't checked lately) would be the first step I think towards supporting your proposal. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Joerg.Schilling at fokus.fraunhofer.de Fri Sep 21 11:56:15 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 21 Sep 2007 20:56:15 +0200 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> Message-ID: <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" <swalker at opensolaris.org> wrote: > > Anyway I think that we need to put our heads together to figure out > > how to expand driver support. > > Before we can get started with adding driver support, we have to know > first what is supported. I believe that we should set up a list with anorder of importance and I believe that this should be layered to e.g. driver groups. I would be happy to see e.g Blue tooth support in my notebook and suspend (or better suspend) support. Then it makes sense to discuss specific hardware that is more important than others. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Darren.Reed at Sun.COM Fri Sep 21 12:55:57 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Fri, 21 Sep 2007 12:55:57 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709210106540.5157@eagle.softorchestra.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <Pine.GSO.4.64.0709210106540.5157@eagle.softorchestra.com> Message-ID: <46F421CD.4000102@Sun.COM> Alan DuBoff wrote: > On Thu, 20 Sep 2007, Darren Reed wrote: > >> Possibly. And having put the problem like this, for better or worse, >> what OpenSolaris needs is a board of directors (or the equivalent): >> a body that can make decisions; not just have discussions via email >> and try to resolve "issues". >> >> How would that be different to the OGB? No open board-discuss >> for starters. Have it function like a real board: meet monthly or >> quarterly, publish the minutes of the meetings, appoint a chairman, >> give the board power to make decisions that change opensolaris >> without needing votes, etc... > > > I thought this was what the OGB was doing...silly me... Maybe they are - there are better things to do than waste time trying to follow ogb-discuss and opensolaris-discuss. Darren From alan.duboff at sun.com Fri Sep 21 13:42:33 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 13:42:33 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F40ED3.70906@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> Message-ID: <Pine.GSO.4.64.0709211331260.8861@konocti> On Fri, 21 Sep 2007, Garrett D'Amore wrote: > This wasn't a complaint. Just a statement... STREAMs programming is > very different from most other kinds of programming, and requires a > unique mindset. As it was, I think Seth probably tried to cover too > much too quickly. Sure, but it was more of an overview session, not a hands on training course on writing device drivers. No fault of Seth for that, yes, there was quite a bit of information to absorb. I know we're in agreement on this. > I believe a series of lectures, perhaps posted on-line, would be > better. If I were designing a kernel class today, I'd break it up into > a series of lectures like this: Sure, I think that I stated something to the safe effect. > *) Simple kernel mode programming/fundamentals. > *) Autoconfiguration. > *) Character and Block Drivers > *) Kernel synchronization and threadng > *) 32- versus 64-bit programming > *) Kernel statistics facility > *) STREAMs drivers > 5) Network drivers > *) SCSI HBA drivers (SCSA) > *) USB drivers > *) Dynamic Reconfiguration > *) Power management > *) Testing and debugging These are all good and it would seem we get back to a similar course that Max Brunning currently teaches at SunU. >> That would be excellent, where would you do that? Not for the summit, >> is it? I mean, that really doesn't fit in with the topics at the >> summit, the summit will most likely be discussing how we could >> incorporate drivers that are not in OpenSolaris today, not write new >> ones, but maybe I'm mistaken. > > I dunno. I was just offering. Its looking like the Summit is going to > be about governance and community topics, and have little to do with > code... to the point that I'm not sure if any actual developers are > going other than myself. I think that was always the intention. Again, I could be wrong. > That bothers me somewhat. If there is a big disconnect between the > engineers doing the work, and the folks who are going to be making > policy, we're setting ourselves up for heartache later. It somehow > doesn't feel right that a bunch of folks who aren't actually involved > intimately with the practice of engineering should be determining > exclusively the processes which engineers must follow. You wouldn't be alone in this feeling, I suspect there are quite a few folks in engineering that are still scratching their heads trying to figure out what exactly Indiana is. From the engineering standpoint, Sun hired Ian and they all of the sudden started hearing things that supposedly they were working on. I believe Ian is trying to do the best he can in the situation he's also been put in, but that doesn't help the communication between those folks involved in that and the actual engineers. > Of course, if this is just a pow-wow of marketing type folks who are > mostly interested in figuring how to *market* OpenSolaris to the rest of > the world, then maybe there's not much point in folks who are mostly > interested in development from attending. I don't think it's fair to call it marketing type folks, there will be quite a few engineers there who are involved in pieces that will be used for Indiana, several of which were already under development before anyone heard light of Indiana. > (And that would, for the most > part, include me. Don't get me wrong, I'm as keen as the next person > to improve OpenSolaris adoption, but talking about web pages and > marketing priorities just doesn't really turn me on.) I don't think it's just about Solaris adoption, I think it's about figuring out a way that those marketing types and engineering types such as yourself to find a way to make all of this work, and be able to create a product that can be distributed in a given timeframe. Again, maybe I'm offbase on this concept as well, I'm just grasping what exactly Indiana is myself, and the definition has been a moving target. > See above. Probably there are a lot of sessions. I don't think we have > any other venue atm, but maybe I'm mistaken? We have Tech Days that travels around, and does present similar session, most of similar length however. This is something that might be worth talking to Chris Atwood about, he's in charge of training at Sun, and offering courses for that purpose (i.e., SunU). -- Alan DuBoff - Solaris x86 IHV/OEM Group From Garrett.Damore at Sun.COM Fri Sep 21 14:15:15 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 14:15:15 -0700 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <Pine.GSO.4.64.0709211331260.8861@konocti> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> Message-ID: <46F43463.9080106@sun.com> > > We have Tech Days that travels around, and does present similar > session, most of similar length however. > > This is something that might be worth talking to Chris Atwood about, > he's in charge of training at Sun, and offering courses for that > purpose (i.e., SunU). One question, and I don't know what the answer is, is how SunU fits in with what I was talking about. SunU charges lots of $$$ for their courses. But you get certified instructors, course materials, and hands-on lab time. What I was talking about was something that would be free, might cover topics that SunU doesn't (such as new information relating specifically to Nevada), and would be downloadable. I don't know how we avoid stepping on toes, etc. Anyway, I'm going to back away from this for now, only because what I'm hearing is a lot of reasons why it won't work, and not enough reason/request/desire to see such a thing work out. I have enough stuff on m plate to do that I know people want not to spend time spinning my wheels on stuff other people aren't interested in. -- Garrett From alan.duboff at sun.com Fri Sep 21 14:41:39 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 14:41:39 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F43463.9080106@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> Message-ID: <Pine.GSO.4.64.0709211434150.8861@konocti> On Fri, 21 Sep 2007, Garrett D'Amore wrote: >> We have Tech Days that travels around, and does present similar >> session, most of similar length however. >> >> This is something that might be worth talking to Chris Atwood about, >> he's in charge of training at Sun, and offering courses for that >> purpose (i.e., SunU). > > One question, and I don't know what the answer is, is how SunU fits in > with what I was talking about. I can't speak for those folks, but I have talked to them. I have pointed out to them that we need a more cost effective solution, to educate the community. I would like to see something creative, like using a black box that we drop ship to a specific location to accomodate folks for such training. Or using universities as a venue where we could do that with an army of laptops. There are various ways we could leverage new ideas and concepts to the old fashioned SunU methodology. My $0.02. > SunU charges lots of $$$ for their courses. But you get certified > instructors, course materials, and hands-on lab time. > > What I was talking about was something that would be free, might cover > topics that SunU doesn't (such as new information relating specifically > to Nevada), and would be downloadable. Yes, we're talking about the same thing, but after I contacted and/or talked to them myself, I felt it would be a better investment of time and $$$s to produce our own content and post it ourselves, and be done with it. > I don't know how we avoid stepping on toes, etc. Anyway, I'm going to > back away from this for now, only because what I'm hearing is a lot of > reasons why it won't work, and not enough reason/request/desire to see > such a thing work out. I have enough stuff on m plate to do that I know > people want not to spend time spinning my wheels on stuff other people > aren't interested in. I think there is interest, just that in the global community we're in now, it needs to be leveraged in a way that can encapsulate those folks and be able to reach them in their specific locales. Providing streaming content is one such way, and one of the more likely ones. The folks that are initmately associated with OpenSolaris.org must not have that on the top of their list, which is understandable, but I think it's a very important aspect for our community. Apple would have had QuickTime content riddled throughout the site had they rolled something out, and as such Apple does think differently, at least than Sun. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Moazam.Raja at Sun.COM Fri Sep 21 14:43:57 2007 From: Moazam.Raja at Sun.COM (Moazam Raja) Date: Fri, 21 Sep 2007 14:43:57 -0700 Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F43463.9080106@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> Message-ID: <46F43B1D.4060609@sun.com> > One question, and I don't know what the answer is, is how SunU fits in > with what I was talking about. > > SunU charges lots of $$$ for their courses. But you get certified > instructors, course materials, and hands-on lab time. > > What I was talking about was something that would be free, might cover > topics that SunU doesn't (such as new information relating specifically > to Nevada), and would be downloadable. > > -- Garrett This sounds like something the local OpenSolaris user groups would be good for (providing 'classes' for free). I'm going to be working on some podcast/video-podcasts geared towards beginner SysAdmins, that might be a start for SVOSUG. -Moazam From storycrafter at gmail.com Fri Sep 21 14:45:08 2007 From: storycrafter at gmail.com (Mark Martin) Date: Fri, 21 Sep 2007 16:45:08 -0500 Subject: [opensolaris-summit] OS Summit topic proposal In-Reply-To: <46F43463.9080106@sun.com> References: <46EB4DE2.8080808@Sun.COM> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> Message-ID: <e40c28290709211445p72deec88i8f20547b1f2293cb@mail.gmail.com> On 9/21/07, Garrett D'Amore <Garrett.Damore at sun.com> wrote: > > > I don't know how we avoid stepping on toes, etc. Anyway, I'm going to > back away from this for now, only because what I'm hearing is a lot of > reasons why it won't work, and not enough reason/request/desire to see > such a thing work out. I have enough stuff on m plate to do that I know > people want not to spend time spinning my wheels on stuff other people > aren't interested in. > > FWIW, I hope this doesn't die completely. I have been lurking in interest because: a) I was hoping there would be some way to consume content from the summit without being able to attend b) That some of that content would be useful to a (potential) contributor outside Sundom c) I might learn more about how to navigate the waters into actually contributing, as I believe I'm only about 10% into that process Broad sips would help me more now than deep pulls from the information hose, and any tidbits I can get from a coding and process standpoint would be 100% useful to me, especially in easily digested streaming video form. I'm hoping Alan's vision comes forward more -- low-res content on YouTube and hi-res on opensolaris.org. To that end, I'd even volunteer time for editing and managing video if that's required. Either way, at some point in my personal future timeline I expect to want to have the in-depth stuff -- the 1 and 2 hour or even 8 hour sessions. I realize that may not come from 15 minute session talks from a summit or TechDay or whatever but hopefully I don't have to go too far away (virtually) from opensolaris.org to find that. And honestly, I don't want to have to go to SunU to get it either unless that's a no-cost proposition. If there's a format, curriculum, or content that can be "leveraged" from there, that seems ostensibly like a win, though. Device driver implementation is of concern for me -- even if I don't have qualifications in that particular area to contribute effectively -- because that's a perceived weakness of OpenSolaris as far as I'm concerned. I'd volunteer what meager resources I could to see that any and all information, especially in streaming media format, came to light. Mark -- ------------------------------------------------------ Born to the false world, the wanderer, Storyteller, The Pied Piper On a quest for immortality Gathering a troop to find the fantasy -- Nightwish -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070921/d5a98bf8/attachment.html> From alan.duboff at sun.com Fri Sep 21 14:57:34 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 14:57:34 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F43B1D.4060609@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> Message-ID: <Pine.GSO.4.64.0709211455010.8861@konocti> On Fri, 21 Sep 2007, Moazam Raja wrote: >> One question, and I don't know what the answer is, is how SunU fits in >> with what I was talking about. >> >> SunU charges lots of $$$ for their courses. But you get certified >> instructors, course materials, and hands-on lab time. >> >> What I was talking about was something that would be free, might cover >> topics that SunU doesn't (such as new information relating specifically >> to Nevada), and would be downloadable. > > >> -- Garrett > > This sounds like something the local OpenSolaris user groups would be > good for (providing 'classes' for free). I'm going to be working on some > podcast/video-podcasts geared towards beginner SysAdmins, that might be > a start for SVOSUG. Absolutely! Seems we do have a location to host our videos, I saw Will Snow at lunch, he manages the opensolaris.org servers. He also handles streaming media, and said that we can host it on media.sun.com (I believe that is the site but it doesn't pull anything up). We can host any content that is created though, Moazam, please feel free to move forward with any ideas you have. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.coopersmith at sun.com Fri Sep 21 15:10:23 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Fri, 21 Sep 2007 15:10:23 -0700 Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <Pine.GSO.4.64.0709211455010.8861@konocti> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> <Pine.GSO.4.64.0709211455010.8861@konocti> Message-ID: <46F4414F.8010405@sun.com> Alan DuBoff wrote: > Absolutely! Seems we do have a location to host our videos, I saw Will > Snow at lunch, he manages the opensolaris.org servers. He also handles > streaming media, and said that we can host it on media.sun.com (I believe > that is the site but it doesn't pull anything up). Try http://mediacast.sun.com/ -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From brandorr at opensolaris.org Fri Sep 21 15:19:37 2007 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 21 Sep 2007 18:19:37 -0400 Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F4414F.8010405@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> <Pine.GSO.4.64.0709211455010.8861@konocti> <46F4414F.8010405@sun.com> Message-ID: <5b5090780709211519y1618dd79l83ffed27721b915@mail.gmail.com> On 9/21/07, Alan Coopersmith <alan.coopersmith at sun.com> wrote: > Alan DuBoff wrote: > > Absolutely! Seems we do have a location to host our videos, I saw Will > > Snow at lunch, he manages the opensolaris.org servers. He also handles > > streaming media, and said that we can host it on media.sun.com (I believe > > that is the site but it doesn't pull anything up). > > Try http://mediacast.sun.com/ Also the documentation community has some videos listed on there homepage: http://www.opensolaris.org/os/community/documentation/ It looks like they are being hosted here: http://frsun.downloads.edgesuite.net/sun/07C00892/media/demos/OpenSolarisDualBoot-Step4-OS-Installation.html -Brian > > -- > -Alan Coopersmith- alan.coopersmith at sun.com > Sun Microsystems, Inc. - X Window System Engineering > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From alan.duboff at sun.com Fri Sep 21 15:24:35 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Fri, 21 Sep 2007 15:24:35 -0700 (PDT) Subject: [opensolaris-summit] OS Summit topic proposal [courses and training] In-Reply-To: <46F4414F.8010405@sun.com> References: <46EB4DE2.8080808@Sun.COM> <46EB67AF.2040806@sun.com> <44285.72.39.216.186.1189873445.squirrel@mail.blastwave.org> <46ECD251.1090409@sun.com> <Pine.GSO.4.64.0709191031530.1532@jazz> <46F32251.7060102@sun.com> <46F3EF9B.90206@sun.com> <Pine.GSO.4.64.0709211052020.5157@eagle.softorchestra.com> <46F40ED3.70906@sun.com> <Pine.GSO.4.64.0709211331260.8861@konocti> <46F43463.9080106@sun.com> <46F43B1D.4060609@sun.com> <Pine.GSO.4.64.0709211455010.8861@konocti> <46F4414F.8010405@sun.com> Message-ID: <Pine.GSO.4.64.0709211517220.8861@konocti> On Fri, 21 Sep 2007, Alan Coopersmith wrote: > Alan DuBoff wrote: >> Absolutely! Seems we do have a location to host our videos, I saw Will >> Snow at lunch, he manages the opensolaris.org servers. He also handles >> streaming media, and said that we can host it on media.sun.com (I >> believe that is the site but it doesn't pull anything up). > > Try http://mediacast.sun.com/ > Yes, that's it Alan, thanks. Will said we can upload stuff there. BTW, a short video on using XRandR would be a good one to have, that's a very handy utility for folks to know about. After we get things going maybe you or Stuart would like to help with that. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Jim.Grisanzio at Sun.COM Fri Sep 21 18:53:20 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Sat, 22 Sep 2007 10:53:20 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <18163.64275.538538.803812@gargle.gargle.HOWL> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> Message-ID: <46F47590.4010309@sun.com> James Carlson wrote: > Stephen Lau writes: >> Darren Reed wrote: >>> To which I'd say that the board must be comprised of only those >>> who commit or make changes to the source code of opensolaris, >>> not those that are only involved in the ARC, website, etc. > [...] >>> Open source projects are fundamentally about code, everything >>> else is just window dressing. The right to be taken seriously >>> should be inline with the contribution to the source code base. >> Which only sets up an even more elitist atmosphere. You would be >> eliminating the valuable contributions of people who are docs people, >> translations, artists, evangelists, etc. > > Indeed. > >> There is nothing inherent about cutting code that somehow makes someone >> qualified to run or lead an open source project or community. > > I agree. I agree as well. Many people have worked hard to build a diverse community open to all. I think we've come a long way, too, given the restrictions and dependencies we've been working with all alone. These things take time, and I think Indiana will help us grow to a new level of diversity. Diversity has been the intention all along from our very earliest planning meetings four years ago. The code is at the center and that first layer of coders is critical as a foundation for the entire community, no question about it. But communities are more than foundations. Also, as we grow in emerging regions around the world, we'll be engaging significantly more non-coders than coders (at least initially). Jim > Perhaps more importantly for this thread, though, I would like some > clarification from Darren on what he thinks are the existing > deficiencies in the OGB. In other words, I can't tell if he's > speaking in generic terms about all open source projects aside from > OpenSolaris, or if he has some clear complaint to be made about the > current membership. The wording seems deliberately inflammatory, > though. > > If he does in fact have a complaint, I'd like to hear it. I *am* an > ARC member, along with one of the other elected OGB members, but that > somehow doesn't keep me (or Alan) from writing code. Am I the > unqualified one due to a lack of political prowess, serving on the > wrong committees, or just for my code? I'd really like to know. > > Who are those drag-inducing people of whom Darren writes? If I'm it, > please make sure I don't end up winning the first round of the vote > next time we do this. ;-} > From Garrett.Damore at Sun.COM Fri Sep 21 19:04:23 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Fri, 21 Sep 2007 19:04:23 -0700 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F47827.5000208@sun.com> Joerg Schilling wrote: > "Shawn Walker" <swalker at opensolaris.org> wrote: > > >>> Anyway I think that we need to put our heads together to figure out >>> how to expand driver support. >>> >> Before we can get started with adding driver support, we have to know >> first what is supported. >> > > I believe that we should set up a list with anorder of importance and I believe > that this should be layered to e.g. driver groups. I would be happy to see e.g > Blue tooth support in my notebook and suspend (or better suspend) support. > Suspend is working just fine now. Its *resume* that is the tricky part. ;-) Seriously, though, I have just been asked to start looking at Suspend/Resume for a certain brand of laptop. Its likely that once that work is done, we'll expand the scope to include support for a variety of other models. Suspend in this case means S3 (suspend to ram) initially. S4 (suspend to disk) is something we still need on x86 as well. (We already have it on SPARC.) > Then it makes sense to discuss specific hardware that is more important than > others. > There will always be favorite hardware that is missing from the list. Making such a list is useful, but shouldn't gate the effort of getting information out to folks who want to help. I can think probably of a half-dozen or so simple driver projects for folks to undertake, if they were so inclined. Right now, the number of people with time and inclination, and ability (or ability to learn) is the limiting factor. Anything that addresses *that* gap is likely to be of the most use. -- Garrett > J?rg > > From Darren.Reed at Sun.COM Fri Sep 21 19:54:41 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Fri, 21 Sep 2007 19:54:41 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <18163.64275.538538.803812@gargle.gargle.HOWL> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> Message-ID: <46F483F1.1050805@Sun.COM> James Carlson wrote: >Stephen Lau writes: > >>Darren Reed wrote: >> >>>To which I'd say that the board must be comprised of only those >>>who commit or make changes to the source code of opensolaris, >>>not those that are only involved in the ARC, website, etc. >>> >[...] > >>>Open source projects are fundamentally about code, everything >>>else is just window dressing. The right to be taken seriously >>>should be inline with the contribution to the source code base. >>> >>Which only sets up an even more elitist atmosphere. You would be >>eliminating the valuable contributions of people who are docs people, >>translations, artists, evangelists, etc. >> > >Indeed. > The suggestion was an off-the-cuff remark. A more accurate pool of people from which to make up a board would be the contributors and core contributors combined as this should (I hope) include those who right documentation, etc. I see no reason why there should be any distinction between core and non-core contributors. But to this day, I've not seen any open source project dominated by "politicians" and thus I'd opine that to suggest that this would happen is not entirely accurate. Or to put it differently, if it did happen within OpenSolaris then it would be the first place it had and would make me wonder more deeply about where OpenSolaris is going. On top of that there are ways and means (like 1 or 2 year limits on terms, combined with limits on how frequently people can come back) to limit the attractiveness of using a position for political gain only. My observation of other projects is that people get "tired" of such positions after a period of time so the other issues become less of a concern than one might expect - especially since these offices are all pro-bono (and should remain that way.) >>There is nothing inherent about cutting code that somehow makes someone >>qualified to run or lead an open source project or community. >> > >I agree. > >Perhaps more importantly for this thread, though, I would like some >clarification from Darren on what he thinks are the existing >deficiencies in the OGB. In other words, I can't tell if he's >speaking in generic terms about all open source projects aside from >OpenSolaris, or if he has some clear complaint to be made about the >current membership. The wording seems deliberately inflammatory, >though. > The OGB doesn't appear to me to be actually doing anything. It appears more more like an arbitration council that also sanctions community decisions about which communities to create and how to make a contributor/core contributor. "So what". In this guise it feels more like a rubber stamp operation than anything meaningful. Darren p.s. Meeting minutes should always include the full name of all those present, not just first names. From swalker at opensolaris.org Fri Sep 21 22:56:39 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 22 Sep 2007 00:56:39 -0500 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <46F47827.5000208@sun.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> <46F47827.5000208@sun.com> Message-ID: <b9c544f0709212256r11508ca1w7a848c049856b602@mail.gmail.com> On 21/09/2007, Garrett D'Amore <Garrett.Damore at sun.com> wrote: > Joerg Schilling wrote: > > "Shawn Walker" <swalker at opensolaris.org> wrote: > > > > > >>> Anyway I think that we need to put our heads together to figure out > >>> how to expand driver support. > >>> > >> Before we can get started with adding driver support, we have to know > >> first what is supported. > >> > > > > I believe that we should set up a list with anorder of importance and I believe > > that this should be layered to e.g. driver groups. I would be happy to see e.g > > Blue tooth support in my notebook and suspend (or better suspend) support. > > > > > Suspend is working just fine now. Its *resume* that is the tricky part. ;-) > > Seriously, though, I have just been asked to start looking at > Suspend/Resume for a certain brand of laptop. Its likely that once that > work is done, we'll expand the scope to include support for a variety of > other models. > > Suspend in this case means S3 (suspend to ram) initially. S4 (suspend > to disk) is something we still need on x86 as well. (We already have it > on SPARC.) > > > Then it makes sense to discuss specific hardware that is more important than > > others. > > > > There will always be favorite hardware that is missing from the list. > Making such a list is useful, but shouldn't gate the effort of getting > information out to folks who want to help. > > I can think probably of a half-dozen or so simple driver projects for > folks to undertake, if they were so inclined. Right now, the number of > people with time and inclination, and ability (or ability to learn) is > the limiting factor. Anything that addresses *that* gap is likely to > be of the most use. I tend to agree; which is why I suggested we need a hardware database of some sort so we can track what does work, or what is supposed to work, along with what needs support. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Joerg.Schilling at fokus.fraunhofer.de Sat Sep 22 03:30:19 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Sat, 22 Sep 2007 12:30:19 +0200 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <46F47827.5000208@sun.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> <46f413cf.CfrdbvxfBo9xg7E+%Joerg.Schilling@fokus.fraunhofer.de> <46F47827.5000208@sun.com> Message-ID: <46f4eebb.GvMw5wwYg+SRiQVQ%Joerg.Schilling@fokus.fraunhofer.de> "Garrett D'Amore" <Garrett.Damore at Sun.COM> wrote: > > I believe that we should set up a list with anorder of importance and I believe > > that this should be layered to e.g. driver groups. I would be happy to see e.g > > Blue tooth support in my notebook and suspend (or better suspend) support. > > > > > Suspend is working just fine now. Its *resume* that is the tricky part. ;-) You don't have a IBM notebook? Suspend _did_ work with Solaris 10 GA, it currently does not work. The resume problem was a frozen clock that could be reanimated with some tricks. > Suspend in this case means S3 (suspend to ram) initially. S4 (suspend > to disk) is something we still need on x86 as well. (We already have it > on SPARC.) Suspend to RAM is the more frequent case that already allows a day or more battery time. > There will always be favorite hardware that is missing from the list. > Making such a list is useful, but shouldn't gate the effort of getting > information out to folks who want to help. We are more than a few people, let us vote.... As long as Sun votes with one Dollar == one vote, some features will never make it into OpenSolaris. Solaris sells to other people than it lives from. > I can think probably of a half-dozen or so simple driver projects for > folks to undertake, if they were so inclined. Right now, the number of > people with time and inclination, and ability (or ability to learn) is > the limiting factor. Anything that addresses *that* gap is likely to > be of the most use. For smaller driver projects, the biggest impact currently is burocracy and the absence of a useful way to integrate code. As we should not impact the quality, I propose a 2 stage system. - A testing source tree allows anyone who has a sufficient credability to add code. If this code causes problems that are not fixed within a specific time, it will beremoved by a gatekeeper. - The primary source is fed by the testing source in case that there was no technical problem for some time and after a legal review was done. - Modifying sources including CLI "enhancements" still needs ARC approvement. - Pure add ons get a reversed ARC system: A sufficient amount of people need to convince the community that the additional code does not fit properly into OpenSolaris. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From shivakumar.gn at gmail.com Sat Sep 22 03:52:42 2007 From: shivakumar.gn at gmail.com (S h i v) Date: Sat, 22 Sep 2007 16:22:42 +0530 Subject: [opensolaris-summit] New topic: Expanding OpenSolaris hardware support, reaching out to hardware vendors and helping them port." In-Reply-To: <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> References: <5b5090780709211133v214dcf53r8f5e10cbef2e78dc@mail.gmail.com> <b9c544f0709211146g34aa655fwcb0ce8b17a3fb989@mail.gmail.com> Message-ID: <319ee2b10709220352h4fb9af78la12855577ba41468@mail.gmail.com> On 9/22/07, Shawn Walker <swalker at opensolaris.org> wrote: > Before we can get started with adding driver support, we have to know > first what is supported. > > I know there is a project in progress to allow the device detection > tool to upload data. > > OpenSolaris needs its own hardware database to be able to support your > proposal before it can move forward. > I see that submission of sddt output results is already there in B72 ISO. Tried submitting, the tool said results successfully submitted to Sun Microsystems Inc. If at all, this needs to be moved to a opensolaris.org database but the feature itself is already there. regards Shiv From Jim.Grisanzio at Sun.COM Sat Sep 22 04:01:00 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Sat, 22 Sep 2007 20:01:00 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F483F1.1050805@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> Message-ID: <46F4F5EC.9090909@sun.com> Darren Reed wrote: > The OGB doesn't appear to me to be actually doing anything. It > appears more more like an arbitration council that also sanctions > community decisions about which communities to create and how to > make a contributor/core contributor. "So what". In this guise > it feels more like a rubber stamp operation than anything meaningful. There was a lot of discussion about this when the initial OGB was writing the Constitution. I think the intent was to distribute leadership and decision making power among the communities and projects and not centralize it in the OGB. I agree with that model because it offers significant incentive and opportunity for participation across the entire community. That one element is what makes the OpenSolaris Constitution so meaningful. The OGB's job was to create and maintain the structure in which a community could grow. They did that. And they did that pretty well, actually. The OGB /represents/ the community, but it doesn't /run/ the community. Operations are distributed by design. The community runs itself via individual community groups and projects. When there are problems or meta issues to consider, the OGB has the authority to step in and mediate and resolve disputes. That seems to be a perfectly reasonable system to me. That's my impression of things, anyway. Jim -- http://blogs.sun.com/jimgris From roland.mainz at nrubsig.org Sun Sep 23 18:03:11 2007 From: roland.mainz at nrubsig.org (Roland Mainz) Date: Mon, 24 Sep 2007 03:03:11 +0200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F16E93.7060801@Sun.COM> Message-ID: <46F70CCF.DD249188@nrubsig.org> Darren.Reed at Sun.COM wrote: > John Plocher wrote: > >Shawn Walker wrote: > >>I tend to agree that I hope that this is > >>more about problem solving than hacking... > > > >I see room for both - especially if we (Sun) can arrange for > >a reasonable set of high end systems to play with. > > > >While some topics (politics/governance/ARCs/...) won't be impacted > >much by the compute infrastructure at the conference, I can see other > >tracks taking a more hands-on dig in and play with code attitude. > > As originally advertised, the event was to be about discussion > relating to the project itself - it wasn't billed as a "hackathon" - > and more specifically, to discuss Indiana. > > If we're going to be more generic and even think about making > it a coding/hacking session then we need to readvertise it as > such. Erm, the idea the behind the requested stuff (short: one server and 2x x86 + 2xSPARC) was only for a few people sticking their heads together on some software problems, including booting custom kernels (e.g. the "server" is userland only (e.g. SunRay, home dirs, editors, compilers, OS/Net build tools etc.) and for jumpstarting the other machines). For a "hackathon" with 100 people we would AFAIK at least need ten times more machines... ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) From roland.mainz at nrubsig.org Sun Sep 23 18:14:08 2007 From: roland.mainz at nrubsig.org (Roland Mainz) Date: Mon, 24 Sep 2007 03:14:08 +0200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F15E8A.70407@sun.com> Message-ID: <46F70F60.662239E7@nrubsig.org> Jim Walker wrote: > John Plocher wrote: > > Shawn Walker wrote: > >> I tend to agree that I hope that this is > >> more about problem solving than hacking... > > > > I see room for both - especially if we (Sun) can arrange for > > a reasonable set of high end systems to play with. > > > > While some topics (politics/governance/ARCs/...) won't be impacted > > much by the compute infrastructure at the conference, I can see other > > tracks taking a more hands-on dig in and play with code attitude. > > > > It would be VERY USEFUL for us all to articulate what exactly we wish > > to get out of the conference - what problems need to be solved that > > would benefit from a concentrated, in person focus? (See the Wiki at > > http://www.genunix.org/wiki/index.php/OpenSolaris_Summit for some > > suggestions) > > If useful, I could use some of the test farm automation > to manage reservation time slots on whatever hardware > we have at the summit. In addition, if we have good > network access we can get to remote test machines. Erm, AFAIK the test farm is non-interactive (e.g. you can't do something like "login and hack the kernel to death"), right ? The idea of the five machines was to have one "central hub" which is user-land only (e.g. nooone except two or three people have "root" access) which handles SunRay services, home dirs (which are exported to the four other machines), jumpstart and has some development tools (compiler, OS/Net build tools etc.) installed and then 2x x86 and 2x SPARC machines to boot custom kernels from the server on demand. This stuff is only thought to give people the opportunity to stick their heads together and try some stuff without having to improvise something with laptops (or worse) ... ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) From roland.mainz at nrubsig.org Sun Sep 23 18:28:00 2007 From: roland.mainz at nrubsig.org (Roland Mainz) Date: Mon, 24 Sep 2007 03:28:00 +0200 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> Message-ID: <46F712A0.443DB9B8@nrubsig.org> Tim Foster wrote: > On Tue, 2007-09-18 at 16:56 -0700, Jesse Silver wrote: > > Server availability- > > Suggestions: > > - Have as many different kinds of machines available as possible, split > > between Sparc and x86. > > - Have at least one central server, stable and powerful, like an M4000 > > with a bunch of storage. > > Why servers ? I thought this was a collective "bang our heads together" > meeting, rather than an actual hackfest ? If if everyone has their noses > buried in laptops for the duration of the weekend, then I'm not sure > we'll have achieved our aims imho. Erm, seem other emails for a response. It's only to prevent the case of having many people around from all around thje globe who may want discuss some practical code stuff and then test or demonstrate it (including custom kernel mods) and then have no machine to do it (or rely on machines which need a day or two to crawl over the OS/Net codebase for a complete rebuild from scratch). --- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;) From phil at bolthole.com Mon Sep 24 09:53:46 2007 From: phil at bolthole.com (Philip Brown) Date: Mon, 24 Sep 2007 09:53:46 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F483F1.1050805@Sun.COM>; from Darren.Reed@Sun.COM on Fri, Sep 21, 2007 at 07:54:41PM -0700 References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> Message-ID: <20070924095346.F14087@bolthole.com> On Fri, Sep 21, 2007 at 07:54:41PM -0700, Darren Reed wrote: > > But to this day, I've not seen any open source project dominated > by "politicians" and thus I'd opine that to suggest that this would > happen is not entirely accurate. I have. The reason that YOU havent, is that those projects FAILED. :-} hence, my concern. > Or to put it differently, if it did happen within OpenSolaris then > it would be the first place it had and would make me wonder more > deeply about where OpenSolaris is going. indeed. From phil at bolthole.com Mon Sep 24 10:09:26 2007 From: phil at bolthole.com (Philip Brown) Date: Mon, 24 Sep 2007 10:09:26 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F47590.4010309@sun.com>; from Jim.Grisanzio@Sun.COM on Sat, Sep 22, 2007 at 10:53:20AM +0900 References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> Message-ID: <20070924100926.G14087@bolthole.com> On Sat, Sep 22, 2007 at 10:53:20AM +0900, Jim Grisanzio wrote: > ... > I agree as well. Many people have worked hard to build a diverse > community open to all. I think we've come a long way, too, given the > restrictions and dependencies we've been working with all alone. These > things take time, and I think Indiana will help us grow to a new level > of diversity. Diversity has been the intention all along from our very > earliest planning meetings four years ago. oooboy.. there's another 10-point buzzword again. When we successfully pair "diversity" with "community", the resulting fusion of synergized energies is certain to enhance the value-add of our open-source paradigm! :-P Open Solaris is not going to do well because of "diversity". It will succeed if we keep good people, motivated to keep doing what they are good at, within overall processes that keep sane and good engineering design as the underlying framework for what they do. Thats why this sub-topic came up. Because good document writers, etc. arent usually particularly gifted at good engineering principles. Whereas good coders... no, make that /excellent coders/... are. From alan.duboff at sun.com Mon Sep 24 11:29:19 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Mon, 24 Sep 2007 11:29:19 -0700 (PDT) Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. In-Reply-To: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> References: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> Message-ID: <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> On Thu, 20 Sep 2007, Brandorr wrote: > I an catching an early flight to San Jose. (landing at 12:30pm). Does > anyone want to try and meet up during the day on Friday? Are you talking about the Fri. before the summit? I might be, not sure, we're a few weeks away. -- Alan DuBoff - Solaris x86 IHV/OEM Group From brandorr at opensolaris.org Mon Sep 24 11:34:46 2007 From: brandorr at opensolaris.org (Brandorr) Date: Mon, 24 Sep 2007 14:34:46 -0400 Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. In-Reply-To: <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> References: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> Message-ID: <5b5090780709241134p4ed1b898pc163db250f3445d2@mail.gmail.com> On 9/24/07, Alan DuBoff <alan.duboff at sun.com> wrote: > On Thu, 20 Sep 2007, Brandorr wrote: > > > I an catching an early flight to San Jose. (landing at 12:30pm). Does > > anyone want to try and meet up during the day on Friday? > > Are you talking about the Fri. before the summit? > > I might be, not sure, we're a few weeks away. > > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > Yes, this is the Friday before. (Oct 12th).. -Brian -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From Sara.Dornsife at Sun.COM Mon Sep 24 12:44:34 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Mon, 24 Sep 2007 14:44:34 -0500 Subject: [opensolaris-summit] Tuesday planning meetings - 8am and 3pm PT 9/25 Message-ID: <46F813A2.8030404@sun.com> Reminder that we will have our next planning meetings tomorrow Tues 9/25 at 8am and 3pm PT. Dial-in information: (866)230-6968 Int'l Access/Caller Paid Dial In Number: (865)544-7856 ACCESS CODE: 2192132 Agenda: Let's get our agenda nailed down Thanks, Sara From Darren.Reed at Sun.COM Mon Sep 24 12:48:37 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Mon, 24 Sep 2007 12:48:37 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F4F5EC.9090909@sun.com> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> Message-ID: <46F81495.7080704@Sun.COM> Jim Grisanzio wrote: >Darren Reed wrote: > > >>The OGB doesn't appear to me to be actually doing anything. It >>appears more more like an arbitration council that also sanctions >>community decisions about which communities to create and how to >>make a contributor/core contributor. "So what". In this guise >>it feels more like a rubber stamp operation than anything meaningful. >> >> > >There was a lot of discussion about this when the initial OGB was >writing the Constitution. I think the intent was to distribute >leadership and decision making power among the communities and projects >and not centralize it in the OGB. I agree with that model because it >offers significant incentive and opportunity for participation across >the entire community. That one element is what makes the OpenSolaris >Constitution so meaningful. > >The OGB's job was to create and maintain the structure in which a >community could grow. They did that. And they did that pretty well, >actually. The OGB /represents/ the community, but it doesn't /run/ the >community. Operations are distributed by design. The community runs >itself via individual community groups and projects. When there are >problems or meta issues to consider, the OGB has the authority to step >in and mediate and resolve disputes. That seems to be a perfectly >reasonable system to me. That's my impression of things, anyway. > > All of which means that if the way OpenSolaris has been constructed is faulty then there's no way to fix it because all that the communities can do is look after themselves, not OpenSolaris as a whole. http://blogs.sun.com/avalon/entry/manufacturing_a_community Darren From Stuart.Kreitman at Sun.COM Mon Sep 24 12:57:41 2007 From: Stuart.Kreitman at Sun.COM (Stuart Kreitman) Date: Mon, 24 Sep 2007 12:57:41 -0700 Subject: [opensolaris-summit] First time out to the area... Will be landing around noon at San Jose Airport. In-Reply-To: <5b5090780709241134p4ed1b898pc163db250f3445d2@mail.gmail.com> References: <5b5090780709200821j2dd55bb4t82e8020680e3c9ae@mail.gmail.com> <Pine.GSO.4.64.0709241128400.5990@eagle.softorchestra.com> <5b5090780709241134p4ed1b898pc163db250f3445d2@mail.gmail.com> Message-ID: <46F816B5.10304@sun.com> Brian: Are you wanting to meet people in San Jose or in Santa Cruz? Since Santa Cruz is only 45 minutes from the Bay Area, I suspect most of us Bay Area (Sun) people will probably drive down in the morning. There's an Open Solaris reception in Santa Cruz Friday night at 7pm. (location?) If there's interest, we could do a parallel happy hour at the same time on this side of the hill. I recommend Fault Line Brewery, in Sunnyvale, just north of the airport. Stuart Brandorr wrote: > On 9/24/07, Alan DuBoff <alan.duboff at sun.com> wrote: > >> On Thu, 20 Sep 2007, Brandorr wrote: >> >> >>> I an catching an early flight to San Jose. (landing at 12:30pm). Does >>> anyone want to try and meet up during the day on Friday? >>> >> Are you talking about the Fri. before the summit? >> >> I might be, not sure, we're a few weeks away. >> >> -- >> >> Alan DuBoff - Solaris x86 IHV/OEM Group >> >> > > Yes, this is the Friday before. (Oct 12th).. > > -Brian > > > From Darren.Reed at Sun.COM Mon Sep 24 16:28:09 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Mon, 24 Sep 2007 16:28:09 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070924200946.GA868815@sun.com> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> Message-ID: <46F84809.500@Sun.COM> Keith M Wesolowski wrote: > ... > >>http://blogs.sun.com/avalon/entry/manufacturing_a_community >> > >I'm going to read this a few more times before responding. You seem >very angry, and some of what you've written appears at first blush to >consist of non sequitors. But I'll keep trying. In the meantime, is >there some concrete problem you're facing that we should be looking >into, or at least know about to have some context. > In short, the OpenSolaris community hasn't built itself or sat down and thought about what structure is best for it. And we;ve now got a structure where various groups are uninterested in change because it might mean that they are "demoted" from being a community because there is a perception that being one is a good thing or means you are important or.... Compare, for example, the decision to have communities for ZFS and SMF with that for security and networking. On the one hand you have two (ongoing) projects within Sun, on the other you have two communities of people with lots of different projects involved. The problem we have is that there's only one level or way to represent a group here: as a community. But who made all of the decisions about this? Sun. Not the OpenSolaris community. Darren From Jim.Grisanzio at Sun.COM Mon Sep 24 16:44:15 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 08:44:15 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070924100926.G14087@bolthole.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> Message-ID: <46F84BCF.1050801@sun.com> Philip Brown wrote: > On Sat, Sep 22, 2007 at 10:53:20AM +0900, Jim Grisanzio wrote: >> ... >> I agree as well. Many people have worked hard to build a diverse >> community open to all. I think we've come a long way, too, given the >> restrictions and dependencies we've been working with all alone. These >> things take time, and I think Indiana will help us grow to a new level >> of diversity. Diversity has been the intention all along from our very >> earliest planning meetings four years ago. > > oooboy.. there's another 10-point buzzword again. You like that, eh? > When we successfully pair "diversity" with "community", the resulting > fusion of synergized energies is certain to enhance the value-add of our > open-source paradigm! > > :-P > > Open Solaris is not going to do well because of "diversity". Really? So we only want coders (as Darren intimated)? And in which geo would those coders live? > It will succeed if we keep good people, motivated to keep doing what they > are good at, within overall processes that keep sane and good engineering > design as the underlying framework for what they do. I basically agree. And I've said that the coders are the core and always will be. But to have a viable global community we also need great diversity in skills, and we need to be able to work across cultures and languages and locations. > Thats why this sub-topic came up. Because good document writers, etc. arent > usually particularly gifted at good engineering principles. Whereas > good coders... no, make that /excellent coders/... are. That's fine. I generally agree. But we need the doc writers too. And the users. And the marketers. And the program managers. And the students. And the professors. And the admins. And the app developers. And the [insert whoever else wants to be a part of the community here]. Jim -- http://blogs.sun.com/jimgris From Jim.Grisanzio at Sun.COM Mon Sep 24 17:23:28 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 09:23:28 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F81495.7080704@Sun.COM> References: <5b5090780709191808j5ff0358fj3935c460a2538b46@mail.gmail.com> <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> Message-ID: <46F85500.2070903@sun.com> Darren Reed wrote: > All of which means that if the way OpenSolaris has been > constructed is faulty then there's no way to fix it because > all that the communities can do is look after themselves, > not OpenSolaris as a whole. No, it doesn't mean that at all. How did you arrive at that conclusion? > http://blogs.sun.com/avalon/entry/manufacturing_a_community Thanks for the reminder. I had been planning to respond to this post. Jim -- http://blogs.sun.com/jimgris From swalker at opensolaris.org Mon Sep 24 17:50:49 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Mon, 24 Sep 2007 19:50:49 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F84809.500@Sun.COM> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> Message-ID: <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> On 24/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > Keith M Wesolowski wrote: > > > ... > > > >>http://blogs.sun.com/avalon/entry/manufacturing_a_community > >> > > > >I'm going to read this a few more times before responding. You seem > >very angry, and some of what you've written appears at first blush to > >consist of non sequitors. But I'll keep trying. In the meantime, is > >there some concrete problem you're facing that we should be looking > >into, or at least know about to have some context. > > > > In short, the OpenSolaris community hasn't built itself or sat > down and thought about what structure is best for it. And we;ve > now got a structure where various groups are uninterested in > change because it might mean that they are "demoted" from > being a community because there is a perception that being one > is a good thing or means you are important or.... > > Compare, for example, the decision to have communities for > ZFS and SMF with that for security and networking. On the > one hand you have two (ongoing) projects within Sun, on the > other you have two communities of people with lots of different > projects involved. The problem we have is that there's only > one level or way to represent a group here: as a community. > > But who made all of the decisions about this? Sun. > Not the OpenSolaris community. The community ratified the constitution, they didn't have to. The community elected the ogb, they didn't have to. You could argue that there needs to be more non-Sun folks that have voting privileges, but those Sun folks are still part of the community. Not only that, it was an absolute disappointment that so few of the people that could vote ended up doing so. Sun hasn't made all the decisions. I'm sure many of us would like to see something better. I'm sure many of us would like the community to live on it's own. However, as you well know, the OpenSolaris community didn't grow in a typical open-source "organic" fashion since what it is centered around was a very mature project already before it became an open one. Given the absolute turmoil and cost to Sun on the inside, I think folks like Jim G. and many others at Sun have done a great job in helping to open internal processes and encourage the growth of a community. Nonetheless, I think everyone can agree that things should and can be improved. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From alan.duboff at sun.com Mon Sep 24 17:57:46 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Mon, 24 Sep 2007 17:57:46 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F84809.500@Sun.COM> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> Message-ID: <Pine.GSO.4.64.0709241749350.26931@konocti> On Mon, 24 Sep 2007, Darren Reed wrote: > In short, the OpenSolaris community hasn't built itself or sat > down and thought about what structure is best for it. And we;ve > now got a structure where various groups are uninterested in > change because it might mean that they are "demoted" from > being a community because there is a perception that being one > is a good thing or means you are important or.... > > Compare, for example, the decision to have communities for > ZFS and SMF with that for security and networking. On the > one hand you have two (ongoing) projects within Sun, on the > other you have two communities of people with lots of different > projects involved. The problem we have is that there's only > one level or way to represent a group here: as a community. > > But who made all of the decisions about this? Sun. > Not the OpenSolaris community. I was talking with John Plocher today, and one of the things that we both agreed with is that we need some way for the community members to understand where and how they can participate, so that we can grow the community as a whole. This would make a great topic for the summit, and possible give some of the community people that are attending, the knowledge and/or understanding on how they can contribute and help. As it is, I think a lot of folks (myself included) are trying to find a place to help, but having a hard time. I know there's a lot of other community people that in the same boat, and it would be good if we could come up with some type of solution where it is known, to be of help. I don't think that the above scenario and/or some of your point were intentional from Sun, even though some exist, it's just the nature of evolution and getting from point A to point B. I don't know how else they would have done things differently without causing too much rocking of the boat. Moving forward is going to need more help and participation from the community, outside of what is ON as we know it today. There's a lot of software that we need to port and or bring over to OpenSolaris, and getting the community members to understand how they can help in that would be most useful, IMO. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Mon Sep 24 18:01:59 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Mon, 24 Sep 2007 18:01:59 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F84BCF.1050801@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> Message-ID: <Pine.GSO.4.64.0709241800240.26931@konocti> On Tue, 25 Sep 2007, Jim Grisanzio wrote: > That's fine. I generally agree. But we need the doc writers too. And the > users. And the marketers. And the program managers. And the students. > And the professors. And the admins. And the app developers. And the > [insert whoever else wants to be a part of the community here]. I agree, we need everyone and everyone should be welcome. FWIW, the current systems does not encourage developers to participate, IMO, it encourages the people that have been doing the work to continue doing so. Many of the community members feel alienated. We need to overcome that hurdle. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Darren.Reed at Sun.COM Mon Sep 24 18:21:22 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Mon, 24 Sep 2007 18:21:22 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070925001601.GB868815@sun.com> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> Message-ID: <46F86292.20607@Sun.COM> Keith M Wesolowski wrote: >On Mon, Sep 24, 2007 at 04:28:09PM -0700, Darren Reed wrote: > > >>In short, the OpenSolaris community hasn't built itself or sat >>down and thought about what structure is best for it. And we;ve >>now got a structure where various groups are uninterested in >>change because it might mean that they are "demoted" from >>being a community because there is a perception that being one >>is a good thing or means you are important or.... >> > >... >Not - as you suggest - because people were afraid of being "demoted", >but because many people felt that their approach to a particular >problem space was radically different from others' and that achieving >consensus within such a framework would be too difficult. In other >words, we left Community Groups as relatively fine-grained entities >precisely because we wanted to help people be more productive. > The other way to look at this is the way in which we achieve things needs to change, or the vehicle to use needs to be different, depending on the need. >>Compare, for example, the decision to have communities for >>ZFS and SMF with that for security and networking. On the >>one hand you have two (ongoing) projects within Sun, on the >>other you have two communities of people with lots of different >>projects involved. The problem we have is that there's only >>one level or way to represent a group here: as a community. >> > >The notion of an "ongoing project" is part of the problem, I think. >In principle, there's no such thing - there are active Projects, there >are completed (or abandoned) Projects, and there are Community Groups. >Logically, if a community forms around a completed Project's artifacts >and - this is the key - wishes to sponsor additional active Projects >to improve and extend that work - then by all means it makes sense to >represent that as a Community Group. Not because it's a reward for >success or as a "promotion," but because that's what it *is*. > By presenting projects at the same level as a community, I believe we've done ourselves a disservice and have possibly made it harder to engage the broaders communities. >>But who made all of the decisions about this? Sun. >>Not the OpenSolaris community. >> > >Really? Here's what actually happened: > ... None of which pertains to the architecting and formation of OpenSolaris as a community, only the formation of the OGB and the constitution. Darren From stevel at sun.com Mon Sep 24 18:59:24 2007 From: stevel at sun.com (Stephen Lau) Date: Mon, 24 Sep 2007 18:59:24 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F86292.20607@Sun.COM> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> <46F86292.20607@Sun.COM> Message-ID: <46F86B7C.70604@sun.com> Darren Reed wrote: > By presenting projects at the same level as a community, > I believe we've done ourselves a disservice and have > possibly made it harder to engage the broaders communities. How are they presented at the same level? >>> But who made all of the decisions about this? Sun. >>> Not the OpenSolaris community. >>> >> >> Really? Here's what actually happened: >> > ... > > None of which pertains to the architecting and formation > of OpenSolaris as a community, only the formation of the > OGB and the constitution. My observation is that architecture and formation of the OpenSolaris community arise from that very same constitution. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From Jim.Grisanzio at Sun.COM Mon Sep 24 23:19:43 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 15:19:43 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <1190698157.1139.7.camel@localhost> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <1190698157.1139.7.camel@localhost> Message-ID: <46F8A87F.2030406@sun.com> Damian Wojslaw wrote: > As one no-coder I have to agree. I did most of Polish ZFS Admin Guide > translation and much of Polish OpenSolaris portal translation. I put > first Polish OpenSolaris conference in motion and hopethat it comes > true. I'm thinking of two more things to make here, so that more people > hear of OpenSolaris. And all that I do is translate docs. I care for > what OpenSolaris/Solaris is and where it's going a great deal and I'd > hate to be excluded from community, or whatever it's called these days. > While I agree that OpenSolaris needs to maintain a high quality of > engineering process and of code, I don't believe that the circle should > be closed for nonprogrammers. For once, receiveng core contributor > status for translation made me even more interesteed in doing next > translations in future. You bring up very good points around communications. In some of the fastest growing markets in the world, there are also some big language barriers. We need to be able to engage large numbers of people (who /want/ to get to know us) in their own language. The translation of documents and the website are critical community-building activities. Spending even a few minutes in China brings this point home real quick. The same in slower growing markets like Korea and Japan. There is probably more English in Eastern Europe, but the point is that things can happen much, much faster if we are multi-lingual. Jim -- Jim Grisanzio http://blogs.sun.com/jimgris From Jim.Grisanzio at Sun.COM Mon Sep 24 23:42:47 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 15:42:47 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709241800240.26931@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> Message-ID: <46F8ADE7.2040707@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Jim Grisanzio wrote: > >> That's fine. I generally agree. But we need the doc writers too. And the >> users. And the marketers. And the program managers. And the students. >> And the professors. And the admins. And the app developers. And the >> [insert whoever else wants to be a part of the community here]. > > I agree, we need everyone and everyone should be welcome. > > FWIW, the current systems does not encourage developers to participate, > IMO, it encourages the people that have been doing the work to continue > doing so. Many of the community members feel alienated. We need to > overcome that hurdle. I would certainly agree that we can do more. That's obvious. But I also see a lot of progress around here as well. I'm just trying to remind everyone that people /are/ participating and we /are/ making progress. I also think it's better to build on what has already been done rather than tearing each other down. Let's cut ourselves some slack from time to time. :) I think a lot of these issues will be better resolved at the Summit when people have a change to meet face-to-face. I'm sorry I'm going to miss that. Oh, well, next time ... Jim -- Jim Grisanzio http://blogs.sun.com/jimgris From alan.duboff at sun.com Tue Sep 25 02:24:13 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 02:24:13 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F8ADE7.2040707@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> Message-ID: <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> On Tue, 25 Sep 2007, Jim Grisanzio wrote: > Alan DuBoff wrote: >> On Tue, 25 Sep 2007, Jim Grisanzio wrote: >> >>> That's fine. I generally agree. But we need the doc writers too. And the >>> users. And the marketers. And the program managers. And the students. >>> And the professors. And the admins. And the app developers. And the >>> [insert whoever else wants to be a part of the community here]. >> >> I agree, we need everyone and everyone should be welcome. >> >> FWIW, the current systems does not encourage developers to participate, >> IMO, it encourages the people that have been doing the work to continue >> doing so. Many of the community members feel alienated. We need to >> overcome that hurdle. > > > I would certainly agree that we can do more. That's obvious. But I also > see a lot of progress around here as well. I'm just trying to remind > everyone that people /are/ participating and we /are/ making progress. I > also think it's better to build on what has already been done rather > than tearing each other down. Let's cut ourselves some slack from time > to time. :) I think a lot of these issues will be better resolved at the > Summit when people have a change to meet face-to-face. I'm sorry I'm > going to miss that. Oh, well, next time ... Jim, Sorry that you won't be at the summit, AFAIK, I will be barring any emergencies. People are participating, agreed, but not in any great capacity. Progress is being made, certainly, and Mercurial will help that to change that even more. I am not faulting anyone with my comments, more so would like to see some type of way that would allow folks to participate and contribute packages of software back to the presumed repositories that will be forming. One very confusing piece is WTF OpenSolaris actually is. To me it means a certain things, and to another person it means something else, as it does another. Sun has done a good job at confusing people with their usage of it. I had a conversation with John Plocher about this a few weeks ago, and we seem to agree that it is not clearly defined what it actually is, and unfortunately it does mean different things to different people. However, with all of this in mind, we will soon have the ability to grow a large repository of open source software, and I think it's important to get people involved with OpenSolaris in this capacity, rather than just working on the kernel pieces (ON, or what I consider to be OpenSolaris at this point, what is available). Consider there are 20,000+ packages that will potentially be available on OpenSolaris, it makes sense to try and organize this piece in a coherent way that will attract and encourage folks to participate. This is very different than the way folks participate today, and potentially where we will need the most help. It would be great if many distributions could build upon this repository as a base, in the same way that folks like Ubuntu leverage the base packages from Debian. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 02:49:30 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 11:49:30 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> Message-ID: <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" <swalker at opensolaris.org> wrote: > > But who made all of the decisions about this? Sun. > > Not the OpenSolaris community. > > The community ratified the constitution, they didn't have to. > > The community elected the ogb, they didn't have to. > > You could argue that there needs to be more non-Sun folks that have > voting privileges, but those Sun folks are still part of the > community. After it looked as it Sun did like to dominate OpenSolaris, I did ask to create a constitution an s government board in November 2004. I did this because I wanted to make sure that we will not get a similar situation as with Linux wehre Torvalds dominates all decisions. There was a hard discussion and I was attacked and told that this will never work. Later (in January 2005) people from Sun started to offer the same as their idea...... Now we have it and my impression is that it's main problem is missing visibility. Missing visibility results in missing interest. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 03:02:11 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 12:02:11 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709241749350.26931@konocti> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> Message-ID: <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> Alan DuBoff <alan.duboff at sun.com> wrote: > I was talking with John Plocher today, and one of the things that we both > agreed with is that we need some way for the community members to > understand where and how they can participate, so that we can grow the > community as a whole. I believe there are two problems: missing visibility of the OGB and missing power of the community. Another big problem is burocracy that makes it hard to contribute and that makes progress slow and this cannot be changed by the OGB. As burocracy is a self serving and self fostering system, only someone from outside that has the needed power may change it.The only such person I am currently aware is Ian Murdock. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 03:16:32 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 12:16:32 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F86292.20607@Sun.COM> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> <46F86292.20607@Sun.COM> Message-ID: <46f8e000.zsYm/LjN85c9owNs%Joerg.Schilling@fokus.fraunhofer.de> Darren Reed <Darren.Reed at Sun.COM> wrote: > By presenting projects at the same level as a community, > I believe we've done ourselves a disservice and have > possibly made it harder to engage the broaders communities. I did not like this disposition from the beginning. It prevented e.g. the creation of a general filesystem community and it artificially limits the power of important people who did not become a "community leader" because of this sructure. I believe we should have an interest based structure where people who are interested in the same topic just communicate with each other. One place where this should already happen is ARC cases. Unfortunately, it is possible to foil the ARC by using "fast tracks" for things that need a detailed discussion. The fact that nobody is able to stop this creates a sense of powerlessness. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From dougs at truemail.co.th Tue Sep 25 03:37:39 2007 From: dougs at truemail.co.th (Doug Scott) Date: Tue, 25 Sep 2007 17:37:39 +0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> Message-ID: <46F8E4F3.4050707@truemail.co.th> Alan DuBoff wrote: > However, with all of this in mind, we will soon have the ability to grow a > large repository of open source software, and I think it's important to > get people involved with OpenSolaris in this capacity, rather than just > working on the kernel pieces (ON, or what I consider to be OpenSolaris at > this point, what is available). Consider there are 20,000+ packages that > will potentially be available on OpenSolaris, it makes sense to try and > organize this piece in a coherent way that will attract and encourage > folks to participate. This is very different than the way folks > participate today, and potentially where we will need the most help. It > would be great if many distributions could build upon this repository as a > base, in the same way that folks like Ubuntu leverage the base packages > from Debian Alan, I very much agree. I think while the ON project is the most important project within OpenSolaris, I think too much "contributing" emphasis is placed on it. I keep reading that people think that it is difficult to contribute to OpenSolaris, and if we limit ourselves to thinking that the ON project = OpenSolaris than this is true. It is actually very easy to contribute if you do not try to work on something that somebody else already doing. As you pointed out that there are thousands of packages and applications which can be ported to OpenSolaris. We should be looking at how we can utilise the efforts from blastwave, pkgsrc and spec-files-extra etc to build a large package repository. This is where I think we actually need the masses to contribute to. Doug From Jim.Grisanzio at Sun.COM Tue Sep 25 05:16:53 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 21:16:53 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <Pine.GSO.4.64.0709250203570.6190@eagle.softorchestra.com> Message-ID: <46F8FC35.2000904@sun.com> Alan DuBoff wrote: > Jim, > > Sorry that you won't be at the summit, AFAIK, I will be barring any > emergencies. Do some surfing for me. :) I'll catch you in January. > People are participating, agreed, but not in any great capacity. I'm thinking of almost 200 putbacks, another 100 or so in the queue, a hand full of ARC cases, the first putbacks from the Intel deal, hundreds of bugs submitted, etc. Also, ksh93 was a nice one. I think there was some nice community work on the specification/selection of Mercurial and the whole conversation around SCM generally and SCM migration. Indiana has generated a lot of participation recently. There's a respectable number of user groups. People are presenting all around the world. There are pretty active university programs in China and India and the US. Lots of projects now, and even some using Mercurial. Docs contributions. Portal website contributions/translations. Entire distros. Etc ... So, with some thought, I bet we could triple that rather random list pretty easily. There is no single source for all the contributions to OpenSolaris, but I would argue that if we spent some time collecting the data we'd find out that it's significant given our current constraints. > Progress > is being made, certainly, and Mercurial will help that to change that even > more. I am not faulting anyone with my comments, I know, sorry about that. I didn't mean to imply that you were criticizing. I just think that we are too hard on ourselves generally. I do, however, understand some of the reasons why, and I understand that much better now that I work remotely. > more so would like to see > some type of way that would allow folks to participate and contribute > packages of software back to the presumed repositories that will be > forming. That sounds very cool. > One very confusing piece is WTF OpenSolaris actually is. To me it means a > certain things, and to another person it means something else, as it does > another. Sun has done a good job at confusing people with their usage of > it. Sure. But again, I'd argue that this is part of the normal birth and growth of a very big project that, by necessity, has come into being in stages over a very long period of time. The confusion is natural. Now, that doesn't mean that we ought to just leave it be. Far from it. We continue to work and complete the project. Honestly, I don't see a problem. I see a work in progress. And I also see Indiana as the next logical step in our evolution. > I had a conversation with John Plocher about this a few weeks ago, and > we seem to agree that it is not clearly defined what it actually is, and > unfortunately it does mean different things to different people. > > However, with all of this in mind, we will soon have the ability to grow a > large repository of open source software, and I think it's important to > get people involved with OpenSolaris in this capacity, rather than just > working on the kernel pieces (ON, or what I consider to be OpenSolaris at > this point, what is available). Consider there are 20,000+ packages that > will potentially be available on OpenSolaris, it makes sense to try and > organize this piece in a coherent way that will attract and encourage > folks to participate. This is very different than the way folks > participate today, and potentially where we will need the most help. It > would be great if many distributions could build upon this repository as a > base, in the same way that folks like Ubuntu leverage the base packages > from Debian. And I think that's great. We'll be able to engage many new levels of developers and users as a result, and participation will significantly increase and diversify. This is a huge opportunity for us, no question about it. Jim -- http://blogs.sun.com/jimgris From Garrett.Damore at Sun.COM Tue Sep 25 05:13:29 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 05:13:29 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f8e000.zsYm/LjN85c9owNs%Joerg.Schilling@fokus.fraunhofer.de> References: <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <20070925001601.GB868815@sun.com> <46F86292.20607@Sun.COM> <46f8e000.zsYm/LjN85c9owNs%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F8FB69.8060302@sun.com> Joerg Schilling wrote: > Darren Reed <Darren.Reed at Sun.COM> wrote: > > >> By presenting projects at the same level as a community, >> I believe we've done ourselves a disservice and have >> possibly made it harder to engage the broaders communities. >> > > I did not like this disposition from the beginning. > > It prevented e.g. the creation of a general filesystem community and it > artificially limits the power of important people who did not become > a "community leader" because of this sructure. > > I believe we should have an interest based structure where people who are > interested in the same topic just communicate with each other. > > One place where this should already happen is ARC cases. Unfortunately, > it is possible to foil the ARC by using "fast tracks" for things that need > a detailed discussion. The fact that nobody is able to stop this creates a > sense of powerlessness. > Actually, that is totally untrue. ARC members can derail any case that they feel needs a full review. They can (and often do) ask for more time, as well, when ongoing discussions need more time to converge. In the past, the ARC members have done this frequently. However, in the one particular case I know about, where you were pretty much the only dissenter, and where everyone else had converged, the membership voted to go ahead and approve the case. Note that in that particular case it was an active decision to approve the fasttrack... there was enough discussion between the membership to ensure that the ARC was pretty much unanimous in the belief that a) the case didn't warrant derailing, b) letting the case run for more time would not have resulted in any further convergence, and c) the case would have been approved if it were a full case. Furthermore, you were not present at the ARC meeting where this was decided, despite the fact that you could have dialed in and that the agenda was public. Penalizing the project team and the ARC because one person in the broader community objects doesn't seem fair. (The cost of effort to go from fasttrack to full case review is fairly high for both the project team and the ARC.) Nobody has absolute veto power. However, the ARC membership (which is only a small handful of technical leaders) can derail the case, if it is warranted. It doesn't happen *often*, but it does happen occasionally. If you're not happy with the way ARC runs its affairs, then I'd recommend that you volunteer as a PSARC intern. Or at least attend a few meetings to see how the process goes, before complaining. -- Garrett > > J?rg > > From Jim.Grisanzio at Sun.COM Tue Sep 25 05:53:48 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Tue, 25 Sep 2007 21:53:48 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F904DC.8010305@sun.com> Joerg Schilling wrote: > After it looked as it Sun did like to dominate OpenSolaris, I did > ask to create a constitution an s government board in November 2004. > I did this because I wanted to make sure that we will not get a similar > situation as with Linux wehre Torvalds dominates all decisions. > > There was a hard discussion and I was attacked and told that this will never > work. > > Later (in January 2005) people from Sun started to offer the same as their > idea...... There were many people who had ideas about boards -- corporate boards, partner/customer boards, community boards, elections, non-elected appointees, no board at all. But you are correct in that we really didn't focus on this to any great extent in our external communications or even with the pilot project until Jan 05 when we released DTrace and announced the CDDL. The intention to form the CAB was part of that 1/05 announcement, and I think we didn't communicate it well at all and we didn't involve the pilot early enough, either. Point taken. I remember that time and I was pretty pissed about it. But I don't think anyone intended to take your ideas. :) > Now we have it and my impression is that it's main problem is missing > visibility. Missing visibility results in missing interest. I think Keith outlined beautifully the CAB-OGB sequence -- all of which was pretty darn visible. I can also remember pinging opensolaris-discuss on various CAB issues and getting very little response. Interest in governance was always rather low. But if that's increasing now than that's cool. Remember, the CAB/OGB was never intended to be a high profile, operational organization for OpenSolaris engineering and community development. It was designed as a bootstrap to get us going with a foundational structure. And it did just that. Future elections will determine how/if its current role changes. Jim -- http://blogs.sun.com/jimgris From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 06:23:13 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 15:23:13 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F904DC.8010305@sun.com> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> <46F904DC.8010305@sun.com> Message-ID: <46f90bc1.ymFwPOaU2f0l+kaH%Joerg.Schilling@fokus.fraunhofer.de> Jim Grisanzio <Jim.Grisanzio at Sun.COM> wrote: > Joerg Schilling wrote: > > > After it looked as it Sun did like to dominate OpenSolaris, I did > > ask to create a constitution an s government board in November 2004. > > I did this because I wanted to make sure that we will not get a similar > > situation as with Linux wehre Torvalds dominates all decisions. > > > > There was a hard discussion and I was attacked and told that this will never > > work. > > > > Later (in January 2005) people from Sun started to offer the same as their > > idea...... > > > There were many people who had ideas about boards -- corporate boards, > partner/customer boards, community boards, elections, non-elected > appointees, no board at all. But you are correct in that we really If this was true, why then was the discussion all against my proposal? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From stevel at sun.com Tue Sep 25 08:25:19 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 08:25:19 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f90bc1.ymFwPOaU2f0l+kaH%Joerg.Schilling@fokus.fraunhofer.de> References: <46F2FE5F.3070508@sun.com> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <b9c544f0709241750n4108e1d2m89c715db4cb622bf@mail.gmail.com> <46f8d9aa.WesTQ/8GOVSXy32j%Joerg.Schilling@fokus.fraunhofer.de> <46F904DC.8010305@sun.com> <46f90bc1.ymFwPOaU2f0l+kaH%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F9285F.7090407@sun.com> Joerg Schilling wrote: > Jim Grisanzio <Jim.Grisanzio at Sun.COM> wrote: > >> Joerg Schilling wrote: >> >>> After it looked as it Sun did like to dominate OpenSolaris, I did >>> ask to create a constitution an s government board in November 2004. >>> I did this because I wanted to make sure that we will not get a similar >>> situation as with Linux wehre Torvalds dominates all decisions. >>> >>> There was a hard discussion and I was attacked and told that this will never >>> work. >>> >>> Later (in January 2005) people from Sun started to offer the same as their >>> idea...... >> >> There were many people who had ideas about boards -- corporate boards, >> partner/customer boards, community boards, elections, non-elected >> appointees, no board at all. But you are correct in that we really > > If this was true, why then was the discussion all against my proposal? I wasn't around for that discussion, so I have no history or context about it.... but more importantly, I don't see what good or use will come out of dredging it up out of the past again. Whoever originated the proposal: kudos to you. We've made progress, and we have a board - let's move forward rather than looking back. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From stevel at sun.com Tue Sep 25 08:32:28 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 08:32:28 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> References: <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <46F92A0C.9090706@sun.com> Joerg Schilling wrote: > Alan DuBoff <alan.duboff at sun.com> wrote: > >> I was talking with John Plocher today, and one of the things that we both >> agreed with is that we need some way for the community members to >> understand where and how they can participate, so that we can grow the >> community as a whole. > > I believe there are two problems: missing visibility of the OGB and missing > power of the community. Why do you perceive the former to be a problem? Visibility of the OGB is a double-edged sword. Yes, we want people to know there is an OGB so that they know there is a path of appeals and such; but we don't want the OGB to be so visible to the point that it is "in-your-face"[1]. The OGB has taken the view that the community is to be self-governing. > Another big problem is burocracy that makes it hard to contribute and that > makes progress slow and this cannot be changed by the OGB. Perhaps. Perhaps not. The topic I suggested for the summit is to identify specific barriers to contribution, so I'd like to hear more details (either now by email, or in a couple of weeks in person) about what specific barriers you see so we can figure out how to break them down. > As burocracy is a self serving and self fostering system, only someone from > outside that has the needed power may change it.The only such person I am > currently aware is Ian Murdock. What is the criteria for being considered "outside"? While Ian clearly has exhibited past leadership, it's not clear to me that he is "outside" (I'm making the assumption that "outside" means external to Sun, apologies if I'm wrong...) or that he has the "needed power." I would argue that people like yourself who have been in the community longer would have just as much credibility and/or ability. cheers, steve [1] http://www.thefreedictionary.com/in-your-face for non-English speakers who may be unfamiliar with the phrase -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From John.Plocher at Sun.COM Tue Sep 25 09:01:40 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 25 Sep 2007 09:01:40 -0700 Subject: [opensolaris-summit] Extended participation... In-Reply-To: <Pine.GSO.4.64.0709241800240.26931@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> Message-ID: <46F930E4.8090708@Sun.Com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Jim Grisanzio wrote: > >> That's fine. I generally agree. But we need the doc writers too. And the >> users. And the marketers. And the program managers. And the students. >> And the professors. And the admins. And the app developers. And the >> [insert whoever else wants to be a part of the community here]. > > I agree, we need everyone and everyone should be welcome. I've been trolling around looking at other FOSS communities, and have noticed that many of them make a distinction between the sites devoted to developers -vs- others. On my mind when I was poking at these sites was AlanB's comment on website-discuss that it was extremely important for all the parts of the OS.o extended site to share the exact same look and feel; however, I didn't see this behavior on these sites. They certainly shared some of the same branding (icons, color schemes...), so they were easily identifiable as being related, but (probably because changeable skins and themes were one of the project's features) there wasn't a slavish obsession with consistency. And it didn't seem to matter. The Gallery photo archive, the E107 CMS, Mozilla/Firefox, Wordpress, etc, each have several "related" web sites, each with a slightly different focus. While some (like e107) seem to have taken this to extremes, the idea that one site does not have to fit all seems to be a good one. (e.g., e107.org, e107coders.org, plugins.e107.org, e107themes.com, e107tutorials.com, e107faq.org ...) All these projects have a similar organization: Their main site is entirely devoted to facilitating downloads, installs and upgrades. (e.g., http://wordpress.org/ ) Once you have installed their particular thing, you get exposed to the next level - themes and skins, plugins and add-on customizations and the like. There were *many* competing sites for these additions, and, for the most part, they all seem to work well together. (e.g., http://www.google.com/search?q=wordpress+themes&aq=t ) You also start getting exposed to the beginnings of "here is how it works under the hood", so you can customize things yourself. Of course this all leads to the developer site(s), where you find the people who are developing the project, skins, themes and plugins... (e.g., http://codex.gallery2.org/Gallery2:Themes ) My point is simply that we need to figure out how to make Alan's website efforts a success in the short term so that we all can move forward with the goal of creating opportunities for others to contribute. Not just in putbacks to ON, but with wiki content, webmastering, repository creation and hosting, doc writing and editing, ARC, Design and Code reviewing and the like. -John From Joerg.Schilling at fokus.fraunhofer.de Tue Sep 25 09:42:48 2007 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Tue, 25 Sep 2007 18:42:48 +0200 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070925164008.GA277983@sun.com> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> Message-ID: <46f93a88.q8BIaA3GH5/e0PGr%Joerg.Schilling@fokus.fraunhofer.de> Keith M Wesolowski <Keith.Wesolowski at sun.com> wrote: > > I believe there are two problems: missing visibility of the OGB and missing > > power of the community. > > Are you concerned about a lack of *transparency* or a lack of > *visibility*? I don't believe the OGB needs to be more visible; if Ok, it is missing visibility. > > Another big problem is burocracy that makes it hard to contribute and that > > makes progress slow and this cannot be changed by the OGB. > > It's not bureaucracy (why does everyone think there's so much of it? > Do you know what the word actually means?) but a lack of adequate As I know what it means, I used it. > > As burocracy is a self serving and self fostering system, only someone from > > outside that has the needed power may change it.The only such person I am > > currently aware is Ian Murdock. > > You're joking, right? No, but it seems that you did not understand me. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From James.Walker at Sun.COM Tue Sep 25 10:35:21 2007 From: James.Walker at Sun.COM (Jim Walker) Date: Tue, 25 Sep 2007 11:35:21 -0600 Subject: [opensolaris-summit] Minutes of the 9/18 Summit Meetings In-Reply-To: <46F70F60.662239E7@nrubsig.org> References: <46F065A2.6010006@sun.com> <1190193394.22489.73.camel@haiiro> <b9c544f0709190630j73aca9f6n1ff5ab40c7b89453@mail.gmail.com> <46F1478E.1080103@Sun.Com> <46F15E8A.70407@sun.com> <46F70F60.662239E7@nrubsig.org> Message-ID: <46F946D9.7060704@sun.com> Roland Mainz wrote: > Erm, AFAIK the test farm is non-interactive (e.g. you can't do something > like "login and hack the kernel to death"), right ? If you mean the test farm we are setting up for OpenSolaris, then it is fully "interactive". It allows you to get access to a nice new test machine that is bare metal with an os build, that you can use to test your opensolaris software. You get root access, so you can hack away. If you want early access to it, signup to be a beta tester: http://opensolaris.org/os/community/testing/testfarm/ > The idea of the five machines was to have one "central hub" which is > user-land only (e.g. nooone except two or three people have "root" > access) which handles SunRay services, home dirs (which are exported to > the four other machines), jumpstart and has some development tools > (compiler, OS/Net build tools etc.) installed and then 2x x86 and 2x > SPARC machines to boot custom kernels from the server on demand. This > stuff is only thought to give people the opportunity to stick their > heads together and try some stuff without having to improvise something > with laptops (or worse) ... Since it looks like the network will be good, I plan on having some test farm machines available remotely for the summit, in addition to whatever we have locally. That said, I agree with many that at this stage in organization development, the summit should focus on discussions that inform the community on where we are at, and where we would like to go, which have the effect of fostering contributions to the overall strategy of OpenSolaris. But, we should have time during mornings, evenings and breaks to work specific hardware and software issues and do installs. And, what you map out above looks good. I don't know who is providing this and what the limits are though. Some virtualization on the machines may make them go farther. Cheers, Jim -- Jim Walker, http://blogs.sun.com/jwalker Sun Microsystems, Software, Solaris QE x77744, 500 Eldorado Blvd, Broomfield CO 80021 From Darren.Reed at Sun.COM Tue Sep 25 12:14:28 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 12:14:28 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <20070925164008.GA277983@sun.com> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> Message-ID: <46F95E14.4000008@Sun.COM> Keith M Wesolowski wrote: >... >It's not bureaucracy (why does everyone think there's so much of it? >Do you know what the word actually means?) but a lack of adequate >infrastructure that makes contributing such a pain. I blame Sun for >this; its insistence on maintaining control of the infrastructure and >its tortuous legal and policy constraints are a main reason more >progress has not been made. That said, anyone could put together a >proposal to move all this outside Sun's control and accelerate the >process, yet no one has. It's unclear whether that's because of >laziness, lack of interest, or lack of means. > > There are a number of things that need to happen here and they've been raised before (at least by myself): 1) move the opensolaris machinery (web server, repository, etc) away from being owned by Sun; 2) to achieve (1), create an opensolaris entity that can own pieces of hardware, etc; 3) in order to fund both (1) and (2), make an opensolaris entity that is a non profit organisation. If we were a true bona-fide opensource project and with a bit of luck, we can get a server into a back-bone ISP's colocation facility using 1RU or less of space. To what extent Sun is required to cooperate here, I don't know. But, now that OpenSolaris is open, if there were enough interested parties we could build a new distribution that we all worked on (with a new name) with the model *we* want and just leave OpenSolaris in Sun's hands. We can still pull in code from OS.o, after it gets mirrored there from nevada. Heck, maybe we participants should just do that anyway, rather than wait for Sun to work out what it wants to do with OS.o? Darren From brandorr at opensolaris.org Tue Sep 25 13:44:47 2007 From: brandorr at opensolaris.org (Brandorr) Date: Tue, 25 Sep 2007 16:44:47 -0400 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F95E14.4000008@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> Message-ID: <5b5090780709251344u5b7221efo1d3b68965a97f53b@mail.gmail.com> On 9/25/07, Darren Reed <Darren.Reed at sun.com> wrote: > Heck, maybe we participants should just do that anyway, rather > than wait for Sun to work out what it wants to do with OS.o? While I understand your frustration at a low level of outside contribution, I don't think Sun is to be blamed. I think it stems from the fact that Solaris is a mature, formerly proprietary, product. Many users of Solaris today don't give a care about any notions of community of contribution. (So long as the get what their support contracts cover) We here at OpenSolaris.org are currently in the process of building a community. This community currently exists mostly of Sun employees, and a few brave new worlders, and a few excited Solaris customers. We are in a bit of a chicken and egg situation. What we need right now is more outside manpower, to build the infrastructure to make it easier for outside contributers to contribute. Right now, one way we can address this, is to do outreach. Currently this is on the agenda for our next user group meeting: Get involved: ------------------ - Mentoring - We are looking at starting a new mentoring program. We are looking for Seniors that would be interested in mentoring Juniors. (Developers and sysadmins). Please email Brian Gupta (brandorr at opensolaris.org) if you are interested. - The OpenSolaris.org Systems administration Community Group is looking to expand membership. syadmin-discuss-subscribe at opensolaris.org - Volunteering to help build a new OpenSolaris user portal, as well as an NYOSUG.org website. (email Brian Gupta (brandorr at opensolaris.org) if you are interested in either.) - We will be launching/announcing a virtual development lab next month, that will allow anyone who is interested in developing OpenSolaris or porting open source apps to OpenSolaris, to gain free access to VMWare ESX Virtual Machines. Let Brian Gupta (brandorr at opensolaris.org) know if you are interested in either helping administer ESX, or if you want access to a VM. (It's not setup yet). - We can use help planning these UG meetings. Mail Isaac or Brian Gupta "Isaac Rozenfeld - NEW YORK" <Isaac.Rozenfeld at sun.com>, Brian Gupta (brandorr at opensolaris.org) - Brian Gupta (brandorr at opensolaris.org) is going to the first OpenSolaris Developers Summit, and will have more to report next month. (http://www.genunix.org/wiki/index.php?title=OpenSolaris_Summit.) - Brian Gupta (brandorr at opensolaris.org) assembled the following new user FAQ: assembling: http://www.genunix.org/wiki/index.php?title=FAQ_Prototype What I would ask each contributing member of OpenSolaris.org, is to find at least one other person that has the ability to contribute, and walk them through their first contribution. (We'd double the number of contributers over night. Also, UserGroups should be actively advertising outside of their own users. e.g. - Reach out to Linux and BSD user groups, as well as local UNIGROUP, Lisa and SAGE chapters. The goal being to increase the number of attendees to your meetings. Basically, I feel that by working around the barriers for now, we will be able to grow the community enough to remove them down the road. (With the added manpower). > Darren > > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ From stevel at sun.com Tue Sep 25 14:48:49 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 14:48:49 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F95E14.4000008@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> Message-ID: <46F98241.8030207@sun.com> Darren Reed wrote: > Keith M Wesolowski wrote: > >> ... >> It's not bureaucracy (why does everyone think there's so much of it? >> Do you know what the word actually means?) but a lack of adequate >> infrastructure that makes contributing such a pain. I blame Sun for >> this; its insistence on maintaining control of the infrastructure and >> its tortuous legal and policy constraints are a main reason more >> progress has not been made. That said, anyone could put together a >> proposal to move all this outside Sun's control and accelerate the >> process, yet no one has. It's unclear whether that's because of >> laziness, lack of interest, or lack of means. >> >> > > There are a number of things that need to happen here and > they've been raised before (at least by myself): > > 1) move the opensolaris machinery (web server, repository, > etc) away from being owned by Sun; While this is a noteworthy long-term goal, I fail to see what problems this solves in the short-term. Who will own them? Who will maintain them? Who will be paid to maintain them? > 2) to achieve (1), create an opensolaris entity that can own > pieces of hardware, etc; > 3) in order to fund both (1) and (2), make an opensolaris > entity that is a non profit organisation. okay, that answers my questions from #1... but... > If we were a true bona-fide opensource project and with a bit > of luck, we can get a server into a back-bone ISP's colocation > facility using 1RU or less of space. There are plenty of open source projects who succeed massively without the need for non-profit foundations. > To what extent Sun is required to cooperate here, I don't know. > > But, now that OpenSolaris is open, if there were enough interested > parties we could build a new distribution that we all worked on > (with a new name) with the model *we* want and just leave > OpenSolaris in Sun's hands. We can still pull in code from OS.o, > after it gets mirrored there from nevada. > > Heck, maybe we participants should just do that anyway, rather > than wait for Sun to work out what it wants to do with OS.o? That's certainly a prerogative that anyone has... after all, the code is open. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From stevel at sun.com Tue Sep 25 15:04:24 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 15:04:24 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F984CE.3000508@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <46F98241.8030207@sun.com> <46F984CE.3000508@Sun.COM> Message-ID: <46F985E8.7020100@sun.com> James C. McPherson wrote: > Stephen Lau wrote: >> Darren Reed wrote: >>> Keith M Wesolowski wrote: >>> >>>> ... >>>> It's not bureaucracy (why does everyone think there's so much of it? >>>> Do you know what the word actually means?) but a lack of adequate >>>> infrastructure that makes contributing such a pain. I blame Sun for >>>> this; its insistence on maintaining control of the infrastructure and >>>> its tortuous legal and policy constraints are a main reason more >>>> progress has not been made. That said, anyone could put together a >>>> proposal to move all this outside Sun's control and accelerate the >>>> process, yet no one has. It's unclear whether that's because of >>>> laziness, lack of interest, or lack of means. >>>> >>>> >>> There are a number of things that need to happen here and >>> they've been raised before (at least by myself): >>> >>> 1) move the opensolaris machinery (web server, repository, >>> etc) away from being owned by Sun; >> >> While this is a noteworthy long-term goal, I fail to see what problems >> this solves in the short-term. >> >> Who will own them? Who will maintain them? Who will be paid to >> maintain them? > > > Who owns, runs and maintains kernel.org? kernel.org isn't much more than a download site. > Dare I suggest slashdot, sourceforge and even OSNews as > examples here? I don't believe /. & osnews establish communities. Sourceforge is probably a more relevant example, but Sourceforge creates micro-communities with hardly any attempt to have cross-project communication. opensolaris.org is attempting to be everything (or at least all-things OpenSolaris related) to everyone, and I would argue that that is probably its fallacy. There is no *single* Linux community site, why should there need to be a single opensolaris.org site? Frankly, I think it'd be good for the health of our community if people did take stuff elsewhere and do interesting things on their own. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From John.Plocher at Sun.COM Tue Sep 25 15:09:55 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 25 Sep 2007 15:09:55 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F983E4.4080804@Sun.COM> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> Message-ID: <46F98733.9020701@Sun.Com> Darren Reed wrote: > ie. Instead of trying to get part of bugster out of Sun, create > a new bug reporting database The dealbreaker for me is that there is no easy way to initially populate this new database with the hundreds of thousands of open bugs/rfes already in bugster, which effectively means that nobody without bugster access will be able to fix/manage bugs, and if they have bugster access, they have no need to use something else... Catch-22. See Machiavelli for more insight. -John From Darren.Reed at Sun.COM Tue Sep 25 15:12:01 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 15:12:01 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F98241.8030207@sun.com> References: <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <46F98241.8030207@sun.com> Message-ID: <46F987B1.4050601@Sun.COM> Stephen Lau wrote: >Darren Reed wrote: > > > ... > >>If we were a true bona-fide opensource project and with a bit >>of luck, we can get a server into a back-bone ISP's colocation >>facility using 1RU or less of space. >> >> > >There are plenty of open source projects who succeed massively without >the need for non-profit foundations. > > True. Outside of the BSDs and a few large proejcts, I'm not sure of any that have pursued this path (strangely, it doesn't come for free.) I see OpenSolaris as being big enough that it is worth considering. It depends on whether the goal of OpenSolaris is to become "business" in the sense that RedHat and others have or the foundations for others, such as BSD does. In this vein, Indiana is a different approach again. It is shaping up to be the first distribution to also come from the base platform vendor. Given that, it is unclear whether becoming non-profit is in the best interests of OpenSolaris. But I will add this: if it does become non-profit then it becomes more attractive for 3rd party people who want to use OpenSolaris to contribute financially to its survival. That should be a carrot hanging from a stick for Sun :) Darren From John.Plocher at Sun.COM Tue Sep 25 13:44:24 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 25 Sep 2007 13:44:24 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> Message-ID: <46F97328.90502@Sun.Com> Peter Tribble wrote: > So how do we encourage communities and individuals to take up the > reins of governance? Not to be flippant, but why would they want to? That is, what problems are they currently having that could be solved by their spending energy on self-governance? My perspective is that the answer is that there are lots of serious problems, none of which can effectively be solved thru self-governance alone. Problem 1: The structure of the meta-community echoes the confusion in people's minds of what exactly OpenSolaris is and should be. Because people are not in agreement with what it is, there is no consensus on how it should be structured. Problem 2: The observable artifacts of the community are still locked away from direct community manipulation: The source tree, the bug database, the web site, the mailing list infrastructure, the ON C-Team and even the ARCs are /ALL/ solely maintainable by a set of under-resourced, under-appreciated and constantly dwindling group of Sun employees. Solve these problems and the community of developers will grow. Ignore them, and it will stagnate and die. -John From Darren.Reed at Sun.COM Tue Sep 25 14:55:48 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 14:55:48 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F97328.90502@Sun.Com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> Message-ID: <46F983E4.4080804@Sun.COM> John Plocher wrote: > Peter Tribble wrote: > >> So how do we encourage communities and individuals to take up the >> reins of governance? > > > Not to be flippant, but why would they want to? > > That is, what problems are they currently having that could > be solved by their spending energy on self-governance? > > My perspective is that the answer is that there are lots of > serious problems, none of which can effectively be solved thru > self-governance alone. > > Problem 1: The structure of the meta-community echoes the > confusion in people's minds of what exactly OpenSolaris is > and should be. Because people are not in agreement with > what it is, there is no consensus on how it should be > structured. > > Problem 2: The observable artifacts of the community are > still locked away from direct community manipulation: The > source tree, the bug database, the web site, the mailing > list infrastructure, the ON C-Team and even the ARCs are > /ALL/ solely maintainable by a set of under-resourced, > under-appreciated and constantly dwindling group of Sun > employees. > > Solve these problems and the community of developers will > grow. Ignore them, and it will stagnate and die. So a question that needs to be asked is, considering where we want to go, is it easier to just build something "new" and let Sun work out how it wants to join up with that, rather than to try and morph what we have into something else? ie. Instead of trying to get part of bugster out of Sun, create a new bug reporting database for OpenSolaris, independant of Sun's internal database, and so on. Darren From alan.coopersmith at sun.com Tue Sep 25 15:05:07 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Tue, 25 Sep 2007 15:05:07 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F983E4.4080804@Sun.COM> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> Message-ID: <46F98613.1060108@sun.com> Darren Reed wrote: > ie. Instead of trying to get part of bugster out of Sun, create > a new bug reporting database for OpenSolaris, independant > of Sun's internal database, That's already the plan, hardly anything new there - the problem isn't deciding to break new ground, it's finding people who want to do more than just complain about the broken bug problem and are willing to put in time to figure out what new system should be used and get it going. If you have the time, please go join the Defect Tracking System (DTS) discussions on tools-discuss. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From stevel at sun.com Tue Sep 25 15:15:23 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 15:15:23 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F983E4.4080804@Sun.COM> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> Message-ID: <46F9887B.5010604@sun.com> Darren Reed wrote: > John Plocher wrote: > >> Peter Tribble wrote: >> >>> So how do we encourage communities and individuals to take up the >>> reins of governance? >> >> >> Not to be flippant, but why would they want to? >> >> That is, what problems are they currently having that could >> be solved by their spending energy on self-governance? >> >> My perspective is that the answer is that there are lots of >> serious problems, none of which can effectively be solved thru >> self-governance alone. >> >> Problem 1: The structure of the meta-community echoes the >> confusion in people's minds of what exactly OpenSolaris is >> and should be. Because people are not in agreement with >> what it is, there is no consensus on how it should be >> structured. >> >> Problem 2: The observable artifacts of the community are >> still locked away from direct community manipulation: The >> source tree, the bug database, the web site, the mailing >> list infrastructure, the ON C-Team and even the ARCs are >> /ALL/ solely maintainable by a set of under-resourced, >> under-appreciated and constantly dwindling group of Sun >> employees. >> >> Solve these problems and the community of developers will >> grow. Ignore them, and it will stagnate and die. > > > So a question that needs to be asked is, considering where > we want to go, is it easier to just build something "new" and > let Sun work out how it wants to join up with that, rather than > to try and morph what we have into something else? > > ie. Instead of trying to get part of bugster out of Sun, create > a new bug reporting database for OpenSolaris, independant > of Sun's internal database, and so on. We've got an existing DTS evaluation going on - it's floundering not because Sun is holding anything back, but because nobody is making the effort to evaluate any of the DTS (or at least they haven't mentioned it publicly). -steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From alan.duboff at sun.com Tue Sep 25 16:08:37 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 16:08:37 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7myvapsae.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Richard Lowe wrote: >> I'm just trying to remind everyone that people /are/ participating >> and we /are/ making progress. > > Slowly, when it comes to development. Very, very slowly. We're > making progress as far as mktg (I guess), and various other things. > Open development? Not even close. > >> I also think it's better to build on what has already been done >> rather than tearing each other down. > > Sure, but pretending the problems don't exist gets us nowhere. It > seems nearly everything I read about opensolaris tells me how peachy > things are, which is, of course, largely crap. Rich, I don't think it's completly fair to say it is not open, not do I think we can just say it hasn't evolved. I do appreciate you giving your honest view though. Please see my blog entry: http://blogs.sun.com/aland/entry/one_package_per_community_member -- Alan DuBoff - Solaris x86 IHV/OEM Group From stevel at sun.com Tue Sep 25 16:28:33 2007 From: stevel at sun.com (Stephen Lau) Date: Tue, 25 Sep 2007 16:28:33 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> Message-ID: <46F999A1.7090900@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Richard Lowe wrote: > >>> I'm just trying to remind everyone that people /are/ participating >>> and we /are/ making progress. >> >> Slowly, when it comes to development. Very, very slowly. We're >> making progress as far as mktg (I guess), and various other things. >> Open development? Not even close. >> >>> I also think it's better to build on what has already been done >>> rather than tearing each other down. >> >> Sure, but pretending the problems don't exist gets us nowhere. It >> seems nearly everything I read about opensolaris tells me how peachy >> things are, which is, of course, largely crap. > > Rich, > > I don't think it's completly fair to say it is not open, not do I think > we can just say it hasn't evolved. I do appreciate you giving your > honest view though. > > Please see my blog entry: > > http://blogs.sun.com/aland/entry/one_package_per_community_member One major problem that I read is: > If we look at the kernel components, do we even want to make it easier for members to contribute? I don't think so, in fact I believe it should be more difficult. To date it has been mostly Sun folks that have contributed the bulk of code in that space, and that should be expected. The engineers that have managed the code over the years are still managing it to date, and this is a good thing from my perspective. They know the code better and have worked on it more than others, it absolutely makes sense for them to do as they have done, wether it's open or not, it's a must to keep those sources well maintained. While I agree with the idea that it should be well maintained, I think perhaps your wording was poor. It came across (to me anyway) that you don't believe non-Sun members should be allow to contribute kernel code? cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From swalker at opensolaris.org Tue Sep 25 16:40:04 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:40:04 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F95E14.4000008@Sun.COM> References: <46F31335.6060700@Sun.COM> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> Message-ID: <b9c544f0709251640n63a77ca6rae158d89d9acffc9@mail.gmail.com> On 25/09/2007, Darren Reed <Darren.Reed at sun.com> wrote: > Keith M Wesolowski wrote: > > >... > >It's not bureaucracy (why does everyone think there's so much of it? > >Do you know what the word actually means?) but a lack of adequate > >infrastructure that makes contributing such a pain. I blame Sun for > >this; its insistence on maintaining control of the infrastructure and > >its tortuous legal and policy constraints are a main reason more > >progress has not been made. That said, anyone could put together a > >proposal to move all this outside Sun's control and accelerate the > >process, yet no one has. It's unclear whether that's because of > >laziness, lack of interest, or lack of means. > > > > > > There are a number of things that need to happen here and > they've been raised before (at least by myself): > > 1) move the opensolaris machinery (web server, repository, > etc) away from being owned by Sun; Sounds great, but who pays for it? What benefit would it have for us? What's stopping someone from starting up a fork and doing this themselves right now? > 2) to achieve (1), create an opensolaris entity that can own > pieces of hardware, etc; > > 3) in order to fund both (1) and (2), make an opensolaris > entity that is a non profit organisation. You talk about project success all the time, but the things you talk about are not the things I think it needs to succeed right now. From my perspective as a indepdenent, non-Sun affiiliated person, your proposals would just burn energy on things now that is better directed at more important issues. I don't think the issues that are facing us right now would be solved by what you are proposing. However, they are all great goals to have for a future date, despite the challenges in achieving them. > If we were a true bona-fide opensource project and with a bit > of luck, we can get a server into a back-bone ISP's colocation > facility using 1RU or less of space. "bona-fide"...interesting terminology. What exactly is the "accepted definition" of "bona-fide opensource project". > But, now that OpenSolaris is open, if there were enough interested > parties we could build a new distribution that we all worked on > (with a new name) with the model *we* want and just leave > OpenSolaris in Sun's hands. We can still pull in code from OS.o, > after it gets mirrored there from nevada. > > Heck, maybe we participants should just do that anyway, rather > than wait for Sun to work out what it wants to do with OS.o? What's stopping someone from doing this now? Personally, I don't see a community division as the "Right Thing To Do" at the moment. I think divided efforts would hurt us even more. I'm fairly frustrated with the progress in certain areas, but I'm not ready to "throw the baby out with the bath water" just yet... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From swalker at opensolaris.org Tue Sep 25 16:49:58 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:49:58 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F985E8.7020100@sun.com> References: <46F31335.6060700@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <46F98241.8030207@sun.com> <46F984CE.3000508@Sun.COM> <46F985E8.7020100@sun.com> Message-ID: <b9c544f0709251649n58c65762rbd8704bcd2e6263c@mail.gmail.com> On 25/09/2007, Stephen Lau <stevel at sun.com> wrote: > James C. McPherson wrote: > > Stephen Lau wrote: > >> Darren Reed wrote: > >>> Keith M Wesolowski wrote: > >>> > >>>> ... > >>>> It's not bureaucracy (why does everyone think there's so much of it? > >>>> Do you know what the word actually means?) but a lack of adequate > >>>> infrastructure that makes contributing such a pain. I blame Sun for > >>>> this; its insistence on maintaining control of the infrastructure and > >>>> its tortuous legal and policy constraints are a main reason more > >>>> progress has not been made. That said, anyone could put together a > >>>> proposal to move all this outside Sun's control and accelerate the > >>>> process, yet no one has. It's unclear whether that's because of > >>>> laziness, lack of interest, or lack of means. > >>>> > >>>> > >>> There are a number of things that need to happen here and > >>> they've been raised before (at least by myself): > >>> > >>> 1) move the opensolaris machinery (web server, repository, > >>> etc) away from being owned by Sun; > >> > >> While this is a noteworthy long-term goal, I fail to see what problems > >> this solves in the short-term. > >> > >> Who will own them? Who will maintain them? Who will be paid to > >> maintain them? > > > > > > Who owns, runs and maintains kernel.org? > > kernel.org isn't much more than a download site. > > > Dare I suggest slashdot, sourceforge and even OSNews as > > examples here? > > I don't believe /. & osnews establish communities. > Sourceforge is probably a more relevant example, but Sourceforge creates > micro-communities with hardly any attempt to have cross-project > communication. > > opensolaris.org is attempting to be everything (or at least all-things > OpenSolaris related) to everyone, and I would argue that that is > probably its fallacy. > > There is no *single* Linux community site, why should there need to be a > single opensolaris.org site? Frankly, I think it'd be good for the > health of our community if people did take stuff elsewhere and do > interesting things on their own. Isn't that part of the "Linux problem" though? Hasn't the "Linux world" (which doesn't have a "unified" place the community can call home) fractured into a "bajillion" communities, websites, and distributions leading to uncoordinated, wasted efforts? I think Ubuntu proved how valuable having a centralized user community, bug tracking, repository, etc. can be. It's the same model that many other projects have as well (FreeBSD, etc.). -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From swalker at opensolaris.org Tue Sep 25 16:51:22 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:51:22 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F98733.9020701@Sun.Com> References: <46F2FE5F.3070508@sun.com> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F98733.9020701@Sun.Com> Message-ID: <b9c544f0709251651t321edf8ct604fd6719adc2659@mail.gmail.com> On 25/09/2007, John Plocher <John.Plocher at sun.com> wrote: > Darren Reed wrote: > > ie. Instead of trying to get part of bugster out of Sun, create > > a new bug reporting database > > The dealbreaker for me is that there is no easy way to initially > populate this new database with the hundreds of thousands of open > bugs/rfes already in bugster, which effectively means that nobody > without bugster access will be able to fix/manage bugs, and if they > have bugster access, they have no need to use something else... Right, but we're always going to have that problem. Sun has proprietary, confidential data in there, and there's no way practical way to scrub those millions of bits of data properly. We may just have to pay this price and move on...the important thing will be whether we can deal with new bugs in both systems going forward. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Darren.Reed at Sun.COM Tue Sep 25 17:07:17 2007 From: Darren.Reed at Sun.COM (Darren Reed) Date: Tue, 25 Sep 2007 17:07:17 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F98733.9020701@Sun.Com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F98733.9020701@Sun.Com> Message-ID: <46F9A2B5.5000509@Sun.COM> John Plocher wrote: > Darren Reed wrote: > >> ie. Instead of trying to get part of bugster out of Sun, create >> a new bug reporting database > > > The dealbreaker for me is that there is no easy way to initially > populate this new database with the hundreds of thousands of open > bugs/rfes already in bugster, which effectively means that nobody > without bugster access will be able to fix/manage bugs, and if they > have bugster access, they have no need to use something else... So OpenSolaris starts with a clean slate. What's wrong with that? :) This can only work in OpenSolaris' favour: it'll have less bugs than Solaris ;-) And if there is a desire to import bugster information later, just reserve some number space for that and start at some other place for OpenSolaris. Darren From alan.duboff at sun.com Tue Sep 25 18:07:35 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 18:07:35 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F999A1.7090900@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> Message-ID: <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Stephen Lau wrote: > One major problem that I read is: >> If we look at the kernel components, do we even want to make it easier for >> members to contribute? I don't think so, in fact I believe it should be >> more difficult. To date it has been mostly Sun folks that have contributed >> the bulk of code in that space, and that should be expected. The engineers >> that have managed the code over the years are still managing it to date, >> and this is a good thing from my perspective. They know the code better >> and have worked on it more than others, it absolutely makes sense for them >> to do as they have done, wether it's open or not, it's a must to keep those >> sources well maintained. > > While I agree with the idea that it should be well maintained, I think > perhaps your wording was poor. It came across (to me anyway) that you don't > believe non-Sun members should be allow to contribute kernel code? Steve, I will edit my blog to reflect what I really meant, that it really is not community or not, it's that the people that understand and know the code should be the ones that maintain it. This is so today with the kernel engineers that are doing just that. In the future, when community people come up to speed, there is no reason they would not fall into the same category. More what I meant was that we just do not see the skills and knowledge in the community to work with and maintain the code in the same capacity. I'm sure that will change over time. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Garrett.Damore at Sun.COM Tue Sep 25 20:45:35 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 20:45:35 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> Message-ID: <46F9D5DF.7030502@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Stephen Lau wrote: > > >> One major problem that I read is: >> >>> If we look at the kernel components, do we even want to make it easier for >>> members to contribute? I don't think so, in fact I believe it should be >>> more difficult. To date it has been mostly Sun folks that have contributed >>> the bulk of code in that space, and that should be expected. The engineers >>> that have managed the code over the years are still managing it to date, >>> and this is a good thing from my perspective. They know the code better >>> and have worked on it more than others, it absolutely makes sense for them >>> to do as they have done, wether it's open or not, it's a must to keep those >>> sources well maintained. >>> >> While I agree with the idea that it should be well maintained, I think >> perhaps your wording was poor. It came across (to me anyway) that you don't >> believe non-Sun members should be allow to contribute kernel code? >> > > Steve, > > I will edit my blog to reflect what I really meant, that it really is not > community or not, it's that the people that understand and know the code > should be the ones that maintain it. This is so today with the kernel > engineers that are doing just that. > > In the future, when community people come up to speed, there is no reason > they would not fall into the same category. More what I meant was that we > just do not see the skills and knowledge in the community to work with and > maintain the code in the same capacity. I'm sure that will change over > time. > Actually, I take pretty strong exception to that fact. Until recently I was such a community member. Other community members are themselves former Sun employees... And still others work at places like Tadpole, where a lot of real kernel work does in fact take place. (To the point that our current cardbus stack is largely derived from original Tadpole work!) Andrew Gallatin, at Myricom for example, has had a lot of positive impact and feedback into what we're doing in the kernel layer. There are undoubtedly many others. To suggest that the community lacks the skill set suggests either a) a level of elitism that I hope we don't exude to the rest of the world, or b) ignorance about what other folks are doing out there. Now, having said all that, we do indeed have a number of processes in place which I don't want want to go away... codereview, test validation, and RTI advocacy are all steps along the way to ensure that the overall quality of what is in Solaris (Open or otherwise) meets some stringent quality considerations. In fact, even the userland code we have in Solaris should meet these guidelines too, its just that (IMO) its easier to find less thorough advocates and code reviewers for userland bits of code. :-) -- Garrett > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From alan.duboff at sun.com Tue Sep 25 21:17:33 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 21:17:33 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F9D5DF.7030502@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> Message-ID: <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Garrett D'Amore wrote: > Actually, I take pretty strong exception to that fact. > > Until recently I was such a community member. And before that you worked at Sun. > Other community members are themselves former Sun employees... Yes, but do we see these folks flocking to contribute code back? Those people, as yourself, know the process and how it works. They have the knowledge and understanding, those are not the people I refer to. Those people as yourself are in a minority and I stated there are a few that do in fact have the skills. Again, you folks are in the minority. > To suggest that the community lacks the skill set suggests either a) a > level of elitism that I hope we don't exude to the rest of the world, or > b) ignorance about what other folks are doing out there. Whatever...split hairs if you like, fine. > Now, having said all that, we do indeed have a number of processes in > place which I don't want want to go away... codereview, test validation, > and RTI advocacy are all steps along the way to ensure that the overall > quality of what is in Solaris (Open or otherwise) meets some stringent > quality considerations. Do you think that folks should be allowed a putback if they don't completely meet the test validation process? -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Tue Sep 25 21:20:17 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 21:20:17 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x74phikrek.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <x74phikrek.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709252117420.7188@eagle.softorchestra.com> On Wed, 26 Sep 2007, Richard Lowe wrote: > For what it's worth, the update pissed me off more than the original > text, for pretty much the reasons you outline below. And why you can have your own blog to say what you like,;-) How would you change the process to work better today? > I most definitely don't want that to go away either, as you know. > Making external development easier is not about removing valuable > process, it's about getting that process to the point it works for > *everyone*. That's a large part of what I think we're making very > little progress on. Sun has provided the information on their process. Have you been able to understand it? How could they make better progress on that? -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Tue Sep 25 23:09:17 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Tue, 25 Sep 2007 23:09:17 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7zlzaj9lv.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> On Wed, 26 Sep 2007, Richard Lowe wrote: >>> Now, having said all that, we do indeed have a number of processes in >>> place which I don't want want to go away... codereview, test validation, >>> and RTI advocacy are all steps along the way to ensure that the overall >>> quality of what is in Solaris (Open or otherwise) meets some stringent >>> quality considerations. >> >> Do you think that folks should be allowed a putback if they don't >> completely meet the test validation process? >> > > Garrett totally didn't say that, he in fact suggested quite the > opposite, very strongly. Rich, Yeah, I know what he said, but sometimes people say something different than they actually do. Since I had been working with him recentely, I just wanted to get clarification from him on that point. I agree with you that he didn't say that, and suggested quite the opposite. I think I stand strongly on the current process and the test validation that has been put in place. Although there is some case to be argued that the test validation shouldn't be so rigorous, the current system in place is what it is, and has been put in place for a reason. When Sun changes that policy, it could change the putback process (at least for some components like device drivers). I was talking with Randy Fishel recently about this, and it's possible in the future that Sun could lighten up on the test validation for components that do not fall into their top tier, but that hasn't happened yet and the process for doing a putback requires quite a rigorous test validation. So far I haven't seen any suggestions from you, just criticism on the process Sun has put in place. Can you elaborate on what exactly Sun could do better? From my viewpoint Sun pushed their process outside the firewall. While it's probably not the easiest process to follow, I'm not sure how else they could have transitioned the code out to the community and still ensure that the process retains the quality they put in place. What can you suggest that would help that better? How would you retain the quality of Solaris as it has been, yet allow the community to participate in the development more so? Your previous comments about the development process not being open must relate to the ON sources that are currently out in the community, after all, there really is nothing more to OpenSolaris at this point aside from some packages on opensolaris.org. But those that have not been putback are not a part of any development process. Can you elaborate on what you would do different in regard to being more open in the development of the current sources? -- Alan DuBoff - Solaris x86 IHV/OEM Group From Jim.Grisanzio at Sun.COM Tue Sep 25 23:30:33 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Wed, 26 Sep 2007 15:30:33 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> Message-ID: <46F9FC89.2030306@sun.com> Peter Tribble wrote: > But does the community know that it's supposed to be governing itself? Many people do, but I'd say most do not. I'd also say that's normal. It's our job to educate people within each Community Group as we go along. Over time if we develop active leaders across the CGs the situation will improve. But as new people become involved, the process of education needs to be constant. Jim -- http://blogs.sun.com/jimgris From Garrett.Damore at Sun.COM Tue Sep 25 23:46:41 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 23:46:41 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> Message-ID: <46FA0051.2010800@sun.com> Alan DuBoff wrote: > On Wed, 26 Sep 2007, Richard Lowe wrote: > > >>>> Now, having said all that, we do indeed have a number of processes in >>>> place which I don't want want to go away... codereview, test validation, >>>> and RTI advocacy are all steps along the way to ensure that the overall >>>> quality of what is in Solaris (Open or otherwise) meets some stringent >>>> quality considerations. >>>> >>> Do you think that folks should be allowed a putback if they don't >>> completely meet the test validation process? >>> >>> >> Garrett totally didn't say that, he in fact suggested quite the >> opposite, very strongly. >> > > Rich, > > Yeah, I know what he said, but sometimes people say something different > than they actually do. Since I had been working with him recentely, I just > wanted to get clarification from him on that point. I agree with you that > For the background here, Alan and I had a disagreement about what the test matrix required when I putback my drivers (afe and mxfe) without performing a certain group of stress tests on all possible platform configurations, as I believed (and still do) that they were redundant when taken against the rest of the tests that had already been performed, and not strictly required for final putback. Anyone else who wants the particulars of the that particular event is welcome to contact me offline. Anyway, that small matter aside, I do believe that a reasonably complete amount of testing is required, and indeed, we have a number of test suites in place to help with that (though they are far from exhaustive.) -- Garrett From Garrett.Damore at Sun.COM Tue Sep 25 23:55:11 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Tue, 25 Sep 2007 23:55:11 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F9887B.5010604@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> Message-ID: <46FA024F.2020002@sun.com> Stephen Lau wrote: > Darren Reed wrote: > >> John Plocher wrote: >> >> >>> Peter Tribble wrote: >>> >>> >>>> So how do we encourage communities and individuals to take up the >>>> reins of governance? >>>> >>> Not to be flippant, but why would they want to? >>> >>> That is, what problems are they currently having that could >>> be solved by their spending energy on self-governance? >>> >>> My perspective is that the answer is that there are lots of >>> serious problems, none of which can effectively be solved thru >>> self-governance alone. >>> >>> Problem 1: The structure of the meta-community echoes the >>> confusion in people's minds of what exactly OpenSolaris is >>> and should be. Because people are not in agreement with >>> what it is, there is no consensus on how it should be >>> structured. >>> >>> Problem 2: The observable artifacts of the community are >>> still locked away from direct community manipulation: The >>> source tree, the bug database, the web site, the mailing >>> list infrastructure, the ON C-Team and even the ARCs are >>> /ALL/ solely maintainable by a set of under-resourced, >>> under-appreciated and constantly dwindling group of Sun >>> employees. >>> >>> Solve these problems and the community of developers will >>> grow. Ignore them, and it will stagnate and die. >>> >> So a question that needs to be asked is, considering where >> we want to go, is it easier to just build something "new" and >> let Sun work out how it wants to join up with that, rather than >> to try and morph what we have into something else? >> >> ie. Instead of trying to get part of bugster out of Sun, create >> a new bug reporting database for OpenSolaris, independant >> of Sun's internal database, and so on. >> > > We've got an existing DTS evaluation going on - it's floundering not > because Sun is holding anything back, but because nobody is making the > effort to evaluate any of the DTS (or at least they haven't mentioned it > publicly). > There is another real problem here. I think the only people who are qualified to "fully" evaluate the bug database are those who are also Sun employees. Why? Because only they really know what the current bugster contains, and what features are needed, and what aren't. Of course, all the Sun employees I know are up to their neck in other work. And it isn't entirely clear that the Sun folks who should be paying for an evaluation of a new DTS have any motivation to do so; after all, bugster works just fine for *their* purposes. Speaking for myself, I don't have the time to do a full evaluation of a DTS. And, I'm not really qualified as a DTS administrator. *HOWEVER*, I know that I would be willing to play in a sandbox that someone else built. And even use it to track real project bugs. I *suspect* that the same is true of other people. So, if someone is energetic enough (a big if, I know) to set up the demo sandbox, then maybe an announcement in a broader context inviting others to participate in the test (as developers, not DTS administrators!) would be helpful. -- Garrett From Jim.Grisanzio at Sun.COM Wed Sep 26 00:06:18 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Wed, 26 Sep 2007 16:06:18 +0900 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7myvapsae.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> Message-ID: <46FA04EA.7060503@sun.com> Richard Lowe wrote: > Sure, but pretending the problems don't exist gets us nowhere. It > seems nearly everything I read about opensolaris tells me how peachy > things are, which is, of course, largely crap. Who's saying nearly everything is peachy? The OpenSolaris lists are filled with strong arguments about all the problems we have to deal with. I don't get the impression things are peachy from a development perspective at all. Things are also not largely crap. I don't support either extreme. Jim -- Jim Grisanzio http://blogs.sun.com/jimgris From alan.duboff at sun.com Wed Sep 26 00:37:45 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 00:37:45 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA0051.2010800@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> Message-ID: <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Garrett D'Amore wrote: > For the background here Wasn't trying to split hairs with you, but you were pretty strong in your statement...<LOL!> > Anyway, that small matter aside, I do believe that a reasonably complete > amount of testing is required, and indeed, we have a number of test suites in > place to help with that (though they are far from exhaustive.) I know you do, and I believe your intentions are quite good. As a case in point, I know you try to follow the DDI strictly. Sun internal drivers seem to have been some of the worst offenders of sticking to the DDI. FWIW, I think it would be worth considering getting drivers put back that didn't meet the rigorous testing, at least in some form of more open system like Indiana is rumored to be. It is still not clear to me what exactly Indiana will include. In regards to skills and/or the offense you took with my comments, which I removed from my blog and made a slight reference change to the previous, let me say this. I have seen several putbacks with massive amounts of files in the short time I've been at Sun. While not common, it's not a surprise to see putbacks that total thousands of files. To work with a set of source and manage such a putback is not for the average developer. It should also be of no surprise that most all developers who have in depth knowledge of Solaris have some ties back to Sun. As a case in point, there are people that could manage putbacks with thousands of files to OpenSolaris that work at VMWare, Google, or Netapp for instance...And in most all cases I bet you can tie them back to Sun. Nothing wrong with that, they are the community now. The point is that there are folks around that have done and could do it again through the community...as long as they followed the process to do so... Those people are in the minority when we look at the community as a whole. One of the things we can fault Sun with is time, it is slow to evolve. I mean that in the sense of getting out of the "Darwin Syndrome" that Andy Tucker used to talk about (coincidentally at VMWare I believe). Seems that Andy projected a 6 month to 1 year duration of the "Darwin Syndrome", and it still feels like we're in it today...:-/ For those that have never heard this term, it was based on Apple's work with the community on Darwin, which was later taken back inside Apple as it failed in the sense of an open source project. Sun has opened some of the processes, in the sense that community folks can attend ARC meetings, for instance...but the bulk of the process still resides inside the SWAN, hence my reference to the "Darwin Syndrome". -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Wed Sep 26 00:47:04 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 00:47:04 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA024F.2020002@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> Message-ID: <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> On Tue, 25 Sep 2007, Garrett D'Amore wrote: > There is another real problem here. I think the only people who are > qualified to "fully" evaluate the bug database are those who are also Sun > employees. Why? Because only they really know what the current bugster > contains, and what features are needed, and what aren't. I just chopped out a bunch of folks on the cc list. But I wanted to ask, what's the difference anyway? I have seen people complain that they can't read the entire bugs on the OpenSolaris bug database. Ok, even if you could read them internally, there are cases where you still need to look at the actual files to fully understand what is being changed. Hypothetically, let's say you created a process that made it so difficult for engineers to do their putback, that they started to slip changes into another change request to get their changes back? This would never happen at Sun, right? My point is that even if you could read the bug database from outside SWAN, in it's whole, it still wouldn't ensure that you could know exactly what a putback did unless you really looked at every single file and evaluated how they relate to the change request anyway. My $0.02. For the record, I'm not claiming this goes on willy nilly, just that in any large set of sources where you have massive amounts of changes happening on a daily basis, there is a higher chance that engineers will integrate more changes to a CR than logical. Maybe it's a part of common habit. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Wed Sep 26 00:55:42 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 00:55:42 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <x7odfpkhh8.fsf@richlowe.net> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> Message-ID: <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> On Wed, 26 Sep 2007, Richard Lowe wrote: > We don't lighten the process, we certainly don't drop any quality > control aspects thereof, we just make it work without the need for > someone on SWAN to hand-hold, go try and find where a bug ended up, > etc, etc. I'm not saying that this shouldn't be the way *everyone* > (to include new hires at Sun) works with our gates at first. I'm > saying there needs to be an alternative. The sponsor process doesn't > scale, and was never intended to. For better or worse, Sun is moving to get outside the SWAN, it's just slow, but as I said in a previous message, I can only fault them on the time it's taking. There is no having your cake and eating it too, if you want to keep the processes in place and not lighten up on anything, it will continue to require a sponsor. The process is really too complicated for the average person to follow without help, IMO. However, testing is the most difficult piece with device drivers for instance. Ok, you don't want to lighten the process, it will continue to take a long time to putback each driver and work through the issues and fix them to be able to meet the rigorous testing that is currently in place. I feel the testing needs to be evaluated in some cases. For instance, testing a NIC driver to be loaded and unloaded thousands of times will probably just tell you if there is a memory leak. There could be better ways to test that, IMO. Loading and unloading a driver is not a common practice on most systems. >> Can you elaborate on what you would do different in regard >> to being more open in the development of the current sources? > > We get to the point where external developers have equal access to the > tools and processes, and useful things can be done without latent > intermediaries. I really don't care what developers inside Sun (or > outside Sun) choose to do. Only that they have the ability to > actually do it themselves, and that the decisions made about changes > made to opensolaris gates are taken by opensolaris, in places > opensolaris can see, rather than by Sun alone. Right, when you can do a putback to OpenSolaris, and Sun can determine if they want to accept that into Solaris, then we will have reached that level. FWIW, we are moving there, it's just moving there slowly. -- Alan DuBoff - Solaris x86 IHV/OEM Group From Sara.Dornsife at Sun.COM Wed Sep 26 06:29:27 2007 From: Sara.Dornsife at Sun.COM (Sara Dornsife) Date: Wed, 26 Sep 2007 08:29:27 -0500 Subject: [opensolaris-summit] Tuesday planning meetings - 8am and 3pm PT 9/25 In-Reply-To: <46F813A2.8030404@sun.com> References: <46F813A2.8030404@sun.com> Message-ID: <46FA5EB7.1030208@sun.com> Notes from today's meeting(s): Attendees: 8:00am Ian Murdock Stuart Kreitman Stefan Teleman Jim Walker Sara Dornsife 3:00pm Ian Murdock Stuart Kreitman Glynn Foster Patrick Finch Shawn Walker John Plocher Jesse Silver Sara Dornsife We decided that based on the topics proposed, we should dedicate all of Saturday to 3 topics. - Packaging System - Installer (including the Distro Constructor) - Modernization Each of these 3 topics will have a 30 minute presentation slot at the beginning of the day to give a high-level overview of the topic and what the topic owner hopes to accomplish during the Summit. My AI on that was to contact the owners and ask them what sub-topics they would be discussing for the rest of the times alloted. Email sent. Stuart asked for projectors for IRC chats to be projected onto a screen. He's going to bring an extra laptop and keep an IRC channel open. We spent some time discussing what our top 3 goals for this Summit were. - Level-setting - what is OpenSolaris? Get everyone in agreement - Packaging system - everyone should leave this Summit with a full understanding of what it is, how to use it, how it is changing the game - Decision-making - we should all have a clear understanding of the decision-making process for OpenSolaris Feel free to insert your own goals here. Glynn is going to formally invite the rest of the OGB. We currently have 3 of the 7 members attending. Glynn has created a schedule and posted it http://www.opensolaris.org/os/project/indiana/summit/ Please take a look and comment. I think it's coming together nicely. For next week, we need owners and secretaries for each section. We also need to make sure that everyone has a ride in from the airport. Please update the wiki if you can pick people up or if you need a ride. http://www.genunix.org/wiki/index.php/OpenSolaris_Summit. As of this morning (9/26) we're at 77 people. Sara Sara Dornsife wrote: > Reminder that we will have our next planning meetings tomorrow Tues 9/25 > at 8am and 3pm PT. > > Dial-in information: > (866)230-6968 > Int'l Access/Caller Paid Dial In Number: (865)544-7856 > ACCESS CODE: 2192132 > > Agenda: > Let's get our agenda nailed down > > Thanks, > Sara > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From alan.coopersmith at sun.com Wed Sep 26 07:52:08 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 07:52:08 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> Message-ID: <46FA7218.50301@sun.com> Alan DuBoff wrote: > What can you suggest that would help that better? How would you retain the > quality of Solaris as it has been, yet allow the community to participate > in the development more so? Your previous comments about the development > process not being open must relate to the ON sources that are currently > out in the community, after all, there really is nothing more to > OpenSolaris at this point aside from some packages on opensolaris.org. You really are hopeless, you know that? I know you'll never believe it for some insane reason, but just in case anyone else reading this thread believes you, there is far more than just ON on OpenSolaris, as is obvious to anyone who opens their eyes and looks. JDS has beat everyone else in putting their master source tree on OpenSolaris.org, where anyone can pull and community members can commit, and other consolidations like X, SFW, Admin, and NWS have their buildable sources out there as well. Is there any point in you continuing to belittle the members of the community and the people at Sun who have worked their asses off making this happen or are you just trying to insult as many people as possible and tear apart the community? -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From Garrett.Damore at Sun.COM Wed Sep 26 08:08:33 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 08:08:33 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> Message-ID: <46FA75F1.9010302@sun.com> Alan DuBoff wrote: > On Tue, 25 Sep 2007, Garrett D'Amore wrote: > >> There is another real problem here. I think the only people who are >> qualified to "fully" evaluate the bug database are those who are also >> Sun employees. Why? Because only they really know what the current >> bugster contains, and what features are needed, and what aren't. > > I just chopped out a bunch of folks on the cc list. > > But I wanted to ask, what's the difference anyway? I have seen people > complain that they can't read the entire bugs on the OpenSolaris bug > database. Ok, even if you could read them internally, there are cases > where you still need to look at the actual files to fully understand > what is being changed. > > Hypothetically, let's say you created a process that made it so > difficult for engineers to do their putback, that they started to slip > changes into another change request to get their changes back? That would be an error. IF the process were so heavy weight, then something is wrong. I've slipped many changes into a single putback ... *but* I try to make sure I have only one CR. (The exception being bugs against a new driver or other software, which I haven't integrated. In that case there is only one CR, which is the integration of the driver in the first place... in that case it makes sense to try to make the first integration as bug-free as you can.) And, FWIW, I think a lot of the RTI advocates would catch this. Most of the advocates I've used in the past have acted as another layer of code review... even though that isn't strictly part of their RTI duties. > > This would never happen at Sun, right? > > My point is that even if you could read the bug database from outside > SWAN, in it's whole, it still wouldn't ensure that you could know > exactly what a putback did unless you really looked at every single > file and evaluated how they relate to the change request anyway. My > $0.02. You're misunderstanding the problem. People use the CR database for a lot more than knowing what was in a putback. It contains customer call records, sample code, attachments, etc. A lot of those use cases happen internally to Sun. I'm not sure I even know what they all are. Someone should involve, for example, RPE and CTE in such an evaluation... they are amongst the heaviest bugster users. > > For the record, I'm not claiming this goes on willy nilly, just that > in any large set of sources where you have massive amounts of changes > happening on a daily basis, there is a higher chance that engineers > will integrate more changes to a CR than logical. Maybe it's a part of > common habit. I don't think so. Certainly I don't see that in the kernel. Codereview and RTI is supposed to protect against it. -- Garrett > > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group From Garrett.Damore at Sun.COM Wed Sep 26 08:18:21 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 08:18:21 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.16644.873233.737411@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> Message-ID: <46FA783D.7050803@sun.com> James Carlson wrote: > Alan DuBoff writes: > >> I know you do, and I believe your intentions are quite good. As a case in >> point, I know you try to follow the DDI strictly. Sun internal drivers >> seem to have been some of the worst offenders of sticking to the DDI. >> > > You seem to say that as though it were a bad thing. > > For drivers that deliver via ON, there's no reason that they need to > stick to the DDI at all. All projects integrating in ON (including > drivers) are able to use ON consolidation private interfaces without > additional review. It's one of the perks of delivering synchronously > with ON. > > The only considerations that matter here are the usual ones, such as > design correctness and code clarity. Adherence to rules that apply > for non-ON code isn't part of the bargain. > > I agree that when driver designs are required to reach outside of the > DDI on a regular basis, then that signals a problem. It most often > signals a problem with the DDI: it means that the functions that the > designers need are not present. It rarely (though perhaps > occasionally) means that the driver writer is doing the wrong thing. > > So, case in point, the GLDv3/Nemo interfaces are not part of the DDI. > Many of our drivers use them anyway. If someone proposed to add a new > Ethernet driver to ON, and that driver used only DDI interfaces (thus > ignoring Nemo), I would certainly consider the driver to be defective. > > Thus, DDI isn't a mandate. > Historically, believe I had the only DDI complaint ethernet drivers that were to be found... either inside ON or outside of ON. I believe that this signaled both deficiencies in ON, and in the various unbundled drivers. I believe that sticking to the DDI, while not a hard requirement for drivers in ON, is a worthwhile cause even for the drivers in ON. 1) It reduces the pain and suffering when non-DDI interfaces change 2) By sticking to documented APIs (or APIs you expect to be documented) you're unlikely to be bitten later by changes in undocumented behavior. 3) By using the DDI, you know that you're using the same APIs that we recommend to external developers. That means you help to uncover any limitations in the DDI. I think if the drivers in ON "had" to use the DDI or go thru the process of getting a waiver/contract, then its likely the DDI would be a lot richer than it is today. (For one, we'd probably have a public API for nexus drivers!) Note that for APIs like GLDv3, kEF, etc. I think use is appropriate... largely because we are using those APIs as we expect an external consumer would, and we're in the process of trying to make them public. This is quite different than drivers that start up their own kernel threads, etc. -- Garrett From Garrett.Damore at Sun.COM Wed Sep 26 08:28:02 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 08:28:02 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA75F1.9010302@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> <46FA75F1.9010302@sun.com> Message-ID: <46FA7A82.7070406@sun.com> Garrett D'Amore wrote: > Alan DuBoff wrote: > >> On Tue, 25 Sep 2007, Garrett D'Amore wrote: >> >> >>> There is another real problem here. I think the only people who are >>> qualified to "fully" evaluate the bug database are those who are also >>> Sun employees. Why? Because only they really know what the current >>> bugster contains, and what features are needed, and what aren't. >>> >> I just chopped out a bunch of folks on the cc list. >> >> But I wanted to ask, what's the difference anyway? I have seen people >> complain that they can't read the entire bugs on the OpenSolaris bug >> database. Ok, even if you could read them internally, there are cases >> where you still need to look at the actual files to fully understand >> what is being changed. >> >> Hypothetically, let's say you created a process that made it so >> difficult for engineers to do their putback, that they started to slip >> changes into another change request to get their changes back? >> > > That would be an error. IF the process were so heavy weight, then > something is wrong. > > I've slipped many changes into a single putback ... *but* I try to make > sure I have only one CR. This didn't come out right. I meant to say, "one CR per bug". So one putback, many bugs, and hence many CRs. Sorry for the confusion. -- Garrett From stevel at sun.com Wed Sep 26 08:42:28 2007 From: stevel at sun.com (Stephen Lau) Date: Wed, 26 Sep 2007 08:42:28 -0700 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA024F.2020002@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> Message-ID: <46FA7DE4.7010208@sun.com> Garrett D'Amore wrote: > Stephen Lau wrote: >> We've got an existing DTS evaluation going on - it's floundering not >> because Sun is holding anything back, but because nobody is making the >> effort to evaluate any of the DTS (or at least they haven't mentioned >> it publicly). >> > > There is another real problem here. I think the only people who are > qualified to "fully" evaluate the bug database are those who are also > Sun employees. Why? Because only they really know what the current > bugster contains, and what features are needed, and what aren't. > > Of course, all the Sun employees I know are up to their neck in other > work. And it isn't entirely clear that the Sun folks who should be > paying for an evaluation of a new DTS have any motivation to do so; > after all, bugster works just fine for *their* purposes. > > Speaking for myself, I don't have the time to do a full evaluation of a > DTS. And, I'm not really qualified as a DTS administrator. *HOWEVER*, > I know that I would be willing to play in a sandbox that someone else > built. And even use it to track real project bugs. I *suspect* that > the same is true of other people. > > So, if someone is energetic enough (a big if, I know) to set up the demo > sandbox, then maybe an announcement in a broader context inviting others > to participate in the test (as developers, not DTS administrators!) > would be helpful. Good points Garrett, but I do think there are still some pieces where non-Sun people should certainly be able to help out. We've outlined a specific list of requirements: http://opensolaris.org/sc/src/website/spec/dts-requirements/d-dts-requirements.txt It should certainly be at least possible for people to help eliminate potential candidates based on whether or not they meet the requirements or not. cheers, steve -- stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development From John.Plocher at Sun.COM Wed Sep 26 08:43:41 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 26 Sep 2007 08:43:41 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> Message-ID: <46FA7E2D.3050508@Sun.Com> Alan DuBoff wrote: > Do you think that folks should be allowed a putback if they don't > completely meet the test validation process? I think this is an example of a myopic viewpoint, and that there are places where this behavior might be perfectly acceptable. Think website wikis, package repository experimentation, prototyping, ... OpenSolaris != the ON Consolidation. The level of detail you (and Garrett) are looking at is really only valid for some segments of the OpenSolaris community. Darren also seems stuck there. There are other segments of the larger OpenSolaris Community where there are different (or even non-existent) test validation processes, places where the bar to participation is intentionally lower. Painting all of OS.o with the "ON brush" does us all a disservice, as does blindly assuming that only ON matters. -John (As I said to Alan in a side conversation, we really should start using the word "OpenSolaris" as an adjective and not a noun. That is, Alan and Garrett are concerned with the OpenSolaris ON Community and associated policies, while others (JimG...) are focused on the OpenSolaris User Groups and Portals; Ian is focused on an OpenSolaris Distro, etc.) From sch at sun.com Wed Sep 26 09:34:16 2007 From: sch at sun.com (Stephen Hahn) Date: Wed, 26 Sep 2007 09:34:16 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> References: <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> Message-ID: <20070926163416.GC6132@eng.sun.com> * Alan DuBoff <alan.duboff at sun.com> [2007-09-26 06:11]: > So far I haven't seen any suggestions from you, just criticism on the > process Sun has put in place. Can you elaborate on what exactly Sun could > do better? I would suggest, Alan, that you haven't looked widely enough. Rich has provided fixes and constructive commentary on tools, processes, and product code, in such a fashion to get Contributor or Core Contributor status in four community groups. He's even given feedback during the review of the DTS requirements... - Stephen -- sch at sun.com http://blogs.sun.com/sch/ From alan.duboff at sun.com Wed Sep 26 10:29:56 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 10:29:56 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.16644.873233.737411@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> Message-ID: <Pine.GSO.4.64.0709261018130.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, James Carlson wrote: > Alan DuBoff writes: >> I know you do, and I believe your intentions are quite good. As a case in >> point, I know you try to follow the DDI strictly. Sun internal drivers >> seem to have been some of the worst offenders of sticking to the DDI. > > You seem to say that as though it were a bad thing. Not a bad thing directly, but if Sun creates and advocates that driver writers adhere to the DDI, but doesn't do it them self, it's a kind of double speak. > So, case in point, the GLDv3/Nemo interfaces are not part of the DDI. > Many of our drivers use them anyway. If someone proposed to add a new > Ethernet driver to ON, and that driver used only DDI interfaces (thus > ignoring Nemo), I would certainly consider the driver to be defective. > > Thus, DDI isn't a mandate. GLDv3 is an edge case right now as the interfaces are going to be public soon. Currently I am working with driver that was written to them, and it requires you copy/use mac.h and mac_ether.h from uts/common/sys/. This is only due to evolution of the development process. That was not what I was referring to in my post per se. This is a valid case for using private interfaces, IMO. It's not that using private interfaces is a bad thing, it's just that the driver stands the chance of breaking if those interfaces change. Any public interfaces are guaranteed to stay stable, I believe. -- Alan DuBoff - Solaris x86 IHV/OEM Group From james.liu at sun.com Wed Sep 26 10:36:10 2007 From: james.liu at sun.com (James C. Liu) Date: Wed, 26 Sep 2007 10:36:10 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> Message-ID: <46FA988A.5020608@sun.com> Alan DuBoff wrote: > There is no having your cake and eating it too, if you want to keep the > processes in place and not lighten up on anything, it will continue to > require a sponsor. The process is really too complicated for the average > person to follow without help, IMO. > Why can't we have our cake and eat it too? We don't need to dump the rigorousness of the existing process. We can still require some sponsor. It may not scale over time and then we'll have to rethink perhaps. But we've had this hallway discussion before, and the solution then was to promote the idea of an "experimental" section for drivers and apps that haven't gone through full testing. Some stuff (like kernel modules) would be available on the install media with the right disclaimers and caveats, and other packages would be available elsewhere. Over time, the idea would be to do the testing and certify code as official and move it over. But that would be transparent to endusers who would already have had access to the pkgs if they needed it. -James Liu - Solaris x86 IHV/OEM Engineering From alan.duboff at sun.com Wed Sep 26 10:51:37 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 10:51:37 -0700 (PDT) Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA7A82.7070406@sun.com> References: <46F2FE5F.3070508@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> <46F983E4.4080804@Sun.COM> <46F9887B.5010604@sun.com> <46FA024F.2020002@sun.com> <Pine.GSO.4.64.0709260038060.7188@eagle.softorchestra.com> <46FA75F1.9010302@sun.com> <46FA7A82.7070406@sun.com> Message-ID: <Pine.GSO.4.64.0709261039310.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, Garrett D'Amore wrote: >> That would be an error. IF the process were so heavy weight, then >> something is wrong. >> >> I've slipped many changes into a single putback ... *but* I try to make >> sure I have only one CR. > > This didn't come out right. I meant to say, "one CR per bug". So one > putback, many bugs, and hence many CRs. Sorry for the confusion. Garrett, I guess my point is that I have heard engineers say they have done that because the process is so difficult. I won't name who, but I think that to some extent this does go on some, today. I was more using this as an example of how the engineers have learned to work around some of the limitation. I don't know that this is a bad thing, just that it can and does happen. It makes the engineer more aware so in the case where something does break, they know they're on the line for it. I wasn't pointing any finger at you or anyone else specific, just pointing out that there are many cases where files are going back on a daily basis. You really need to look at the changes to fully understand what is going back. I agree with what you said about workarounds, examples, contact info, etc...Sun is moving towards getting the bug database out to the community though, time being the only real process holding it up. I was also more using that as a case where the community claims foul, that Sun has not opened up the bug database. I was only pointing out that it wouldn't magically help people understand every CR that gets putback, they would still need to look at and evaluate the files and fixes to fully understand each case. Again, I don't know how you could easily transition such a large code base to a community and not cause some issues as the bug database, unless it just all magically happened together at the same time (source code management, bug database, sources, ARCs, etc...). In the case of Solaris, as it was, there was just too much process for that to happen cleanly. As it is they did a good job, IMO, and our community is moving towards a pure, open model. -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.duboff at sun.com Wed Sep 26 11:02:28 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 11:02:28 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA7E2D.3050508@Sun.Com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> Message-ID: <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, John Plocher wrote: > Alan DuBoff wrote: >> Do you think that folks should be allowed a putback if they don't >> completely meet the test validation process? > > > I think this is an example of a myopic viewpoint, and that there > are places where this behavior might be perfectly acceptable. Think > website wikis, package repository experimentation, prototyping, ... > > OpenSolaris != the ON Consolidation. Let me clear something up, before AlanC's jugular explodes.;-) I realize that OpenSolaris is not just ON, however today it pretty much is from a sources standpoint. When someone refers to OpenSolaris today, that is what is there. > The level of detail you (and Garrett) are looking at is really only > valid for some segments of the OpenSolaris community. Darren also > seems stuck there. There are other segments of the larger OpenSolaris > Community where there are different (or even non-existent) test > validation processes, places where the bar to participation is > intentionally lower. Agreed. However, today what we have are ON sources that primarily comprise what is OpenSolaris. This is not to say we are limited to this or that other pieces will not be added, I am just pointing out where we are today. > (As I said to Alan in a side conversation, we really should start > using the word "OpenSolaris" as an adjective and not a noun. That > is, Alan and Garrett are concerned with the OpenSolaris ON Community > and associated policies, while others (JimG...) are focused on the > OpenSolaris User Groups and Portals; Ian is focused on an OpenSolaris > Distro, etc.) And add AlanC to the mix, he has his view also, as others do. My point was that it's been confusing at best to figure out what OpenSolaris really is for some, like me for instance. In speaking for myself, and only myself, I can say that I have found it confusing what OpenSolaris really is. To further complicate that, marketing has also been unclear on how they intend to use the OpenSolaris name them self. It has been unclear to me anyway. -- Alan DuBoff - Solaris x86 IHV/OEM Group From randy.fishel at sun.com Wed Sep 26 11:21:41 2007 From: randy.fishel at sun.com (Randy Fishel) Date: Wed, 26 Sep 2007 11:21:41 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FA783D.7050803@sun.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> Message-ID: <Pine.GSO.4.64.0709261050440.2186@grimmy> On Wed, 26 Sep 2007, Garrett D'Amore wrote: > I believe that sticking to the DDI, while not a hard requirement for > drivers in ON, is a worthwhile cause even for the drivers in ON. I would argue that it must be a hard requirement, and for the reasons you state below. > > 1) It reduces the pain and suffering when non-DDI interfaces change > 2) By sticking to documented APIs (or APIs you expect to be documented) > you're unlikely to be bitten later by changes in undocumented behavior. > 3) By using the DDI, you know that you're using the same APIs that we > recommend to external developers. That means you help to uncover any > limitations in the DDI. > > I think if the drivers in ON "had" to use the DDI or go thru the process > of getting a waiver/contract, then its likely the DDI would be a lot > richer than it is today. (For one, we'd probably have a public API for > nexus drivers!) > > Note that for APIs like GLDv3, kEF, etc. I think use is appropriate... > largely because we are using those APIs as we expect an external > consumer would, and we're in the process of trying to make them public. > This is quite different than drivers that start up their own kernel > threads, etc. > > -- Garrett We do make the requirement of drivers to specify all the interfaces they expect to import and export. We don't make them list the DDI/DKI, as that is a set of implied interfaces. They must, however, specify if they intend to use GLDvX, kEF, and even FMA. If a driver is using any other interface, it should have listed it, and would be in violation of its ARC case if it didn't (IMHO, they should even list consolidation private interfaces). The problem, with this "requirement", is that we have no way to validate or enforce that only the ARC'd API's have been used. We leave it to the developer (in theory, checked by the RTI advocate) to appropriately use the API's. But if you know of a driver that uses a non-public API, and it doesn't have an ARC case stating this use (so all the GLDv3 drivers should be OK), then you can file a bug on the driver. Just maybe one of the things we need to think about is some set of driver validation tools. ---- Randy From alan.duboff at sun.com Wed Sep 26 11:25:26 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 11:25:26 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FA988A.5020608@sun.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> <46FA988A.5020608@sun.com> Message-ID: <Pine.GSO.4.64.0709261118000.7356@eagle.softorchestra.com> On Wed, 26 Sep 2007, James C. Liu wrote: > > Alan DuBoff wrote: > >> There is no having your cake and eating it too, if you want to keep the >> processes in place and not lighten up on anything, it will continue to >> require a sponsor. The process is really too complicated for the average >> person to follow without help, IMO. >> > > Why can't we have our cake and eat it too? We don't need to dump > the rigorousness of the existing process. We can still require some > sponsor. It may not scale over time and then we'll have to rethink > perhaps. But we've had this hallway discussion before, and the > solution then was to promote the idea of an "experimental" section > for drivers and apps that haven't gone through full testing. Some > stuff (like kernel modules) would be available on the install media > with the right disclaimers and caveats, and other packages would be > available elsewhere. Over time, the idea would be to do the testing > and certify code as official and move it over. But that would > be transparent to endusers who would already have had access to the > pkgs if they needed it. > > -James Liu - Solaris x86 IHV/OEM Engineering > James, Yes, this would allow for drivers to be included that haven't passed the rigorous testing. However, the OP stated that they didn't want to lighten up on testing, or any current requirements. I was only pointing out that while it doesn't need to be mandatory for them to have a sponsor (today they are required to have one), it's all in their best interest. If there was no sponsor requirement today, it would be more difficult for folks to grasp the process. I think the sponsor idea is to assist more than anything else, at least from my view. If all community members understood the putback process well enough there would be no need for a sponsor, they would be able to do everything. Oh, and if all process was outside the SWAN, that would also help and where I believe we're headed. -- Alan DuBoff - Solaris x86 IHV/OEM Group From John.Plocher at Sun.COM Wed Sep 26 11:32:50 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 26 Sep 2007 11:32:50 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> Message-ID: <46FAA5D2.20300@Sun.Com> Alan DuBoff wrote: >> OpenSolaris != the ON Consolidation. > > Let me clear something up, before AlanC's jugular explodes.;-) > > I realize that OpenSolaris is not just ON, however today it pretty much > is from a sources standpoint. When someone refers to OpenSolaris today, > that is what is there. You seem to be showing either an amazing degree of either ignorance or intentional disrespect for those who have built out the non-ON parts of the community. It may be true that when /YOU/ say "OpenSolaris", /YOU/ mean ON, but please don't generalize your feelings without some serious data to back up your claims. Some conversations don't work well in email; this seems to have become one of them. Why don't we let this topic die? -John From Garrett.Damore at Sun.COM Wed Sep 26 11:30:03 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 11:30:03 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <Pine.GSO.4.64.0709261050440.2186@grimmy> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <Pine.GSO.4.64.0709261050440.2186@grimmy> Message-ID: <46FAA52B.6090403@sun.com> Randy Fishel wrote: > On Wed, 26 Sep 2007, Garrett D'Amore wrote: > > >> I believe that sticking to the DDI, while not a hard requirement for >> drivers in ON, is a worthwhile cause even for the drivers in ON. >> > > I would argue that it must be a hard requirement, and for the > reasons you state below. > > >> 1) It reduces the pain and suffering when non-DDI interfaces change >> 2) By sticking to documented APIs (or APIs you expect to be documented) >> you're unlikely to be bitten later by changes in undocumented behavior. >> 3) By using the DDI, you know that you're using the same APIs that we >> recommend to external developers. That means you help to uncover any >> limitations in the DDI. >> >> I think if the drivers in ON "had" to use the DDI or go thru the process >> of getting a waiver/contract, then its likely the DDI would be a lot >> richer than it is today. (For one, we'd probably have a public API for >> nexus drivers!) >> >> Note that for APIs like GLDv3, kEF, etc. I think use is appropriate... >> largely because we are using those APIs as we expect an external >> consumer would, and we're in the process of trying to make them public. >> This is quite different than drivers that start up their own kernel >> threads, etc. >> >> -- Garrett >> > > > We do make the requirement of drivers to specify all the interfaces > they expect to import and export. We don't make them list the > DDI/DKI, as that is a set of implied interfaces. They must, however, > specify if they intend to use GLDvX, kEF, and even FMA. If a driver > is using any other interface, it should have listed it, and would be > in violation of its ARC case if it didn't (IMHO, they should even list > consolidation private interfaces). > > The problem, with this "requirement", is that we have no way to > validate or enforce that only the ARC'd API's have been used. We > leave it to the developer (in theory, checked by the RTI advocate) to > appropriately use the API's. But if you know of a driver that uses a > non-public API, and it doesn't have an ARC case stating this use (so > all the GLDv3 drivers should be OK), then you can file a bug on the > driver. > > > Just maybe one of the things we need to think about is some set of > driver validation tools. > There used to be one.. .it was called "ddict". I sometimes believe I'm the only person on the planet who ever actually used "ddict". -- Garrett > > > ---- Randy > From Garrett.Damore at Sun.COM Wed Sep 26 11:44:01 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 11:44:01 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261118000.7356@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> <46FA988A.5020608@sun.com> <Pine.GSO.4.64.0709261118000.7356@eagle.softorchestra.com> Message-ID: <46FAA871.1070707@sun.com> Alan DuBoff wrote: > On Wed, 26 Sep 2007, James C. Liu wrote: > > >> Alan DuBoff wrote: >> >> >>> There is no having your cake and eating it too, if you want to keep the >>> processes in place and not lighten up on anything, it will continue to >>> require a sponsor. The process is really too complicated for the average >>> person to follow without help, IMO. >>> >>> >> Why can't we have our cake and eat it too? We don't need to dump >> the rigorousness of the existing process. We can still require some >> sponsor. It may not scale over time and then we'll have to rethink >> perhaps. But we've had this hallway discussion before, and the >> solution then was to promote the idea of an "experimental" section >> for drivers and apps that haven't gone through full testing. Some >> stuff (like kernel modules) would be available on the install media >> with the right disclaimers and caveats, and other packages would be >> available elsewhere. Over time, the idea would be to do the testing >> and certify code as official and move it over. But that would >> be transparent to endusers who would already have had access to the >> pkgs if they needed it. >> >> -James Liu - Solaris x86 IHV/OEM Engineering >> >> > > James, > > Yes, this would allow for drivers to be included that haven't passed the > rigorous testing. However, the OP stated that they didn't want to lighten > up on testing, or any current requirements. I was only pointing out that > while it doesn't need to be mandatory for them to have a sponsor (today > they are required to have one), it's all in their best interest. > > If there was no sponsor requirement today, it would be more difficult for > folks to grasp the process. I think the sponsor idea is to assist more > than anything else, at least from my view. If all community members > understood the putback process well enough there would be no need for a > sponsor, they would be able to do everything. Oh, and if all process was > outside the SWAN, that would also help and where I believe we're headed. > Let me state this without reservation: * I believe that all device drivers, and any other kernel software, should pass a fairly standardized set of test requirements. * I do not believe that "waiving" test failures is a reasonable thing to do, *unless* it can be shown that the failures meet one of a *very* small set of criteria: 1) The failure is due to a bug or bad assumption in the test suite. (Yes, I've had a few of these.) And the driver is performing properly. (Demonstrated by hand outside of the test suite, if necessary.) 2) The failure is benign and only shows up under unrealistic test scenarios. For example, a network driver that fails due to performance limitations when pushing out 1-byte UDP packets at 200Kpps/sec might be reasonable... as long as once the test concluded the NIC resumed behaving normally without administrator intervention. In case 1, a bug is warranted. In case 2, it may or may not be warranted, depending on the actual severity and likelihood of anyone ever hitting it. (Some of the test scenarios are very, very demanding, and very, very strict... e.g. dropping even a single ethernet packet can cause some of the NICDRV tests to fail!) There may be other cases than the three above, but I definitely think they should be examined closely before any waiver would ever be granted. -- Garrett > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From phil at bolthole.com Wed Sep 26 11:51:01 2007 From: phil at bolthole.com (Philip Brown) Date: Wed, 26 Sep 2007 11:51:01 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com>; from alan.duboff@sun.com on Wed, Sep 26, 2007 at 12:55:42AM -0700 References: <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <x7odfpkhh8.fsf@richlowe.net> <Pine.GSO.4.64.0709260048210.7188@eagle.softorchestra.com> Message-ID: <20070926115101.A91989@bolthole.com> Putting my dusty "Driver developer" hat on for a second.. .:-) On Wed, Sep 26, 2007 at 12:55:42AM -0700, Alan DuBoff wrote: > ... > I feel the testing needs to be evaluated in some cases. For instance, > testing a NIC driver to be loaded and unloaded thousands of times will > probably just tell you if there is a memory leak. There could be better > ways to test that, IMO. Loading and unloading a driver is not a common > practice on most systems. Actually, that sounds like a GOOD TEST to me, that should be kept in! I've occasionally had bugs in drivers of mine, that really only manifested themselves in a comprehensible way, at driver load time. Very rare ones, to be sure. But then, that's a core component of driver testing: working out all the "very rare" cases, and making them bulletproof. From Garrett.Damore at Sun.COM Wed Sep 26 11:46:57 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 11:46:57 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.42089.235484.84538@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <Pine.GSO.4.64.0709261050440.2186@grimmy> <18170.42089.235484.84538@gargle.gargle.HOWL> Message-ID: <46FAA921.5070107@sun.com> James Carlson wrote: > Randy Fishel writes: > >> We do make the requirement of drivers to specify all the interfaces >> they expect to import and export. We don't make them list the >> DDI/DKI, as that is a set of implied interfaces. They must, however, >> specify if they intend to use GLDvX, kEF, and even FMA. If a driver >> is using any other interface, it should have listed it, and would be >> in violation of its ARC case if it didn't (IMHO, they should even list >> consolidation private interfaces). >> > > You're welcome to that opinion, and you can certainly demand it in any > reviews you do yourself, but it's not actually an architectural > requirement. > > It'd be possible to set some sort of OpenSolaris-wide policy here. We > could have an ARC case that specifies what things must be "declared" > and how they're declared. Such a thing (other than the existing > interface taxonomy) does not currently exist for the special case of > drivers. A fast-track specifying this (and the detailed rationale > behind it) might be helpful. > +1. I certainly never heard, or believed, that imported ON consolidation private interfaces required to be called out in ARC reviews. I always figured at that point, they sort of feel underneath ARC's radar. (Now importing project-private interfaces, or using ON private APIs from unbundled drivers, is a totally different story.) -- Garrett From phil at bolthole.com Wed Sep 26 12:03:06 2007 From: phil at bolthole.com (Philip Brown) Date: Wed, 26 Sep 2007 12:03:06 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FA783D.7050803@sun.com>; from Garrett.Damore@Sun.COM on Wed, Sep 26, 2007 at 08:18:21AM -0700 References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> Message-ID: <20070926120306.B91989@bolthole.com> On Wed, Sep 26, 2007 at 08:18:21AM -0700, Garrett D'Amore wrote: > James Carlson wrote: > >.... > > Thus, DDI isn't a mandate. > > > > Historically, believe I had the only DDI complaint ethernet drivers that > were to be found... either inside ON or outside of ON. I believe that > this signaled both deficiencies in ON, and in the various unbundled drivers. > > I believe that sticking to the DDI, while not a hard requirement for > drivers in ON, is a worthwhile cause even for the drivers in ON. Gaaaah... DDI *should be* a mandate, even for drivers in ON. Why? because it ensures that the DDI level stuff, stays actually USEFUL. Multiple times, I think, I was trying to write a driver, and looking through the DDI for nice clean ways to do it.. and, there just wasnt a way to do it! There were existing sun drivers that did similar things, and they got the job done just fine... by going around DDI. But for external people, "you cant get there from here". So, the answer to the long-standing question/gripe of, "why are there so few drivers by non-sun people out there?" would be at least in part, "Well, gee, becuase sun didnt provide non-src-privileged people a means of writing them!" So, I doubly underscore the later paragraph from Garrett: > 3) By using the DDI, you know that you're using the same APIs that we > recommend to external developers. That means you help to uncover any > limitations in the DDI. Even in this modern day and age, where people now have "access to the source"... there is something very positive to be said about having an official DDI for things, that must be adhered to.. because then, (ideally), the interfaces would actually get DOCUMENTED...? Believe it or not, open source developers DO actually appreciate documentation (even if they rarely write any themselves ;-) From alan.coopersmith at sun.com Wed Sep 26 12:07:05 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 12:07:05 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> Message-ID: <46FAADD9.4090509@sun.com> Alan DuBoff wrote: > I realize that OpenSolaris is not just ON, however today it pretty much is > from a sources standpoint. Except that JDS's sources are about twice the size of ON's last I heard (19 million lines for JDS vs. 10 million for ON), and there's a huge amount of source from all the other consolidations. For regular Solaris, ON is only around 20% of the software - for OpenSolaris, it's probably closer to 30-40%, since we have some large chunks of Solaris like CDE that aren't released. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From swalker at opensolaris.org Tue Sep 25 16:43:12 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:43:12 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> References: <46F2FE5F.3070508@sun.com> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> Message-ID: <b9c544f0709251643o15ad5a8fraeaa174bf4ce0d32@mail.gmail.com> On 25/09/2007, Peter Tribble <peter.tribble at gmail.com> wrote: > On 9/25/07, Stephen Lau <stevel at sun.com> wrote: > > Joerg Schilling wrote: > > > > > > I believe there are two problems: missing visibility of the OGB and missing > > > power of the community. > > > > Why do you perceive the former to be a problem? Visibility of the OGB > > is a double-edged sword. Yes, we want people to know there is an OGB so > > that they know there is a path of appeals and such; but we don't want > > the OGB to be so visible to the point that it is "in-your-face"[1]. The > > OGB has taken the view that the community is to be self-governing. > > But does the community know that it's supposed to be governing itself? Given our voter turnout, and the problems we had in getting leaders to communicate about their community, one could conclude they are strangely not aware (or apathetic). > I suspect that most members of the community simply don't understand > what powers, rights, and responsibilities they have, nor how to exercise > them. I think part of the problem is that most people's eyes are going to glaze over once you start talking about those things. Unless you can sum them all up in a few bullet lists, most folks are going to ignore them... > So how do we encourage communities and individuals to take up the > reins of governance? A better user-based web portal with tools to ease that process with short-and-to-the-point summaries would go a long way... -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From swalker at opensolaris.org Tue Sep 25 16:45:04 2007 From: swalker at opensolaris.org (Shawn Walker) Date: Tue, 25 Sep 2007 18:45:04 -0500 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <46F97328.90502@Sun.Com> References: <46F2FE5F.3070508@sun.com> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <46F92A0C.9090706@sun.com> <df1347730709251328t4fe360b5k6b772d3c8aa409a@mail.gmail.com> <46F97328.90502@Sun.Com> Message-ID: <b9c544f0709251645h65020ed2rcb7d9cdfe1e1d736@mail.gmail.com> On 25/09/2007, John Plocher <John.Plocher at sun.com> wrote: > Peter Tribble wrote: > > So how do we encourage communities and individuals to take up the > > reins of governance? > > Not to be flippant, but why would they want to? > > That is, what problems are they currently having that could > be solved by their spending energy on self-governance? > > My perspective is that the answer is that there are lots of > serious problems, none of which can effectively be solved thru > self-governance alone. > > Problem 1: The structure of the meta-community echoes the > confusion in people's minds of what exactly OpenSolaris is > and should be. Because people are not in agreement with > what it is, there is no consensus on how it should be > structured. +1 "Where there is no vision the people perish..." or something like that... > Problem 2: The observable artifacts of the community are > still locked away from direct community manipulation: The > source tree, the bug database, the web site, the mailing > list infrastructure, the ON C-Team and even the ARCs are > /ALL/ solely maintainable by a set of under-resourced, > under-appreciated and constantly dwindling group of Sun > employees. +1 Power to the people > Solve these problems and the community of developers will > grow. Ignore them, and it will stagnate and die. Agreed. -- Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/ "Beware of bugs in the above code; I have only proved it correct, not tried it. " --Donald Knuth From Garrett.Damore at Sun.COM Wed Sep 26 12:12:45 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 12:12:45 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <20070926120306.B91989@bolthole.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> Message-ID: <46FAAF2D.4030901@sun.com> So, now my question for PSARC is, if someone (say me, or Philip) wanted to start up a case to make this a new requirement going forward (and presumably we could allow for drivers to waive the requirement, but they should need to be justified) wanted to sponsor a case, does this fall under the realm of what would be called a fasttrack? Such a proposal would itself be very light weight, but I'm not at all convinced it would be non-controversial enough to qualify for less than a full case. If nothing else, it gives PSARC a chance to express a stronger opinion. I'm willing to author the original document (fasttrack or otherwise), although if its a full case then I need a sponsor on PSARC, I believe. -- Garrett Philip Brown wrote: > On Wed, Sep 26, 2007 at 08:18:21AM -0700, Garrett D'Amore wrote: > >> James Carlson wrote: >> >>> .... >>> Thus, DDI isn't a mandate. >>> >>> >> Historically, believe I had the only DDI complaint ethernet drivers that >> were to be found... either inside ON or outside of ON. I believe that >> this signaled both deficiencies in ON, and in the various unbundled drivers. >> >> I believe that sticking to the DDI, while not a hard requirement for >> drivers in ON, is a worthwhile cause even for the drivers in ON. >> > > Gaaaah... > > DDI *should be* a mandate, even for drivers in ON. > > Why? because it ensures that the DDI level stuff, stays actually USEFUL. > > Multiple times, I think, I was trying to write a driver, and looking > through the DDI for nice clean ways to do it.. > and, there just wasnt a way to do it! > There were existing sun drivers that did similar things, and they got the > job done just fine... by going around DDI. > But for external people, "you cant get there from here". > > > So, the answer to the long-standing question/gripe of, > "why are there so few drivers by non-sun people out there?" > would be at least in part, > > "Well, gee, becuase sun didnt provide non-src-privileged people a means of > writing them!" > > > So, I doubly underscore the later paragraph from Garrett: > > > >> 3) By using the DDI, you know that you're using the same APIs that we >> recommend to external developers. That means you help to uncover any >> limitations in the DDI. >> > > > > Even in this modern day and age, where people now have "access to the > source"... there is something very positive to be said about having an > official DDI for things, that must be adhered to.. because then, > (ideally), the interfaces would actually get DOCUMENTED...? > > Believe it or not, open source developers DO actually appreciate > documentation (even if they rarely write any themselves ;-) > > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit > From phil at bolthole.com Wed Sep 26 12:37:23 2007 From: phil at bolthole.com (Philip Brown) Date: Wed, 26 Sep 2007 12:37:23 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAAF2D.4030901@sun.com>; from Garrett.Damore@Sun.COM on Wed, Sep 26, 2007 at 12:12:45PM -0700 References: <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> Message-ID: <20070926123723.D5803@bolthole.com> On Wed, Sep 26, 2007 at 12:12:45PM -0700, Garrett D'Amore wrote: > So, now my question for PSARC is, if someone (say me, or Philip) wanted > to start up a case to make this a new requirement going forward (and > presumably we could allow for drivers to waive the requirement, but they > should need to be justified) wanted to sponsor a case, does this fall > under the realm of what would be called a fasttrack? > > Such a proposal would itself be very light weight, but I'm not at all > convinced it would be non-controversial enough to qualify for less than > a full case. If nothing else, it gives PSARC a chance to express a > stronger opinion. for the record.. I believe I actually raised this proposal in front of a few high level people, a few years back (and I think Stephen Hahn was present at that meeting, too ;-) [*waves to Stephen*] I even had the stipulation of, "requirement could be waived, but had to justify" then too. but it still got shot down, in the meeting. From randy.fishel at sun.com Wed Sep 26 12:39:02 2007 From: randy.fishel at sun.com (Randy Fishel) Date: Wed, 26 Sep 2007 12:39:02 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18170.42089.235484.84538@gargle.gargle.HOWL> References: <46F2F56F.70706@Sun.COM> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <Pine.GSO.4.64.0709261050440.2186@grimmy> <18170.42089.235484.84538@gargle.gargle.HOWL> Message-ID: <Pine.GSO.4.64.0709261237320.8725@grimmy> (resend) > Randy Fishel writes: > > We do make the requirement of drivers to specify all the interfaces > > they expect to import and export. We don't make them list the > > DDI/DKI, as that is a set of implied interfaces. They must, however, > > specify if they intend to use GLDvX, kEF, and even FMA. If a driver > > is using any other interface, it should have listed it, and would be > > in violation of its ARC case if it didn't (IMHO, they should even list > > consolidation private interfaces). > > You're welcome to that opinion, and you can certainly demand it in any > reviews you do yourself, but it's not actually an architectural > requirement. Agreed, it is my opinion, and I try to work that into my reviews of drivers. I would, however, like it to be more an architectual requirement, and would be willing to make some arguments or proposals in the near future (it would be real nice to get my current project into Solaris). > > It'd be possible to set some sort of OpenSolaris-wide policy here. We > could have an ARC case that specifies what things must be "declared" > and how they're declared. Such a thing (other than the existing > interface taxonomy) does not currently exist for the special case of > drivers. A fast-track specifying this (and the detailed rationale > behind it) might be helpful. > As Garrett, I am in agreement, and am more than willing to work with and/or sponsor such an effort (though again, after I integrate my current set of bits). Garrett- you in for a collaboration here? ---- Randy From alan.duboff at sun.com Wed Sep 26 15:06:47 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 15:06:47 -0700 (PDT) Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <46FAADD9.4090509@sun.com> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> <46FAADD9.4090509@sun.com> Message-ID: <Pine.GSO.4.64.0709261458090.17209@konocti> On Wed, 26 Sep 2007, Alan Coopersmith wrote: > Alan DuBoff wrote: >> I realize that OpenSolaris is not just ON, however today it pretty much >> is from a sources standpoint. > > Except that JDS's sources are about twice the size of ON's last I heard > (19 million lines for JDS vs. 10 million for ON), and there's a huge > amount of source from all the other consolidations. > > For regular Solaris, ON is only around 20% of the software - for > OpenSolaris, it's probably closer to 30-40%, since we have some large > chunks of Solaris like CDE that aren't released. Alan, Not wanting to debate what OpenSolaris actually is, but it's interesting that you can take a set of sources that were already open sourced by other folks, include them on the OpenSolaris website and they become one and the same. Maybe this is one large distinction between (Open)Solaris and Linux, because Linux is really just a kernel and the other software is the system, although the majority of folks refer to the entire system when they say "Linux". I read a day or two ago (from you I believe) that Sun will not be calling the desktop JDS anymore and that it would be calling it GNOME. I remember GNOME as an open source project before OpenSolaris was created. Sun didn't do anything different with GNOME to include it with OpenSolaris, yet somehow it has become a part of it all. If Sun would clearly define how OpenSolaris is determined, there wouldn't be various views of it, including me, you, John, Jim, or others... >From the above I suspect we're gonna be using IBM's ancient definition of MLOC to determine what is a part of OpenSolaris.:-/ -- Alan DuBoff - Solaris x86 IHV/OEM Group From alan.coopersmith at sun.com Wed Sep 26 16:08:14 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 16:08:14 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAE504.6040108@Sun.COM> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> Message-ID: <46FAE65E.1040003@sun.com> Joseph Kowalski wrote: > Where "lsarc" could be what I believe are the interested parties. > Somethings may be interesting to {psarc, lsarc, wsarc}, but not (for > example) fwarc. Are drivers ever reviewed by any arc other than PSARC? I'm not sure LSARC, WSARC or FWARC need to be involved with this. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From David.Clack at Sun.COM Wed Sep 26 17:13:11 2007 From: David.Clack at Sun.COM (David Clack) Date: Wed, 26 Sep 2007 17:13:11 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261458090.17209@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> <46FAADD9.4090509@sun.com> <Pine.GSO.4.64.0709261458090.17209@konocti> Message-ID: <1190851991.27503.70.camel@localhost> On Wed, 2007-09-26 at 15:06 -0700, Alan DuBoff wrote: > On Wed, 26 Sep 2007, Alan Coopersmith wrote: > > > Alan DuBoff wrote: > >> I realize that OpenSolaris is not just ON, however today it pretty much > >> is from a sources standpoint. > > > > Except that JDS's sources are about twice the size of ON's last I heard > > (19 million lines for JDS vs. 10 million for ON), and there's a huge > > amount of source from all the other consolidations. > > > > For regular Solaris, ON is only around 20% of the software - for > > OpenSolaris, it's probably closer to 30-40%, since we have some large > > chunks of Solaris like CDE that aren't released. > > Alan, > > Not wanting to debate what OpenSolaris actually is, but it's interesting > that you can take a set of sources that were already open sourced by other > folks, include them on the OpenSolaris website and they become one and the > same. Maybe this is one large distinction between (Open)Solaris and Linux, > because Linux is really just a kernel and the other software is the > system, although the majority of folks refer to the entire system when > they say "Linux". > > I read a day or two ago (from you I believe) that Sun will not be calling > the desktop JDS anymore and that it would be calling it GNOME. I remember > GNOME as an open source project before OpenSolaris was created. Sun didn't > do anything different with GNOME to include it with OpenSolaris, yet > somehow it has become a part of it all. Not true, take a look at svn co svn +ssh://anon at svn.opensolaris.org/svn/jds/spec-files/trunk We patch it to make it run on our kernel, fix some bugs, add trusted Solaris extensions Lots of the good original work for parts of Gnome are done by Sun JDS people. I hate it when people put down the hard work of our people. > If Sun would clearly define how > OpenSolaris is determined, there wouldn't be various views of it, > including me, you, John, Jim, or others... > > >From the above I suspect we're gonna be using IBM's ancient definition of > MLOC to determine what is a part of OpenSolaris.:-/ > > -- > > Alan DuBoff - Solaris x86 IHV/OEM Group > _______________________________________________ > opensolaris-summit mailing list > opensolaris-summit at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/opensolaris-summit David Clack Solaris X86 Evangelist Senior Systems Engineer OEM Software Sales Sun Microsystems 642, Chinook Ave SE, Ocean Shores, WA, USA, 98569 Phone +1-360-289-2158 Fax +1-360-289-2091 Mobile +1-206-265-1904 Email David.Clack at Sun.COM -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/opensolaris-summit/attachments/20070926/cc5dc1a9/attachment.html> From alan.coopersmith at sun.com Wed Sep 26 17:21:16 2007 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Wed, 26 Sep 2007 17:21:16 -0700 Subject: [opensolaris-summit] My comments (very subjective) on proposed Summit topics In-Reply-To: <Pine.GSO.4.64.0709261458090.17209@konocti> References: <46F2F56F.70706@Sun.COM> <20070920155952.A17130@bolthole.com> <46F2FE5F.3070508@sun.com> <20070920162935.B17130@bolthole.com> <46F308F4.8050101@Sun.COM> <20070920172025.D17130@bolthole.com> <46F31335.6060700@Sun.COM> <46F31530.9020102@sun.com> <18163.64275.538538.803812@gargle.gargle.HOWL> <46F47590.4010309@sun.com> <20070924100926.G14087@bolthole.com> <46F84BCF.1050801@sun.com> <Pine.GSO.4.64.0709241800240.26931@konocti> <46F8ADE7.2040707@sun.com> <x7myvapsae.fsf@richlowe.net> <Pine.GSO.4.64.0709251605500.7188@eagle.softorchestra.com> <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <46FA7E2D.3050508@Sun.Com> <Pine.GSO.4.64.0709261052530.7356@eagle.softorchestra.com> <46FAADD9.4090509@sun.com> <Pine.GSO.4.64.0709261458090.17209@konocti> Message-ID: <46FAF77C.10502@sun.com> Alan DuBoff wrote: > Not wanting to debate what OpenSolaris actually is, but it's interesting > that you can take a set of sources that were already open sourced by > other folks, include them on the OpenSolaris website and they become one > and the same. You prefer 100% original work then? Look no further than the Storage community, with their non-ON releases such as QFS & SAM - they even have a simple diagram, easily colored to show how much of their OpenSolaris work is outside of ON: http://www.opensolaris.org/os/community/storage/ But then, ON doesn't count if you're looking for 100% original - how much of it came from BSD's open source releases? -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From alan.duboff at sun.com Wed Sep 26 17:31:12 2007 From: alan.duboff at sun.com (Alan DuBoff) Date: Wed, 26 Sep 2007 17:31:12 -0700 (PDT) Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAE65E.1040003@sun.com> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> <46FAE65E.1040003@sun.com> Message-ID: <Pine.GSO.4.64.0709261729580.17209@konocti> On Wed, 26 Sep 2007, Alan Coopersmith wrote: > Are drivers ever reviewed by any arc other than PSARC? I'm not sure > LSARC, WSARC or FWARC need to be involved with this. I don't think so, but they are supposed to be by c-team. PSARC seems more interested in the interfaces and/or wether new ones are required or if the driver uses existing interfaces. -- Alan DuBoff - Solaris x86 IHV/OEM Group From John.Plocher at Sun.COM Wed Sep 26 17:43:19 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 26 Sep 2007 17:43:19 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAE504.6040108@Sun.COM> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> Message-ID: <46FAFCA7.2090701@Sun.Com> Joseph Kowalski wrote: > I'd assume that: > To: psarc-ext > CC: lsarc > CC: wsarc > ... > CC: <submitter> > > Should be OK. (John?) The problem is that external people can't send mail to WSARC and LSARC aliases directly, so those ARC members would be cut off from all external replies. The "best" answer right now is to create the case as a PSARC case, put the wsarc and lsarc aliases in the IAM:Interest list and send the mail to PSARC-ext only, letting the mail subsystem handle the notification and inclusion of the other ARCs. -John From Garrett.Damore at Sun.COM Wed Sep 26 21:44:40 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 21:44:40 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAFCA7.2090701@Sun.Com> References: <200709262251.l8QMpNWc016463@marduk.eng.sun.com> <46FAE504.6040108@Sun.COM> <46FAFCA7.2090701@Sun.Com> Message-ID: <46FB3538.4000302@sun.com> John Plocher wrote: > Joseph Kowalski wrote: > >> I'd assume that: >> To: psarc-ext >> CC: lsarc >> CC: wsarc >> ... >> CC: <submitter> >> >> Should be OK. (John?) > > The problem is that external people can't send mail to WSARC and LSARC > aliases directly, so those ARC members would be cut off from all external > replies. > > The "best" answer right now is to create the case as a PSARC case, > put the wsarc and lsarc aliases in the IAM:Interest list and send > the mail to PSARC-ext only, letting the mail subsystem handle the > notification and inclusion of the other ARCs. I really believe that this is only a PSARC issue. I doubt very much that other ARCs look at software that imports both the DDI and non-public kernel APIs. And note that I'm really only talking about *device drivers*.... not other kernel software. -- Garrett > > -John > > > From Garrett.Damore at Sun.COM Wed Sep 26 21:46:25 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 21:46:25 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FAFE9F.50809@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> Message-ID: <46FB35A1.2030101@sun.com> Artem Kachitchkine wrote: > > Why is this an ARC issue? Drivers don't usually go through ARC. it > should probably be on the ON C-Team Driver Checklist. Add DDICT to the > ON tool set and integrate it with wx. What I'm talking about is an ARC requirement to *disclose* non-DDI interface consumption by drivers, and to strongly discourage the practice of going outside of the DDI for device drivers. I think that *is* an issue for PSARC at least. (We're talking about interfaces here, not implementation.) -- Garrett > > -Artem From Garrett.Damore at Sun.COM Wed Sep 26 22:06:56 2007 From: Garrett.Damore at Sun.COM (Garrett D'Amore) Date: Wed, 26 Sep 2007 22:06:56 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB38F2.5090604@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> Message-ID: <46FB3A70.2020906@sun.com> Artem Kachitchkine wrote: > >> What I'm talking about is an ARC requirement to *disclose* non-DDI >> interface consumption by drivers, and to strongly discourage the >> practice of going outside of the DDI for device drivers. I think >> that *is* an issue for PSARC at least. > > How do you envision to enforce this in practice? Should C-team > communicate back to PSARC? It used to be mostly one-way, making it > bounce back and forth is thickening the process, while we're trying to > thin it, to encourage external contributors. > > > (We're talking about interfaces here, not implementation.) > > In this particular case, the line is unclear. A minor bugfix can turn > a DDI compliant driver into a non-compliant one. In my mind it's more > implementation (think "lint clean") than architecture. > > -Artem As a policy, I'm not sure how *hard and fast* the rule needs to be. I certainly think that when a case for a new driver goes before ARC, it should list the APIs it is using. This may mean that driver projects which were able to skip ARC altogether (although a poor choice, IMO, self-review should be used instead in such cases, IMO) now need to expose themselves to ARC. In and of itself, that's not too bad. And some APIs, like the GLDv3, should probably not cause a case to do more than self-review, so the cost of effort is minimal in such cases. Bug fixes that import APIs that the driver didn't have before.... I'm not so sure about. I'd hope that a fasttrack is not too burdensome. And again, maybe even some would qualify for self-review. Certainly the transition in going from DDI compliant to non-DDI-compliant is not one that I would want to just sweep under the rug. Better to announce that transition up front, IMO. (And if going non-DDI compliant is required to fix a bug, my first question would be "why?") -- Garrett From Artem.Kachitchkin at Sun.COM Wed Sep 26 22:00:34 2007 From: Artem.Kachitchkin at Sun.COM (Artem Kachitchkine) Date: Wed, 26 Sep 2007 22:00:34 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB35A1.2030101@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> Message-ID: <46FB38F2.5090604@sun.com> > What I'm talking about is an ARC requirement to *disclose* non-DDI > interface consumption by drivers, and to strongly discourage the > practice of going outside of the DDI for device drivers. I think that > *is* an issue for PSARC at least. How do you envision to enforce this in practice? Should C-team communicate back to PSARC? It used to be mostly one-way, making it bounce back and forth is thickening the process, while we're trying to thin it, to encourage external contributors. > (We're talking about interfaces here, not implementation.) In this particular case, the line is unclear. A minor bugfix can turn a DDI compliant driver into a non-compliant one. In my mind it's more implementation (think "lint clean") than architecture. -Artem From Artem.Kachitchkin at Sun.COM Wed Sep 26 22:17:07 2007 From: Artem.Kachitchkin at Sun.COM (Artem Kachitchkine) Date: Wed, 26 Sep 2007 22:17:07 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3A70.2020906@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> Message-ID: <46FB3CD3.5000909@sun.com> > Certainly the transition in going from DDI compliant to > non-DDI-compliant is not one that I would want to just sweep under the > rug. Better to announce that transition up front, IMO. (And if going > non-DDI compliant is required to fix a bug, my first question would be > "why?") I'm not disagreeing this that. I'm just wondering why ARC should be the one asking that question and not C-team/CRT. -Artem From Jim.Grisanzio at Sun.COM Thu Sep 27 06:56:57 2007 From: Jim.Grisanzio at Sun.COM (Jim Grisanzio) Date: Thu, 27 Sep 2007 22:56:57 +0900 Subject: [opensolaris-summit] [ogb-discuss] My comments (very subjective) on proposed Summit topics In-Reply-To: <5b5090780709251344u5b7221efo1d3b68965a97f53b@mail.gmail.com> References: <46F31335.6060700@Sun.COM> <46F483F1.1050805@Sun.COM> <46F4F5EC.9090909@sun.com> <46F81495.7080704@Sun.COM> <20070924200946.GA868815@sun.com> <46F84809.500@Sun.COM> <Pine.GSO.4.64.0709241749350.26931@konocti> <46f8dca3.xDcZ/DgOcTnPG6W6%Joerg.Schilling@fokus.fraunhofer.de> <20070925164008.GA277983@sun.com> <46F95E14.4000008@Sun.COM> <5b5090780709251344u5b7221efo1d3b68965a97f53b@mail.gmail.com> Message-ID: <46FBB6A9.8050307@sun.com> Brandorr wrote: > On 9/25/07, Darren Reed <Darren.Reed at sun.com> wrote: >> Heck, maybe we participants should just do that anyway, rather >> than wait for Sun to work out what it wants to do with OS.o? > > While I understand your frustration at a low level of outside > contribution, I don't think Sun is to be blamed. I think it stems from > the fact that Solaris is a mature, formerly proprietary, product. Many > users of Solaris today don't give a care about any notions of > community of contribution. (So long as the get what their support > contracts cover) > > We here at OpenSolaris.org are currently in the process of building a > community. This community currently exists mostly of Sun employees, > and a few brave new worlders, and a few excited Solaris customers. > > We are in a bit of a chicken and egg situation. What we need right now > is more outside manpower, to build the infrastructure to make it > easier for outside contributers to contribute. > > Right now, one way we can address this, is to do outreach. Currently > this is on the agenda for our next user group meeting: This is a very nice list below. Perhaps you can post these in Advocacy? I'm sure there are many contribution docs within various community groups and projects, but here are two general contribution docs: http://opensolaris.org/os/community/advocacy/help/ http://opensolaris.org/os/communities/participation/ Jim > Get involved: > ------------------ > - Mentoring - We are looking at starting a new mentoring program. We > are looking for Seniors that would be interested in mentoring Juniors. > (Developers and sysadmins). Please email Brian Gupta > (brandorr at opensolaris.org) if you are interested. > - The OpenSolaris.org Systems administration Community Group is > looking to expand membership. > syadmin-discuss-subscribe at opensolaris.org > - Volunteering to help build a new OpenSolaris user portal, as well as > an NYOSUG.org website. (email Brian Gupta (brandorr at opensolaris.org) > if you are interested in either.) > - We will be launching/announcing a virtual development lab next > month, that will allow anyone who is interested in developing > OpenSolaris or porting open source apps to OpenSolaris, to gain free > access to VMWare ESX Virtual Machines. Let Brian Gupta > (brandorr at opensolaris.org) know if you are interested > in either helping administer ESX, or if you want access to a VM. (It's > not setup yet). > - We can use help planning these UG meetings. Mail Isaac or Brian Gupta > "Isaac Rozenfeld - NEW YORK" <Isaac.Rozenfeld at sun.com>, Brian Gupta > (brandorr at opensolaris.org) > - Brian Gupta (brandorr at opensolaris.org) is going to the first > OpenSolaris Developers Summit, and will have more to report next > month. > (http://www.genunix.org/wiki/index.php?title=OpenSolaris_Summit.) > - Brian Gupta (brandorr at opensolaris.org) assembled the following new > user FAQ: assembling: > http://www.genunix.org/wiki/index.php?title=FAQ_Prototype > > What I would ask each contributing member of OpenSolaris.org, is to > find at least one other person that has the ability to contribute, > and walk them through their first contribution. (We'd double the > number of contributers over night. > > Also, UserGroups should be actively advertising outside of their own > users. e.g. - Reach out to Linux and BSD user groups, as well as local > UNIGROUP, Lisa and SAGE chapters. The goal being to increase the > number of attendees to your meetings. > > Basically, I feel that by working around the barriers for now, we will > be able to grow the community enough to remove them down the road. > (With the added manpower). From John.Plocher at Sun.COM Thu Sep 27 09:08:29 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 27 Sep 2007 09:08:29 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB35A1.2030101@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> Message-ID: <46FBD57D.1020100@Sun.Com> Garrett D'Amore wrote: > Artem Kachitchkine wrote: >> Why is this an ARC issue? > I think that > *is* an issue for PSARC at least. This issue is on the boundry between the "systems" architecture of OpenSolaris and the design/implementation of the ON subsystem. IFF device drivers were an implementation detail of the OpenSolaris ON consolidation alone, then the ARCs would not really care about it, since they could safely delegate the whole shebang to the people managing ON. If ON wished to do a global refactoring or flag day or whatever, they /could/ also go thru and update every single driver to match, and all such changes would be invisible to everyone outside of ON. Of course, the above assumption is false. Drivers are written outside of the ON consolidation; there is even early pie-in-the-sky talk about refactoring the existing ON to potentially separate out things like drivers into their own consolidation(s). This makes the driver interface to ON part of the systems integration conversation - it no longer is simply an implementation detail of ON. -John From John.Plocher at Sun.COM Thu Sep 27 09:14:53 2007 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 27 Sep 2007 09:14:53 -0700 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3CD3.5000909@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> <46FB3CD3.5000909@sun.com> Message-ID: <46FBD6FD.60003@Sun.Com> Artem Kachitchkine wrote: > I'm not disagreeing this that. I'm just wondering why ARC should be the > one asking that question and not C-team/CRT. One reason may be that the ON C-Team right now is a Sun-only closed entity, this is the /Open/Solaris community, and the ARC has more visibility and/or transparency in the community than does the C-Team. Another is that not all drivers from the community are going to be of interest to the Sun ON C-Team... -John From james.d.carlson at Sun.COM Thu Sep 27 05:33:00 2007 From: james.d.carlson at Sun.COM (James Carlson) Date: Thu, 27 Sep 2007 08:33:00 -0400 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3CD3.5000909@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> <46FB3CD3.5000909@sun.com> Message-ID: <18171.41724.616477.425290@gargle.gargle.HOWL> Artem Kachitchkine writes: > > Certainly the transition in going from DDI compliant to > > non-DDI-compliant is not one that I would want to just sweep under the > > rug. Better to announce that transition up front, IMO. (And if going > > non-DDI compliant is required to fix a bug, my first question would be > > "why?") > > I'm not disagreeing this that. I'm just wondering why ARC should be the > one asking that question and not C-team/CRT. I think you might be confusing policy with enforcement. The requirement itself (if we were to approve one) is a system architectural issue. We would be saying that, despite the clear indications to the contrary in the interface taxonomy and existing ARC practice, projects of a certain type may not actually use Consolidation Private interfaces from their own consolidation without additional review. That's the ARC issue. _Enforcement_ of these requirements is strictly in the hands of the applicable C-teams. Just as with our other rules, it's up to the C-team to make sure that the project team has properly complied with all applicable review requirements, architectural restrictions, and the like. Yes, if I were on the C-team, I'd be advocating a new checklist item and probably some new local policies to make sure that we catch any mistakes that might be made. Those things aren't architectural, though. Only the policy (disallowing unreviewed use of private interfaces, and thus mandating architectural review) itself is. -- James Carlson, Solaris Networking <james.d.carlson at sun.com> Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677 From sommerfeld at sun.com Thu Sep 27 05:30:36 2007 From: sommerfeld at sun.com (Bill Sommerfeld) Date: Thu, 27 Sep 2007 08:30:36 -0400 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB3CD3.5000909@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <46FB3A70.2020906@sun.com> <46FB3CD3.5000909@sun.com> Message-ID: <1190896236.5428.26.camel@localhost> On Wed, 2007-09-26 at 22:17 -0700, Artem Kachitchkine wrote: > I'm not disagreeing this that. I'm just wondering why ARC should be the > one asking that question and not C-team/CRT. If our goal is to have a DDI that is sufficient for all types of drivers then the ARC needs to know about the holes in it that are found in practice. The c-team/crt is the last line of defense as always and would in fact be the ones asking the question if the project team hadn't figured it out for themselves. When the answer is "no, we couldn't get it done within the DDI", it's a deficiency in the DDI and (in a world where the DDI is supposed to be truly complete) the ARC really needs to know about this. - Bill From james.d.carlson at Sun.COM Thu Sep 27 05:34:53 2007 From: james.d.carlson at Sun.COM (James Carlson) Date: Thu, 27 Sep 2007 08:34:53 -0400 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB38F2.5090604@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> Message-ID: <18171.41837.171584.86169@gargle.gargle.HOWL> Artem Kachitchkine writes: > > (We're talking about interfaces here, not implementation.) > > In this particular case, the line is unclear. A minor bugfix can turn a > DDI compliant driver into a non-compliant one. In my mind it's more > implementation (think "lint clean") than architecture. That has always been true. Even a one-line bug fix may well have important architectural implications, just as a multi-kiloline change might not have any architectural issues worth reviewing. You can't determine ARC issues by running "wc -l". -- James Carlson, Solaris Networking <james.d.carlson at sun.com> Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677 From jek3 at sun.com Thu Sep 27 10:53:25 2007 From: jek3 at sun.com (Joseph Kowalski) Date: Thu, 27 Sep 2007 07:53:25 -1000 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <46FB35A1.2030101@sun.com> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> Message-ID: <46FBEE15.5010101@sun.com> Garrett D'Amore wrote: > Artem Kachitchkine wrote: >> >> Why is this an ARC issue? Drivers don't usually go through ARC. it >> should probably be on the ON C-Team Driver Checklist. Add DDICT to >> the ON tool set and integrate it with wx. > > What I'm talking about is an ARC requirement to *disclose* non-DDI > interface consumption by drivers, and to strongly discourage the > practice of going outside of the DDI for device drivers. I think that > *is* an issue for PSARC at least. (We're talking about interfaces > here, not implementation.) Sorry, I was suggesting methods for the general case. Yea, this is pretty much a "PSARC only" issue. Right now, drivers often don't need ARC involvement (beyond the "register the name" paperwork - closed approved automatic). I think the "closed approved" path is only appropriate (already) for DDI-compliant drivers... if <DDI-compliant> -> closed approved automatic else -> fast-track (or greater) I think the only thing we need to do is to advertise and enforce the above, existing rule. The question is "how to enforce". I really don't have a suggestion for that. (I suspect somebody will!) - jek3 From jek3 at sun.com Thu Sep 27 11:07:58 2007 From: jek3 at sun.com (Joseph Kowalski) Date: Thu, 27 Sep 2007 08:07:58 -1000 Subject: [opensolaris-summit] driver limits [was Re: My comments (very subjective) on proposed Summit topics] In-Reply-To: <18171.41837.171584.86169@gargle.gargle.HOWL> References: <46F999A1.7090900@sun.com> <Pine.GSO.4.64.0709251805100.7188@eagle.softorchestra.com> <46F9D5DF.7030502@sun.com> <Pine.GSO.4.64.0709252112140.7188@eagle.softorchestra.com> <x7zlzaj9lv.fsf@richlowe.net> <Pine.GSO.4.64.0709252248500.7188@eagle.softorchestra.com> <46FA0051.2010800@sun.com> <Pine.GSO.4.64.0709252354430.7188@eagle.softorchestra.com> <18170.16644.873233.737411@gargle.gargle.HOWL> <46FA783D.7050803@sun.com> <20070926120306.B91989@bolthole.com> <46FAAF2D.4030901@sun.com> <46FAFE9F.50809@sun.com> <46FB35A1.2030101@sun.com> <46FB38F2.5090604@sun.com> <18171.41837.171584.86169@gargle.gargle.HOWL> Message-ID: <46FBF17E.1060400@sun.com> James Carlson wrote: > You can't determine ARC issues by running "wc -l". > Poster boy: The DVD worth of materials for SUS6 was a fast-track. - jek3 From Jesse.Silver at Sun.COM Sat Sep 29 13:32:46 2007 From: Jesse.Silver at Sun.COM (Jesse Silver) Date: Sat, 29 Sep 2007 13:32:46 -0700 Subject: [opensolaris-summit] Hotel Reservations Message-ID: <46FEB66E.30109@sun.com> To those in need of hotel accommodations: The Summit is closing in fast. If you need a hotel room and have yet to reserve one, please call the Coast Santa Cruz Hotel immediately at 831-426-4330. Reference "Sun Microsystems" to get our special rate of $189/night. The rooms WILL BE FILLED by the end of this week, so please call now to secure one for yourself. If you have any trouble with the reservations people, or are told that the room block is full, please call me, Jesse Silver, at 877-868-6937. Thanks, and we're looking forward to a great summit, Jesse