From rac at eastlink.ca Mon Feb 11 23:44:25 2008 From: rac at eastlink.ca (MC) Date: Mon, 11 Feb 2008 23:44:25 PST Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions Message-ID: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> > With the three years of experience since then, Sun > has decided that it > would be better for both the community and for its > own business, if > distros were allowed to use the OpenSolaris trademark > so that they could > form an ecosystem of compatible software package and > services. > > To jumpstart this ecosystem, Sun has decided to fund > Project Indiana to > build a base distro, which it will name OpenSolaris, > and to allow others > to use the OpenSolaris trademark in approved ways in > the name or > marketing of their distros. I think this is a great reply. Things can move quickly now that everyone can start working from this new starting point. I suggest a new clear branding initiative spearheaded by Sun. OpenSolaris (the distro) OpenSolaris Codebase (the new name for the entire OpenSolaris code base) For other-distro compatibility certification and trademark usage, I see four likely situations to address: 1) Branding OpenSolaris with the certification brand (what to do here comes after having decided what happens below) 2) Branding other distros built using the OpenSolaris distro constructor as compatible and "built with"[1] certification brands 3) Branding other distros that use the OpenSolaris Codebase as "compatible"[1] certification brand 4) Branding a distro that uses OpenSolaris code but has not passed the certification process[1] [1] The compatibility criteria that must be met to gain use of the compatibility certification brands has yet to be decided. Below is my discussion about these situations. See the summary at the bottom to skip the details. Re: #2 and #3: After thinking about it, I don't see a benefit for Sun or anyone else in having both the "Built with [distro constructor]" and "Compatible with" brands. I see no reason to give the use of distro constructor such importance. I think a rigorous set of requirements to meet to be "OpenSolaris Compatible" is sufficient, and perhaps using the distro constructor would buy you an easier path through the certification process. Note that "Compatible with OpenSolaris" and "OpenSolaris Compatible" are synonym trademarks-brands here. So in this new world I'm suggesting, the recent NexentaCore press release could have said something like "NexentaCore is a free and open source operating system combining the OpenSolaris Codebase with GNU application userland. It is OpenSolaris and Linux 2.4 compatible." (the breakup of the "OpenSolaris Compatible" trademark is on purpose to suggest a real world use IF Sun is kind and not too tough on enforcement of the exact brand). The NexentaCore website and docs and installer could then splash the "OpenSolaris Compatible" logo everywhere ala "Intel Inside". Re: #4: It is easy to say "Built on/with OpenSolaris". Either you don't claim that as a trademark now and you let both incompatible and compatible distros use it, or you claim it now and decide what it means. I suggest claiming it as a synonym for OpenSolaris Compatible. Then to deal with incompatible distros, you let them use a new brand-trademark: "Built with the OpenSolaris Codebase". So you get the OpenSolaris brand out there even on incompatible distros. To prevent confusion you A) make the branding art significantly different, B) strongly discourage the use of "Built on/with OpenSolaris" in favor of "OpenSolaris Compatible" and "Compatible with OpenSolaris", and C) keep an eye on incompatible distros to make sure they follow the rules. OR you could do none of this and let the dice roll. But whatever you do I'd consider heading these issues off at the pass and control that fire before it starts. This leaves #1: Branding the OpenSolaris distro (note: that is a term like the "Linux kernel"). Do you give it "OpenSolaris Compatible" and "Built with the OpenSolaris Codebase"? Confusion here (incompatible distros get "Built with" and compatible ones with "Compatible"?) is a good reason to kill off or control the use of "Built with" before it starts. Summary AKA My actions if I represented Sun. Let any distro that passes the certification tests use "OpenSolaris Compatible" and "Compatible with OpenSolaris". Play hardball with anyone running an incompatible distro and let them use no OpenSolaris trademark except "OpenSolaris Codebase". Reserve the "Built with" and "Built on" permutations to control their use as situations demand. "OpenSolaris is now officially an OpenSolaris Compatible distribution built with the OpenSolaris Codebase." From John.Plocher at Sun.COM Tue Feb 12 14:40:22 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 12 Feb 2008 14:40:22 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <27261788.1202802295282.JavaMail.Twebapp@oss-app1> References: <27261788.1202802295282.JavaMail.Twebapp@oss-app1> Message-ID: <47B22056.9040909@Sun.Com> [Followups set to trademark-policy-dev at opensolaris.org -John] MC wrote: > I think this is a great reply. Things can move quickly now that everyone > can start working from this new starting point. I agree - and have updated the Trademark and Branding wiki page to reflect my understanding of both Bill's and your comments. http://www.genunix.org/wiki/index.php/Trademark_usage_and_Branding_guideline Feedback, updates and comments welcome here or directly on the wiki pages themselves. > 2) Branding other distros built using the OpenSolaris distro constructor > as compatible and "built with"[1] certification brands > 3) Branding other distros that use the OpenSolaris Codebase as "compatible" As you suggest, I'm not sure if there is a difference here, or if it matters. The "built with some distro constructor", "built with some binary set of packages", and other similar metrics all start to sound like mechanisms to get to a goal, rather than the goal itself. That is, if the goal is an ecosystem of compatible distros linked together thru a shared set of application repositories, does it really matter what specific mechanism a given distro actually uses, as long as the result (interoperability and compatibility) is maintained? IMO, the branding needs to capture the successful attainment of the goal while being agnostic about the details of how the goal was reached - after all, the mechanism will evolve over time as it gets improved, won't it? -John From James.Walker at Sun.COM Tue Feb 12 22:49:01 2008 From: James.Walker at Sun.COM (Jim Walker) Date: Tue, 12 Feb 2008 23:49:01 -0700 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> Message-ID: <47B292DD.7060407@sun.com> MC wrote: > Summary AKA My actions if I represented Sun. > > Let any distro that passes the certification tests use "OpenSolaris > Compatible" and "Compatible with OpenSolaris". Play hardball with > anyone running an incompatible distro and let them use no OpenSolaris > trademark except "OpenSolaris Codebase". Reserve the "Built with" > and "Built on" permutations to control their use as situations > demand. > Now we just need to determine what is the minimum set of code from the OpenSolaris Codebase needed to define an OpenSolaris Distribution? Once we have this we can talk about certification. Cheers, Jim From John.Plocher at Sun.COM Tue Feb 12 23:10:02 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 12 Feb 2008 23:10:02 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <9797729.1202878928208.JavaMail.Twebapp@oss-app1> References: <9797729.1202878928208.JavaMail.Twebapp@oss-app1> Message-ID: <47B297CA.1040901@Sun.Com> [lets move this to the trademark and branding discussion alias, since almost everyone here is subscribed there as well. -John] Rainer Heilke wrote: >>From Sun's text: > > "...and to allow others to use the OpenSolaris trademark in approved > ways in the name or marketing of their distros." > > They have opened a forum for dialog. ... I've updated the T&B wiki to reflect this update, the existing T&B stuff from Sara (enthusiast buttons, fair use...) and references to the other various places T&B are mentioned - the OS.o charter, Sun's T&B policy, etc. Please read over what is there with an eye towards answering the question "what more do we wish this document to say?" One thing that has come out of the OGB/BillF dialog is that I now know all the Sun people who are involved in making T&B decisions about OpenSolaris, and we all are starting to understand and value all the various perspectives that are in play. Some of these people can't easily contribute directly to this forum, but they all are committed to working to make this exercise work in some sort of win-win way for both Sun and the OS.o community. Yes, there are some things that Sun can't change, others that aren't easy to change, and still others that Sun won't change. Having said that, though, I believe that there are positive things we *can* do together. I'd ask that, whether we individually agree or disagree with Sun's position statement, we commit to working together to make what we do have work for us all. http://www.genunix.org/wiki/index.php/Trademark_usage_and_Branding_guideline -John From John.Plocher at Sun.COM Tue Feb 12 23:20:11 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 12 Feb 2008 23:20:11 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B292DD.7060407@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B292DD.7060407@sun.com> Message-ID: <47B29A2B.2040807@Sun.Com> Jim Walker wrote: > Now we just need to determine what is the minimum set of code from > the OpenSolaris Codebase needed to define an OpenSolaris Distribution? > > Once we have this we can talk about certification. If the goal is along the lines of "being able to download and run anything in an OpenSolaris application repository", then we *could* try the existence proof of letting the community of users determine whether or not a distro actually does this. It should be relatively easy to spot the failures :-) They will be the ones that can't download stuff from repositories, ones that are not binary compatible because someone hacked libc, etc. Going down the road of defining a Java community style reference platform and conformance test suite seems way too resource intensive to be of much immediate value... Yeah, I know - this is all heresy to someone as involved with testing as you are :-) -John From James.Walker at Sun.COM Wed Feb 13 01:08:37 2008 From: James.Walker at Sun.COM (Jim Walker) Date: Wed, 13 Feb 2008 02:08:37 -0700 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B29A2B.2040807@Sun.Com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B292DD.7060407@sun.com> <47B29A2B.2040807@Sun.Com> Message-ID: <47B2B395.2000900@sun.com> John Plocher wrote: > Jim Walker wrote: >> Now we just need to determine what is the minimum set of code from >> the OpenSolaris Codebase needed to define an OpenSolaris Distribution? >> >> Once we have this we can talk about certification. > > > If the goal is along the lines of "being able to download and run > anything in an OpenSolaris application repository", then we *could* > try the existence proof of letting the community of users determine > whether or not a distro actually does this. > > It should be relatively easy to spot the failures :-) They will > be the ones that can't download stuff from repositories, ones that > are not binary compatible because someone hacked libc, etc. > Going down the road of defining a Java community style reference > platform and conformance test suite seems way too resource intensive > to be of much immediate value... > > Yeah, I know - this is all heresy to someone as involved with > testing as you are :-) Actually, I'm not a major proponent of a Comprehensive OpenSolaris Certification Test Suite, even though it would guarantee employment to many test engineers long into the future. I don't think it would be as Agile as a package certification model which would encompass both application and core OpenSolaris packages, whatever "core" ends up being. We could have certified and uncertified packages. The certified ones would have some type of Software Quality Assurance process associated with them. By the way, an objective customer bug vote method like you describe, would be a good addition to the SQA process. In addition, maybe the "core" packages could be dynamic, such that the current top 10% in terms of usage or inclusion in distributions defines them. I love adaptive systems. Certification at the distribution level is a great quest for a knight who likes to fight battles for the love of fighting, but I think there will be too much variance at that level to have one standard that we can all agree on, and an arbitrary standard is worse than no standard in terms of the cost on people and business and time. Of course this doesn't preclude organizations like Sun to invest billions of dollars to establish and maintain their own distribution quality level, since in the final analysis, the burden of proof remains with the distribution owner. They are the ones that have to answer to their customers regardless of what certification they pass, and all of us benefit when they answer in a good way by making yet another improvement to our common code base. Cheers, Jim From Joerg.Schilling at fokus.fraunhofer.de Wed Feb 13 07:59:27 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Wed, 13 Feb 2008 16:59:27 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> Message-ID: <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> MC wrote: > I suggest a new clear branding initiative spearheaded by Sun. > > OpenSolaris (the distro) > OpenSolaris Codebase (the new name for the entire OpenSolaris code base) If you use the same name for a code base and a distro, you confuse people. 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 Eric.Boutilier at Sun.COM Wed Feb 13 09:15:26 2008 From: Eric.Boutilier at Sun.COM (Eric Boutilier) Date: Wed, 13 Feb 2008 11:15:26 -0600 (CST) Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B2B395.2000900@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B292DD.7060407@sun.com> <47B29A2B.2040807@Sun.Com> <47B2B395.2000900@sun.com> Message-ID: +1 On Wed, 13 Feb 2008, Jim Walker wrote: > > Actually, I'm not a major proponent of a Comprehensive OpenSolaris > Certification Test Suite, even though it would guarantee employment > to many test engineers long into the future. I don't think it would > be as Agile as a package certification model which would encompass > both application and core OpenSolaris packages, whatever "core" > ends up being. We could have certified and uncertified packages. > The certified ones would have some type of Software Quality > Assurance process associated with them. By the way, an objective > customer bug vote method like you describe, would be a good addition > to the SQA process. In addition, maybe the "core" packages could > be dynamic, such that the current top 10% in terms of usage or > inclusion in distributions defines them. I love adaptive systems. > > Certification at the distribution level is a great quest for a knight > who likes to fight battles for the love of fighting, but I think > there will be too much variance at that level to have one standard > that we can all agree on, and an arbitrary standard is worse than no > standard in terms of the cost on people and business and time. > > Of course this doesn't preclude organizations like Sun to invest > billions of dollars to establish and maintain their own distribution > quality level, since in the final analysis, the burden of proof remains > with the distribution owner. They are the ones that have to answer > to their customers regardless of what certification they pass, and > all of us benefit when they answer in a good way by making yet another > improvement to our common code base. > > Cheers, > Jim > From Alan.Coopersmith at Sun.COM Wed Feb 13 09:18:57 2008 From: Alan.Coopersmith at Sun.COM (Alan Coopersmith) Date: Wed, 13 Feb 2008 09:18:57 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47B32681.4060609@sun.com> Joerg Schilling wrote: > MC wrote: > >> I suggest a new clear branding initiative spearheaded by Sun. >> >> OpenSolaris (the distro) >> OpenSolaris Codebase (the new name for the entire OpenSolaris code base) > > If you use the same name for a code base and a distro, you confuse people. Seems to work for the BSD's. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From Joerg.Schilling at fokus.fraunhofer.de Wed Feb 13 09:22:12 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Wed, 13 Feb 2008 18:22:12 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B32681.4060609@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47B32681.4060609@sun.com> Message-ID: <47b32744.fSGg7JnygmWWV3Kj%Joerg.Schilling@fokus.fraunhofer.de> Alan Coopersmith wrote: > Joerg Schilling wrote: > > MC wrote: > > > >> I suggest a new clear branding initiative spearheaded by Sun. > >> > >> OpenSolaris (the distro) > >> OpenSolaris Codebase (the new name for the entire OpenSolaris code base) > > > > If you use the same name for a code base and a distro, you confuse people. > > Seems to work for the BSD's. But they have only one name per fork. 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 swalker at opensolaris.org Wed Feb 13 09:22:13 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Wed, 13 Feb 2008 11:22:13 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Feb 13, 2008 9:59 AM, Joerg Schilling wrote: > MC wrote: > > > I suggest a new clear branding initiative spearheaded by Sun. > > > > OpenSolaris (the distro) > > OpenSolaris Codebase (the new name for the entire OpenSolaris code base) > > If you use the same name for a code base and a distro, you confuse people. Only for those easily confused. Many, many, projects use the same name for their "distributed software" as their codebase. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Wed Feb 13 10:01:02 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Wed, 13 Feb 2008 10:01:02 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b32744.fSGg7JnygmWWV3Kj%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47B32681.4060609@sun.com> <47b32744.fSGg7JnygmWWV3Kj%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47B3305E.7000001@Sun.Com> Joerg Schilling wrote: > But they have only one name per fork. So, let's not fork. Seriously. Yes, it is nice and easy to fork off an independent distro, but that road leads to distro hell for users. It is obviously much harder to work together with others to build a 'better something' in common, especially when we all seem to have an excess of thumbs, cliques that don't talk to each other, hidden and not so hidden agendas, language, culture and philosophy barriers, stubbornness and wounded pride. In spite of all the valid and reasonable reasons to fragment and go it alone, I believe we need to take a big breath and commit to finding ways to rise above it all and work together. -John From Alan.Coopersmith at Sun.COM Wed Feb 13 11:10:43 2008 From: Alan.Coopersmith at Sun.COM (Alan Coopersmith) Date: Wed, 13 Feb 2008 11:10:43 -0800 Subject: [trademark-policy-dev] [osol-discuss] [indiana-discuss] [advocacy-discuss] OpenSolaris Developer Preview 2 Available In-Reply-To: <47b32e26.Z3pmjd3rxRZpksgy%Joerg.Schilling@fokus.fraunhofer.de> References: <47B24DE3.60102@sun.com> <47b3248f.G10i6Wtx14PjVvq6%Joerg.Schilling@fokus.fraunhofer.de> <47B32DAF.4000003@sun.com> <47b32e26.Z3pmjd3rxRZpksgy%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47B340B3.50001@sun.com> Joerg Schilling wrote: > BTW: I vote against allowing to replace /sbin/sh by something different than a > 100% compatible Bourne shell for distributions that like to use the OpenSolaris > trademark. Discussions of what properties distros must have to use the OpenSolaris trademark are being held on trademark-policy-dev at opensolaris.org, and not any of the other mailing lists you cc'ed - take it there or expect them to not see it. (Followups set to there.) -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From swalker at opensolaris.org Wed Feb 13 16:01:24 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Wed, 13 Feb 2008 18:01:24 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B2B395.2000900@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B292DD.7060407@sun.com> <47B29A2B.2040807@Sun.Com> <47B2B395.2000900@sun.com> Message-ID: On Feb 13, 2008 3:08 AM, Jim Walker wrote: > to many test engineers long into the future. I don't think it would > be as Agile as a package certification model which would encompass > both application and core OpenSolaris packages, whatever "core" > ends up being. We could have certified and uncertified packages. > The certified ones would have some type of Software Quality > Assurance process associated with them. By the way, an objective > customer bug vote method like you describe, would be a good addition > to the SQA process. In addition, maybe the "core" packages could > be dynamic, such that the current top 10% in terms of usage or > inclusion in distributions defines them. I love adaptive systems. I might note that Ubuntu takes a similar approach by having a "supported" package repository and unsupported ones. > Certification at the distribution level is a great quest for a knight > who likes to fight battles for the love of fighting, but I think > there will be too much variance at that level to have one standard > that we can all agree on, and an arbitrary standard is worse than no > standard in terms of the cost on people and business and time. I'm not convinced of that just yet; I think there's room for compromise here. Maybe there's too much in a distribution the size of say, Solaris Express, to certify. However, in a distribution that fits on a CD (such as Project Indiana), there may be close to the right amount. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From mb1x at gmx.com Wed Feb 13 16:49:22 2008 From: mb1x at gmx.com (Martin Bochnig) Date: Thu, 14 Feb 2008 01:49:22 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B292DD.7060407@sun.com> <47B29A2B.2040807@Sun.Com> <47B2B395.2000900@sun.com> Message-ID: <20080214004922.285300@gmx.net> > I might note that Ubuntu takes a similar approach by having a > "supported" package repository and unsupported ones. [...] > I'm not convinced of that just yet; I think there's room for compromise > here. > > Maybe there's too much in a distribution the size of say, Solaris > Express, to certify. However, in a distribution that fits on a CD > (such as Project Indiana), there may be close to the right amount. What does it have to do with the 700MB size of a typical CD? Conary itself would definitely fit onto the existing LiveCD at almost no cost (in addition to pkgadd and IPS, itis only a few megabytes). I have apologized (and hereby would like to do so again) to Dave Miner and the IPS team for my initial unfriendly mails (there had been a misunderstanding) and we are looking forward to find the best approach (IMHO: Maybe one should merge the best out of both worlds?). This will be discussed on the public lists and I am looking forward to it. But in advance you can expect us from opensolaris's conary project to release testing packages, diffs, a src tree and Mini-HOWTO. p.s. As there isn't a conary-eval discuss list yet, and as it does matter to the pkg- and install- communities, I have added those two. Sorry for cross-posting. Till then, thank you. http://www.opensolaris.org/os/project/conary-eval/ From brandorr at opensolaris.org Wed Feb 13 20:46:41 2008 From: brandorr at opensolaris.org (Brandorr) Date: Wed, 13 Feb 2008 23:46:41 -0500 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> Many open source projects rename the codebase/project when they run into conflict with a commercial entity over a trademark. Even though I had grown fond of the name OpenSolaris, as it invokes ties to it's ancestry in Solaris, we really don't have a choice now but to rename the project. (Because Sun is limiting use of the name.) As an advocate of all the distros and their common code base, I feel that a common name to tie them all together is called for. (Now that Sun has decided to confuse the issue. Frankly it is not a legal issue, other open source projects protect their trademarks without making a distro. e.g. - Linux, has a trademark foundation protecting it's trademark. http://www.linuxmark.org/ Arguing that the law requires to take the actions they did, is not accurate, and is deliberately confusing the issue. Sun had many choices, least of which being not using the OpenSolaris trademark in a way that was incongruous with it's originally intended uses. -Brian On Feb 13, 2008 10:59 AM, Joerg Schilling < Joerg.Schilling at fokus.fraunhofer.de> wrote: > MC wrote: > > > I suggest a new clear branding initiative spearheaded by Sun. > > > > OpenSolaris (the distro) > > OpenSolaris Codebase (the new name for the entire OpenSolaris code base) > > If you use the same name for a code base and a distro, you confuse people. > > 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 > _______________________________________________ > trademark-policy-dev mailing list > trademark-policy-dev at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/trademark-policy-dev > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ http://www.genunix.org/wiki/index.php/OpenSolaris_New_User_FAQ -------------- next part -------------- An HTML attachment was scrubbed... URL: From swalker at opensolaris.org Wed Feb 13 20:54:13 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Wed, 13 Feb 2008 22:54:13 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> Message-ID: On Feb 13, 2008 10:46 PM, Brandorr wrote: > Many open source projects rename the codebase/project when they run into > conflict with a commercial entity over a trademark. Even though I had grown > fond of the name OpenSolaris, as it invokes ties to it's ancestry in > Solaris, we really don't have a choice now but to rename the project. > (Because Sun is limiting use of the name.) Sun has little choice in limiting usage of the name, especially since the Solaris trademark is contained within the OpenSolaris trademark. They have a duty to their shareholders to protect the usage of their trademarks vigorously and under the terms of trademark law. > As an advocate of all the distros and their common code base, I feel that a > common name to tie them all together is called for. (Now that Sun has > decided to confuse the issue. Frankly it is not a legal issue, other open > source projects protect their trademarks without making a distro. e.g. - > Linux, has a trademark foundation protecting it's trademark. > > http://www.linuxmark.org/ > > Arguing that the law requires to take the actions they did, is not accurate, The action of using their name? No. Protecting their trademark and having very specific guidelines for its use? Yes. Again, the situation with "Linux" is different. There is no Linux product. There is a Linux trademark that is allowed to be used by anyone that incorporate the Linux product into their own. The Solaris trademark however, is different because it presents both a product and a codebase (indirectly) through the OpenSolaris trademark. I don't understand why people attempt to constantly compare the Linux trademark situation as being equivalent to that of the Solaris/OpenSolaris one; it is not equivalent by any stretch of the imagination. Linux is a kernel; Solaris/OpenSolaris is a complete operating system and product. The Linux foundation does not directly market a commercial offering; Sun does. > and is deliberately confusing the issue. Sun had many choices, least of > which being not using the OpenSolaris trademark in a way that was > incongruous with it's originally intended uses. Who intended what? Can anyone say here that Sun originally intended to *never* use the trademark for a distribution? I don't think so. Maybe it wasn't in the initial plans, but it seems illogical to me to conclude "never." -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Alan.Coopersmith at Sun.COM Wed Feb 13 21:03:25 2008 From: Alan.Coopersmith at Sun.COM (Alan Coopersmith) Date: Wed, 13 Feb 2008 21:03:25 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> Message-ID: <47B3CB9D.8010109@sun.com> I think you misunderstand the argument - Sun did not use the OpenSolaris name on Indiana to protect the mark - it's keeping final say over other users to protect their mark. Sun has always limited use of the OpenSolaris name this way, and with these moves is allowing more uses than have been allowed in the past - previously no distro could use it, now they've granted use of the name to Indiana already, and asked for the community to join in developing guidelines to determine how other distros can use it as well. -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering Brandorr wrote: > Many open source projects rename the codebase/project when they run into > conflict with a commercial entity over a trademark. Even though I had > grown fond of the name OpenSolaris, as it invokes ties to it's ancestry > in Solaris, we really don't have a choice now but to rename the project. > (Because Sun is limiting use of the name.) > > As an advocate of all the distros and their common code base, I feel > that a common name to tie them all together is called for. (Now that Sun > has decided to confuse the issue. Frankly it is not a legal issue, other > open source projects protect their trademarks without making a distro. > e.g. - Linux, has a trademark foundation protecting it's trademark. > > http://www.linuxmark.org/ > > Arguing that the law requires to take the actions they did, is not > accurate, and is deliberately confusing the issue. Sun had many choices, > least of which being not using the OpenSolaris trademark in a way that > was incongruous with it's originally intended uses. > > -Brian > > On Feb 13, 2008 10:59 AM, Joerg Schilling > > wrote: > > MC > wrote: > > > I suggest a new clear branding initiative spearheaded by Sun. > > > > OpenSolaris (the distro) > > OpenSolaris Codebase (the new name for the entire OpenSolaris code > base) > > If you use the same name for a code base and a distro, you confuse > people. > > 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 > _______________________________________________ > trademark-policy-dev mailing list > trademark-policy-dev at opensolaris.org > > http://mail.opensolaris.org/mailman/listinfo/trademark-policy-dev > > > > > -- > - Brian Gupta > > http://opensolaris.org/os/project/nycosug/ > > http://www.genunix.org/wiki/index.php/OpenSolaris_New_User_FAQ > > > ------------------------------------------------------------------------ > > _______________________________________________ > trademark-policy-dev mailing list > trademark-policy-dev at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/trademark-policy-dev From webmink at sun.com Thu Feb 14 00:36:37 2008 From: webmink at sun.com (Simon Phipps) Date: Thu, 14 Feb 2008 08:36:37 +0000 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> Message-ID: <6FB26979-599B-41A8-836A-9C22C0434DAE@sun.com> On Feb 14, 2008, at 04:46, Brandorr wrote: > Frankly it is not a legal issue, other open source projects protect > their trademarks without making a distro. e.g. - Linux, has a > trademark foundation protecting it's trademark. If you think the situation with Linux (or indeed other projects) is that simple you have never spoken with a trademark lawyer. Tiki and I consulted with all the largest projects at OSCON last year and found most are very concerned about trademarks. > Arguing that the law requires to take the actions they did, is not > accurate, and is deliberately confusing the issue. While your grievance is justified, arguing as you are oversimplifies the issue. The fact is, the way the OpenSolaris mark is used ultimately has to be not just determined but controlled by Sun in a way I suspect you will never find acceptable. The discussions I've been party to over the last month or so make it clear to me that a fine-grained control over use of the mark is inevitable. I continue to argue for each of those uses to be steered by input here in the Advocacy community, but we'll never be able to call the shots over what uses are permitted, only over how they are bounded. And that's because of trademark law. As Alan C points out, when it comes to use connected with distros, we are now invited by Sun to propose the boundaries for how distros associate "OpenSolaris" with their own distro name within the boundaries Sun has specified. That has not been the case before, and I suggest we positively seize the opportunity by requesting the uses we want and collaborating to create rules Sun can endorse. S. From James.Walker at Sun.COM Thu Feb 14 04:32:50 2008 From: James.Walker at Sun.COM (Jim Walker) Date: Thu, 14 Feb 2008 05:32:50 -0700 Subject: [trademark-policy-dev] Distribution Certification In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B292DD.7060407@sun.com> <47B29A2B.2040807@Sun.Com> <47B2B395.2000900@sun.com> Message-ID: <47B434F2.4040402@sun.com> Shawn Walker wrote: >> Certification at the distribution level is a great quest for a knight >> who likes to fight battles for the love of fighting, but I think >> there will be too much variance at that level to have one standard >> that we can all agree on, and an arbitrary standard is worse than no >> standard in terms of the cost on people and business and time. > > I'm not convinced of that just yet; I think there's room for compromise here. > > Maybe there's too much in a distribution the size of say, Solaris > Express, to certify. However, in a distribution that fits on a CD > (such as Project Indiana), there may be close to the right amount. There is always room for compromise. It's clear, one size won't fit all. I think it will be possible to certify packages, and once we define what the core OpenSolaris packages are, we will be able to clearly identify OpenSolaris distributions. But, certifying distributions will be more elusive. We could have prospective distribution owners submit their bits to the community for user acceptance testing (beta testing). A combination of package certification and community voting may be a first step to qualify distributions so they can get the OpenSolaris trademark. Most likely we will end up with multiple levels of qualification. Knowing what motivates people to create a distribution will help us better meet their needs. Here are some examples: 1. Distributions created to advance a person or group of people professionally (ie. creating a working distribution is a great thing to put on a resume). 2. Distributions created to provide new functionality for the good of humankind (ie. no direct monetary compensation, but a lot of good will). 3. Distributions created to provide specific marketable applications and services in support of a business (ie. money). 4. Distributions created to provide multiple marketable applications and services in support of a business (ie. big money). All these distributions are valid and should be provided for, but distributions that are economically driven (3 and 4) may have more influence on the process, since people's financial well being depends on them. In addition, money (like time and space) creates more options. Cheers, Jim From Joerg.Schilling at fokus.fraunhofer.de Thu Feb 14 05:58:42 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 14 Feb 2008 14:58:42 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> Message-ID: <47b44912.YNoiMdNL7gsYHuVD%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > Again, the situation with "Linux" is different. There is no Linux > product. There is a Linux trademark that is allowed to be used by > anyone that incorporate the Linux product into their own. There is no difference. The community voted against having a product called "OpenSolaris". There is a preview project called Indiana with a minor name bug that still could be fixed. 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 swalker at opensolaris.org Thu Feb 14 07:04:41 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 14 Feb 2008 09:04:41 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b44912.YNoiMdNL7gsYHuVD%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47b44912.YNoiMdNL7gsYHuVD%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Thu, Feb 14, 2008 at 7:58 AM, Joerg Schilling wrote: > "Shawn Walker" wrote: > > > > Again, the situation with "Linux" is different. There is no Linux > > product. There is a Linux trademark that is allowed to be used by > > anyone that incorporate the Linux product into their own. > > There is no difference. The community voted against having a product called > "OpenSolaris". There was no such community vote and the community doesn't have control over that. It is Sun's trademark; not ours, we have only been given permission to use it. The only thing that I am aware of happening is our elected board expressing their disagreement with the decision. Even if you count that as the "community vote" -- it doesn't matter. The community has no right to vote on things it doesn't (and should not in my belief) control. > There is a preview project called Indiana with a minor name bug that still > could be fixed. To say it is a bug is denying Sun's ownership of the trademark. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Joerg.Schilling at fokus.fraunhofer.de Thu Feb 14 08:48:54 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 14 Feb 2008 17:48:54 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > > > I suggest a new clear branding initiative spearheaded by Sun. > > > > > > OpenSolaris (the distro) > > > OpenSolaris Codebase (the new name for the entire OpenSolaris code base) > > > > If you use the same name for a code base and a distro, you confuse people. > > Only for those easily confused. > > Many, many, projects use the same name for their "distributed > software" as their codebase. OpenSolaris is not a Project but the codebase for many projects. This is why we cannot have a distro called "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 swalker at opensolaris.org Thu Feb 14 09:02:54 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 14 Feb 2008 11:02:54 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Thu, Feb 14, 2008 at 10:48 AM, Joerg Schilling wrote: > "Shawn Walker" wrote: > > > > > I suggest a new clear branding initiative spearheaded by Sun. > > > > > > > > OpenSolaris (the distro) > > > > OpenSolaris Codebase (the new name for the entire OpenSolaris code base) > > > > > > If you use the same name for a code base and a distro, you confuse people. > > > > Only for those easily confused. > > > > Many, many, projects use the same name for their "distributed > > software" as their codebase. > > OpenSolaris is not a Project but the codebase for many projects. > This is why we cannot have a distro called "OpenSolaris". Sorry, I don't see it that way. There is no rule and nothing preventing a distribution called OpenSolaris. I believe it is what many users want, and it is their expectation. I fervently believe in an official distribution. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Thu Feb 14 09:24:51 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 14 Feb 2008 09:24:51 -0800 Subject: [trademark-policy-dev] What is wrong about having a distro called OpenSolaris? In-Reply-To: <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47B47963.4080005@Sun.Com> Joerg Schilling wrote: > OpenSolaris is not a Project but the codebase for many projects. > This is why we cannot have a distro called "OpenSolaris". Why is this a bad thing? IMO, it is not very confusing to be told that the OpenSolaris distro is a product that is made up out of the OpenSolaris codebase developed by the OpenSolaris Community over on a web site called OpenSolaris.org. Rather than simply asserting that this is bad, it would be helpful and probably a lot more actionable to enumerate the problems that doing so would cause and/or the things that you would like to do that would be made more difficult. "If there is a distro called OpenSolaris, then _[this problem]_". "I'm doing _[something]_, and I want to be able to use an OpenSolaris trademark to _[whatever]_." Fill in the blanks. Come up with a list of perceived problems and unmet needs. Once we know about them, we can try and address them. -John From peter.tribble at gmail.com Thu Feb 14 12:57:21 2008 From: peter.tribble at gmail.com (Peter Tribble) Date: Thu, 14 Feb 2008 20:57:21 +0000 Subject: [trademark-policy-dev] What is wrong about having a distro called OpenSolaris? In-Reply-To: <47B47963.4080005@Sun.Com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> <47B47963.4080005@Sun.Com> Message-ID: On Thu, Feb 14, 2008 at 5:24 PM, John Plocher wrote: > Joerg Schilling wrote: > > OpenSolaris is not a Project but the codebase for many projects. > > This is why we cannot have a distro called "OpenSolaris". > > Why is this a bad thing? IMO, it is not very confusing to be told > that the OpenSolaris distro is a product that is made up out of the > OpenSolaris codebase developed by the OpenSolaris Community over on > a web site called OpenSolaris.org. A minor issue is that we've spent some time attempting to explain to people that this isn't the case. And while I support the general idea that there should be an OpenSolaris distribution, I would argue that it is the OpenSolaris community (or its elected representatives, the OGB) that should make such a decision, rather than having it forced on it from outside. And that, furthermore, if there were such an OpenSolaris distribution then it should be up to the OpenSolaris community (or its elected representatives, the OGB) to decide what that distribution should look like (or maybe just choose one of those available to us). Even if Indiana == OpenSolaris were the right decision, having it forced on the community is bound to create bad feeling and opposition. I'm not convinced that Indiana is that OpenSolaris distribution. If that positioning were dropped, then Indiana would be free to stand on its own as a truly remarkable piece of work, and the divisiveness that has ensued would not come about. And who knows, the community might then be able to adopt it as its own. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ From peter.tribble at gmail.com Thu Feb 14 14:14:52 2008 From: peter.tribble at gmail.com (Peter Tribble) Date: Thu, 14 Feb 2008 22:14:52 +0000 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B3CB9D.8010109@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> Message-ID: On Thu, Feb 14, 2008 at 5:03 AM, Alan Coopersmith wrote: > I think you misunderstand the argument - Sun did not use the OpenSolaris > name on Indiana to protect the mark - it's keeping final say over other > users to protect their mark. Sun has always limited use of the OpenSolaris > name this way, and with these moves is allowing more uses than have been > allowed in the past - previously no distro could use it, now they've granted > use of the name to Indiana already, and asked for the community to join in > developing guidelines to determine how other distros can use it as well. It's the double standard here that bothers many - that Indiana is allowed to use the name before the the guidelines establishing the use of the mark have been developed, and that Indiana may not be subject to those guidelines. So, a question: will Indiana be subject to the same rules regarding the use of the mark as other distros, or will it be above them? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ From John.Plocher at Sun.COM Thu Feb 14 14:40:47 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 14 Feb 2008 14:40:47 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> Message-ID: <47B4C36F.5040700@Sun.Com> Peter Tribble wrote: > So, a question: will Indiana be subject to the same rules regarding the > use of the mark as other distros, or will it be above them? I believe (and thus, this is only my opinion about something that has not yet reached closure within the community) that there will be some sort of "by definition" connection between the rules and the indiana effort. I base this entirely on Sun's choice of words: > Sun has decided to fund Project Indiana to build a base distro, > which it will name OpenSolaris Unless y'all setp up and provide a set of needs/requirements, alternate language, metrics and plans, there is a strong possibility that the rules might end up being as simplistic as OpenSolaris is the distro produced by the Indiana Project, which is based on the IPS package system and repositories filled with apps that can be installed using IPS. and "XXX - OpenSolaris Compatible" means that it uses the IPS packaging system and can install and run applications downloaded from the various IPS repositories. -John P.S., I'm still waiting for Joerg to join the Indiana project and get his stuff included as part of the prototyping efforts there. After all, as a simple prototype, it apparently doesn't have the same pesky ARC requirements that ON does :-) This is a not-so-subtle way of saying that y'all can argue all you want about the name, or you can simply join the project and influence it into becoming what you think it should become. Myself, I'm off on a long weekend trip to Baja California Sur for some whale petting and wild tequila monster hunting. Talk to y'all again next week. http://groups.yahoo.com/group/flybaja2008/ -John From swalker at opensolaris.org Thu Feb 14 22:44:40 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 15 Feb 2008 00:44:40 -0600 Subject: [trademark-policy-dev] What is wrong about having a distro called OpenSolaris? In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> <47B47963.4080005@Sun.Com> Message-ID: On Thu, Feb 14, 2008 at 2:57 PM, Peter Tribble wrote: > On Thu, Feb 14, 2008 at 5:24 PM, John Plocher wrote: > > Joerg Schilling wrote: > > > OpenSolaris is not a Project but the codebase for many projects. > > > This is why we cannot have a distro called "OpenSolaris". > > > > Why is this a bad thing? IMO, it is not very confusing to be told > > that the OpenSolaris distro is a product that is made up out of the > > OpenSolaris codebase developed by the OpenSolaris Community over on > > a web site called OpenSolaris.org. > > A minor issue is that we've spent some time attempting to > explain to people that this isn't the case. > > And while I support the general idea that there should be an OpenSolaris > distribution, I would argue that it is the OpenSolaris community (or its > elected representatives, the OGB) that should make such a decision, > rather than having it forced on it from outside. And that, furthermore, if > there were such an OpenSolaris distribution then it should be up to the > OpenSolaris community (or its elected representatives, the OGB) to > decide what that distribution should look like (or maybe just choose one > of those available to us). I don't believe such a thing will ever work. I think Debian is proof that a completely democracy-based process to a distribution leads to a stagnant distribution that few people want to use and a bad reputation among companies. Despite Debian's vaunted stability, they have the least ISV support of any well-known GNU/Linux distribution. I strongly believe that you must have a corporation helping drive a project: Ubuntu, Fedora, etc. are all proof of that to me. > Even if Indiana == OpenSolaris were the right decision, having it > forced on the community is bound to create bad feeling and opposition. > > I'm not convinced that Indiana is that OpenSolaris distribution. If that > positioning were dropped, then Indiana would be free to stand on its > own as a truly remarkable piece of work, and the divisiveness that has > ensued would not come about. And who knows, the community might > then be able to adopt it as its own. I am. Look at the press. Long before Sun ever announced or decided that Indiana was OpenSolaris, the press was already calling it OpenSolaris, and a preview of what was to come in Solaris. Many users and technology enthusiasts are excited, nay elated, at seeing the fruits of years of labour from many hard-working engineers presented in Indiana. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Joerg.Schilling at fokus.fraunhofer.de Fri Feb 15 03:11:36 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 Feb 2008 12:11:36 +0100 Subject: [trademark-policy-dev] What is wrong about having a distro called OpenSolaris? In-Reply-To: <47B47963.4080005@Sun.Com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> <47B47963.4080005@Sun.Com> Message-ID: <47b57368.bi4YI6fE1Pw+m+ns%Joerg.Schilling@fokus.fraunhofer.de> John Plocher wrote: > Joerg Schilling wrote: > > OpenSolaris is not a Project but the codebase for many projects. > > This is why we cannot have a distro called "OpenSolaris". > > > Why is this a bad thing? IMO, it is not very confusing to be told > that the OpenSolaris distro is a product that is made up out of the > OpenSolaris codebase developed by the OpenSolaris Community over on > a web site called OpenSolaris.org. Well, it is so obvious that I thought I don't need to explain. If Sun really starts to call som distro "OpenSolaris" it undermines future Community decsions like e.g. the definition of what a compatible distro means. I really hope Sun is not going to burn the name "OpenSolaris".... note that it would be hard to explain that a distro called "OpenSolaris" is not "OpenSolaris-compatible". 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 Fri Feb 15 05:32:09 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 Feb 2008 14:32:09 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> Message-ID: <47b59459.frg2t/OxHVj1rdqa%Joerg.Schilling@fokus.fraunhofer.de> "Peter Tribble" wrote: > So, a question: will Indiana be subject to the same rules regarding the > use of the mark as other distros, or will it be above them? If it continues to use the name "OpenSolaris", it is obvious that it will/must ignore other rules too. Maybe we ned to find a new name for OpenSolaris because Sun is plannung to burn the old one.... 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 Fri Feb 15 05:47:36 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 Feb 2008 14:47:36 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B4C36F.5040700@Sun.Com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> Message-ID: <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> John Plocher wrote: > Peter Tribble wrote: > > Sun has decided to fund Project Indiana to build a base distro, > > which it will name OpenSolaris The last information I reveived from Sun is that Sun will never fund a freely redistributable OpenSolaris based distro. > OpenSolaris is the distro produced by the Indiana Project, which is > based on the IPS package system and repositories filled with apps > that can be installed using IPS. Wrong: the community decided not to use the name "OpenSolaris" for a specific distro in order to avoid confusion. > and > > "XXX - OpenSolaris Compatible" means that it uses the IPS packaging > system and can install and run applications downloaded from the > various IPS repositories. Wrong too: _We_ the OpenSolaris community did not yet decide about the rules for calling a distro "compatible". We, the OpenSolaris community not decide whether IPS plays a role in a "compatible OpenSolaris distribution". > > P.S., I'm still waiting for Joerg to join the Indiana project and > get his stuff included as part of the prototyping efforts there. I am still waiting for at least one Sun person to join the SchilliX project in order to express interest in collaboration. 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 ceri at submonkey.net Fri Feb 15 05:54:33 2008 From: ceri at submonkey.net (Ceri Davies) Date: Fri, 15 Feb 2008 13:54:33 +0000 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <20080215135433.GL99425@submonkey.net> On Fri, Feb 15, 2008 at 02:47:36PM +0100, Joerg Schilling wrote: > John Plocher wrote: > > P.S., I'm still waiting for Joerg to join the Indiana project and > > get his stuff included as part of the prototyping efforts there. > > I am still waiting for at least one Sun person to join the SchilliX project in > order to express interest in collaboration. That's not how it works. You're a downstream user of the OpenSolaris code, it's therefore your responsibility to collaborate with the OpenSolaris project. Nobody expects the GCC folks to join every project that uses their compiler, yet they're happy to fix problems that are submitted to them properly. Ceri -- That must be wonderful! I don't understand it at all. -- Moliere -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available URL: From swalker at opensolaris.org Fri Feb 15 06:41:18 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 15 Feb 2008 08:41:18 -0600 Subject: [trademark-policy-dev] What is wrong about having a distro called OpenSolaris? In-Reply-To: <47b57368.bi4YI6fE1Pw+m+ns%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> <47B47963.4080005@Sun.Com> <47b57368.bi4YI6fE1Pw+m+ns%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Fri, Feb 15, 2008 at 5:11 AM, Joerg Schilling wrote: > John Plocher wrote: > > > Joerg Schilling wrote: > > > OpenSolaris is not a Project but the codebase for many projects. > > > This is why we cannot have a distro called "OpenSolaris". > > > > > > Why is this a bad thing? IMO, it is not very confusing to be told > > that the OpenSolaris distro is a product that is made up out of the > > OpenSolaris codebase developed by the OpenSolaris Community over on > > a web site called OpenSolaris.org. > > Well, it is so obvious that I thought I don't need to explain. > > If Sun really starts to call som distro "OpenSolaris" it undermines > future Community decsions like e.g. the definition of what a compatible > distro means. Actually, it doesn't. Because even if Indiana is never released, Sun can still define whatever rules they want for compatibility as the trademark holder. They can say "to call your distribution OpenSolaris-compatible you must: a, b, c." > I really hope Sun is not going to burn the name "OpenSolaris".... I'm pretty sure they're not going to; they're going to ensure that it remains valuable. > note that it would be hard to explain that a distro called "OpenSolaris" > is not "OpenSolaris-compatible". I don't think that can happen. I think that if the community wants to develop their own compatibility standard, they're going to have to come up with their own name. I also think that argument doesn't hold anyway for *all* targets, since as others have pointed before, compatibility between an embedded environment and a desktop environment is difficult at best. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Joerg.Schilling at fokus.fraunhofer.de Fri Feb 15 07:16:11 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 Feb 2008 16:16:11 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <20080215135433.GL99425@submonkey.net> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> Message-ID: <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> Ceri Davies wrote: > On Fri, Feb 15, 2008 at 02:47:36PM +0100, Joerg Schilling wrote: > > John Plocher wrote: > > > > P.S., I'm still waiting for Joerg to join the Indiana project and > > > get his stuff included as part of the prototyping efforts there. > > > > I am still waiting for at least one Sun person to join the SchilliX project in > > order to express interest in collaboration. > > That's not how it works. You're a downstream user of the OpenSolaris > code, it's therefore your responsibility to collaborate with the > OpenSolaris project. Nobody expects the GCC folks to join every project It seems that you did not understad the background: Sun is the downstream user of the project/idea of a redistributable OpenSolaris based distribution. I invited Sun in June 2005 and I will see whether there is interest for collaboration. On the other side, I am collaborating with the OpenSolaris code base although at Sun's side there seems to be obly limited 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 Fri Feb 15 07:26:29 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 Feb 2008 16:26:29 +0100 Subject: [trademark-policy-dev] What is wrong about having a distro called OpenSolaris? In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <47b470f6.NfSs27wulDjn5jVd%Joerg.Schilling@fokus.fraunhofer.de> <47B47963.4080005@Sun.Com> <47b57368.bi4YI6fE1Pw+m+ns%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47b5af25.wgRz7V1sz9E5DDLp%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > > Well, it is so obvious that I thought I don't need to explain. > > > > If Sun really starts to call som distro "OpenSolaris" it undermines > > future Community decsions like e.g. the definition of what a compatible > > distro means. > > Actually, it doesn't. Because even if Indiana is never released, Sun > can still define whatever rules they want for compatibility as the > trademark holder. It is not important what Sun as trademark holder _could_ do.... Important is that Sun in 2004 claimed to target on complete opennesss. If this is no longer true, we (the community) need to fork just in order to be able to be able to decide on the project. If Sun is still interested in a truely open "OpenSolaris", Sun needs to honor the decisions from the community. This of course includes that Sun cannot make decisions that anticipate yet to do decisions of the community. If "OpenSolaris" is a truely open project, then the community needs to be able to decide on compatibility rules. 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 suhail_stk at hotmail.com Fri Feb 15 06:02:04 2008 From: suhail_stk at hotmail.com (suhail ansari) Date: Fri, 15 Feb 2008 06:02:04 PST Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> Message-ID: <29050663.1203084203374.JavaMail.Twebapp@oss-app1> I totally agree with you. Because successful open-source project sometime face fragmentation problem, standardization is very important for any open-source project. The ideas that you have suggested will surely help in this aspect. From alan.coopersmith at sun.com Fri Feb 15 08:01:45 2008 From: alan.coopersmith at sun.com (Alan Coopersmith) Date: Fri, 15 Feb 2008 08:01:45 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47B5B769.9080409@sun.com> Joerg Schilling wrote: > John Plocher wrote: > >> Peter Tribble wrote: > >>> Sun has decided to fund Project Indiana to build a base distro, >>> which it will name OpenSolaris > > The last information I reveived from Sun is that Sun will never fund a > freely redistributable OpenSolaris based distro. Then you have failed to receive the information Sun has been providing for the last year, including at the OpenSolaris Developer Summit, where we all saw you in the room when Sun employees talked about doing exactly that. We cannot help you if you refuse to listen. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From swalker at opensolaris.org Fri Feb 15 08:31:31 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 15 Feb 2008 10:31:31 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Fri, Feb 15, 2008 at 7:47 AM, Joerg Schilling wrote: > John Plocher wrote: > > > > Peter Tribble wrote: > > > > Sun has decided to fund Project Indiana to build a base distro, > > > which it will name OpenSolaris > > The last information I reveived from Sun is that Sun will never fund a > freely redistributable OpenSolaris based distro. Then you received the wrong information. Indiana is a completely, freely, redistributable OpenSolaris-based distribution. > > > > OpenSolaris is the distro produced by the Indiana Project, which is > > based on the IPS package system and repositories filled with apps > > that can be installed using IPS. > > Wrong: the community decided not to use the name "OpenSolaris" for a specific > distro in order to avoid confusion. Really? When did we vote on that. I don't remember it... -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Fri Feb 15 08:32:19 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 15 Feb 2008 10:32:19 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Fri, Feb 15, 2008 at 9:16 AM, Joerg Schilling wrote: > Ceri Davies wrote: > > > On Fri, Feb 15, 2008 at 02:47:36PM +0100, Joerg Schilling wrote: > > > John Plocher wrote: > > > > > > P.S., I'm still waiting for Joerg to join the Indiana project and > > > > get his stuff included as part of the prototyping efforts there. > > > > > > I am still waiting for at least one Sun person to join the SchilliX project in > > > order to express interest in collaboration. > > > > That's not how it works. You're a downstream user of the OpenSolaris > > code, it's therefore your responsibility to collaborate with the > > OpenSolaris project. Nobody expects the GCC folks to join every project > > > It seems that you did not understad the background: > > Sun is the downstream user of the project/idea of a redistributable OpenSolaris > based distribution. I invited Sun in June 2005 and I will see whether there is > interest for collaboration. > > On the other side, I am collaborating with the OpenSolaris code base although > at Sun's side there seems to be obly limited interest. Downstream applies to code; not ideas. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 15 08:55:44 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 15 Feb 2008 08:55:44 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b59459.frg2t/OxHVj1rdqa%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47b59459.frg2t/OxHVj1rdqa%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On 15-Feb-08, at 5:32 AM, Joerg Schilling wrote: > "Peter Tribble" wrote: > >> So, a question: will Indiana be subject to the same rules regarding >> the >> use of the mark as other distros, or will it be above them? > > If it continues to use the name "OpenSolaris", it is obvious that it > will/must > ignore other rules too. > > Maybe we ned to find a new name for OpenSolaris because Sun is > plannung to burn > the old one.... +1 Independence from Sun is key, as is apparent by recent SMI decisions that ignore community wishes -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part URL: From Joerg.Schilling at fokus.fraunhofer.de Fri Feb 15 09:14:02 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 Feb 2008 18:14:02 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B5B769.9080409@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <47B5B769.9080409@sun.com> Message-ID: <47b5c85a.rsCfN954eqRKZEMp%Joerg.Schilling@fokus.fraunhofer.de> Alan Coopersmith wrote: > We cannot help you if you refuse to listen. Sometimes, it would help a lot of Sun did listen to the community.... 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 Fri Feb 15 09:22:49 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 Feb 2008 18:22:49 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > On Fri, Feb 15, 2008 at 9:16 AM, Joerg Schilling > > It seems that you did not understad the background: > > > > Sun is the downstream user of the project/idea of a redistributable OpenSolaris > > based distribution. I invited Sun in June 2005 and I will see whether there is > > interest for collaboration. > > > > On the other side, I am collaborating with the OpenSolaris code base although > > at Sun's side there seems to be obly limited interest. > > Downstream applies to code; not ideas. Do you have an idea what fair use of ideas is? I believe not 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 swalker at opensolaris.org Fri Feb 15 10:00:59 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 15 Feb 2008 12:00:59 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Fri, Feb 15, 2008 at 11:22 AM, Joerg Schilling wrote: > "Shawn Walker" wrote: > > > On Fri, Feb 15, 2008 at 9:16 AM, Joerg Schilling > > > > > It seems that you did not understad the background: > > > > > > Sun is the downstream user of the project/idea of a redistributable OpenSolaris > > > based distribution. I invited Sun in June 2005 and I will see whether there is > > > interest for collaboration. > > > > > > On the other side, I am collaborating with the OpenSolaris code base although > > > at Sun's side there seems to be obly limited interest. > > > > Downstream applies to code; not ideas. > > Do you have an idea what fair use of ideas is? I believe not The idea of a redistributable opensolaris isn't unique. It's something that was asked for from day one. You deserve to be commended for being the first to do it, but that doesn't mean that you have exclusive rights to it. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Fri Feb 15 13:01:11 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 15 Feb 2008 15:01:11 -0600 Subject: [trademark-policy-dev] Distribution Certification In-Reply-To: <47B434F2.4040402@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B292DD.7060407@sun.com> <47B29A2B.2040807@Sun.Com> <47B2B395.2000900@sun.com> <47B434F2.4040402@sun.com> Message-ID: On Thu, Feb 14, 2008 at 6:32 AM, Jim Walker wrote: > Shawn Walker wrote: > >> Certification at the distribution level is a great quest for a knight > >> who likes to fight battles for the love of fighting, but I think > >> there will be too much variance at that level to have one standard > >> that we can all agree on, and an arbitrary standard is worse than no > >> standard in terms of the cost on people and business and time. > > > > I'm not convinced of that just yet; I think there's room for compromise here. > > > > Maybe there's too much in a distribution the size of say, Solaris > > Express, to certify. However, in a distribution that fits on a CD > > (such as Project Indiana), there may be close to the right amount. > > There is always room for compromise. > > It's clear, one size won't fit all. I think it will be possible to > certify packages, and once we define what the core OpenSolaris > packages are, we will be able to clearly identify OpenSolaris > distributions. But, certifying distributions will be more elusive. > We could have prospective distribution owners submit their bits > to the community for user acceptance testing (beta testing). That's basically what I was trying to say. Make a distribution out of a set of certified (supported) packages and *poof* you have a certified distribution :) If I remember correctly, this is what Ubuntu does. Everything they distribute is in the "supported" package repository (there are additional packages in the supported repository that aren't part of the distribution CD). > A combination of package certification and community voting may > be a first step to qualify distributions so they can get the > OpenSolaris trademark. Most likely we will end up with multiple > levels of qualification. Yep. I hope that the trademark-policy project can work with the new Distribution CG. > Knowing what motivates people to create a distribution will help us > better meet their needs. Here are some examples: > > 1. Distributions created to advance a person or group of people > professionally (ie. creating a working distribution is a great > thing to put on a resume). > > 2. Distributions created to provide new functionality for the good > of humankind (ie. no direct monetary compensation, but a lot of > good will). > > 3. Distributions created to provide specific marketable applications > and services in support of a business (ie. money). > > 4. Distributions created to provide multiple marketable applications > and services in support of a business (ie. big money). > > All these distributions are valid and should be provided for, but > distributions that are economically driven (3 and 4) may have more > influence on the process, since people's financial well being depends > on them. In addition, money (like time and space) creates more options. Completely agree. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From peter.tribble at gmail.com Sat Feb 16 15:03:05 2008 From: peter.tribble at gmail.com (Peter Tribble) Date: Sat, 16 Feb 2008 23:03:05 +0000 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47B4C36F.5040700@Sun.Com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> Message-ID: On Feb 14, 2008 10:40 PM, John Plocher wrote: > Peter Tribble wrote: > > So, a question: will Indiana be subject to the same rules regarding the > > use of the mark as other distros, or will it be above them? > > I believe (and thus, this is only my opinion about something > that has not yet reached closure within the community) that > there will be some sort of "by definition" connection between > the rules and the indiana effort. I base this entirely on > Sun's choice of words: > > > Sun has decided to fund Project Indiana to build a base distro, > > which it will name OpenSolaris So you believe that Indiana will be above the rules? > Unless y'all setp up and provide a set of needs/requirements, alternate > language, metrics and plans, there is a strong possibility that the rules > might end up being as simplistic as > > OpenSolaris is the distro produced by the Indiana Project, which is > based on the IPS package system and repositories filled with apps > that can be installed using IPS. Hang on, isn't OpenSolaris a community and a place for collaboration and conversation around OpenSolaris technology? That's where arbitrarily calling a distribution "OpenSolaris" causes trouble - it creates confusion and misunderstanding as to what the thing is. Is it the community, the technology, a specific distribution, or a family of distributions? This is especially germane when the distribution arbitrarily anointed with the name isn't necessarily accepted as having the right to use the name by the other things that use the name. (I note that nothing in your definition says that the OpenSolaris distribution has any relationship to the OpenSolaris community, or that it uses OpenSolaris technology, or that it has any relationship to the important object known as Solaris. Based on that definition, OpenSolaris could be a rebranded Linux distro.) > and > > "XXX - OpenSolaris Compatible" means that it uses the IPS packaging > system and can install and run applications downloaded from the > various IPS repositories. If that were a definition of Indiana compatibility, it would make a lot of sense. That's the sort of definition that users and ISVs want. However, as a definition of OpenSolaris compatibility, it makes no sense, because a consumer of software with the mark can easily understand the compatibility bit, but then has no idea whether the compatibility refers to the technology, a specific distribution, or any of a family of distributions. That assumes that we carry on as we are. And attempting to have more than one thing called OpenSolaris, with the things sometimes at each others throats (I'm not sure I could describe the present conflagration as anything but), can only devalue OpenSolaris and its use as a trademark, and Solaris by association. An alternative vision, in which the naming of Indiana as OpenSolaris would work just fine, would be to throw away the current OpenSolaris community and project and replace it with one who's remit was to develop the one true distribution. Do we want to do that? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ From John.Plocher at Sun.COM Tue Feb 19 10:40:58 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Tue, 19 Feb 2008 10:40:58 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> Message-ID: <47BB22BA.3030106@Sun.Com> Peter Tribble wrote: +-- John Plocher quoted William Franklin as saying | Sun has decided to fund Project Indiana to build a base distro, | which it will name OpenSolaris +--- > So you believe that Indiana will be above the rules? My own _belief_ is that Sun intends to write the rules in such a way as to make the output of the Indiana project in some way *be* the definition of "OpenSolaris" - an existence proof, if you will. The key for Sun seems to be the desire to create an ecosystem that is larger than simply "one distro among many incompatible but similar distros". What does this imply? I don't know for sure, but I can make some guesses and assumptions - although beware - as my perspective and opinion, they might actually be completely wrong: It seems that Sun wishes us (the T&B project) to create and propose trademark policies that Sun[1] will influence, adopt and publish that will dictate how others can use the various taglines and derivatives of Sun's OpenSolaris trademark. As such, the policies will dictate how others can use Sun's trademarks - of course, Sun itself will probably reserve the right to allow other uses as it sees fit, and whatever Sun does with its marks itself is, by legal definition, always OK by the rules. It seems that Sun[1] has decided that Shillix, Nexenta[2], Belenix, SXCE and SXDE are NOT going to be chosen to be the distro that will be called OpenSolaris. Sorry Joerg, Erast and Moinak, but this point doesn't seem to be negotiable. It seems that Sun[1] wishes to carve out uses of the OpenSolaris trademark that *can* be applied to those "other" distros, if they wish to use it and if they qualify. Sun has set out some of the boundaries for such "rules", but clearly intends for the community to have significant input into their formation. Nowhere in any of this do I see any indication that Sun is restricting who can work on Indiana or the T&B efforts, what can go into them, etc. If you don't like what they are today, or where they are going, then get involved and change them from within. Not to belabor the point, there seem to be three options here: 1) cry and whine and throw a tantrum about how poor the interactions were/are/will be between Sun and the community while ignoring all the places where the interactions are working just fine, or 2) go join the Indiana project *NOW*, become a contributer, and inject your vision, code and passion to transform it into the OpenSolaris Distro you need it to be, and stay active in this T&B project and inject your vision, text and passion to create the "rules" you want to have. or 3) global thermonuclear war - blow up the community, fork off into two or more camps, or otherwise create the very distro-hell that we don't want because we can't or won't work together as a community[3]. To me, the right choice is obvious. Hopefully it is to you as well, and I pray that we both choose the same "obvious" option! ____ [1] When I say Sun, I specifically mean William Franklin (as messenger) and the executives who are making the decisions that he is being asked to convey to the OGB. This is not some amorphous and unaccountable entity, but a specific set of individuals who have chosen to use Bill as their spokesperson. [2] Personally, when I see the list of things that are planned and/or desired for Indiana, and I look over at what Nexenta has done, I sigh and shake my head... [3] Am I frustrated, dismayed and angry at how this all has played out? Hell yes! In this area, I find myself in complete agreement with Al and John and Martin (see http://mail.opensolaris.org/pipermail/ogb-discuss/2008-February/004721.html) > ... This is especially germane when the > distribution arbitrarily anointed with the name isn't necessarily > accepted as having the right to use the name by the other things > that use the name. It seems painfully obvious at this point in time that the decision isn't in the hands of "the other things that use the name". > An alternative vision, in which the naming of Indiana as OpenSolaris > would work just fine, would be to throw away the current OpenSolaris > community and project and replace it with one who's remit was to > develop the one true distribution. Do we want to do that? This certainly seems to be something that should be discussed as part of the upcoming OGB elections - it is too important of a discussion to have it get lost as part of a massive thread that nobody is following anymore... (sigh, as if anyone will read this far :-) -John From Joerg.Schilling at fokus.fraunhofer.de Thu Feb 21 07:40:10 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 Feb 2008 16:40:10 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47bd9b5a.ux3iFhGWtwQ8p8J3%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > On Fri, Feb 15, 2008 at 7:47 AM, Joerg Schilling > wrote: > > John Plocher wrote: > > > > > > > Peter Tribble wrote: > > > > > > Sun has decided to fund Project Indiana to build a base distro, > > > > which it will name OpenSolaris > > > > The last information I reveived from Sun is that Sun will never fund a > > freely redistributable OpenSolaris based distro. > > Then you received the wrong information. Indiana is a completely, > freely, redistributable OpenSolaris-based distribution. If I received wrong information than you seem to imply that Sun like to help funding SchilliX. Please tell us more... 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 swalker at opensolaris.org Thu Feb 21 08:03:55 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 10:03:55 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47bd9b5a.ux3iFhGWtwQ8p8J3%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <47bd9b5a.ux3iFhGWtwQ8p8J3%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Thu, Feb 21, 2008 at 9:40 AM, Joerg Schilling wrote: > "Shawn Walker" wrote: > > > > On Fri, Feb 15, 2008 at 7:47 AM, Joerg Schilling > > wrote: > > > John Plocher wrote: > > > > > > > > > > Peter Tribble wrote: > > > > > > > > Sun has decided to fund Project Indiana to build a base distro, > > > > > which it will name OpenSolaris > > > > > > The last information I reveived from Sun is that Sun will never fund a > > > freely redistributable OpenSolaris based distro. > > > > Then you received the wrong information. Indiana is a completely, > > freely, redistributable OpenSolaris-based distribution. > > If I received wrong information than you seem to imply that Sun like to > help funding SchilliX. Please tell us more... Why would they do that and how do you get such an implication? -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Joerg.Schilling at fokus.fraunhofer.de Thu Feb 21 08:16:43 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 Feb 2008 17:16:43 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > On Fri, Feb 15, 2008 at 11:22 AM, Joerg Schilling > wrote: > > "Shawn Walker" wrote: > > > Downstream applies to code; not ideas. > > > > Do you have an idea what fair use of ideas is? I believe not > > The idea of a redistributable opensolaris isn't unique. It's something > that was asked for from day one. It is something that was available from day one +3 If Sun is interested in such a beast, why was there no collaboration from Sun's side? 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 swalker at opensolaris.org Thu Feb 21 08:24:45 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 10:24:45 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Thu, Feb 21, 2008 at 10:16 AM, Joerg Schilling wrote: > "Shawn Walker" wrote: > > > On Fri, Feb 15, 2008 at 11:22 AM, Joerg Schilling > > wrote: > > > "Shawn Walker" wrote: > > > > > > Downstream applies to code; not ideas. > > > > > > Do you have an idea what fair use of ideas is? I believe not > > > > The idea of a redistributable opensolaris isn't unique. It's something > > that was asked for from day one. > > It is something that was available from day one +3 No, I'm fairly certain the opensolaris.org website was up before there was a freely redistributable version of Solaris/OpenSolaris. > If Sun is interested in such a beast, why was there no collaboration from Sun's > side? Why does Sun have to collaborate? Why does anyone have to? If people want to collaborate; they will. As for Sun's reasons; you need to ask Sun. No matter what there reasons were, does it matter? There is nothing wrong with choosing independent action. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Joerg.Schilling at fokus.fraunhofer.de Thu Feb 21 08:46:09 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 Feb 2008 17:46:09 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47bdaad1.z5cPPiy99ZKsaC2p%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > On Thu, Feb 21, 2008 at 10:16 AM, Joerg Schilling > wrote: > > "Shawn Walker" wrote: > > > > > On Fri, Feb 15, 2008 at 11:22 AM, Joerg Schilling > > > wrote: > > > > "Shawn Walker" wrote: > > > > > > > > > Downstream applies to code; not ideas. > > > > > > > > Do you have an idea what fair use of ideas is? I believe not > > > > > > The idea of a redistributable opensolaris isn't unique. It's something > > > that was asked for from day one. > > > > It is something that was available from day one +3 > > No, I'm fairly certain the opensolaris.org website was up before there > was a freely redistributable version of Solaris/OpenSolaris. Of course, it was up since ~ September 14th 2004, but you did not have access to it and at the time SchilliX came out, there was still absolutely no support for community activities like SchilliX. The SchilliX has been registered in March 2005 at berlios.de for this reason. > > If Sun is interested in such a beast, why was there no collaboration from Sun's > > side? > > Why does Sun have to collaborate? Why does anyone have to? > > If people want to collaborate; they will. It is bad practice not to collaborate but to take _many_ if not all ideas. This kind of habbit harms credibility. 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 John.Plocher at Sun.COM Thu Feb 21 08:54:57 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 08:54:57 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47BDACE1.2080506@Sun.Com> [Followups to /dev/null] Joerg Schilling wrote: > If Sun is interested in such a beast, why was there no collaboration from Sun's > side? Because Sun isn't a single entity, because Sun screwed up, because Sun wasn't ready, because Sun was still trying to figure out what it needed to do first, because Sun was busy with all the other 27 million things it had to do just to launch the community, because the people at Sun are human, and thus prone to make mistakes and change their minds, because you make collaborating with yourself unreasonably difficult, because Sun makes collaborating with itself unreasonably difficult, because the people in charge of Solaris today are not the same ones who were in charge when the community was launched, because we learned something in the last few years, etc. Pick whatever excuse/rationalization/explanation that you feel fits. Then get over it and move on. -John From Joerg.Schilling at fokus.fraunhofer.de Thu Feb 21 09:01:23 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 Feb 2008 18:01:23 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <47bd9b5a.ux3iFhGWtwQ8p8J3%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47bdae63.b7Jqp9ZYiTcNwp09%Joerg.Schilling@fokus.fraunhofer.de> "Shawn Walker" wrote: > > > > > Peter Tribble wrote: > > > > > > > > > > Sun has decided to fund Project Indiana to build a base distro, > > > > > > which it will name OpenSolaris > > > > > > > > The last information I reveived from Sun is that Sun will never fund a > > > > freely redistributable OpenSolaris based distro. > > > > > > Then you received the wrong information. Indiana is a completely, > > > freely, redistributable OpenSolaris-based distribution. > > > > If I received wrong information than you seem to imply that Sun like to > > help funding SchilliX. Please tell us more... > > Why would they do that and how do you get such an implication? I am just following your claim that I did receive wrong information. 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 Thu Feb 21 09:05:45 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 Feb 2008 18:05:45 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47BDACE1.2080506@Sun.Com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> <47BDACE1.2080506@Sun.Com> Message-ID: <47bdaf69.rZPYnmHKc7ofkwbw%Joerg.Schilling@fokus.fraunhofer.de> John Plocher wrote: > Joerg Schilling wrote: > > If Sun is interested in such a beast, why was there no collaboration from Sun's > > side? > > > Because Sun isn't a single entity, because Sun screwed up, because Sun wasn't > ready, because Sun was still trying to figure out what it needed to do first, If Sun wasn't ready, why was there no change in mind after Sun became ready? 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 John.Plocher at Sun.COM Thu Feb 21 09:31:35 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 09:31:35 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft Message-ID: <47BDB577.6040203@Sun.Com> http://www.genunix.org/wiki/index.php/Trademark_usage_and_Branding_guideline I would like to wrap this project up in time for the upcoming election. Towards that end, I would like to solicit a final round of suggestions, disagreements, rationalizations and use cases, so that we can come to closure on this draft, discuss it with Sun's trademark lawyers, and present it as a formal proposal to the ARC community and OGB. The deadline for Submissions for Constitutional Amendments and other Ballot Questions is Mon Mar 3rd (24:00 hrs). I'd for the ARC and OGB to be able to discuss this before that deadline, which means they must have it in their hands by Wed, Feb 27. This leaves us less than a week to be done. Since there has been almost no argument over the current draft wording, I feel that this is well within our ability. I will integrate comments and suggestions into the document as I receive them, while pushing the discussion to reach closure/agreement as needed. IF YOU HAVE CONCERNS OR OBJECTIONS TO THE CURRENT WORDING, please speak up NOW with alternatives. If you feel that you may need to vote "-1", *PLEASE* articulate your rationale for doing so NOW so that we can try to resolve your problem and work together to deliver a policy that meets all our needs. Lets aim for Sunday, Feb 24 to have a final draft; I will work with Sun's trademark lawyers on Monday, republish a final draft with any changes, and call a project-level vote (majority, with required rationale for -1 votes) at the end of the day Monday, with the assumption that an approved document will be sent to the OGB Tuesday night, for discussion at the weekly ARC and OGB meetings on Wednesday. Yes this is tight, but I think we can do it. -John -------------------------------------------- Current wording of T&B guidelines Draft 0.10 -------------------------------------------- > OpenSolaris Trademark Usage Guidelines - version WORKING DRAFT 0.10 > > NOT APPROVED FOR USE > Draft 0.10 - Updated to reflect Sun's policies regarding the OpenSolaris brand. > purple text is derived from Bill Franklin's email to the OGB > green text is from the existing OS.o trademark FAQ > black text is from the community, and can be edited at will. I've tried to phrase these things as "might" to distinguish them from the things that are harder for us to change. > > > Context > > * Sun Trademark Policy > * The OpenSolaris Community Charter says: > > * However, nothing in this charter shall be construed so as to > confer to the OGB: (a) any title or right under copyright, > patent, trademark, or other intellectual property law; (b) > control of or interest in any asset, tangible or intangible, of > Sun Microsystems, Inc. or any of its subsidiaries; (c) control of > or interest in Sun Microsystems, Inc. or any of its subsidiaries. > > * What are the permitted uses of the OpenSolaris trademark? (From the > Trademark FAQ) > > * First, the OpenSolaris Fan Buttons and Banners are available now. > * Second, Fair Use of the OpenSolaris trademark is, of course, > allowed. > * Third, we are creating a program that will let you communicate > that your derived work includes OpenSolaris technology. > * We are also planning to create buttons to communicate that a > website or system is running the OpenSolaris technology. > > Description > > This project is chartered to work with Sun (the trademark owner) to help > them develop a Trademark usage and Branding policy that covers the use of > the OpenSolaris Trademark. > > ISSUES > > ISSUE At this point in time, nevada is what we have, and indiana is an > undefined future possibility. How can people today use the OpenSolaris > marks without implying that they are using/doing indiana? > > Trademark Ownership > > Solaris and OpenSolaris are trademarks of Sun Microsystems, Inc., and the > Solaris brand is reserved for Sun's enterprise product. > > Approved OpenSolaris Brand Uses > > Fair Use > > Fair use of the OpenSolaris trademark is, of course, allowed. For example, > you can talk about the OpenSolaris project and refer to it in your > writing, in your blogs, etc. And the growing collection of OpenSolaris > User Groups can use the OpenSolaris trademark in their User Group names, > as long as they comply with Sun trademark policies for User Groups. > > Descriptions of Your Products > > You may use a separate, explanatory phrase or tagline after your product > name that describes how your product relates to one of Sun's. (See Sun > Trademark Policy) The potential brandings below are taglines with specific > meanings. > > OpenSolaris Fan Buttons and Banners > > The fan buttons and banners make it easy for you to endorse and show your > support for the OpenSolaris community and technology on your websites. > There is no registration program; the fan buttons are easy to get. There > are Usage Guidelines and a Trademark License that govern the usage of > these buttons and banners. > > Potential Brandings > > OpenSolaris > > * Sun has decided to fund Project Indiana to build a base distro, which > it will call OpenSolaris. > > ___ - Built on/for OpenSolaris and ___ - Compatible with OpenSolaris > > Compatibility is at the core of the high level OpenSolaris brand. The > intent of this branding level is to indicate that a given distro (or a > packaged application?) is closely aligned with the ecosystem of compatible > software packages and services being created around the OpenSolaris base > distro. > > TODO: Sun seems to want to use the branding to support the concept of > "allowed to use the OpenSolaris trademark so that they could form an > ecosystem of compatible software package and services"; the earlier > discussions around a "core" and/or "test suites" are valuable, but > probably too low level. (See the parking lot issues on the discussion > page). > > * Might be used by derivatives of OpenSolaris that are "closely > aligned". > * Might need some sort of license (clik thru?) (what about attribution?) > * Derivatives of the code base that are not "closely aligned with the > ecosystem" are not intended to qualify to use this branding. > * Do we need/want a tagline for components and repositories to label > themselves as "built for OpenSolaris"? > * Might be a mechanism to tag packages that were intended or certified > or ??? for use with OpenSolaris. > > ___ - Built with OpenSolaris Technology, -or- Built on the OpenSolaris > Codebase > > This tagline might be used to deal with incompatible distros and other > components that use the OpenSolaris codebase but do not produce an > environment that is closely aligned... as above. > > __person__ - OpenSolaris Certified > > * Might be used for training and skills certification of individuals > > Logos > > * (may be part of the enthusiasts stuff) there is a need for icons and > logos that can be used for for grub boot, VM desktop icons, > favicon/web as well as "powered by" and "runs on" usage. > > Other branding guidelines may be developed in the future > > * Legally, for the purposes of many nations' trademark laws, all other > uses of the OpenSolaris trademark need to be reserved by Sun, but > terms for use are negotiable. Specifically, nothing here precludes the > development of additional guidelines and subsequent adoption by Sun as > such needs present themseves. > * It might be a good thing to stake out some of these other items - > appliances and embedded, technology reuse (i.e., Apple's and BSD's > reuse of DTrace and ZFS...), etc. > > Branding and endorsement may or may not be apropriate in all situations > > * Any use of Sun's trademarks on a product or service requires a > license. > * The community should really encourage attribution back to the > OpenSolaris community... > * Fair Use of the OpenSolaris brand does not include or imply any > commitment by Sun to provide support, service, indemnification, or > updates. From swalker at opensolaris.org Thu Feb 21 09:50:23 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 11:50:23 -0600 Subject: [trademark-policy-dev] "Collaboration" Re: Sun's Responses to the OpenSolaris Trademark Questions Message-ID: On Thu, Feb 21, 2008 at 11:01 AM, Joerg Schilling wrote: > "Shawn Walker" wrote: > > > > > > Peter Tribble wrote: > > > > > > > > > > > > Sun has decided to fund Project Indiana to build a base distro, > > > > > > > which it will name OpenSolaris > > > > > > > > > > The last information I reveived from Sun is that Sun will never fund a > > > > > freely redistributable OpenSolaris based distro. > > > > > > > > Then you received the wrong information. Indiana is a completely, > > > > freely, redistributable OpenSolaris-based distribution. > > > > > > If I received wrong information than you seem to imply that Sun like to > > > help funding SchilliX. Please tell us more... > > > > Why would they do that and how do you get such an implication? > > I am just following your claim that I did receive wrong information. No, I'm asking why you believe that Sun wanting to do something means that they should, could, or would fund SchilliX? Why does Sun have to do anything? Haven't they already "helped fund SchilliX" and every other OpenSolaris distribution by spending millions on opening the source for Solaris? -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Alan.Coopersmith at Sun.COM Thu Feb 21 09:51:31 2008 From: Alan.Coopersmith at Sun.COM (Alan Coopersmith) Date: Thu, 21 Feb 2008 09:51:31 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47bdaf69.rZPYnmHKc7ofkwbw%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> <47BDACE1.2080506@Sun.Com> <47bdaf69.rZPYnmHKc7ofkwbw%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47BDBA23.3090508@sun.com> Joerg Schilling wrote: > John Plocher wrote: > >> Joerg Schilling wrote: >>> If Sun is interested in such a beast, why was there no collaboration from Sun's >>> side? >> >> Because Sun isn't a single entity, because Sun screwed up, because Sun wasn't >> ready, because Sun was still trying to figure out what it needed to do first, > > If Sun wasn't ready, why was there no change in mind after Sun became ready? It's inexplicable, so please stop trying to get explanations - none of us here can provide anything beyond "Because that's not what the people in Sun management who were creating the distro plan decided to do." Figuring out what went wrong there is not the point of this forum - figuring out what the policy should be for allowing distros to use the OpenSolaris trademark is the point of this forum. -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From swalker at opensolaris.org Thu Feb 21 09:51:57 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 11:51:57 -0600 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47bdaad1.z5cPPiy99ZKsaC2p%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <20080215135433.GL99425@submonkey.net> <47b5acbb.lpcb9MA+ehgJFQO6%Joerg.Schilling@fokus.fraunhofer.de> <47b5ca69.BQOveAbFU55gA48O%Joerg.Schilling@fokus.fraunhofer.de> <47bda3eb.ErfDoWum15qhxd3p%Joerg.Schilling@fokus.fraunhofer.de> <47bdaad1.z5cPPiy99ZKsaC2p%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Thu, Feb 21, 2008 at 10:46 AM, Joerg Schilling wrote: > "Shawn Walker" wrote: > > > On Thu, Feb 21, 2008 at 10:16 AM, Joerg Schilling > > wrote: > > > "Shawn Walker" wrote: > > > > > > > On Fri, Feb 15, 2008 at 11:22 AM, Joerg Schilling > > > > wrote: > > > > > "Shawn Walker" wrote: > > > > > > > > > > > > Downstream applies to code; not ideas. > > > > > > > > > > Do you have an idea what fair use of ideas is? I believe not > > > > > > > > The idea of a redistributable opensolaris isn't unique. It's something > > > > that was asked for from day one. > > > > > > It is something that was available from day one +3 > > > > No, I'm fairly certain the opensolaris.org website was up before there > > was a freely redistributable version of Solaris/OpenSolaris. > > Of course, it was up since ~ September 14th 2004, but you did not have access > to it and at the time SchilliX came out, there was still absolutely no support > for community activities like SchilliX. > > The SchilliX has been registered in March 2005 at berlios.de for this reason. Nice backstory, but I don't see how it relates. The point stands that people wanted a freely redistributable version of Solaris/OpenSolaris *years ago*. The idea is not new. > > > If Sun is interested in such a beast, why was there no collaboration from Sun's > > > side? > > > > Why does Sun have to collaborate? Why does anyone have to? > > > > If people want to collaborate; they will. > > It is bad practice not to collaborate but to take _many_ if not all ideas. What ideas did they take? Can you state with specific clarity which ideas were "taken" from you? I haven't seen any so far. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Alan.Coopersmith at Sun.COM Thu Feb 21 09:56:01 2008 From: Alan.Coopersmith at Sun.COM (Alan Coopersmith) Date: Thu, 21 Feb 2008 09:56:01 -0800 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47bdae63.b7Jqp9ZYiTcNwp09%Joerg.Schilling@fokus.fraunhofer.de> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <47bd9b5a.ux3iFhGWtwQ8p8J3%Joerg.Schilling@fokus.fraunhofer.de> <47bdae63.b7Jqp9ZYiTcNwp09%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <47BDBB31.4040204@sun.com> Joerg Schilling wrote: > "Shawn Walker" wrote: > >>> > > > Peter Tribble wrote: >>> > > >>> > > > > Sun has decided to fund Project Indiana to build a base distro, >>> > > > > which it will name OpenSolaris >>> > > >>> > > The last information I reveived from Sun is that Sun will never fund a >>> > > freely redistributable OpenSolaris based distro. >>> > >>> > Then you received the wrong information. Indiana is a completely, >>> > freely, redistributable OpenSolaris-based distribution. >>> >>> If I received wrong information than you seem to imply that Sun like to >>> help funding SchilliX. Please tell us more... >> Why would they do that and how do you get such an implication? > > I am just following your claim that I did receive wrong information. The wrong information was "Sun will never fund a freely redistributable OpenSolaris based distro." That is clearly wrong, because they are funding Indiana. That has nothing to do with funding SchilliX, because SchilliX is not the only possible freely redistributable OpenSolaris based distro - it is possible to fund Indiana to make a freely redistributable OpenSolaris based distro without funding SchilliX to do the same, as clearly proven by the last 6-9 months of history. The OpenSolaris world has distros other than SchilliX, and not everything is about you. That said, funding SchilliX is off-topic for this forum, since it has nothing to do with developing a policy for use of the OpenSolaris trademark by distros unless you want the policy to say that only distros funded by Sun will be allowed to use the trademark, in which case I can promise you that suggestion will be voted against by at least this OGB member (as well as pretty much everyone else I assume). -- -Alan Coopersmith- alan.coopersmith at sun.com Sun Microsystems, Inc. - X Window System Engineering From Joerg.Schilling at fokus.fraunhofer.de Thu Feb 21 10:11:20 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 Feb 2008 19:11:20 +0100 Subject: [trademark-policy-dev] Sun's Responses to the OpenSolaris Trademark Questions In-Reply-To: <47BDBB31.4040204@sun.com> References: <20855286.1202802295294.JavaMail.Twebapp@oss-app1> <47b313df.VUjM+zDOnIU22hd9%Joerg.Schilling@fokus.fraunhofer.de> <5b5090780802132046h3e6c8a70nf10c409fdfd69dc3@mail.gmail.com> <47B3CB9D.8010109@sun.com> <47B4C36F.5040700@Sun.Com> <47b597f8.1C4eD1u5CXNjro5O%Joerg.Schilling@fokus.fraunhofer.de> <47bd9b5a.ux3iFhGWtwQ8p8J3%Joerg.Schilling@fokus.fraunhofer.de> <47bdae63.b7Jqp9ZYiTcNwp09%Joerg.Schilling@fokus.fraunhofer.de> <47BDBB31.4040204@sun.com> Message-ID: <47bdbec8.qmnpEjG/etLk0AHv%Joerg.Schilling@fokus.fraunhofer.de> Alan Coopersmith wrote: > Joerg Schilling wrote: ... > Indiana to make a freely redistributable OpenSolaris based distro > without funding SchilliX to do the same, as clearly proven by the last > 6-9 months of history. The OpenSolaris world has distros other than > SchilliX, and not everything is about you. The OpenSolaris surce would not exist without the Sun employees but the OpenSolaris success would not exist without the community outside Sun. If you don't get this, I see no way to continue this conversation. 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 justin at erenkrantz.com Thu Feb 21 11:12:18 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Thu, 21 Feb 2008 11:12:18 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BDB577.6040203@Sun.Com> References: <47BDB577.6040203@Sun.Com> Message-ID: <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> On Thu, Feb 21, 2008 at 9:31 AM, John Plocher wrote: Comments inline. ... > > Approved OpenSolaris Brand Uses > > > > Fair Use > > > > Fair use of the OpenSolaris trademark is, of course, allowed. For example, > > you can talk about the OpenSolaris project and refer to it in your > > writing, in your blogs, etc. And the growing collection of OpenSolaris > > User Groups can use the OpenSolaris trademark in their User Group names, > > as long as they comply with Sun trademark policies for User Groups. I think "User groups" should be a separate sub-section (ie parallel to Fair Use/Description of Your Products/etc.). Their usage is not intended to be covered under the fair use doctrine, but under a separate trademark policy. > > Descriptions of Your Products > > > > You may use a separate, explanatory phrase or tagline after your product > > name that describes how your product relates to one of Sun's. (See Sun > > Trademark Policy) The potential brandings below are taglines with specific > > meanings. > > > > OpenSolaris Fan Buttons and Banners > > > > The fan buttons and banners make it easy for you to endorse and show your > > support for the OpenSolaris community and technology on your websites. > > There is no registration program; the fan buttons are easy to get. There > > are Usage Guidelines and a Trademark License that govern the usage of > > these buttons and banners. > > > > Potential Brandings > > > > OpenSolaris > > > > * Sun has decided to fund Project Indiana to build a base distro, which > > it will call OpenSolaris. I think the sub-section title is a bit misleading. I'd personally rephrase the title of this section as "Sun's OpenSolaris Distribution" or "Distribution name: OpenSolaris" or something similar. The problem with this title is the conflation of the name and the community. To also minimize confusion, it should also be noted here that, AIUI, Sun's position is that it does not intend anyone else except for itself to ever use the bare name OpenSolaris to refer to a complete distribution. (This is to forestall anyone asking this.) > > ___ - Built on/for OpenSolaris and ___ - Compatible with OpenSolaris Again, I think something explicitly indicating that we're referring to Third-party usage would be helpful for readers. > > Compatibility is at the core of the high level OpenSolaris brand. The > > intent of this branding level is to indicate that a given distro (or a > > packaged application?) is closely aligned with the ecosystem of compatible > > software packages and services being created around the OpenSolaris base > > distro. > > > > TODO: Sun seems to want to use the branding to support the concept of > > "allowed to use the OpenSolaris trademark so that they could form an > > ecosystem of compatible software package and services"; the earlier > > discussions around a "core" and/or "test suites" are valuable, but > > probably too low level. (See the parking lot issues on the discussion > > page). > > > > * Might be used by derivatives of OpenSolaris that are "closely > > aligned". > > * Might need some sort of license (clik thru?) (what about attribution?) > > * Derivatives of the code base that are not "closely aligned with the > > ecosystem" are not intended to qualify to use this branding. > > * Do we need/want a tagline for components and repositories to label > > themselves as "built for OpenSolaris"? > > * Might be a mechanism to tag packages that were intended or certified > > or ??? for use with OpenSolaris. > > > > ___ - Built with OpenSolaris Technology, -or- Built on the OpenSolaris > > Codebase > > > > This tagline might be used to deal with incompatible distros and other > > components that use the OpenSolaris codebase but do not produce an > > environment that is closely aligned... as above. I think this category is attempting to be too precise; plus it's very confusing with respect to the first item in the list. I'm not sure there's a real need to differentiate between the two cases. For example, Indiana and Nexenta and SchilliX won't be 100% compatible with each other because of different shells, layouts, packaging systems, etc. Yes, they might be binary compatible, but that'd be it. Why draw a line? If it's not Indiana, then it has to be "Built On" or "Powered By" or whatever. I'd recommend just tossing this. > > __person__ - OpenSolaris Certified > > > > * Might be used for training and skills certification of individuals Is this really intended? Does either the OGB or Sun directly plan on doing certification trainings? And, again, due to the overall naming, I would think OpenSolaris certification could only mean Indiana and not some other distribution. > > Logos > > > > * (may be part of the enthusiasts stuff) there is a need for icons and > > logos that can be used for for grub boot, VM desktop icons, > > favicon/web as well as "powered by" and "runs on" usage. > > > > Other branding guidelines may be developed in the future > > > > * Legally, for the purposes of many nations' trademark laws, all other > > uses of the OpenSolaris trademark need to be reserved by Sun, but > > terms for use are negotiable. Specifically, nothing here precludes the > > development of additional guidelines and subsequent adoption by Sun as > > such needs present themseves. > > * It might be a good thing to stake out some of these other items - > > appliances and embedded, technology reuse (i.e., Apple's and BSD's > > reuse of DTrace and ZFS...), etc. If (when?) a community group develops a new project that was not wholly originated by Sun, what will the policies be? Imagine that if DTrace or ZFS were started as part of the broader OpenSolaris, but that FreeBSD chose to implement that code - how can they describe the functionality? I've not been quite clear whether Sun is happy with the names dtrace and ZFS being used by other OSes. In fact given the bullet points below, I guess they probably aren't. > > Branding and endorsement may or may not be apropriate in all situations > > > > * Any use of Sun's trademarks on a product or service requires a > > license. > > * The community should really encourage attribution back to the > > OpenSolaris community... > > * Fair Use of the OpenSolaris brand does not include or imply any > > commitment by Sun to provide support, service, indemnification, or > > updates. Anyway, I hope you find these comments constructive. -- justin From John.Plocher at Sun.COM Thu Feb 21 13:12:23 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 13:12:23 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> Message-ID: <47BDE937.5060406@Sun.Com> Justin Erenkrantz wrote: > Anyway, I hope you find these comments constructive. I did, and I thank you for them. I incorporated them into draft 0.11. http://www.genunix.org/wiki/index.php/Trademark_usage_and_Branding_guideline -John From johnsonnenschein at gmail.com Thu Feb 21 19:41:30 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 19:41:30 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BDE937.5060406@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> Message-ID: <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> Forgive me if I've come to the table late, but it seems this: "Any use of Sun's trademarks on a product or service requires a license. " would imply that should Nexenta become magically compatible enough to use the name OpenSolaris at all, they would be ineligible by virtue that they could not sell support for their distro or products related to it? Not that there's a damned thing we can do about it, I'm just looking for clarification here -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part URL: From swalker at opensolaris.org Thu Feb 21 19:44:05 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 21:44:05 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> Message-ID: On Thu, Feb 21, 2008 at 9:41 PM, John Sonnenschein wrote: > Forgive me if I've come to the table late, but it seems this: > > "Any use of Sun's trademarks on a product or service requires a > license. " > would imply that should Nexenta become magically compatible enough to > use the name OpenSolaris at all, they would be ineligible by virtue > that they could not sell support for their distro or products related > to it? I think there's an unwritten assumption that fair use doesn't need to be explicitly documented. However, perhaps some examples of fair use should be included so that there is no confusion. In other words, Nexenta can use the name as they are at the moment (I believe) under the terms of fair use. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Thu Feb 21 19:47:36 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 19:47:36 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> Message-ID: <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> On 21-Feb-08, at 7:44 PM, Shawn Walker wrote: > On Thu, Feb 21, 2008 at 9:41 PM, John Sonnenschein > wrote: >> Forgive me if I've come to the table late, but it seems this: >> >> "Any use of Sun's trademarks on a product or service requires a >> license. " > >> would imply that should Nexenta become magically compatible enough to >> use the name OpenSolaris at all, they would be ineligible by virtue >> that they could not sell support for their distro or products related >> to it? > > I think there's an unwritten assumption that fair use doesn't need to > be explicitly documented. > > However, perhaps some examples of fair use should be included so that > there is no confusion. > > In other words, Nexenta can use the name as they are at the moment (I > believe) under the terms of fair use. Are you certain of that? Even if they wanted to sell their distro? My problem here is that if someone like Oracle wanted to run off with the code and make their own distro ( Such as "Unbreakable Linux" ) I think it would be good for OpenSolaris if they could tag the name at the end of it. Quite aside from whether it's good for Oracle or not, they could sell " Oracle TransactOS " and that does nothing to help us, but if they sold " Oracle TransactOS - Built on OpenSolaris " that's free advertising for us. cheers -John -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part URL: From johnsonnenschein at gmail.com Thu Feb 21 19:50:40 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 19:50:40 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BDE937.5060406@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> Message-ID: <40EE52E1-0E31-4D22-A9AB-F2E28DDBB202@gmail.com> Also to note, what can we do about OS.o sub-projects that want to court the favour of other projects ( like dtrace & apple, zfs & freebsd, etc ) IPS - An OpenSolaris Technology ? would this be acceptable? -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part URL: From swalker at opensolaris.org Thu Feb 21 19:52:11 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 21:52:11 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> Message-ID: On Thu, Feb 21, 2008 at 9:47 PM, John Sonnenschein wrote: > > On 21-Feb-08, at 7:44 PM, Shawn Walker wrote: > > > On Thu, Feb 21, 2008 at 9:41 PM, John Sonnenschein > > wrote: > >> Forgive me if I've come to the table late, but it seems this: > >> > >> "Any use of Sun's trademarks on a product or service requires a > >> license. " > > > >> would imply that should Nexenta become magically compatible enough to > >> use the name OpenSolaris at all, they would be ineligible by virtue > >> that they could not sell support for their distro or products related > >> to it? > > > > I think there's an unwritten assumption that fair use doesn't need to > > be explicitly documented. > > > > However, perhaps some examples of fair use should be included so that > > there is no confusion. > > > > In other words, Nexenta can use the name as they are at the moment (I > > believe) under the terms of fair use. > > Are you certain of that? Even if they wanted to sell their distro? Yes, because people from Sun have already stated this; and they are already selling a commercial product right now. See NexentaStor. > My problem here is that if someone like Oracle wanted to run off with > the code and make their own distro ( Such as "Unbreakable Linux" ) I > think it would be good for OpenSolaris if they could tag the name at > the end of it. Quite aside from whether it's good for Oracle or not, > they could sell " Oracle TransactOS " and that does nothing to help > us, but if they sold " Oracle TransactOS - Built on OpenSolaris " > that's free advertising for us. That usage is different from what Nexenta is currently doing. I was only talking about their current usage. In other words, it's fair game to describe the origins of your product in marketing material, etc. from what I've been told by a Sun employee. However, they couldn't call their product "OpenSolaris XYZ!" or "XYZ: an OpenSolaris product". -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Thu Feb 21 19:54:05 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 21:54:05 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> Message-ID: On Thu, Feb 21, 2008 at 9:52 PM, Shawn Walker wrote: > However, they couldn't call their product "OpenSolaris XYZ!" or "XYZ: > an OpenSolaris product". This obviously only applies under the current usage guidelines. Under the new proposed guidelines, they would be able to use the name, but they would have to comply with the guidelines first. So it's actually a better situation than the current one. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Thu Feb 21 19:56:54 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 19:56:54 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> Message-ID: <30C92F2A-6930-40FC-9E5C-334DA5B2CF3A@gmail.com> On 21-Feb-08, at 7:54 PM, Shawn Walker wrote: > On Thu, Feb 21, 2008 at 9:52 PM, Shawn Walker > wrote: >> However, they couldn't call their product "OpenSolaris XYZ!" or "XYZ: >> an OpenSolaris product". > > This obviously only applies under the current usage guidelines. > > Under the new proposed guidelines, they would be able to use the name, > but they would have to comply with the guidelines first. > > So it's actually a better situation than the current one. I'm still uncertain that they would be able to use it in a commercial context ( i'm talking about the new guidelines, talking about what they could do before is unhelpful) , and that's where I'm stuck right now. It's rather late, so any trademark lawyers are likely off doing more interesting things, but I think this should be revisited in the morning -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part URL: From swalker at opensolaris.org Thu Feb 21 20:13:11 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 22:13:11 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <30C92F2A-6930-40FC-9E5C-334DA5B2CF3A@gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> <30C92F2A-6930-40FC-9E5C-334DA5B2CF3A@gmail.com> Message-ID: On Thu, Feb 21, 2008 at 9:56 PM, John Sonnenschein wrote: > > On 21-Feb-08, at 7:54 PM, Shawn Walker wrote: > > > On Thu, Feb 21, 2008 at 9:52 PM, Shawn Walker > > wrote: > >> However, they couldn't call their product "OpenSolaris XYZ!" or "XYZ: > >> an OpenSolaris product". > > > > This obviously only applies under the current usage guidelines. > > > > Under the new proposed guidelines, they would be able to use the name, > > but they would have to comply with the guidelines first. > > > > So it's actually a better situation than the current one. > > > I'm still uncertain that they would be able to use it in a commercial > context ( i'm talking about the new guidelines, talking about what > they could do before is unhelpful) , and that's where I'm stuck right > now. Sun may choose to restrict it in that fashion, but if they do, that is far more restrictive than the Linux trademark usage. It doesn't make much sense for Sun to claim that the rules are for quality control, and then to make it apparent that they are not only about quality, but commercial control. However, we come again to the horrific choice of trademark by Sun. Since OpenSolaris contains the Solaris trademark Sun may have no choice here but to prevent commercial usage to prevent marketplace confusion. Ugh. > It's rather late, so any trademark lawyers are likely off doing more > interesting things, but I think this should be revisited in the morning Yeah. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Thu Feb 21 20:37:53 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 20:37:53 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> Message-ID: <47BE51A1.90703@Sun.Com> John Sonnenschein wrote: > Forgive me if I've come to the table late, but it seems this: > "Any use of Sun's trademarks on a product or service requires a license. " > > would imply that should Nexenta become magically compatible enough to > use the name OpenSolaris at all, they would be ineligible by virtue that > they could not sell support for their distro or products related to it? > > Not that there's a damned thing we can do about it, I'm just looking for > clarification here I *think* that what is meant here is that Erast simply would need to click (OK, I accept) on a web page that specified the proper usage terms for the branding he wished to use. Should Nexenta start using the mark in unapproved ways, this gives Sun a legally acceptable way of showing that they properly communicated the terms of use to Erast and that he accepted them. I *do not think* that this implies in any way any sort of license to use the product itself. See the fan button click-thru license page for an example: http://www.opensolaris.org/os/about/buttons/ -John From John.Plocher at Sun.COM Thu Feb 21 20:41:48 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 20:41:48 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> Message-ID: <47BE528C.2050508@Sun.Com> Shawn Walker wrote: > from what I've been told by a Sun employee. Please look at the T&B section on fair use as well as the reference to Sun's trademark and branding FAQ and let me/us know if they answer your questions about fair use. I don't wish to copy/paste that sort of stuff into this document if I can help it - IMHO, simpler is better, especially with legal-like stuff like this. -John From johnsonnenschein at gmail.com Thu Feb 21 20:46:36 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 20:46:36 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BE51A1.90703@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> Message-ID: On 21-Feb-08, at 8:37 PM, John Plocher wrote: > John Sonnenschein wrote: >> Forgive me if I've come to the table late, but it seems this: >> "Any use of Sun's trademarks on a product or service requires a >> license. " >> would imply that should Nexenta become magically compatible enough >> to use the name OpenSolaris at all, they would be ineligible by >> virtue that they could not sell support for their distro or >> products related to it? >> Not that there's a damned thing we can do about it, I'm just >> looking for clarification here > > I *think* that what is meant here is that Erast simply would need > to click (OK, I accept) on a web page that specified the proper > usage terms for the branding he wished to use. Should Nexenta > start using the mark in unapproved ways, this gives Sun a legally > acceptable way of showing that they properly communicated the > terms of use to Erast and that he accepted them. > > I *do not think* that this implies in any way any sort of license to > use the product itself. > > See the fan button click-thru license page for an example: > http://www.opensolaris.org/os/about/buttons/ Sorry, perhaps i am unclear. I'm asking if some community group wanted to start selling an OS based on opensolaris, would they be permitted to do that while retaining the right to use the name... that is, would "Oracle UnbreakableOS - built on OpenSolaris" be a permissible product for them to sell, for money ( assuming of course they comply with the copyright licensing ) ? From swalker at opensolaris.org Thu Feb 21 20:48:13 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Thu, 21 Feb 2008 22:48:13 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BE528C.2050508@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> <47BE528C.2050508@Sun.Com> Message-ID: On Thu, Feb 21, 2008 at 10:41 PM, John Plocher wrote: > Shawn Walker wrote: > > from what I've been told by a Sun employee. > > > Please look at the T&B section on fair use as well as the reference to Sun's > trademark and branding FAQ and let me/us know if they answer your questions > about fair use. I don't wish to copy/paste that sort of stuff into this > document if I can help it - IMHO, simpler is better, especially with legal-like > stuff like this. That's fine, and would be my preference. Maybe a "references" section with a link to something explaining fair use would suffice though? -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Thu Feb 21 20:49:05 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 20:49:05 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <40EE52E1-0E31-4D22-A9AB-F2E28DDBB202@gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <40EE52E1-0E31-4D22-A9AB-F2E28DDBB202@gmail.com> Message-ID: <029B92BB-16F3-482B-A545-3072E72D1789@gmail.com> On 21-Feb-08, at 7:50 PM, John Sonnenschein wrote: > Also to note, what can we do about OS.o sub-projects that want to > court the favour of other projects ( like dtrace & apple, zfs & > freebsd, etc ) > > IPS - An OpenSolaris Technology > > ? would this be acceptable? I think a standard ought to be set that would allow this usage. Perhaps any project initiated on OpenSolaris.org intended for inclusion on any distribution that fits the " - built with opensolaris technology " moniker ? From John.Plocher at Sun.COM Thu Feb 21 20:53:23 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 20:53:23 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> Message-ID: <47BE5543.9060809@Sun.Com> John Sonnenschein wrote: > Sorry, perhaps i am unclear. I'm asking if some community group wanted > to start selling an OS based on opensolaris, would they be permitted to > do that while retaining the right to use the name... that is, would > "Oracle UnbreakableOS - built on OpenSolaris" be a permissible product > for them to sell, for money ( assuming of course they comply with the > copyright licensing ) ? s/copyright/trademark/ From my discussions with "the executives" and "the lawyers", I get the impression that not only is this acceptable, it is also expected and encouraged. (It is also probably impossible to prevent, since if the distro meets the quality criteria, it is my understanding that Sun has to allow the usage) -John From John.Plocher at Sun.COM Thu Feb 21 20:56:14 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 20:56:14 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <91358224-E7E0-4FEB-8224-F360763B6A15@gmail.com> <47BE528C.2050508@Sun.Com> Message-ID: <47BE55EE.1010304@Sun.Com> Shawn Walker wrote: > That's fine, and would be my preference. Maybe a "references" section > with a link to something explaining fair use would suffice though? Done -John From John.Plocher at Sun.COM Thu Feb 21 20:59:23 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 20:59:23 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <029B92BB-16F3-482B-A545-3072E72D1789@gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <40EE52E1-0E31-4D22-A9AB-F2E28DDBB202@gmail.com> <029B92BB-16F3-482B-A545-3072E72D1789@gmail.com> Message-ID: <47BE56AB.2070007@Sun.Com> John Sonnenschein wrote: > I think a standard ought to be set that would allow this usage. How about: ===___ - An OpenSolaris Technology=== * This tagline might be used to indicate that the project/feature was initiated on or obtained from the OpenSolaris.org community. From johnsonnenschein at gmail.com Thu Feb 21 20:59:23 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 20:59:23 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BE5543.9060809@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> Message-ID: On 21-Feb-08, at 8:53 PM, John Plocher wrote: > John Sonnenschein wrote: >> Sorry, perhaps i am unclear. I'm asking if some community group >> wanted to start selling an OS based on opensolaris, would they be >> permitted to do that while retaining the right to use the name... >> that is, would "Oracle UnbreakableOS - built on OpenSolaris" be a >> permissible product for them to sell, for money ( assuming of >> course they comply with the copyright licensing ) ? > > s/copyright/trademark/ No, i mean the copyright licensing... as in the CDDL. Unimportant > From my discussions with "the executives" and "the lawyers", I get the > impression that not only is this acceptable, it is also expected and > encouraged. (It is also probably impossible to prevent, since if the > distro meets the quality criteria, it is my understanding that Sun > has to allow the usage I think this ought to be clarified in no uncertain terms, because I get the impression that the name isn't up for grabs for commercial uses, only for community/nonprofit uses. Particularly commercial uses that compete directly with Sun's products ( that would be, an operating system ) From johnsonnenschein at gmail.com Thu Feb 21 21:01:34 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Thu, 21 Feb 2008 21:01:34 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BE56AB.2070007@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <40EE52E1-0E31-4D22-A9AB-F2E28DDBB202@gmail.com> <029B92BB-16F3-482B-A545-3072E72D1789@gmail.com> <47BE56AB.2070007@Sun.Com> Message-ID: +1 On 21-Feb-08, at 8:59 PM, John Plocher wrote: > John Sonnenschein wrote: >> I think a standard ought to be set that would allow this usage. > > How about: > > ===___ - An OpenSolaris Technology=== > * This tagline might be used to indicate that the project/feature was > initiated on or obtained from the OpenSolaris.org community. > > From John.Plocher at Sun.COM Thu Feb 21 21:05:20 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Thu, 21 Feb 2008 21:05:20 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> Message-ID: <47BE5810.7090404@Sun.Com> John Sonnenschein wrote: > I think this ought to be clarified in no uncertain terms, Added to the ISSUES list in the document. -John From shivakumar.gn at gmail.com Thu Feb 21 23:33:37 2008 From: shivakumar.gn at gmail.com (S h i v) Date: Fri, 22 Feb 2008 13:03:37 +0530 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BE5543.9060809@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> Message-ID: <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> On Fri, Feb 22, 2008 at 10:23 AM, John Plocher wrote: > John Sonnenschein wrote: > > Sorry, perhaps i am unclear. I'm asking if some community group wanted > > to start selling an OS based on opensolaris, would they be permitted to > > do that while retaining the right to use the name... that is, would > > "Oracle UnbreakableOS - built on OpenSolaris" be a permissible product > > for them to sell, for money ( assuming of course they comply with the > > copyright licensing ) ? > > s/copyright/trademark/ > > > From my discussions with "the executives" and "the lawyers", I get the > impression that not only is this acceptable, it is also expected and > encouraged. (It is also probably impossible to prevent, since if the > distro meets the quality criteria, it is my understanding that Sun > has to allow the usage) > Good to know. Also my +1 to all comments by Justin. Can you please number the issues (even if there is only one currently) so that referring to it is easier. ISSUE At this point in time, nevada is what we have, and indiana is an undefined future possibility. How can people today use the OpenSolaris marks without implying that they are using/doing indiana? Given that Indiana will be referred to as OpenSolaris. "People using OpenSolaris marks without implying they are using Indiana" will remain a proposition that cannot be satisfied. The trademarks are to cover use of a mark to imply origin and endorsement. When a user downloads something called OpenSolaris, then it is implied that it is from the OpenSolaris community. The interests of the community and the distro gets tied together. Community using the trademark at the same time not to imply Indiana will only lead to confusion. There is an issue with the issue! Usage of logos need to be made a more liberal and meaningful for community purposes. Below sentence is not suitable for community purposes "The OpenSolaris Community Project logo requires permission to use and for you to sign a license to post. Click on that image to access requirements information and the license agreement." http://www.opensolaris.org/os/about/buttons Terms [1] and [2] in the license agreement are not suitable for the community. [1] The OpenSolaris Technology Fan Button or Banner must be configured on your Web site as an active link back to the OpenSolaris technology web page at http://opensolaris.org. [2] The OpenSolaris Technology Fan Buttons or Banner is intended solely for use and display in HTML on the World Wide Web. Do not use it in advertising/promotional materials, collateral, products, labels, packaging or any other printed material." Html is not the only way to use a button! Also have a look at "Please, No Modifications (thanks!)" section of http://www.opensolaris.org/os/about/buttons page. A community project should seek art-work and submissions so that it can be approved and made available instead of the message provided at the link. The logos can definitely do with some face-lift and community contribution while being in line with the trademark policy. It is good to seek submissions here as well. Due to some of the reasons above we have not been able to use it in our UG. (How do I use a log in a flyer that I send across about a UG program!) regards Shiv From swalker at opensolaris.org Fri Feb 22 07:46:35 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 09:46:35 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:33 AM, S h i v wrote: > ISSUE At this point in time, nevada is what we have, and indiana is an > undefined future possibility. How can people today use the OpenSolaris > marks without implying that they are using/doing indiana? I disagree that it implies Indiana. Remember that Indiana is just a prototype and will eventually disappear leaving only what we have already today. While it is true that many of the things learned in Indiana will get applied to Solaris/OpenSolaris, it is really no different than any other project that could affect it in the same way. > Usage of logos need to be made a more liberal and meaningful for > community purposes. This needs explanation. > Below sentence is not suitable for community purposes Well, that's sad, but unfortunately, it is necessary. Remember that OpenSolaris is Sun's trademark, and they are going to be marketing things using it. As a result, they have to have quality control procedures in place. You can ask for more specific uses, but I don't think it is reasonable to expect liberal permission to do whatever you want with it. A license of some form is going to be necssary. > Also have a look at "Please, No Modifications (thanks!)" section of > http://www.opensolaris.org/os/about/buttons page. > A community project should seek art-work and submissions so that it > can be approved and made available instead of the message provided at > the link. You can discuss this with Advocacy. I believe they have the ability to get such materials approved by Marketing. > The logos can definitely do with some face-lift and community > contribution while being in line with the trademark policy. It is good > to seek submissions here as well. Unfortunately, I don't think you understand the need a company has to protect trademarks. They can't allow anyone to use the trademark in any way possible and still have it be commercially viable. > Due to some of the reasons above we have not been able to use it in > our UG. (How do I use a log in a flyer that I send across about a UG > program!) I know other UGs have been using it, so perhaps there just needs to be some clarification. I know based on my conversations with Sun folks that they fully intend UGs to be able to ues the logo, etc. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From Marty.Duey at Sun.COM Fri Feb 22 08:33:46 2008 From: Marty.Duey at Sun.COM (Marty Duey) Date: Fri, 22 Feb 2008 09:33:46 -0700 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BE56AB.2070007@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <40EE52E1-0E31-4D22-A9AB-F2E28DDBB202@gmail.com> <029B92BB-16F3-482B-A545-3072E72D1789@gmail.com> <47BE56AB.2070007@Sun.Com> Message-ID: <47BEF96A.6090502@sun.com> I'd replace "might" with "can". Marty On 2/21/2008 9:59 PM, John Plocher wrote: > John Sonnenschein wrote: >> I think a standard ought to be set that would allow this usage. > > How about: > > ===___ - An OpenSolaris Technology=== > * This tagline might be used to indicate that the project/feature was > initiated on or obtained from the OpenSolaris.org community. > > > _______________________________________________ > trademark-policy-dev mailing list > trademark-policy-dev at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/trademark-policy-dev From rich.teer at rite-group.com Fri Feb 22 09:12:23 2008 From: rich.teer at rite-group.com (Rich Teer) Date: Fri, 22 Feb 2008 09:12:23 -0800 (PST) Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> Message-ID: On Fri, 22 Feb 2008, Shawn Walker wrote: > I disagree that it implies Indiana. Remember that Indiana is just a > prototype and will eventually disappear leaving only what we have > already today. If that was the case, Sun wouldn't have made such a big deal about using the name OpenSolaris for Indiana. -- Rich Teer, SCSA, SCNA, SCSECA, OGB member CEO, My Online Home Inventory URLs: http://www.rite-group.com/rich http://www.linkedin.com/in/richteer http://www.myonlinehomeinventory.com From swalker at opensolaris.org Fri Feb 22 09:31:34 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 11:31:34 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 11:12 AM, Rich Teer wrote: > On Fri, 22 Feb 2008, Shawn Walker wrote: > > > I disagree that it implies Indiana. Remember that Indiana is just a > > prototype and will eventually disappear leaving only what we have > > already today. > > If that was the case, Sun wouldn't have made such a big deal about > using the name OpenSolaris for Indiana. I said *eventually* :) Since Indiana is Solaris Next, then inevitably, Indiana will *eventually* disappear and become Solaris/OpenSolaris; thus OpenSolaris will no longer represent Indiana but merely the codebase as it will be. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Fri Feb 22 10:16:37 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 12:16:37 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 11:31 AM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 11:12 AM, Rich Teer wrote: > > On Fri, 22 Feb 2008, Shawn Walker wrote: > > > > > I disagree that it implies Indiana. Remember that Indiana is just a > > > prototype and will eventually disappear leaving only what we have > > > already today. > > > > If that was the case, Sun wouldn't have made such a big deal about > > using the name OpenSolaris for Indiana. > > I said *eventually* :) > > Since Indiana is Solaris Next, then inevitably, Indiana will > *eventually* disappear and become Solaris/OpenSolaris; thus > OpenSolaris will no longer represent Indiana but merely the codebase > as it will be. I just want to be clear that this is based on a posting made by Ian on advocacy-discuss on Sun, 17 Feb 2008 15:33:08 -0500: "It's true, OpenSolaris (Indiana) will replace Solaris Express Developer Edition after SXDE 1/08." -Ian Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 10:30:38 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 13:30:38 -0500 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> Message-ID: <241540330802221030v57b38798ocbc4888ac2e5b67b@mail.gmail.com> On Fri, Feb 22, 2008 at 1:16 PM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 11:31 AM, Shawn Walker wrote: > > On Fri, Feb 22, 2008 at 11:12 AM, Rich Teer wrote: > > > On Fri, 22 Feb 2008, Shawn Walker wrote: > > > > > > > I disagree that it implies Indiana. Remember that Indiana is just a > > > > prototype and will eventually disappear leaving only what we have > > > > already today. > > > > > > If that was the case, Sun wouldn't have made such a big deal about > > > using the name OpenSolaris for Indiana. > > > > I said *eventually* :) > > > > Since Indiana is Solaris Next, then inevitably, Indiana will > > *eventually* disappear and become Solaris/OpenSolaris; thus > > OpenSolaris will no longer represent Indiana but merely the codebase > > as it will be. > > I just want to be clear that this is based on a posting made by Ian on > advocacy-discuss on Sun, 17 Feb 2008 15:33:08 -0500: > > "It's true, OpenSolaris (Indiana) will replace Solaris Express Developer > Edition after SXDE 1/08." -Ian SXDE has always been just secret code for "old version of SXCE" so I'll shed no tears about that, but it would be tragic to see SXCE go -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From John.Plocher at Sun.COM Fri Feb 22 10:31:46 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 10:31:46 -0800 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BEF96A.6090502@sun.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <40EE52E1-0E31-4D22-A9AB-F2E28DDBB202@gmail.com> <029B92BB-16F3-482B-A545-3072E72D1789@gmail.com> <47BE56AB.2070007@Sun.Com> <47BEF96A.6090502@sun.com> Message-ID: <47BF1512.5050700@Sun.Com> Marty Duey wrote: > I'd replace "might" with "can". I don't want to confuse the existing forms (fan buttons, fair use...) with the new ones we are inventing. I deliberately used the future conditional forms to signify that these are under discussion and not yet agreed upon. Once they are "accepted", I will change them to the "can" form. -John From swalker at opensolaris.org Fri Feb 22 10:33:47 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 12:33:47 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <241540330802221030v57b38798ocbc4888ac2e5b67b@mail.gmail.com> References: <47BDB577.6040203@Sun.Com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> <241540330802221030v57b38798ocbc4888ac2e5b67b@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 12:30 PM, John Sonnenschein wrote: > On Fri, Feb 22, 2008 at 1:16 PM, Shawn Walker wrote: > > On Fri, Feb 22, 2008 at 11:31 AM, Shawn Walker wrote: > > > On Fri, Feb 22, 2008 at 11:12 AM, Rich Teer wrote: > > > > On Fri, 22 Feb 2008, Shawn Walker wrote: > > > > > > > > > I disagree that it implies Indiana. Remember that Indiana is just a > > > > > prototype and will eventually disappear leaving only what we have > > > > > already today. > > > > > > > > If that was the case, Sun wouldn't have made such a big deal about > > > > using the name OpenSolaris for Indiana. > > > > > > I said *eventually* :) > > > > > > Since Indiana is Solaris Next, then inevitably, Indiana will > > > *eventually* disappear and become Solaris/OpenSolaris; thus > > > OpenSolaris will no longer represent Indiana but merely the codebase > > > as it will be. > > > > I just want to be clear that this is based on a posting made by Ian on > > advocacy-discuss on Sun, 17 Feb 2008 15:33:08 -0500: > > > > "It's true, OpenSolaris (Indiana) will replace Solaris Express Developer > > Edition after SXDE 1/08." -Ian > > SXDE has always been just secret code for "old version of SXCE" so > I'll shed no tears about that, but it would be tragic to see SXCE go >From what was posted on advocacy-discuss the other day, it appears SXCE will remain for quite some time until release trains merge again. Sun still has to build that internally to support Solaris in their day-to-day business, so I imagine they will keep releasing it. However, I don't know which way they will go. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Fri Feb 22 10:53:39 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 10:53:39 -0800 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> References: <20080222042039.127020@gmx.net> <47BE541E.8090500@Sun.Com> <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> Message-ID: <47BF1A33.6060808@Sun.Com> [redirecting to the T&B discussion alias, sorry I didn't catch this earlier...] John Sonnenschein wrote: > But still, If someone wanted a different design paradigm altogether, One of the paradigms of agile programming is starting with the simplest solution. Extra functionality can then be added later. I don't believe that the T&B stuff we have now precludes even the extreme innovation you postulate because A) the policy can evolve, and B) the experience I have with ~20 years of Solaris shows that extreme innovation can still happen in environments that value backwards compatibility. If you still see a problem, please suggest wording changes that would better deal with this situation. -John From johnsonnenschein at gmail.com Fri Feb 22 10:58:08 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 13:58:08 -0500 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF1A33.6060808@Sun.Com> References: <47BE541E.8090500@Sun.Com> <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> Message-ID: <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> On Fri, Feb 22, 2008 at 1:53 PM, John Plocher wrote: > [redirecting to the T&B discussion alias, sorry I didn't catch this earlier...] > > > John Sonnenschein wrote: > > But still, If someone wanted a different design paradigm altogether, > > One of the paradigms of agile programming is starting with the simplest > solution. Extra functionality can then be added later. I don't believe > that the T&B stuff we have now precludes even the extreme innovation you > postulate because A) the policy can evolve, and B) the experience I have > with ~20 years of Solaris shows that extreme innovation can still happen > in environments that value backwards compatibility. > > If you still see a problem, please suggest wording changes that would better > deal with this situation. It's not a problem of wording, it's a problem of how strict the metric for "compatibility" and "sameness" to carry the "- an OpenSolaris distro" tag will be. Does changing the packaging system change it? the windowing system? the userspace? the order of directories in $PATH? the desktop background? -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From John.Plocher at Sun.COM Fri Feb 22 11:03:34 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 11:03:34 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> References: <20080222042039.127020@gmx.net> <47BE541E.8090500@Sun.Com> <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> Message-ID: <47BF1C86.8070704@Sun.Com> [sorry for the double followup] John Sonnenschein wrote: > But still, If someone wanted a different design paradigm altogether, [duh, I missed your point - I reacted to your gen-too example assuming that you were testing the bounds of the "what is compatible" definition rather than "what if I don't wish to play the compatible game".] IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff is to build a bigger picture. Things that work to improve and enlarge that bigger picture will probably always be easy to add (to the system and to the T&B policy). And, as I said before, experience shows that it is indeed possible to do significant innovation while retaining compatibility. (c.f., zfs, dtrace, ...) OTOH, things that break apart or fragment the ecosystem won't. If you want to go off on your own, and not play with the ecosystem/community, then the ecosystem/community has no desire or obligation to let you claim you are still a part of the ecosystem/community. This assertion is at the heart of the T&B policy, and is intentional. -John From johnsonnenschein at gmail.com Fri Feb 22 11:07:41 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 14:07:41 -0500 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <47BF1C86.8070704@Sun.Com> References: <47BE541E.8090500@Sun.Com> <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> Message-ID: <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > [sorry for the double followup] > > John Sonnenschein wrote: > > But still, If someone wanted a different design paradigm altogether, > > [duh, I missed your point - I reacted to your gen-too example assuming that > you were testing the bounds of the "what is compatible" definition rather > than "what if I don't wish to play the compatible game".] > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > is to build a bigger picture. Things that work to improve and enlarge > that bigger picture will probably always be easy to add (to the system and > to the T&B policy). And, as I said before, experience shows that it is > indeed possible to do significant innovation while retaining compatibility. > (c.f., zfs, dtrace, ...) > > OTOH, things that break apart or fragment the ecosystem won't. If you > want to go off on your own, and not play with the ecosystem/community, > then the ecosystem/community has no desire or obligation to let you claim > you are still a part of the ecosystem/community. > > This assertion is at the heart of the T&B policy, and is intentional. So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... etc every month, moreso than any other group, but choose not to support IPS packages for ideological or technological reasons ( let's assume for argument's sakes that they find the something about the core fundamentals of IPS unworkable, more than simply iteratively fixing could ever solve), choosing instead to support .ebuilds or Bochnig's conary thing, you don't consider them part of the community then? -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 11:22:38 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:22:38 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein wrote: > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > [sorry for the double followup] > > > > John Sonnenschein wrote: > > > But still, If someone wanted a different design paradigm altogether, > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > you were testing the bounds of the "what is compatible" definition rather > > than "what if I don't wish to play the compatible game".] > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > is to build a bigger picture. Things that work to improve and enlarge > > that bigger picture will probably always be easy to add (to the system and > > to the T&B policy). And, as I said before, experience shows that it is > > indeed possible to do significant innovation while retaining compatibility. > > (c.f., zfs, dtrace, ...) > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > want to go off on your own, and not play with the ecosystem/community, > > then the ecosystem/community has no desire or obligation to let you claim > > you are still a part of the ecosystem/community. > > > > This assertion is at the heart of the T&B policy, and is intentional. > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > etc every month, moreso than any other group, but choose not to > support IPS packages for ideological or technological reasons ( let's > assume for argument's sakes that they find the something about the > core fundamentals of IPS unworkable, more than simply iteratively > fixing could ever solve), choosing instead to support .ebuilds or > Bochnig's conary thing, you don't consider them part of the community > then? No, they are choosing to remove themselves from the ecosystem; not the T&B policy. Besides, choosing to not support the official package format wouldn't be very good for users so I would question how much they really want to support the community. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Fri Feb 22 11:26:16 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 11:26:16 -0800 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> References: <47BE541E.8090500@Sun.Com> <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> Message-ID: <47BF21D8.9090205@Sun.Com> John Sonnenschein wrote: > It's not a problem of wording, it's a problem of how strict the metric > for "compatibility" and "sameness" to carry the "- an OpenSolaris > distro" tag will be. The metric has evolved since this T&B discussion started - it used to be "some set of packages" or "passes some runtime test". The current thoughts are more pragmatic: If the goal is to have a large ecosystem, and the ecosystem is defined as the ability to create packages, put them in repositories and have compatible systems be able to download, install and use them, then the metric becomes more of an existence proof - if you can do all this, then you are part of the ecosystem. It is my understanding that the distro built on OS.o out of OS.o bits from the OS.o repository will be called OpenSolaris. It is my hope that the only difference between the above and the version of OpenSolaris that Sun ships will be the choice of which packages from the repository Sun chooses to pre-cache on their release media. > Does changing the packaging system change it? the windowing system? > the userspace? the order of directories in $PATH? the desktop > background? It is my understanding that if the changes are done on OS.o as part of the Indiana effort (or its successors), then they *are* the definition of OpenSolaris. What is missing in the T&B doc is a metric that tells us when the changes you made cross the line from "being OpenSolaris" to "being based on OpenSolaris". My simplistic assumption was that if you make your own distro (instead of simply rebuilding/redistributing the OS.o one), then you are "built on". While that may be sufficient for a first pass, I can see a point where we (the os.o community) would wish to have multiple distros along the lines of kbuntu and edubuntu.... -John From johnsonnenschein at gmail.com Fri Feb 22 11:26:24 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 14:26:24 -0500 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> Message-ID: <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> On Fri, Feb 22, 2008 at 2:22 PM, Shawn Walker wrote: > > On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein > wrote: > > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > > [sorry for the double followup] > > > > > > John Sonnenschein wrote: > > > > But still, If someone wanted a different design paradigm altogether, > > > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > > you were testing the bounds of the "what is compatible" definition rather > > > than "what if I don't wish to play the compatible game".] > > > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > > is to build a bigger picture. Things that work to improve and enlarge > > > that bigger picture will probably always be easy to add (to the system and > > > to the T&B policy). And, as I said before, experience shows that it is > > > indeed possible to do significant innovation while retaining compatibility. > > > (c.f., zfs, dtrace, ...) > > > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > > want to go off on your own, and not play with the ecosystem/community, > > > then the ecosystem/community has no desire or obligation to let you claim > > > you are still a part of the ecosystem/community. > > > > > > This assertion is at the heart of the T&B policy, and is intentional. > > > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > > etc every month, moreso than any other group, but choose not to > > support IPS packages for ideological or technological reasons ( let's > > assume for argument's sakes that they find the something about the > > core fundamentals of IPS unworkable, more than simply iteratively > > fixing could ever solve), choosing instead to support .ebuilds or > > Bochnig's conary thing, you don't consider them part of the community > > then? > > No, they are choosing to remove themselves from the ecosystem; not the > T&B policy. > > Besides, choosing to not support the official package format wouldn't > be very good for users so I would question how much they really want > to support the community. Even if hypothetically the official package format sucks by design and there's no real way to fix it except by throwing it out, except that won't happen because of pigheadedness on the part of the official package-format-choosing-group? Seems that it would help users to be slightly incompatible there... API and ABI compatibility are far more important in this case than package compatibility -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From brandorr at opensolaris.org Fri Feb 22 11:27:54 2008 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 22 Feb 2008 14:27:54 -0500 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> Message-ID: <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> John, Try not to get confused. It doesn't matter. You can't "change" anything under the new policy. As I understand, you must use the Sun compiled binaries. If you build something from source and it barks the same and walks the same, it still can't be "OpenSolaris". The policy seems very clear. OpenSolaris is now going to be a Sun product. Think RedHat Enterprise Linux vs. CentOS Linux. They are effectively identical, and built from the same source code. OpenSolaris is no longer the independent community project that Sun claimed they were trying to create and keep separate from their commercial product Solaris. OpenSolaris now is their commercial product, with Solaris being deprecated for use with legacy customers. Actually, I wonder. Is using Sun supplied binaries mandatory? Can individual community members compile an IPS distro from the opensolaris.org source tree and call it xyz OpenSolaris? (This last question is for the room.) Cheers, -Brian On Fri, Feb 22, 2008 at 1:58 PM, John Sonnenschein wrote: > On Fri, Feb 22, 2008 at 1:53 PM, John Plocher wrote: > > [redirecting to the T&B discussion alias, sorry I didn't catch this earlier...] > > > > > > John Sonnenschein wrote: > > > But still, If someone wanted a different design paradigm altogether, > > > > One of the paradigms of agile programming is starting with the simplest > > solution. Extra functionality can then be added later. I don't believe > > that the T&B stuff we have now precludes even the extreme innovation you > > postulate because A) the policy can evolve, and B) the experience I have > > with ~20 years of Solaris shows that extreme innovation can still happen > > in environments that value backwards compatibility. > > > > If you still see a problem, please suggest wording changes that would better > > deal with this situation. > > It's not a problem of wording, it's a problem of how strict the metric > for "compatibility" and "sameness" to carry the "- an OpenSolaris > distro" tag will be. > > Does changing the packaging system change it? the windowing system? > the userspace? the order of directories in $PATH? the desktop > background? > > > -- > PGP Public Key 0x437AF1A1 > Available on hkp://pgp.mit.edu > _______________________________________________ > > > trademark-policy-dev mailing list > trademark-policy-dev at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/trademark-policy-dev > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ http://www.genunix.org/wiki/index.php/OpenSolaris_New_User_FAQ From johnsonnenschein at gmail.com Fri Feb 22 11:35:35 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 14:35:35 -0500 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF21D8.9090205@Sun.Com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <47BF21D8.9090205@Sun.Com> Message-ID: <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> Okay, another question then. What subset of packages are required to qualify for compatibility? Say for example I build a distro that only includes SUNWips, SUNWos-net ( or whatever ON will be called ) and the rest are pulled from another repo ( blastwave, for example ). And lets say this distro builds on top of a whole slew of weird packages that are mostly incompatible with anything that indiana includes ( say... the freebsd userspace ) compiled with a compiler that makes ABI compatibility an impossibility ( like gcc 2.95 ) Is it still " BreakFOO - an OpenSolaris Distribution " ? technically you can use IPS to pull in all the indiana stuff and make it compatible, and it includes a subset of packages that indiana does, but in and of itself it is not. Or will the whole of indiana be required for compatibility ( that is, KIndiana is an impossibility, because it doesn't include GNOME ) ? On Fri, Feb 22, 2008 at 2:26 PM, John Plocher wrote: > John Sonnenschein wrote: > > It's not a problem of wording, it's a problem of how strict the metric > > for "compatibility" and "sameness" to carry the "- an OpenSolaris > > distro" tag will be. > > The metric has evolved since this T&B discussion started - it used > to be "some set of packages" or "passes some runtime test". > > The current thoughts are more pragmatic: If the goal is to have a > large ecosystem, and the ecosystem is defined as the ability to > create packages, put them in repositories and have compatible systems > be able to download, install and use them, then the metric becomes > more of an existence proof - if you can do all this, then you are > part of the ecosystem. > > It is my understanding that the distro built on OS.o out of OS.o bits > from the OS.o repository will be called OpenSolaris. > > It is my hope that the only difference between the above and the > version of OpenSolaris that Sun ships will be the choice of which > packages from the repository Sun chooses to pre-cache on their > release media. > > > > > Does changing the packaging system change it? the windowing system? > > the userspace? the order of directories in $PATH? the desktop > > background? > > > It is my understanding that if the changes are done on OS.o as part > of the Indiana effort (or its successors), then they *are* the definition > of OpenSolaris. > > What is missing in the T&B doc is a metric that tells us when the changes > you made cross the line from "being OpenSolaris" to "being based on > OpenSolaris". My simplistic assumption was that if you make your > own distro (instead of simply rebuilding/redistributing the OS.o one), > then you are "built on". While that may be sufficient for a first pass, > I can see a point where we (the os.o community) would wish to have > multiple distros along the lines of kbuntu and edubuntu.... > > -John > > > -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From justin at erenkrantz.com Fri Feb 22 11:37:32 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Fri, 22 Feb 2008 11:37:32 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> Message-ID: <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> On Fri, Feb 22, 2008 at 11:22 AM, Shawn Walker wrote: > Besides, choosing to not support the official package format wouldn't > be very good for users so I would question how much they really want > to support the community. I wouldn't equate having a different packaging format as being not supportive of the community. I realize Indiana developers would certainly want to enforce that, but IPS is new and I think it's reasonable if a distribution were to use the old .pkg format, or .deb's or .rpm's or ports or whatnot and still be considered part of the OpenSolaris ecosystem. If a line has to be drawn, I'd draw the line at binary and API compatibility - but certainly packaging format isn't where I think the line should be. My $.02. -- justin From swalker at opensolaris.org Fri Feb 22 11:37:26 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:37:26 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:26 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 2:22 PM, Shawn Walker wrote: > > > > On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein > > wrote: > > > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > > > [sorry for the double followup] > > > > > > > > John Sonnenschein wrote: > > > > > But still, If someone wanted a different design paradigm altogether, > > > > > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > > > you were testing the bounds of the "what is compatible" definition rather > > > > than "what if I don't wish to play the compatible game".] > > > > > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > > > is to build a bigger picture. Things that work to improve and enlarge > > > > that bigger picture will probably always be easy to add (to the system and > > > > to the T&B policy). And, as I said before, experience shows that it is > > > > indeed possible to do significant innovation while retaining compatibility. > > > > (c.f., zfs, dtrace, ...) > > > > > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > > > want to go off on your own, and not play with the ecosystem/community, > > > > then the ecosystem/community has no desire or obligation to let you claim > > > > you are still a part of the ecosystem/community. > > > > > > > > This assertion is at the heart of the T&B policy, and is intentional. > > > > > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > > > etc every month, moreso than any other group, but choose not to > > > support IPS packages for ideological or technological reasons ( let's > > > assume for argument's sakes that they find the something about the > > > core fundamentals of IPS unworkable, more than simply iteratively > > > fixing could ever solve), choosing instead to support .ebuilds or > > > Bochnig's conary thing, you don't consider them part of the community > > > then? > > > > No, they are choosing to remove themselves from the ecosystem; not the > > T&B policy. > > > > Besides, choosing to not support the official package format wouldn't > > be very good for users so I would question how much they really want > > to support the community. > > Even if hypothetically the official package format sucks by design and > there's no real way to fix it except by throwing it out, except that > won't happen because of pigheadedness on the part of the official > package-format-choosing-group? Seems that it would help users to be > slightly incompatible there... API and ABI compatibility are far more > important in this case than package compatibility Are you really putting so little faith in the members (engineers) of this community? Because that's what your hypothetically saying. You're saying, "what if hypothetically speaking, all the engineers produce something that sucks by design and there's no way to fix it except by throwing it out." Creating hypothetical problems that are unlikely to exist will only cause us to precious time. If you really want to apply that hypothetical situation, then it could apply to *any* project here. ON, GNOME, Installer, etc. the list goes on. Where do you draw the hypothetical line for hypothetical problems to figure out hypothetical solutions? My brain is ready to explode :) Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Fri Feb 22 11:41:16 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:41:16 -0600 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:27 PM, Brandorr wrote: > John, > > Try not to get confused. It doesn't matter. You can't "change" > anything under the new policy. As I understand, you must use the Sun > compiled binaries. If you build something from source and it barks the > same and walks the same, it still can't be "OpenSolaris". The policy > seems very clear. >From a very simplistic level, how else do you guarantee compatibility until we have a gargantuan test suite? There is no way to tell software vendors that *yes, this beyond a doubt will work the same*. > Actually, I wonder. Is using Sun supplied binaries mandatory? Can That has not been established yet, so this could be "much do about nothing." I believe it still needs discussion to accommodate inevitable ports to other architectures. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Fri Feb 22 11:41:38 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 11:41:38 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> References: <47BE541E.8090500@Sun.Com> <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> Message-ID: <47BF2572.5080204@Sun.Com> John Sonnenschein wrote: > So, if a subgroup... choose not to > support IPS packages ... you don't consider them part of the community > then? I'd leave that decision up to their sponsoring CG. I think it extremely unlikely that any subgroup would be allowed to integrate into ON, for example, without first having addressed the packaging issue - and if they don't have the appropriate color of packages, the ON CG won't let them integrate. If there is a technological reason that prevents the subteam from packaging their stuff, that is a bug, and the bug needs to get fixed, even if it means evolving the packaging system. Once the bug is fixed, the integration obstacle goes away. If you simply don't like the packaging system, the productive answer is to join Danek and Stephen and Bart's IPS project and make it /better/, however you may define /better/. Alternatively, you could work with Brian on coronary, or you could roll your own gen-three system, and, if it/they were accepted by the ON, admin and install CG/Consolidations, you (and we) all win. -John From johnsonnenschein at gmail.com Fri Feb 22 11:42:50 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 14:42:50 -0500 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: References: <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> Message-ID: <241540330802221142s34de758eob986ca524ec9a2f1@mail.gmail.com> On Fri, Feb 22, 2008 at 2:37 PM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 1:26 PM, John Sonnenschein > > > wrote: > > > > On Fri, Feb 22, 2008 at 2:22 PM, Shawn Walker wrote: > > > > > > On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein > > > wrote: > > > > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > > > > [sorry for the double followup] > > > > > > > > > > John Sonnenschein wrote: > > > > > > But still, If someone wanted a different design paradigm altogether, > > > > > > > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > > > > you were testing the bounds of the "what is compatible" definition rather > > > > > than "what if I don't wish to play the compatible game".] > > > > > > > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > > > > is to build a bigger picture. Things that work to improve and enlarge > > > > > that bigger picture will probably always be easy to add (to the system and > > > > > to the T&B policy). And, as I said before, experience shows that it is > > > > > indeed possible to do significant innovation while retaining compatibility. > > > > > (c.f., zfs, dtrace, ...) > > > > > > > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > > > > want to go off on your own, and not play with the ecosystem/community, > > > > > then the ecosystem/community has no desire or obligation to let you claim > > > > > you are still a part of the ecosystem/community. > > > > > > > > > > This assertion is at the heart of the T&B policy, and is intentional. > > > > > > > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > > > > etc every month, moreso than any other group, but choose not to > > > > support IPS packages for ideological or technological reasons ( let's > > > > assume for argument's sakes that they find the something about the > > > > core fundamentals of IPS unworkable, more than simply iteratively > > > > fixing could ever solve), choosing instead to support .ebuilds or > > > > Bochnig's conary thing, you don't consider them part of the community > > > > then? > > > > > > No, they are choosing to remove themselves from the ecosystem; not the > > > T&B policy. > > > > > > Besides, choosing to not support the official package format wouldn't > > > be very good for users so I would question how much they really want > > > to support the community. > > > > Even if hypothetically the official package format sucks by design and > > there's no real way to fix it except by throwing it out, except that > > won't happen because of pigheadedness on the part of the official > > package-format-choosing-group? Seems that it would help users to be > > slightly incompatible there... API and ABI compatibility are far more > > important in this case than package compatibility > > Are you really putting so little faith in the members (engineers) of > this community? > > Because that's what your hypothetically saying. > > You're saying, "what if hypothetically speaking, all the engineers > produce something that sucks by design and there's no way to fix it > except by throwing it out." > > Creating hypothetical problems that are unlikely to exist will only > cause us to precious time. > > If you really want to apply that hypothetical situation, then it could > apply to *any* project here. ON, GNOME, Installer, etc. the list goes > on. > > Where do you draw the hypothetical line for hypothetical problems to > figure out hypothetical solutions? You are attacking me on the basis of whether I have faith in the engineers or not, whereas I am trying to clarify what is meant by certain terminology by using hypothetical scenarios. I have plenty of faith in the engineers and I don't think this scenario is likely. I would like to be more clear on the terminology that is being bandied about as if everyone understands it and is on the same page and has the same values and ideals, which may or may not be true, but the work of a programmer, like the work of a trademark lawyer, is one of precision, so it is important to be as precise as possible when throwing terms around. -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 11:44:04 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:44:04 -0600 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <47BF21D8.9090205@Sun.Com> <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:35 PM, John Sonnenschein wrote: > Or will the whole of indiana be required for compatibility ( that is, > KIndiana is an impossibility, because it doesn't include GNOME ) ? I believe Ian stated explicitly at the conference that he wants to see this happen; so I imagine simply providing additional (but not replacement) packages won't affect anything. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Fri Feb 22 11:46:47 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:46:47 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221142s34de758eob986ca524ec9a2f1@mail.gmail.com> References: <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> <241540330802221142s34de758eob986ca524ec9a2f1@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:42 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 2:37 PM, Shawn Walker wrote: > > On Fri, Feb 22, 2008 at 1:26 PM, John Sonnenschein > > > > > > wrote: > > > > > > On Fri, Feb 22, 2008 at 2:22 PM, Shawn Walker wrote: > > > > > > > > On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein > > > > wrote: > > > > > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > > > > > [sorry for the double followup] > > > > > > > > > > > > John Sonnenschein wrote: > > > > > > > But still, If someone wanted a different design paradigm altogether, > > > > > > > > > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > > > > > you were testing the bounds of the "what is compatible" definition rather > > > > > > than "what if I don't wish to play the compatible game".] > > > > > > > > > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > > > > > is to build a bigger picture. Things that work to improve and enlarge > > > > > > that bigger picture will probably always be easy to add (to the system and > > > > > > to the T&B policy). And, as I said before, experience shows that it is > > > > > > indeed possible to do significant innovation while retaining compatibility. > > > > > > (c.f., zfs, dtrace, ...) > > > > > > > > > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > > > > > want to go off on your own, and not play with the ecosystem/community, > > > > > > then the ecosystem/community has no desire or obligation to let you claim > > > > > > you are still a part of the ecosystem/community. > > > > > > > > > > > > This assertion is at the heart of the T&B policy, and is intentional. > > > > > > > > > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > > > > > etc every month, moreso than any other group, but choose not to > > > > > support IPS packages for ideological or technological reasons ( let's > > > > > assume for argument's sakes that they find the something about the > > > > > core fundamentals of IPS unworkable, more than simply iteratively > > > > > fixing could ever solve), choosing instead to support .ebuilds or > > > > > Bochnig's conary thing, you don't consider them part of the community > > > > > then? > > > > > > > > No, they are choosing to remove themselves from the ecosystem; not the > > > > T&B policy. > > > > > > > > Besides, choosing to not support the official package format wouldn't > > > > be very good for users so I would question how much they really want > > > > to support the community. > > > > > > Even if hypothetically the official package format sucks by design and > > > there's no real way to fix it except by throwing it out, except that > > > won't happen because of pigheadedness on the part of the official > > > package-format-choosing-group? Seems that it would help users to be > > > slightly incompatible there... API and ABI compatibility are far more > > > important in this case than package compatibility > > > > Are you really putting so little faith in the members (engineers) of > > this community? > > > > Because that's what your hypothetically saying. > > > > You're saying, "what if hypothetically speaking, all the engineers > > produce something that sucks by design and there's no way to fix it > > except by throwing it out." > > > > Creating hypothetical problems that are unlikely to exist will only > > cause us to precious time. > > > > If you really want to apply that hypothetical situation, then it could > > apply to *any* project here. ON, GNOME, Installer, etc. the list goes > > on. > > > > Where do you draw the hypothetical line for hypothetical problems to > > figure out hypothetical solutions? > > You are attacking me on the basis of whether I have faith in the > engineers or not, whereas I am trying to clarify what is meant by > certain terminology by using hypothetical scenarios. I have plenty of > faith in the engineers and I don't think this scenario is likely. No, I am not attacking you. I am providing critical input to your hypothetical situation. Regardless of how you want to view it, my statement and yours end up being fairly equivalent. If you believe that such a packaging system will be integrated in such a hypothetical state, then you are also saying you don't believe in members of this community to produce something good. > I would like to be more clear on the terminology that is being bandied > about as if everyone understands it and is on the same page and has > the same values and ideals, which may or may not be true, but the work > of a programmer, like the work of a trademark lawyer, is one of > precision, so it is important to be as precise as possible when > throwing terms around. I think statements that question the integrity of the community are unnecessary. You need to assume certain basic good about folks and deal with problems only when they arise. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 11:47:55 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 14:47:55 -0500 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <47BF21D8.9090205@Sun.Com> <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> Message-ID: <241540330802221147xbf5cecana2dfe7a2433b750@mail.gmail.com> what if CKSodduserspace is just an additional package? What if I build a distro on it that doesn't include SUNWnormaluserpsace ? it is the same parallel as the distro that includes SUNWkde but not SUNWgnome. The GUI is replaced. Which case is allowed to use the "- an OpenSolaris Distribution " moniker, and on what grounds? On Fri, Feb 22, 2008 at 2:44 PM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 1:35 PM, John Sonnenschein > wrote: > > Or will the whole of indiana be required for compatibility ( that is, > > KIndiana is an impossibility, because it doesn't include GNOME ) ? > > I believe Ian stated explicitly at the conference that he wants to see > this happen; so I imagine simply providing additional (but not > replacement) packages won't affect anything. > > > > -- > Shawn Walker, Software and Systems Analyst > http://binarycrusader.blogspot.com/ > > "To err is human -- and to blame it on a computer is even more so." - > Robert Orben > -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 11:46:47 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:46:47 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221142s34de758eob986ca524ec9a2f1@mail.gmail.com> References: <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> <241540330802221142s34de758eob986ca524ec9a2f1@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:42 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 2:37 PM, Shawn Walker wrote: > > On Fri, Feb 22, 2008 at 1:26 PM, John Sonnenschein > > > > > > wrote: > > > > > > On Fri, Feb 22, 2008 at 2:22 PM, Shawn Walker wrote: > > > > > > > > On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein > > > > wrote: > > > > > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > > > > > [sorry for the double followup] > > > > > > > > > > > > John Sonnenschein wrote: > > > > > > > But still, If someone wanted a different design paradigm altogether, > > > > > > > > > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > > > > > you were testing the bounds of the "what is compatible" definition rather > > > > > > than "what if I don't wish to play the compatible game".] > > > > > > > > > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > > > > > is to build a bigger picture. Things that work to improve and enlarge > > > > > > that bigger picture will probably always be easy to add (to the system and > > > > > > to the T&B policy). And, as I said before, experience shows that it is > > > > > > indeed possible to do significant innovation while retaining compatibility. > > > > > > (c.f., zfs, dtrace, ...) > > > > > > > > > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > > > > > want to go off on your own, and not play with the ecosystem/community, > > > > > > then the ecosystem/community has no desire or obligation to let you claim > > > > > > you are still a part of the ecosystem/community. > > > > > > > > > > > > This assertion is at the heart of the T&B policy, and is intentional. > > > > > > > > > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > > > > > etc every month, moreso than any other group, but choose not to > > > > > support IPS packages for ideological or technological reasons ( let's > > > > > assume for argument's sakes that they find the something about the > > > > > core fundamentals of IPS unworkable, more than simply iteratively > > > > > fixing could ever solve), choosing instead to support .ebuilds or > > > > > Bochnig's conary thing, you don't consider them part of the community > > > > > then? > > > > > > > > No, they are choosing to remove themselves from the ecosystem; not the > > > > T&B policy. > > > > > > > > Besides, choosing to not support the official package format wouldn't > > > > be very good for users so I would question how much they really want > > > > to support the community. > > > > > > Even if hypothetically the official package format sucks by design and > > > there's no real way to fix it except by throwing it out, except that > > > won't happen because of pigheadedness on the part of the official > > > package-format-choosing-group? Seems that it would help users to be > > > slightly incompatible there... API and ABI compatibility are far more > > > important in this case than package compatibility > > > > Are you really putting so little faith in the members (engineers) of > > this community? > > > > Because that's what your hypothetically saying. > > > > You're saying, "what if hypothetically speaking, all the engineers > > produce something that sucks by design and there's no way to fix it > > except by throwing it out." > > > > Creating hypothetical problems that are unlikely to exist will only > > cause us to precious time. > > > > If you really want to apply that hypothetical situation, then it could > > apply to *any* project here. ON, GNOME, Installer, etc. the list goes > > on. > > > > Where do you draw the hypothetical line for hypothetical problems to > > figure out hypothetical solutions? > > You are attacking me on the basis of whether I have faith in the > engineers or not, whereas I am trying to clarify what is meant by > certain terminology by using hypothetical scenarios. I have plenty of > faith in the engineers and I don't think this scenario is likely. No, I am not attacking you. I am providing critical input to your hypothetical situation. Regardless of how you want to view it, my statement and yours end up being fairly equivalent. If you believe that such a packaging system will be integrated in such a hypothetical state, then you are also saying you don't believe in members of this community to produce something good. > I would like to be more clear on the terminology that is being bandied > about as if everyone understands it and is on the same page and has > the same values and ideals, which may or may not be true, but the work > of a programmer, like the work of a trademark lawyer, is one of > precision, so it is important to be as precise as possible when > throwing terms around. I think statements that question the integrity of the community are unnecessary. You need to assume certain basic good about folks and deal with problems only when they arise. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 11:49:52 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 14:49:52 -0500 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: References: <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> <241540330802221142s34de758eob986ca524ec9a2f1@mail.gmail.com> Message-ID: <241540330802221149w54928ab7uf38e323cba567958@mail.gmail.com> On Fri, Feb 22, 2008 at 2:46 PM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 1:42 PM, John Sonnenschein > > > wrote: > > > > On Fri, Feb 22, 2008 at 2:37 PM, Shawn Walker wrote: > > > On Fri, Feb 22, 2008 at 1:26 PM, John Sonnenschein > > > > > > > > > wrote: > > > > > > > > On Fri, Feb 22, 2008 at 2:22 PM, Shawn Walker wrote: > > > > > > > > > > On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein > > > > > wrote: > > > > > > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > > > > > > [sorry for the double followup] > > > > > > > > > > > > > > John Sonnenschein wrote: > > > > > > > > But still, If someone wanted a different design paradigm altogether, > > > > > > > > > > > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > > > > > > you were testing the bounds of the "what is compatible" definition rather > > > > > > > than "what if I don't wish to play the compatible game".] > > > > > > > > > > > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > > > > > > is to build a bigger picture. Things that work to improve and enlarge > > > > > > > that bigger picture will probably always be easy to add (to the system and > > > > > > > to the T&B policy). And, as I said before, experience shows that it is > > > > > > > indeed possible to do significant innovation while retaining compatibility. > > > > > > > (c.f., zfs, dtrace, ...) > > > > > > > > > > > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > > > > > > want to go off on your own, and not play with the ecosystem/community, > > > > > > > then the ecosystem/community has no desire or obligation to let you claim > > > > > > > you are still a part of the ecosystem/community. > > > > > > > > > > > > > > This assertion is at the heart of the T&B policy, and is intentional. > > > > > > > > > > > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > > > > > > etc every month, moreso than any other group, but choose not to > > > > > > support IPS packages for ideological or technological reasons ( let's > > > > > > assume for argument's sakes that they find the something about the > > > > > > core fundamentals of IPS unworkable, more than simply iteratively > > > > > > fixing could ever solve), choosing instead to support .ebuilds or > > > > > > Bochnig's conary thing, you don't consider them part of the community > > > > > > then? > > > > > > > > > > No, they are choosing to remove themselves from the ecosystem; not the > > > > > T&B policy. > > > > > > > > > > Besides, choosing to not support the official package format wouldn't > > > > > be very good for users so I would question how much they really want > > > > > to support the community. > > > > > > > > Even if hypothetically the official package format sucks by design and > > > > there's no real way to fix it except by throwing it out, except that > > > > won't happen because of pigheadedness on the part of the official > > > > package-format-choosing-group? Seems that it would help users to be > > > > slightly incompatible there... API and ABI compatibility are far more > > > > important in this case than package compatibility > > > > > > Are you really putting so little faith in the members (engineers) of > > > this community? > > > > > > Because that's what your hypothetically saying. > > > > > > You're saying, "what if hypothetically speaking, all the engineers > > > produce something that sucks by design and there's no way to fix it > > > except by throwing it out." > > > > > > Creating hypothetical problems that are unlikely to exist will only > > > cause us to precious time. > > > > > > If you really want to apply that hypothetical situation, then it could > > > apply to *any* project here. ON, GNOME, Installer, etc. the list goes > > > on. > > > > > > Where do you draw the hypothetical line for hypothetical problems to > > > figure out hypothetical solutions? > > > > You are attacking me on the basis of whether I have faith in the > > engineers or not, whereas I am trying to clarify what is meant by > > certain terminology by using hypothetical scenarios. I have plenty of > > faith in the engineers and I don't think this scenario is likely. > > No, I am not attacking you. I am providing critical input to your > hypothetical situation. > > Regardless of how you want to view it, my statement and yours end up > being fairly equivalent. > > If you believe that such a packaging system will be integrated in such > a hypothetical state, then you are also saying you don't believe in > members of this community to produce something good. > > > > I would like to be more clear on the terminology that is being bandied > > about as if everyone understands it and is on the same page and has > > the same values and ideals, which may or may not be true, but the work > > of a programmer, like the work of a trademark lawyer, is one of > > precision, so it is important to be as precise as possible when > > throwing terms around. > > I think statements that question the integrity of the community are > unnecessary. You need to assume certain basic good about folks and > deal with problems only when they arise. I do assume basic goods about the community, but if recent events are any indication you must deal with potential problems well before they arise, because if they're bad, they don't get dealt with... you just get told to "get over it and move on" -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 11:57:00 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:57:00 -0600 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221147xbf5cecana2dfe7a2433b750@mail.gmail.com> References: <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <47BF21D8.9090205@Sun.Com> <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> <241540330802221147xbf5cecana2dfe7a2433b750@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:47 PM, John Sonnenschein wrote: > what if CKSodduserspace is just an additional package? What if I build > a distro on it that doesn't include SUNWnormaluserpsace ? Remember that the distribution is defined by the repositories. As long as those repositories are still accessible, I would imagine it could be still considered compatible. However, the definition of this compatibility is still too nebulous for me to say with certainty. > it is the same parallel as the distro that includes SUNWkde but not > SUNWgnome. The GUI is replaced. The GUI being replaced doesn't break anything because packages installing will trigger the automatic installation of any dependent packages. > Which case is allowed to use the "- an OpenSolaris Distribution " > moniker, and on what grounds? I believe that if you go back in our mail archive you can answer that question with help from the current T&B proposal draft. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Fri Feb 22 11:57:28 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 11:57:28 -0800 Subject: [trademark-policy-dev] Brandorr: recompiling "OpenSolaris" from the sources? In-Reply-To: <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> Message-ID: <47BF2928.6070707@Sun.Com> [moved to T&B alias only] Brandorr wrote: > John, > > Try not to get confused. It doesn't matter. You can't "change" > anything under the new policy. As I understand, you must use the Sun > compiled binaries. If you build something from source and it barks the > same and walks the same, it still can't be "OpenSolaris". The policy > seems very clear. Where do you get this understanding from? While you and I had discussions similar to what you assert several months ago, they were only and obviously brainstorming ideas. I understand you are disappointed and bitter about many things; that is no reason to represent opinions and ideas as facts. > Actually, I wonder. Is using Sun supplied binaries mandatory? Can > individual community members compile an IPS distro from the > opensolaris.org source tree Good question. I assume so, and the wording of the T&B policy currently is directed at that goal. Nevertheless, I'll add your question to the list I need to resolve. > ...and call it xyz OpenSolaris? (This last > question is for the room.) The room does not need to answer - the lawyers already have: http://www.sun.com/policies/trademarks/ Sun trademarks cannot be included in the name of any third party product, technology, program or service. This includes free and educational materials, open source distributions, and the titles of informational web sites. This is simply a generalization of US and international trademark law; you can not use other people's trademarks in the name of your product. -John From swalker at opensolaris.org Fri Feb 22 11:57:50 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 13:57:50 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221149w54928ab7uf38e323cba567958@mail.gmail.com> References: <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <241540330802221126u58188a13m7a57e90f020298a2@mail.gmail.com> <241540330802221142s34de758eob986ca524ec9a2f1@mail.gmail.com> <241540330802221149w54928ab7uf38e323cba567958@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 1:49 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 2:46 PM, Shawn Walker wrote: > > On Fri, Feb 22, 2008 at 1:42 PM, John Sonnenschein > > > > > > wrote: > > > > > > On Fri, Feb 22, 2008 at 2:37 PM, Shawn Walker wrote: > > > > On Fri, Feb 22, 2008 at 1:26 PM, John Sonnenschein > > > > > > > > > > > > wrote: > > > > > > > > > > On Fri, Feb 22, 2008 at 2:22 PM, Shawn Walker wrote: > > > > > > > > > > > > On Fri, Feb 22, 2008 at 1:07 PM, John Sonnenschein > > > > > > wrote: > > > > > > > On Fri, Feb 22, 2008 at 2:03 PM, John Plocher wrote: > > > > > > > > [sorry for the double followup] > > > > > > > > > > > > > > > > John Sonnenschein wrote: > > > > > > > > > But still, If someone wanted a different design paradigm altogether, > > > > > > > > > > > > > > > > [duh, I missed your point - I reacted to your gen-too example assuming that > > > > > > > > you were testing the bounds of the "what is compatible" definition rather > > > > > > > > than "what if I don't wish to play the compatible game".] > > > > > > > > > > > > > > > > IMO, the goal of the OpenSolaris brand-enforced ecosystem of compatible stuff > > > > > > > > is to build a bigger picture. Things that work to improve and enlarge > > > > > > > > that bigger picture will probably always be easy to add (to the system and > > > > > > > > to the T&B policy). And, as I said before, experience shows that it is > > > > > > > > indeed possible to do significant innovation while retaining compatibility. > > > > > > > > (c.f., zfs, dtrace, ...) > > > > > > > > > > > > > > > > OTOH, things that break apart or fragment the ecosystem won't. If you > > > > > > > > want to go off on your own, and not play with the ecosystem/community, > > > > > > > > then the ecosystem/community has no desire or obligation to let you claim > > > > > > > > you are still a part of the ecosystem/community. > > > > > > > > > > > > > > > > This assertion is at the heart of the T&B policy, and is intentional. > > > > > > > > > > > > > > So, if a subgroup has hundreds of putbacks to ON, JDS, Xorg, caiman... > > > > > > > etc every month, moreso than any other group, but choose not to > > > > > > > support IPS packages for ideological or technological reasons ( let's > > > > > > > assume for argument's sakes that they find the something about the > > > > > > > core fundamentals of IPS unworkable, more than simply iteratively > > > > > > > fixing could ever solve), choosing instead to support .ebuilds or > > > > > > > Bochnig's conary thing, you don't consider them part of the community > > > > > > > then? > > > > > > > > > > > > No, they are choosing to remove themselves from the ecosystem; not the > > > > > > T&B policy. > > > > > > > > > > > > Besides, choosing to not support the official package format wouldn't > > > > > > be very good for users so I would question how much they really want > > > > > > to support the community. > > > > > > > > > > Even if hypothetically the official package format sucks by design and > > > > > there's no real way to fix it except by throwing it out, except that > > > > > won't happen because of pigheadedness on the part of the official > > > > > package-format-choosing-group? Seems that it would help users to be > > > > > slightly incompatible there... API and ABI compatibility are far more > > > > > important in this case than package compatibility > > > > > > > > Are you really putting so little faith in the members (engineers) of > > > > this community? > > > > > > > > Because that's what your hypothetically saying. > > > > > > > > You're saying, "what if hypothetically speaking, all the engineers > > > > produce something that sucks by design and there's no way to fix it > > > > except by throwing it out." > > > > > > > > Creating hypothetical problems that are unlikely to exist will only > > > > cause us to precious time. > > > > > > > > If you really want to apply that hypothetical situation, then it could > > > > apply to *any* project here. ON, GNOME, Installer, etc. the list goes > > > > on. > > > > > > > > Where do you draw the hypothetical line for hypothetical problems to > > > > figure out hypothetical solutions? > > > > > > You are attacking me on the basis of whether I have faith in the > > > engineers or not, whereas I am trying to clarify what is meant by > > > certain terminology by using hypothetical scenarios. I have plenty of > > > faith in the engineers and I don't think this scenario is likely. > > > > No, I am not attacking you. I am providing critical input to your > > hypothetical situation. > > > > Regardless of how you want to view it, my statement and yours end up > > being fairly equivalent. > > > > If you believe that such a packaging system will be integrated in such > > a hypothetical state, then you are also saying you don't believe in > > members of this community to produce something good. > > > > > > > I would like to be more clear on the terminology that is being bandied > > > about as if everyone understands it and is on the same page and has > > > the same values and ideals, which may or may not be true, but the work > > > of a programmer, like the work of a trademark lawyer, is one of > > > precision, so it is important to be as precise as possible when > > > throwing terms around. > > > > I think statements that question the integrity of the community are > > unnecessary. You need to assume certain basic good about folks and > > deal with problems only when they arise. > > I do assume basic goods about the community, but if recent events are > any indication you must deal with potential problems well before they > arise, because if they're bad, they don't get dealt with... you just > get told to "get over it and move on" Then we'll never get anywhere. The list of potential problems is hypothetically endless. A certain element of trust is required for people to collaborate. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From John.Plocher at Sun.COM Fri Feb 22 12:15:31 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 12:15:31 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> Message-ID: <47BF2D63.8080107@Sun.Com> Justin Erenkrantz wrote: > I'd draw the line at binary and API compatibility - but > certainly packaging format isn't where I think the line should be. What kind of ecosystem do you have if there isn't compatibility at the packaging and application distribution level? You don't have an ecosystem - you simply have a set of distros that each require the ISVs to produce a slightly different product for each of them. In my mind, that's a failure. -John From brandorr at opensolaris.org Fri Feb 22 12:15:29 2008 From: brandorr at opensolaris.org (Brandorr) Date: Fri, 22 Feb 2008 15:15:29 -0500 Subject: [trademark-policy-dev] Brandorr: recompiling "OpenSolaris" from the sources? In-Reply-To: <47BF2928.6070707@Sun.Com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> Message-ID: <5b5090780802221215k4fa35802id7212c3f25e1bd44@mail.gmail.com> Thanks John.. It wasn't anything that you said that lead me to this. It was the Sun marketing collateral that is being handed out through the UG channels. (I can see if we can track it down, but it's not really that important, if you are going to clarify whether Sun built binaries are required for licensed use of the trademark). Cheers, Brian On Fri, Feb 22, 2008 at 2:57 PM, John Plocher wrote: > [moved to T&B alias only] > > Brandorr wrote: > > John, > > > > Try not to get confused. It doesn't matter. You can't "change" > > anything under the new policy. As I understand, you must use the Sun > > compiled binaries. If you build something from source and it barks the > > same and walks the same, it still can't be "OpenSolaris". The policy > > seems very clear. > > Where do you get this understanding from? While you and I had discussions > similar to what you assert several months ago, they were only and obviously > brainstorming ideas. > > I understand you are disappointed and bitter about many things; that is > no reason to represent opinions and ideas as facts. > > > Actually, I wonder. Is using Sun supplied binaries mandatory? Can > > individual community members compile an IPS distro from the > > opensolaris.org source tree > > Good question. I assume so, and the wording of the T&B policy > currently is directed at that goal. Nevertheless, I'll add your > question to the list I need to resolve. > > > ...and call it xyz OpenSolaris? (This last > > question is for the room.) > > The room does not need to answer - the lawyers already have: > http://www.sun.com/policies/trademarks/ > > Sun trademarks cannot be included in the name of any third party > product, technology, program or service. This includes free and > educational materials, open source distributions, and the titles > of informational web sites. > > This is simply a generalization of US and international trademark law; > you can not use other people's trademarks in the name of your product. > > -John > > -- - Brian Gupta http://opensolaris.org/os/project/nycosug/ http://www.genunix.org/wiki/index.php/OpenSolaris_New_User_FAQ From johnsonnenschein at gmail.com Fri Feb 22 13:00:17 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 16:00:17 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF3567.4090606@opensolaris.org> References: <47bdbd2a.3qu+nYoaaRBkRGtl%Joerg.Schilling@fokus.fraunhofer.de> <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF3567.4090606@opensolaris.org> Message-ID: <241540330802221300hc6d8e2fw95a00dec1b723718@mail.gmail.com> On Fri, Feb 22, 2008 at 3:49 PM, Garrett D'Amore wrote: > > John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > >> Joerg Schilling wrote: > >> > I have no problems is a distribution made from exactly the bits in > >> > "The OpenSolaris repository" (no less but definitely also no more) > >> > is called "OpenSolaris". > >> > >> I added this to the metric associated with "OpenSolaris": > >> > >> Distribution name: OpenSolaris > >> > >> * Sun has decided to fund Project Indiana to build a base distro, which it > >> will call OpenSolaris. > >> * Potential Metric: The various consolidations that make up the OpenSolaris > >> community will publish their {packages, bits, stuff} in an OpenSolaris > >> Repository, which will be used by Indiana to create the distro called > >> OpenSolaris. > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > >> repository" (no less but definitely also no more)? > >> o This would seem to require that either > >> A) Sun would need to not include things not on OS.o in their > >> product, or > >> B) Other distros could be supersets of the OS.o stuff and still > >> call themselves OpenSolaris. > >> > >> Does this capture your intent? > >> > > > > It certainly does not capture mine. The sticking point is the last > > one, that is, KIndiana is precluded from existing by virtue of the > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > for compatibility are built from a minimalist core ( ie, no GUI save > > for twm ), I don't see how a superset is workable > > > > A few thoughts. > > 1) This discussion should be removed from ogb-discuss. It has nothing > to do with Governance, I think. At least not until a specific proposal > is brought for a vote. You're right. moved to trademark-policy-dev > 2) As far as ISVs go, part of compatibility means making sure that core > libraries and infrastructure are present. A KDE-based distro, might > create problems if it does not also include the libraries that ISVs are > likely to have used from Gnome. (E.g., imagine if Opera uses GTK and > gnome libraries.... will their product work on your hypothetical > KIndiana?) So, from a strict compatibility point of view, at a > minimum, I think the programming APIs all need to be included. > Otherwise, "compatibility" is meaningless/useless (IMO). So then yes, KIndiana for all intents & purposes is impossible, unless it also includes all the libraries that Indiana proper does, no matter how irrelevant ( evolution-pim-libs ). It is essentially then just indiana with kde tacked on to it, rather than the Ubuntu model where the core is the same, but you can jettison unneeded desktop environments to make an ubuntu distribution ( KUbuntu does not include GNOME, but you can install it after the fact ) -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From Joerg.Schilling at fokus.fraunhofer.de Fri Feb 22 13:04:24 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 22 Feb 2008 22:04:24 +0100 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <241540330802221030v57b38798ocbc4888ac2e5b67b@mail.gmail.com> References: <47BDB577.6040203@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> <241540330802221030v57b38798ocbc4888ac2e5b67b@mail.gmail.com> Message-ID: <47bf38d8.Gp0zK9UCRcWJpXwh%Joerg.Schilling@fokus.fraunhofer.de> "John Sonnenschein" wrote: > > "It's true, OpenSolaris (Indiana) will replace Solaris Express Developer > > Edition after SXDE 1/08." -Ian > > SXDE has always been just secret code for "old version of SXCE" so > I'll shed no tears about that, but it would be tragic to see SXCE go Correct and it would be more than tragic as it would kill base for the current development. You still needs bits from SXCE to be able to compile ON. 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 johnsonnenschein at gmail.com Fri Feb 22 13:06:01 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 16:06:01 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221300hc6d8e2fw95a00dec1b723718@mail.gmail.com> References: <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF3567.4090606@opensolaris.org> <241540330802221300hc6d8e2fw95a00dec1b723718@mail.gmail.com> Message-ID: <241540330802221306i1f36a8f3w8036414d765a5bdd@mail.gmail.com> On Fri, Feb 22, 2008 at 4:00 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 3:49 PM, Garrett D'Amore > wrote: > > > > John Sonnenschein wrote: > > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > > > >> Joerg Schilling wrote: > > >> > I have no problems is a distribution made from exactly the bits in > > >> > "The OpenSolaris repository" (no less but definitely also no more) > > >> > is called "OpenSolaris". > > >> > > >> I added this to the metric associated with "OpenSolaris": > > >> > > >> Distribution name: OpenSolaris > > >> > > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > >> will call OpenSolaris. > > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > >> Repository, which will be used by Indiana to create the distro called > > >> OpenSolaris. > > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > >> repository" (no less but definitely also no more)? > > >> o This would seem to require that either > > >> A) Sun would need to not include things not on OS.o in their > > >> product, or > > >> B) Other distros could be supersets of the OS.o stuff and still > > >> call themselves OpenSolaris. > > >> > > >> Does this capture your intent? > > >> > > > > > > It certainly does not capture mine. The sticking point is the last > > > one, that is, KIndiana is precluded from existing by virtue of the > > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > > for compatibility are built from a minimalist core ( ie, no GUI save > > > for twm ), I don't see how a superset is workable > > > > > > > A few thoughts. > > > > 1) This discussion should be removed from ogb-discuss. It has nothing > > to do with Governance, I think. At least not until a specific proposal > > is brought for a vote. > > You're right. moved to trademark-policy-dev > > > > 2) As far as ISVs go, part of compatibility means making sure that core > > libraries and infrastructure are present. A KDE-based distro, might > > create problems if it does not also include the libraries that ISVs are > > likely to have used from Gnome. (E.g., imagine if Opera uses GTK and > > gnome libraries.... will their product work on your hypothetical > > KIndiana?) So, from a strict compatibility point of view, at a > > minimum, I think the programming APIs all need to be included. > > Otherwise, "compatibility" is meaningless/useless (IMO). > > So then yes, KIndiana for all intents & purposes is impossible, unless > it also includes all the libraries that Indiana proper does, no matter > how irrelevant ( evolution-pim-libs ). It is essentially then just > indiana with kde tacked on to it, rather than the Ubuntu model where > the core is the same, but you can jettison unneeded desktop > environments to make an ubuntu distribution ( KUbuntu does not include > GNOME, but you can install it after the fact ) Sorry for the double post, but to add to this, I don't personally see a problem with something resembling the current ARC model with different guarantees of stability ( in this case, compatibility as well ) for different portions, so long as those guarantees are posted somewhere visible. ie, an ISV that chooses to take advantage of gnome-libs is not guaranteed to work across all "- an OpenSolaris Distribution" distros, but other libs ( dbus ? ), they are -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 13:23:15 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 15:23:15 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221306i1f36a8f3w8036414d765a5bdd@mail.gmail.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF3567.4090606@opensolaris.org> <241540330802221300hc6d8e2fw95a00dec1b723718@mail.gmail.com> <241540330802221306i1f36a8f3w8036414d765a5bdd@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 3:06 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 4:00 PM, John Sonnenschein > wrote: > > > > On Fri, Feb 22, 2008 at 3:49 PM, Garrett D'Amore > > wrote: > > > > > > John Sonnenschein wrote: > > > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > > > > > >> Joerg Schilling wrote: > > > >> > I have no problems is a distribution made from exactly the bits in > > > >> > "The OpenSolaris repository" (no less but definitely also no more) > > > >> > is called "OpenSolaris". > > > >> > > > >> I added this to the metric associated with "OpenSolaris": > > > >> > > > >> Distribution name: OpenSolaris > > > >> > > > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > > >> will call OpenSolaris. > > > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > > >> Repository, which will be used by Indiana to create the distro called > > > >> OpenSolaris. > > > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > > >> repository" (no less but definitely also no more)? > > > >> o This would seem to require that either > > > >> A) Sun would need to not include things not on OS.o in their > > > >> product, or > > > >> B) Other distros could be supersets of the OS.o stuff and still > > > >> call themselves OpenSolaris. > > > >> > > > >> Does this capture your intent? > > > >> > > > > > > > > It certainly does not capture mine. The sticking point is the last > > > > one, that is, KIndiana is precluded from existing by virtue of the > > > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > > > for compatibility are built from a minimalist core ( ie, no GUI save > > > > for twm ), I don't see how a superset is workable > > > > > > > > > > A few thoughts. > > > > > > 1) This discussion should be removed from ogb-discuss. It has nothing > > > to do with Governance, I think. At least not until a specific proposal > > > is brought for a vote. > > > > You're right. moved to trademark-policy-dev > > > > > > > 2) As far as ISVs go, part of compatibility means making sure that core > > > libraries and infrastructure are present. A KDE-based distro, might > > > create problems if it does not also include the libraries that ISVs are > > > likely to have used from Gnome. (E.g., imagine if Opera uses GTK and > > > gnome libraries.... will their product work on your hypothetical > > > KIndiana?) So, from a strict compatibility point of view, at a > > > minimum, I think the programming APIs all need to be included. > > > Otherwise, "compatibility" is meaningless/useless (IMO). > > > > So then yes, KIndiana for all intents & purposes is impossible, unless > > it also includes all the libraries that Indiana proper does, no matter > > how irrelevant ( evolution-pim-libs ). It is essentially then just > > indiana with kde tacked on to it, rather than the Ubuntu model where > > the core is the same, but you can jettison unneeded desktop > > environments to make an ubuntu distribution ( KUbuntu does not include > > GNOME, but you can install it after the fact ) > > Sorry for the double post, but to add to this, I don't personally see > a problem with something resembling the current ARC model with > different guarantees of stability ( in this case, compatibility as > well ) for different portions, so long as those guarantees are posted > somewhere visible. ie, an ISV that chooses to take advantage of > gnome-libs is not guaranteed to work across all "- an OpenSolaris > Distribution" distros, but other libs ( dbus ? ), they are That's just it. Users want to be able to user their applications regardless of distribution; just like the mainstream operating systems. That's a big problem for the GNU/Linux world right now and something worth ensuring doesn't happen here. When an application tries to install that requires gnome-libs, I imagine the package system will kick in and install the proper libraries, so even if you are running KDE it will still work. Now what is debatable is whether those libraries or some portion of them always has to be present for compatibility. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 13:24:41 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 16:24:41 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF3567.4090606@opensolaris.org> <241540330802221300hc6d8e2fw95a00dec1b723718@mail.gmail.com> <241540330802221306i1f36a8f3w8036414d765a5bdd@mail.gmail.com> Message-ID: <241540330802221324t31fe9634h2dd50eeb2e759035@mail.gmail.com> On Fri, Feb 22, 2008 at 4:23 PM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 3:06 PM, John Sonnenschein > > > wrote: > > > > On Fri, Feb 22, 2008 at 4:00 PM, John Sonnenschein > > wrote: > > > > > > On Fri, Feb 22, 2008 at 3:49 PM, Garrett D'Amore > > > wrote: > > > > > > > > John Sonnenschein wrote: > > > > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > > > > > > > >> Joerg Schilling wrote: > > > > >> > I have no problems is a distribution made from exactly the bits in > > > > >> > "The OpenSolaris repository" (no less but definitely also no more) > > > > >> > is called "OpenSolaris". > > > > >> > > > > >> I added this to the metric associated with "OpenSolaris": > > > > >> > > > > >> Distribution name: OpenSolaris > > > > >> > > > > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > > > >> will call OpenSolaris. > > > > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > > > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > > > >> Repository, which will be used by Indiana to create the distro called > > > > >> OpenSolaris. > > > > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > > > >> repository" (no less but definitely also no more)? > > > > >> o This would seem to require that either > > > > >> A) Sun would need to not include things not on OS.o in their > > > > >> product, or > > > > >> B) Other distros could be supersets of the OS.o stuff and still > > > > >> call themselves OpenSolaris. > > > > >> > > > > >> Does this capture your intent? > > > > >> > > > > > > > > > > It certainly does not capture mine. The sticking point is the last > > > > > one, that is, KIndiana is precluded from existing by virtue of the > > > > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > > > > for compatibility are built from a minimalist core ( ie, no GUI save > > > > > for twm ), I don't see how a superset is workable > > > > > > > > > > > > > A few thoughts. > > > > > > > > 1) This discussion should be removed from ogb-discuss. It has nothing > > > > to do with Governance, I think. At least not until a specific proposal > > > > is brought for a vote. > > > > > > You're right. moved to trademark-policy-dev > > > > > > > > > > 2) As far as ISVs go, part of compatibility means making sure that core > > > > libraries and infrastructure are present. A KDE-based distro, might > > > > create problems if it does not also include the libraries that ISVs are > > > > likely to have used from Gnome. (E.g., imagine if Opera uses GTK and > > > > gnome libraries.... will their product work on your hypothetical > > > > KIndiana?) So, from a strict compatibility point of view, at a > > > > minimum, I think the programming APIs all need to be included. > > > > Otherwise, "compatibility" is meaningless/useless (IMO). > > > > > > So then yes, KIndiana for all intents & purposes is impossible, unless > > > it also includes all the libraries that Indiana proper does, no matter > > > how irrelevant ( evolution-pim-libs ). It is essentially then just > > > indiana with kde tacked on to it, rather than the Ubuntu model where > > > the core is the same, but you can jettison unneeded desktop > > > environments to make an ubuntu distribution ( KUbuntu does not include > > > GNOME, but you can install it after the fact ) > > > > Sorry for the double post, but to add to this, I don't personally see > > a problem with something resembling the current ARC model with > > different guarantees of stability ( in this case, compatibility as > > well ) for different portions, so long as those guarantees are posted > > somewhere visible. ie, an ISV that chooses to take advantage of > > gnome-libs is not guaranteed to work across all "- an OpenSolaris > > Distribution" distros, but other libs ( dbus ? ), they are > > That's just it. Users want to be able to user their applications > regardless of distribution; just like the mainstream operating > systems. > > That's a big problem for the GNU/Linux world right now and something > worth ensuring doesn't happen here. > > When an application tries to install that requires gnome-libs, I > imagine the package system will kick in and install the proper > libraries, so even if you are running KDE it will still work. That is something that I think would be okay. -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 13:24:26 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 15:24:26 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <47BF2D63.8080107@Sun.Com> References: <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> Message-ID: On Fri, Feb 22, 2008 at 2:15 PM, John Plocher wrote: > Justin Erenkrantz wrote: > > I'd draw the line at binary and API compatibility - but > > certainly packaging format isn't where I think the line should be. > > > What kind of ecosystem do you have if there isn't compatibility at > the packaging and application distribution level? > > You don't have an ecosystem - you simply have a set of distros > that each require the ISVs to produce a slightly different > product for each of them. In my mind, that's a failure. ...and quite similar to the problem that exists in the GNU/Linux world today. I don't want to have to package my software (as an ISV) in five different flavours. I want to be able to produce them in one format and know that on all "compatible" distributions it should work. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From justin at erenkrantz.com Fri Feb 22 13:52:34 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Fri, 22 Feb 2008 13:52:34 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <47BF2D63.8080107@Sun.Com> References: <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> Message-ID: <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> On Fri, Feb 22, 2008 at 12:15 PM, John Plocher wrote: > What kind of ecosystem do you have if there isn't compatibility at > the packaging and application distribution level? But, why use the trademark policy to enforce what should be a technical decision? It's quite clear that this community has existed for a long time without a clear packaging and application distribution system - other than pkg which relatively few third-party Solaris application use anyway. > You don't have an ecosystem - you simply have a set of distros > that each require the ISVs to produce a slightly different > product for each of them. In my mind, that's a failure. There are plenty of tools out there (such as alien) that translate between package formats - that's a technical problem and one that is clearly addressable. There's no doubt that there is a Linux ecosystem even though there are different packaging formats. However, forbidding a third-party distribution any access to the OpenSolaris name because they believe that another packaging system is superior is a dubious position. I believe this is legislating a technical decision by using trademark law to win the argument by fiat. If IPS is technically superior, it'll obviously win out - but if the OpenSolaris community precludes any other experimentation, we'll simply never know that. Mac OS X happily co-exists with Fink and MacPorts. Why should OpenSolaris be different? It's quite clear that Indiana is experimenting with different strategies (such as removing compatibility with Solaris scripts because of altered PATH and shells) - why exclude others that same right? But, furthermore and more importantly, doing so in the trademark policy seems short-sighted. -- justin From swalker at opensolaris.org Fri Feb 22 14:03:20 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 16:03:20 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> References: <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 3:52 PM, Justin Erenkrantz wrote: > On Fri, Feb 22, 2008 at 12:15 PM, John Plocher wrote: > > What kind of ecosystem do you have if there isn't compatibility at > > the packaging and application distribution level? > > But, why use the trademark policy to enforce what should be a > technical decision? It's quite clear that this community has existed > for a long time without a clear packaging and application distribution > system - other than pkg which relatively few third-party Solaris > application use anyway. Because the name implies something. That's why. As for the requirement existing previously; the right to use the name in the proposed ways didn't exist previously. As such, these new requirements grant *additional* privileges that were not available before. When users see a specific moniker that implies compatibility by use of the trademark; they will be able to have some confidence that their applications, etc. will "just work." > > You don't have an ecosystem - you simply have a set of distros > > that each require the ISVs to produce a slightly different > > product for each of them. In my mind, that's a failure. > > There are plenty of tools out there (such as alien) that translate > between package formats - that's a technical problem and one that is > clearly addressable. There's no doubt that there is a Linux ecosystem > even though there are different packaging formats. However, > forbidding a third-party distribution any access to the OpenSolaris > name because they believe that another packaging system is superior is > a dubious position. I believe this is legislating a technical > decision by using trademark law to win the argument by fiat. If IPS > is technically superior, it'll obviously win out - but if the > OpenSolaris community precludes any other experimentation, we'll > simply never know that. Translating package formats is not good enough and if you use the packages to help enforce compatibility you must have packaging be part of that. > Mac OS X happily co-exists with Fink and MacPorts. Why should > OpenSolaris be different? It's quite clear that Indiana is Because Mac OS X isn't OpenSolaris/Solaris? One could argue why not do things like Windows, or IRIX, or a host of other operating systems. Mac OS X users have a very different set of needs than OpenSolaris/Solaris users. > - why exclude others that same right? But, furthermore and more > importantly, doing so in the trademark policy seems short-sighted. -- Because that right breaks the user's right to use their system peaceably. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 14:11:17 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 17:11:17 -0500 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: References: <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> Message-ID: <241540330802221411w2dcd93aen1e73aeaf382982c1@mail.gmail.com> On Fri, Feb 22, 2008 at 5:03 PM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 3:52 PM, Justin Erenkrantz > wrote: > > On Fri, Feb 22, 2008 at 12:15 PM, John Plocher wrote: > > > What kind of ecosystem do you have if there isn't compatibility at > > > the packaging and application distribution level? > > > > But, why use the trademark policy to enforce what should be a > > technical decision? It's quite clear that this community has existed > > for a long time without a clear packaging and application distribution > > system - other than pkg which relatively few third-party Solaris > > application use anyway. > > Because the name implies something. That's why. But in other discussions you claim it's "just a name"... > As for the requirement existing previously; the right to use the name > in the proposed ways didn't exist previously. > > As such, these new requirements grant *additional* privileges that > were not available before. > > When users see a specific moniker that implies compatibility by use of > the trademark; they will be able to have some confidence that their > applications, etc. will "just work." > > > > > You don't have an ecosystem - you simply have a set of distros > > > that each require the ISVs to produce a slightly different > > > product for each of them. In my mind, that's a failure. > > > > There are plenty of tools out there (such as alien) that translate > > between package formats - that's a technical problem and one that is > > clearly addressable. There's no doubt that there is a Linux ecosystem > > even though there are different packaging formats. However, > > forbidding a third-party distribution any access to the OpenSolaris > > name because they believe that another packaging system is superior is > > a dubious position. I believe this is legislating a technical > > decision by using trademark law to win the argument by fiat. If IPS > > is technically superior, it'll obviously win out - but if the > > OpenSolaris community precludes any other experimentation, we'll > > simply never know that. > > Translating package formats is not good enough and if you use the > packages to help enforce compatibility you must have packaging be part > of that. But why should you use the packages to enforce compatibility? Compatibility != sameness . Sameness = compatibility, but it's a non-reflexive relation. > > > Mac OS X happily co-exists with Fink and MacPorts. Why should > > OpenSolaris be different? It's quite clear that Indiana is > > Because Mac OS X isn't OpenSolaris/Solaris? > > One could argue why not do things like Windows, or IRIX, or a host of > other operating systems. or like linux... *cough* $PATH *cough* > Mac OS X users have a very different set of needs than > OpenSolaris/Solaris users. > > > > - why exclude others that same right? But, furthermore and more > > importantly, doing so in the trademark policy seems short-sighted. -- > > Because that right breaks the user's right to use their system peaceably. so the ends justify the means... -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 14:28:51 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 16:28:51 -0600 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <241540330802221411w2dcd93aen1e73aeaf382982c1@mail.gmail.com> References: <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> <241540330802221411w2dcd93aen1e73aeaf382982c1@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 4:11 PM, John Sonnenschein wrote: > On Fri, Feb 22, 2008 at 5:03 PM, Shawn Walker wrote: > > On Fri, Feb 22, 2008 at 3:52 PM, Justin Erenkrantz > > wrote: > > > On Fri, Feb 22, 2008 at 12:15 PM, John Plocher wrote: > > > > What kind of ecosystem do you have if there isn't compatibility at > > > > the packaging and application distribution level? > > > > > > But, why use the trademark policy to enforce what should be a > > > technical decision? It's quite clear that this community has existed > > > for a long time without a clear packaging and application distribution > > > system - other than pkg which relatively few third-party Solaris > > > application use anyway. > > > > Because the name implies something. That's why. > > But in other discussions you claim it's "just a name"... Now you're omitting the context of what I said which gives it meaning; without that context it's meaningless. The context was whether it was worth the current flame wars, provocations, etc. It is not. This is about compatibility and trademarks in general which is an entirely different thing. > > As for the requirement existing previously; the right to use the name > > in the proposed ways didn't exist previously. > > > > As such, these new requirements grant *additional* privileges that > > were not available before. > > > > When users see a specific moniker that implies compatibility by use of > > the trademark; they will be able to have some confidence that their > > applications, etc. will "just work." > > > > > > > > You don't have an ecosystem - you simply have a set of distros > > > > that each require the ISVs to produce a slightly different > > > > product for each of them. In my mind, that's a failure. > > > > > > There are plenty of tools out there (such as alien) that translate > > > between package formats - that's a technical problem and one that is > > > clearly addressable. There's no doubt that there is a Linux ecosystem > > > even though there are different packaging formats. However, > > > forbidding a third-party distribution any access to the OpenSolaris > > > name because they believe that another packaging system is superior is > > > a dubious position. I believe this is legislating a technical > > > decision by using trademark law to win the argument by fiat. If IPS > > > is technically superior, it'll obviously win out - but if the > > > OpenSolaris community precludes any other experimentation, we'll > > > simply never know that. > > > > Translating package formats is not good enough and if you use the > > packages to help enforce compatibility you must have packaging be part > > of that. > > But why should you use the packages to enforce compatibility? > Compatibility != sameness . Sameness = compatibility, but it's a > non-reflexive relation. Until we have a gigantic test suite, that's the only way we can truly guarantee compatibility. Just saying that "it should be compatible" isn't good enough. > > > Mac OS X happily co-exists with Fink and MacPorts. Why should > > > OpenSolaris be different? It's quite clear that Indiana is > > > > Because Mac OS X isn't OpenSolaris/Solaris? > > > > One could argue why not do things like Windows, or IRIX, or a host of > > other operating systems. > > or like linux... *cough* $PATH *cough* We're talking about ABI and packaging; not paths. Let's take the best things, not the worst :) > > Mac OS X users have a very different set of needs than > > OpenSolaris/Solaris users. > > > > > > > - why exclude others that same right? But, furthermore and more > > > importantly, doing so in the trademark policy seems short-sighted. -- > > > > Because that right breaks the user's right to use their system peaceably. > > so the ends justify the means... So slay me for caring more about users than about egos. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 14:51:12 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 17:51:12 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> Message-ID: <241540330802221451k1c57a3a6x3867fcfef91d2651@mail.gmail.com> On Fri, Feb 22, 2008 at 5:47 PM, Mark Martin wrote: > On Fri, Feb 22, 2008 at 4:14 PM, John Plocher wrote: > > > > Maybe we really do end up with > > > > OpenSolaris (the minimal core thing) > > > > OpenSolaris by Sun > > > > OpenSolaris with KDE > > > > OpenSolaris for Dummies > > > > and > > Solaris 11 - based on OpenSolaris > > > > And Plocher Model Train Appliance 2009 - based on OpenSolaris yes, well... this much was implied already > (which is possibly NOT based on the minimal core thing). > > I'm keeping in mind the Polaris effort and recent discussions around a > set-top box "sized" distro. All the more reason to define the minimalist core + repo (by the way, most of the intent of this reply was to try to take it off ogb-discuss and on to trademark-policy-dev ) -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 14:53:51 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 16:53:51 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> Message-ID: [moving to trademark-policy-dev] On Fri, Feb 22, 2008 at 4:45 PM, John Sonnenschein wrote: > On Fri, Feb 22, 2008 at 5:37 PM, Shawn Walker wrote: > > > > On Fri, Feb 22, 2008 at 4:14 PM, John Plocher wrote: > > > John Sonnenschein wrote: > > > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > > > > >> Distribution name: OpenSolaris > > > >> > > > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > > >> will call OpenSolaris. > > > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > > >> Repository, which will be used by Indiana to create the distro called > > > >> OpenSolaris. > > > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > > >> repository" (no less but definitely also no more)? > > > >> o This would seem to require that either > > > >> A) Sun would need to not include things not on OS.o in their > > > >> product, or > > > >> B) Other distros could be supersets of the OS.o stuff and still > > > >> call themselves OpenSolaris. > > > >> > > > >> Does this capture your intent? > > > > > > > > It certainly does not capture mine. The sticking point is the last > > > > one, that is, KIndiana is precluded from existing by virtue of the > > > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > > > for compatibility are built from a minimalist core ( ie, no GUI save > > > > for twm ), I don't see how a superset is workable > > > > > > > > > > > > > Thanks - this actually helps clarify something for me :-) > > > > > > "It Seams To Me" that the things that are produced by Indiana > > > really need to be that minimalist core (i.e., enough to > > > boot into an installer that is network aware in whatever > > > locale the user needs) AND a set of repositories where > > > the various packages are to be found. That minimalist core > > > is then called OpenSolaris. > > > > > > The branding for OpenSolaris would then need to allow for > > > superset distros, so that Sun can create a distro that has > > > GNOME and NetBeans and Java, and you can create a distro that > > > has KDE instead. > > > > > > On a technical level, the only difference between the two distros > > > would seem to be A) the recipe of packages to be installed > > > by default, and B) the set of packages pre-cached on the > > > distribution media (so that the default recipe can be installed > > > without requiring network access...) > > > > > > As Shawn said, because of IPS/Repos, the ecosystem is > > > available for getting new applications AND for satisfying > > > their install-time dependencies. This would mean that any > > > OpenSolaris-core containing distro could ultimately run any > > > Built-for-OpenSolaris application. > > > > Putting on my ISV hat for a moment; I think you have to draw the line somewhere. > > > > That is, as an application developer, I want to be able to distribute > > my applications for any OpenSolaris user without triggering a > > multi-gigabyte download. > > > > That means that the core should not be so minimalist that to use > > almost any basic application that most of a repository needs to be > > installed. > > > > For example, I distribute an Adventure Game Runtime engine for Ubuntu > > users that relies on all of the libraries being fetched from apt-get, > > etc. > > > > Most users were happy because they either: > > > > 1) already had those libraries installed > > > > 2) weren't bothered by extra downloads from the repository > > > > However, some users were bothered at having to download a lot more, > > since the < 1mb download suddenly turned into > 10mb. > > > > My gut tells me that you have to pick a desktop and a set of libraries > > to part of core compatibility and that those should be installed by > > default. > > > > OS X, Windows, etc. are successful, in part, precisely because > > developers can depend upon a rich set of APIs always being available > > on a user's system. > > > > On OS X, for example, you can always depend on the "Core" libraries > > being present for audio, graphics, etc.. > > > > On Windows, you can always count on GDI, DirectDraw, DirectSound, and > > international text services > > > > I guess what I'm getting at is that developers should be able to > > distribute their applications in such a manner that if they stick with > > a "core set" of dependencies, the user won't have to download anything > > additional and they will get a fully functional application. > > > > That is my ultimate personal goal in this compatibility thing. > > > > Does that mean that certain libraries and environments might get > > favoured somewhat over another? You're darn right. But that is a > > benefit to users and ISVs. > > > > Cheers, > > And then you have to pick a desktop, and then you have to pick some > RDBMSes and then you pick... blah blah, before you know it we've got a > 3-DVD pile of software as the core distro. I didn't say everything but the kitchen sink. For example, standardise on SDL for a basic graphics, input, etc. library Standardise on OpenAL for 3D Audio Standardise on OSS for sound Obviously there is room for compromise here. For example, we could standardise on requiring Qt and Gtk to both be installed since both are desktop-agnostic. > It is up to the distribution maintainer to choose a rich set of > software to include in order that his distro doesn't feel like "the > one where installing basic software takes a long time". As far as > compatibility in terms of the trademark goes, however, the fact that > they can pull from the same repo is sufficient IMO. > > Once again, +1 to Plocher's initial idea. I cannot support such an Idea; and such an idea would be useless to most ISVs in my experience. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 14:59:32 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 17:59:32 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> Message-ID: <241540330802221459v360229fi10af6bf3b2f98295@mail.gmail.com> On Fri, Feb 22, 2008 at 5:53 PM, Shawn Walker wrote: > [moving to trademark-policy-dev] > > > > On Fri, Feb 22, 2008 at 4:45 PM, John Sonnenschein > wrote: > > On Fri, Feb 22, 2008 at 5:37 PM, Shawn Walker wrote: > > > > > > On Fri, Feb 22, 2008 at 4:14 PM, John Plocher wrote: > > > > John Sonnenschein wrote: > > > > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > > > > > > >> Distribution name: OpenSolaris > > > > >> > > > > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > > > >> will call OpenSolaris. > > > > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > > > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > > > >> Repository, which will be used by Indiana to create the distro called > > > > >> OpenSolaris. > > > > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > > > >> repository" (no less but definitely also no more)? > > > > >> o This would seem to require that either > > > > >> A) Sun would need to not include things not on OS.o in their > > > > >> product, or > > > > >> B) Other distros could be supersets of the OS.o stuff and still > > > > >> call themselves OpenSolaris. > > > > >> > > > > >> Does this capture your intent? > > > > > > > > > > It certainly does not capture mine. The sticking point is the last > > > > > one, that is, KIndiana is precluded from existing by virtue of the > > > > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > > > > for compatibility are built from a minimalist core ( ie, no GUI save > > > > > for twm ), I don't see how a superset is workable > > > > > > > > > > > > > > > > > Thanks - this actually helps clarify something for me :-) > > > > > > > > "It Seams To Me" that the things that are produced by Indiana > > > > really need to be that minimalist core (i.e., enough to > > > > boot into an installer that is network aware in whatever > > > > locale the user needs) AND a set of repositories where > > > > the various packages are to be found. That minimalist core > > > > is then called OpenSolaris. > > > > > > > > The branding for OpenSolaris would then need to allow for > > > > superset distros, so that Sun can create a distro that has > > > > GNOME and NetBeans and Java, and you can create a distro that > > > > has KDE instead. > > > > > > > > On a technical level, the only difference between the two distros > > > > would seem to be A) the recipe of packages to be installed > > > > by default, and B) the set of packages pre-cached on the > > > > distribution media (so that the default recipe can be installed > > > > without requiring network access...) > > > > > > > > As Shawn said, because of IPS/Repos, the ecosystem is > > > > available for getting new applications AND for satisfying > > > > their install-time dependencies. This would mean that any > > > > OpenSolaris-core containing distro could ultimately run any > > > > Built-for-OpenSolaris application. > > > > > > Putting on my ISV hat for a moment; I think you have to draw the line somewhere. > > > > > > That is, as an application developer, I want to be able to distribute > > > my applications for any OpenSolaris user without triggering a > > > multi-gigabyte download. > > > > > > That means that the core should not be so minimalist that to use > > > almost any basic application that most of a repository needs to be > > > installed. > > > > > > For example, I distribute an Adventure Game Runtime engine for Ubuntu > > > users that relies on all of the libraries being fetched from apt-get, > > > etc. > > > > > > Most users were happy because they either: > > > > > > 1) already had those libraries installed > > > > > > 2) weren't bothered by extra downloads from the repository > > > > > > However, some users were bothered at having to download a lot more, > > > since the < 1mb download suddenly turned into > 10mb. > > > > > > My gut tells me that you have to pick a desktop and a set of libraries > > > to part of core compatibility and that those should be installed by > > > default. > > > > > > OS X, Windows, etc. are successful, in part, precisely because > > > developers can depend upon a rich set of APIs always being available > > > on a user's system. > > > > > > On OS X, for example, you can always depend on the "Core" libraries > > > being present for audio, graphics, etc.. > > > > > > On Windows, you can always count on GDI, DirectDraw, DirectSound, and > > > international text services > > > > > > I guess what I'm getting at is that developers should be able to > > > distribute their applications in such a manner that if they stick with > > > a "core set" of dependencies, the user won't have to download anything > > > additional and they will get a fully functional application. > > > > > > That is my ultimate personal goal in this compatibility thing. > > > > > > Does that mean that certain libraries and environments might get > > > favoured somewhat over another? You're darn right. But that is a > > > benefit to users and ISVs. > > > > > > Cheers, > > > > And then you have to pick a desktop, and then you have to pick some > > RDBMSes and then you pick... blah blah, before you know it we've got a > > 3-DVD pile of software as the core distro. > > I didn't say everything but the kitchen sink. > > For example, standardise on SDL for a basic graphics, input, etc. library > > Standardise on OpenAL for 3D Audio > > Standardise on OSS for sound > > Obviously there is room for compromise here. > > For example, we could standardise on requiring Qt and Gtk to both be > installed since both are desktop-agnostic. more shovelware... > > It is up to the distribution maintainer to choose a rich set of > > software to include in order that his distro doesn't feel like "the > > one where installing basic software takes a long time". As far as > > compatibility in terms of the trademark goes, however, the fact that > > they can pull from the same repo is sufficient IMO. > > > > Once again, +1 to Plocher's initial idea. > > I cannot support such an Idea; and such an idea would be useless to > most ISVs in my experience. Why? it will not break their program, it may just take longer to install on minimalist distros. I think that's an acceptable tradeoff. ISV's are happy ( their program runs everywhere ), distro maintainers are happy ( they don't have GNOME shoved down their throats ) -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Fri Feb 22 15:04:19 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 17:04:19 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <5c902b9e0802221448u7156d3daw4b583ac89231c7fc@mail.gmail.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <5c902b9e0802221448u7156d3daw4b583ac89231c7fc@mail.gmail.com> Message-ID: [moving to trademark-policy-dev] On Fri, Feb 22, 2008 at 4:48 PM, Justin Erenkrantz wrote: > On Fri, Feb 22, 2008 at 2:37 PM, Shawn Walker wrote: > > > Does that mean that certain libraries and environments might get > > favoured somewhat over another? You're darn right. But that is a > > benefit to users and ISVs. > > I think this level of micromanagement is dangerous as it borders on > practically forbidding any OpenSolaris-compatible distributions except > Indiana. That may actually be what the community wants, but then > let's not pretend that we encourage third-party distributions nor do > we encourage any real experimentation in the overall distribution. > Note that Indiana continues to portray itself as a grand experiment - > but I fear we're going to find ourselves painted in a corner if no one > else can try the exact same things Indiana is doing right now. As a > hypothetical, if we treated Solaris 10 as the 'base', would your > vision permit Indiana to be doing the things it's doing? I have a > very hard time believing that experiments such as Indiana would fit in > with your world view. -- justin Sorry, but I can't agree at all. I also don't see it as "micromanagement." This is basic OS platform 101. Every other OS does it; even the LSB for Linux establishes certain libraries, APIs, etc. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Fri Feb 22 15:05:47 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 17:05:47 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221459v360229fi10af6bf3b2f98295@mail.gmail.com> References: <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <241540330802221459v360229fi10af6bf3b2f98295@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 4:59 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 5:53 PM, Shawn Walker wrote: > > [moving to trademark-policy-dev] > > > > > > > > On Fri, Feb 22, 2008 at 4:45 PM, John Sonnenschein > > wrote: > > > On Fri, Feb 22, 2008 at 5:37 PM, Shawn Walker wrote: > > > > > > > > On Fri, Feb 22, 2008 at 4:14 PM, John Plocher wrote: > > > > > John Sonnenschein wrote: > > > > > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > > > > > > > > >> Distribution name: OpenSolaris > > > > > >> > > > > > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > > > > >> will call OpenSolaris. > > > > > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > > > > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > > > > >> Repository, which will be used by Indiana to create the distro called > > > > > >> OpenSolaris. > > > > > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > > > > >> repository" (no less but definitely also no more)? > > > > > >> o This would seem to require that either > > > > > >> A) Sun would need to not include things not on OS.o in their > > > > > >> product, or > > > > > >> B) Other distros could be supersets of the OS.o stuff and still > > > > > >> call themselves OpenSolaris. > > > > > >> > > > > > >> Does this capture your intent? > > > > > > > > > > > > It certainly does not capture mine. The sticking point is the last > > > > > > one, that is, KIndiana is precluded from existing by virtue of the > > > > > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > > > > > for compatibility are built from a minimalist core ( ie, no GUI save > > > > > > for twm ), I don't see how a superset is workable > > > > > > > > > > > > > > > > > > > > > Thanks - this actually helps clarify something for me :-) > > > > > > > > > > "It Seams To Me" that the things that are produced by Indiana > > > > > really need to be that minimalist core (i.e., enough to > > > > > boot into an installer that is network aware in whatever > > > > > locale the user needs) AND a set of repositories where > > > > > the various packages are to be found. That minimalist core > > > > > is then called OpenSolaris. > > > > > > > > > > The branding for OpenSolaris would then need to allow for > > > > > superset distros, so that Sun can create a distro that has > > > > > GNOME and NetBeans and Java, and you can create a distro that > > > > > has KDE instead. > > > > > > > > > > On a technical level, the only difference between the two distros > > > > > would seem to be A) the recipe of packages to be installed > > > > > by default, and B) the set of packages pre-cached on the > > > > > distribution media (so that the default recipe can be installed > > > > > without requiring network access...) > > > > > > > > > > As Shawn said, because of IPS/Repos, the ecosystem is > > > > > available for getting new applications AND for satisfying > > > > > their install-time dependencies. This would mean that any > > > > > OpenSolaris-core containing distro could ultimately run any > > > > > Built-for-OpenSolaris application. > > > > > > > > Putting on my ISV hat for a moment; I think you have to draw the line somewhere. > > > > > > > > That is, as an application developer, I want to be able to distribute > > > > my applications for any OpenSolaris user without triggering a > > > > multi-gigabyte download. > > > > > > > > That means that the core should not be so minimalist that to use > > > > almost any basic application that most of a repository needs to be > > > > installed. > > > > > > > > For example, I distribute an Adventure Game Runtime engine for Ubuntu > > > > users that relies on all of the libraries being fetched from apt-get, > > > > etc. > > > > > > > > Most users were happy because they either: > > > > > > > > 1) already had those libraries installed > > > > > > > > 2) weren't bothered by extra downloads from the repository > > > > > > > > However, some users were bothered at having to download a lot more, > > > > since the < 1mb download suddenly turned into > 10mb. > > > > > > > > My gut tells me that you have to pick a desktop and a set of libraries > > > > to part of core compatibility and that those should be installed by > > > > default. > > > > > > > > OS X, Windows, etc. are successful, in part, precisely because > > > > developers can depend upon a rich set of APIs always being available > > > > on a user's system. > > > > > > > > On OS X, for example, you can always depend on the "Core" libraries > > > > being present for audio, graphics, etc.. > > > > > > > > On Windows, you can always count on GDI, DirectDraw, DirectSound, and > > > > international text services > > > > > > > > I guess what I'm getting at is that developers should be able to > > > > distribute their applications in such a manner that if they stick with > > > > a "core set" of dependencies, the user won't have to download anything > > > > additional and they will get a fully functional application. > > > > > > > > That is my ultimate personal goal in this compatibility thing. > > > > > > > > Does that mean that certain libraries and environments might get > > > > favoured somewhat over another? You're darn right. But that is a > > > > benefit to users and ISVs. > > > > > > > > Cheers, > > > > > > And then you have to pick a desktop, and then you have to pick some > > > RDBMSes and then you pick... blah blah, before you know it we've got a > > > 3-DVD pile of software as the core distro. > > > > I didn't say everything but the kitchen sink. > > > > For example, standardise on SDL for a basic graphics, input, etc. library > > > > Standardise on OpenAL for 3D Audio > > > > Standardise on OSS for sound > > > > Obviously there is room for compromise here. > > > > For example, we could standardise on requiring Qt and Gtk to both be > > installed since both are desktop-agnostic. > > more shovelware... Explain. I'm not familiar with this apparently derogatory term. > > > It is up to the distribution maintainer to choose a rich set of > > > software to include in order that his distro doesn't feel like "the > > > one where installing basic software takes a long time". As far as > > > compatibility in terms of the trademark goes, however, the fact that > > > they can pull from the same repo is sufficient IMO. > > > > > > Once again, +1 to Plocher's initial idea. > > > > I cannot support such an Idea; and such an idea would be useless to > > most ISVs in my experience. > > Why? it will not break their program, it may just take longer to > install on minimalist distros. I think that's an acceptable tradeoff. > ISV's are happy ( their program runs everywhere ), distro maintainers > are happy ( they don't have GNOME shoved down their throats ) Why not? To provide a better user experience. That's why. An integrated system always provides a better system than just a bunch of pieces that later are put together. An integrated system is why Solaris is a better OS than GNU/Linux IMO. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Fri Feb 22 15:07:09 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Fri, 22 Feb 2008 17:07:09 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF5438.6090102@opensolaris.org> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> Message-ID: [moved to trademark-policy-dev] On Fri, Feb 22, 2008 at 5:01 PM, Garrett D'Amore wrote: > > John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 5:37 PM, Shawn Walker wrote: > > > >> On Fri, Feb 22, 2008 at 4:14 PM, John Plocher wrote: > >> > John Sonnenschein wrote: > >> > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > >> > > >> > >> Distribution name: OpenSolaris > >> > >> > >> > >> * Sun has decided to fund Project Indiana to build a base distro, which it > >> > >> will call OpenSolaris. > >> > >> * Potential Metric: The various consolidations that make up the OpenSolaris > >> > >> community will publish their {packages, bits, stuff} in an OpenSolaris > >> > >> Repository, which will be used by Indiana to create the distro called > >> > >> OpenSolaris. > >> > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > >> > >> repository" (no less but definitely also no more)? > >> > >> o This would seem to require that either > >> > >> A) Sun would need to not include things not on OS.o in their > >> > >> product, or > >> > >> B) Other distros could be supersets of the OS.o stuff and still > >> > >> call themselves OpenSolaris. > >> > >> > >> > >> Does this capture your intent? > >> > > > >> > > It certainly does not capture mine. The sticking point is the last > >> > > one, that is, KIndiana is precluded from existing by virtue of the > >> > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > >> > > for compatibility are built from a minimalist core ( ie, no GUI save > >> > > for twm ), I don't see how a superset is workable > >> > > > >> > > >> > > >> > Thanks - this actually helps clarify something for me :-) > >> > > >> > "It Seams To Me" that the things that are produced by Indiana > >> > really need to be that minimalist core (i.e., enough to > >> > boot into an installer that is network aware in whatever > >> > locale the user needs) AND a set of repositories where > >> > the various packages are to be found. That minimalist core > >> > is then called OpenSolaris. > >> > > >> > The branding for OpenSolaris would then need to allow for > >> > superset distros, so that Sun can create a distro that has > >> > GNOME and NetBeans and Java, and you can create a distro that > >> > has KDE instead. > >> > > >> > On a technical level, the only difference between the two distros > >> > would seem to be A) the recipe of packages to be installed > >> > by default, and B) the set of packages pre-cached on the > >> > distribution media (so that the default recipe can be installed > >> > without requiring network access...) > >> > > >> > As Shawn said, because of IPS/Repos, the ecosystem is > >> > available for getting new applications AND for satisfying > >> > their install-time dependencies. This would mean that any > >> > OpenSolaris-core containing distro could ultimately run any > >> > Built-for-OpenSolaris application. > >> > >> Putting on my ISV hat for a moment; I think you have to draw the line somewhere. > >> > >> That is, as an application developer, I want to be able to distribute > >> my applications for any OpenSolaris user without triggering a > >> multi-gigabyte download. > >> > >> That means that the core should not be so minimalist that to use > >> almost any basic application that most of a repository needs to be > >> installed. > >> > >> For example, I distribute an Adventure Game Runtime engine for Ubuntu > >> users that relies on all of the libraries being fetched from apt-get, > >> etc. > >> > >> Most users were happy because they either: > >> > >> 1) already had those libraries installed > >> > >> 2) weren't bothered by extra downloads from the repository > >> > >> However, some users were bothered at having to download a lot more, > >> since the < 1mb download suddenly turned into > 10mb. > >> > >> My gut tells me that you have to pick a desktop and a set of libraries > >> to part of core compatibility and that those should be installed by > >> default. > >> > >> OS X, Windows, etc. are successful, in part, precisely because > >> developers can depend upon a rich set of APIs always being available > >> on a user's system. > >> > >> On OS X, for example, you can always depend on the "Core" libraries > >> being present for audio, graphics, etc.. > >> > >> On Windows, you can always count on GDI, DirectDraw, DirectSound, and > >> international text services > >> > >> I guess what I'm getting at is that developers should be able to > >> distribute their applications in such a manner that if they stick with > >> a "core set" of dependencies, the user won't have to download anything > >> additional and they will get a fully functional application. > >> > >> That is my ultimate personal goal in this compatibility thing. > >> > >> Does that mean that certain libraries and environments might get > >> favoured somewhat over another? You're darn right. But that is a > >> benefit to users and ISVs. > >> > >> Cheers, > >> > > > > And then you have to pick a desktop, and then you have to pick some > > RDBMSes and then you pick... blah blah, before you know it we've got a > > 3-DVD pile of software as the core distro. > > > > It is up to the distribution maintainer to choose a rich set of > > software to include in order that his distro doesn't feel like "the > > one where installing basic software takes a long time". As far as > > compatibility in terms of the trademark goes, however, the fact that > > they can pull from the same repo is sufficient IMO. > > > > Once again, +1 to Plocher's initial idea. > > > > I agree that 3 DVDs shouldn't be necessary. But I also think a fairly > rich "core" is useful. > > One of the problems I have with the current trend is sticking > *everything* into the WOS. Automake is probably as good a poster child > as anything else. ISVs don't need that for software delivery. > > Or, to pick another example, the fact that we seem to have a bajillion > alternatives -- do we really need to ship 3 or 4 SCMs? (Or do they > really need to be part of the "compatibility core".) > > I think it would be useful for someone to look at pruning down what is > in the WOS to just the stuff that folks are likely to need to *use* 3rd > party software. (Not to *develop* it!) Then where "conflicts" exist > (e.g. different MUAs or different font libraries, or whatever!) pick > just *one* that gets to be called "core". All the other options then > need not be part of the requirements for a "compatible" branding, and it > can then be possible to give ISVs good "advice" about what APIs and > other foundation software they should be building on top of if they want > to retain maximum compatibility with OpenSolaris distributions. That sums up my view fairly nicely. ISVs and users both will have a better experience when there is a rich, core set of APIs that they can depend on. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Fri Feb 22 15:14:30 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 18:14:30 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> Message-ID: <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> On Fri, Feb 22, 2008 at 6:07 PM, Shawn Walker wrote: > [moved to trademark-policy-dev] > > On Fri, Feb 22, 2008 at 5:01 PM, Garrett D'Amore > wrote: > > > > John Sonnenschein wrote: > > > On Fri, Feb 22, 2008 at 5:37 PM, Shawn Walker wrote: > > > > > >> On Fri, Feb 22, 2008 at 4:14 PM, John Plocher wrote: > > >> > John Sonnenschein wrote: > > >> > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > >> > > > >> > >> Distribution name: OpenSolaris > > >> > >> > > >> > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > >> > >> will call OpenSolaris. > > >> > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > >> > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > >> > >> Repository, which will be used by Indiana to create the distro called > > >> > >> OpenSolaris. > > >> > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > >> > >> repository" (no less but definitely also no more)? > > >> > >> o This would seem to require that either > > >> > >> A) Sun would need to not include things not on OS.o in their > > >> > >> product, or > > >> > >> B) Other distros could be supersets of the OS.o stuff and still > > >> > >> call themselves OpenSolaris. > > >> > >> > > >> > >> Does this capture your intent? > > >> > > > > >> > > It certainly does not capture mine. The sticking point is the last > > >> > > one, that is, KIndiana is precluded from existing by virtue of the > > >> > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > >> > > for compatibility are built from a minimalist core ( ie, no GUI save > > >> > > for twm ), I don't see how a superset is workable > > >> > > > > >> > > > >> > > > >> > Thanks - this actually helps clarify something for me :-) > > >> > > > >> > "It Seams To Me" that the things that are produced by Indiana > > >> > really need to be that minimalist core (i.e., enough to > > >> > boot into an installer that is network aware in whatever > > >> > locale the user needs) AND a set of repositories where > > >> > the various packages are to be found. That minimalist core > > >> > is then called OpenSolaris. > > >> > > > >> > The branding for OpenSolaris would then need to allow for > > >> > superset distros, so that Sun can create a distro that has > > >> > GNOME and NetBeans and Java, and you can create a distro that > > >> > has KDE instead. > > >> > > > >> > On a technical level, the only difference between the two distros > > >> > would seem to be A) the recipe of packages to be installed > > >> > by default, and B) the set of packages pre-cached on the > > >> > distribution media (so that the default recipe can be installed > > >> > without requiring network access...) > > >> > > > >> > As Shawn said, because of IPS/Repos, the ecosystem is > > >> > available for getting new applications AND for satisfying > > >> > their install-time dependencies. This would mean that any > > >> > OpenSolaris-core containing distro could ultimately run any > > >> > Built-for-OpenSolaris application. > > >> > > >> Putting on my ISV hat for a moment; I think you have to draw the line somewhere. > > >> > > >> That is, as an application developer, I want to be able to distribute > > >> my applications for any OpenSolaris user without triggering a > > >> multi-gigabyte download. > > >> > > >> That means that the core should not be so minimalist that to use > > >> almost any basic application that most of a repository needs to be > > >> installed. > > >> > > >> For example, I distribute an Adventure Game Runtime engine for Ubuntu > > >> users that relies on all of the libraries being fetched from apt-get, > > >> etc. > > >> > > >> Most users were happy because they either: > > >> > > >> 1) already had those libraries installed > > >> > > >> 2) weren't bothered by extra downloads from the repository > > >> > > >> However, some users were bothered at having to download a lot more, > > >> since the < 1mb download suddenly turned into > 10mb. > > >> > > >> My gut tells me that you have to pick a desktop and a set of libraries > > >> to part of core compatibility and that those should be installed by > > >> default. > > >> > > >> OS X, Windows, etc. are successful, in part, precisely because > > >> developers can depend upon a rich set of APIs always being available > > >> on a user's system. > > >> > > >> On OS X, for example, you can always depend on the "Core" libraries > > >> being present for audio, graphics, etc.. > > >> > > >> On Windows, you can always count on GDI, DirectDraw, DirectSound, and > > >> international text services > > >> > > >> I guess what I'm getting at is that developers should be able to > > >> distribute their applications in such a manner that if they stick with > > >> a "core set" of dependencies, the user won't have to download anything > > >> additional and they will get a fully functional application. > > >> > > >> That is my ultimate personal goal in this compatibility thing. > > >> > > >> Does that mean that certain libraries and environments might get > > >> favoured somewhat over another? You're darn right. But that is a > > >> benefit to users and ISVs. > > >> > > >> Cheers, > > >> > > > > > > And then you have to pick a desktop, and then you have to pick some > > > RDBMSes and then you pick... blah blah, before you know it we've got a > > > 3-DVD pile of software as the core distro. > > > > > > It is up to the distribution maintainer to choose a rich set of > > > software to include in order that his distro doesn't feel like "the > > > one where installing basic software takes a long time". As far as > > > compatibility in terms of the trademark goes, however, the fact that > > > they can pull from the same repo is sufficient IMO. > > > > > > Once again, +1 to Plocher's initial idea. > > > > > > > I agree that 3 DVDs shouldn't be necessary. But I also think a fairly > > rich "core" is useful. > > > > One of the problems I have with the current trend is sticking > > *everything* into the WOS. Automake is probably as good a poster child > > as anything else. ISVs don't need that for software delivery. > > > > Or, to pick another example, the fact that we seem to have a bajillion > > alternatives -- do we really need to ship 3 or 4 SCMs? (Or do they > > really need to be part of the "compatibility core".) > > > > I think it would be useful for someone to look at pruning down what is > > in the WOS to just the stuff that folks are likely to need to *use* 3rd > > party software. (Not to *develop* it!) Then where "conflicts" exist > > (e.g. different MUAs or different font libraries, or whatever!) pick > > just *one* that gets to be called "core". All the other options then > > need not be part of the requirements for a "compatible" branding, and it > > can then be possible to give ISVs good "advice" about what APIs and > > other foundation software they should be building on top of if they want > > to retain maximum compatibility with OpenSolaris distributions. > > That sums up my view fairly nicely. ISVs and users both will have a > better experience when there is a rich, core set of APIs that they can > depend on. And that's great and all, Sun OpenSolaris ( indiana ) should totally ship with all that, and they should be in the repo so all "- An OpenSolaris Distribution " distros can use it We're talking about what the core that the trademark requires though, so I'm not convinced that forcing certain decisions that impact more than just one distribution should be taken lightly like that. Once again for the folks in the back of the room. If it is in the repo, then it can be installed with an ISV's software. It may just take longer to install on fluxDiana - An OpenSolaris Distribution, and the users of fluxDiana should know that. -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From John.Plocher at Sun.COM Fri Feb 22 15:23:59 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 15:23:59 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> References: <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> Message-ID: <47BF598F.6080702@Sun.Com> Justin Erenkrantz wrote: > On Fri, Feb 22, 2008 at 12:15 PM, John Plocher wrote: >> What kind of ecosystem do you have if there isn't compatibility at >> the packaging and application distribution level? > > But, why use the trademark policy to enforce what should be a > technical decision? Because there doesn't seem to be any other reasonable and implementable way to define compatibility at this level. If you wish to claim you are part of the OpenSolaris ecosystem, you need to /be/ a part of it - whatever that ecosystem is. > There are plenty of tools out there (such as alien) that translate > between package formats - that's a technical problem and one that is > clearly addressable. There's no doubt that there is a Linux ecosystem Addressable from a package consumer perspective, maybe. But not from a package producer one - if I use an opensolaris system to build my app, I should also be able to package it and install it into a repository for others to use. If your distro doesn't have the packaging tools to let me do that, I'm no longer able to interact with the OpenSolaris ecosystem. From an ISV perspective, there isn't a single Linux ecosystem, there several, maybe even many. The choice an ISV has is 'what to do?'; the answer tends to be "redhat or SuSe or maybe we should just not bother". > even though there are different packaging formats. However, > forbidding a third-party distribution any access to the OpenSolaris > name because they believe that another packaging system is superior is > a dubious position. I believe this is legislating a technical The T&B guidelines don't actually mention IPS; they simply say that OpenSolaris will be whatever the Indiana project actually produces. Right now, their prototypes are using IPS, but they are still only prototypes. All I can say is get involved and cause them to change if you don't like what is being done. (I don't have any power over the decision to anoint Indiana; all I can do is work on delivering a T&B policy that doesn't suck...) > But, furthermore and more > importantly, doing so in the trademark policy seems short-sighted. -- I've tried to find a better way of meeting the goal of a compatible ecosystem; this is the least restrictive way I've been able to come up with. As it is aparently lacking in the area of packaging, do you have a better/different metric in mind that we could use instead? -John From John.Plocher at Sun.COM Fri Feb 22 15:37:10 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 15:37:10 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> Message-ID: <47BF5CA6.6050009@Sun.Com> Shawn Walker wrote: > what is in core... IMO, this is what would distinguish the various distros - how "complete" they were for the types of things you do. If you were the "I want it all there" type of guy, you'd gravitate towards one of those 3-DVD-Set distros; on the other hand, if you had a nice fat internet pipe, you might rather use a LIVE-USB-Stick distro, or even one of those micro-CD found in the latest issue of OpenSolaris for Dummies :-) I agree with Justin - this level of detail is way out of scope , even though someone will need to address it before we actually have such a beast. -John From justin at erenkrantz.com Fri Feb 22 15:42:58 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Fri, 22 Feb 2008 15:42:58 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <5c902b9e0802221448u7156d3daw4b583ac89231c7fc@mail.gmail.com> Message-ID: <5c902b9e0802221542m1f996d4g59e4f81b6bbbb8e9@mail.gmail.com> On Fri, Feb 22, 2008 at 3:04 PM, Shawn Walker wrote: > This is basic OS platform 101. Every other OS does it; No, they simply don't. > even the LSB > for Linux establishes certain libraries, APIs, etc. And the LSB exactly demonstrates this. Through the use of technical means (ie alien and other tools; plus restrictions in the LSB not to use RH-specific RPM features), Debian can be LSB compliant even though it uses a different packaging scheme. Would you be satisfied granting the rights to the "Built on OpenSolaris" name if the third-party distribution came with a tool that converted from IPS into whatever format it deems best-suited for its users? -- justin From johnsonnenschein at gmail.com Fri Feb 22 15:52:40 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 18:52:40 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF5CA6.6050009@Sun.Com> References: <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5CA6.6050009@Sun.Com> Message-ID: <241540330802221552u320f6968r626a4b2e02519953@mail.gmail.com> On Fri, Feb 22, 2008 at 6:37 PM, John Plocher wrote: > Shawn Walker wrote: > > what is in core... > > IMO, this is what would distinguish the various distros - how "complete" > they were for the types of things you do. > > If you were the "I want it all there" type of guy, you'd gravitate > towards one of those 3-DVD-Set distros; on the other hand, if you had > a nice fat internet pipe, you might rather use a LIVE-USB-Stick distro, > or even one of those micro-CD found in the latest issue of OpenSolaris for > Dummies :-) > > I agree with Justin - this level of detail is way out of scope , > even though someone will need to address it before we actually have > such a beast. I don't know that it is out of scope here, seeing as how we're trying to flesh out a sane trademark policy such that a distro can use the name. Hence why I liked your proposal, it sets a provision for compatibility that all distributions can use ( ie, the presence of the package repo ) without artificially shovelling extra software in to the distribution if it is not strictly necessary. An ISV can then, if they really absolutely want to have a super-fast install on all distributions no matter what else, include the required packages on their distribution medium ( CDROM ), and installing it would pull the packages on the system if they don't already exist. If the ISV is happy to assume packages exist, that's fine too, because the packaging system will pull them off the repo if the ISV guessed wrong. -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From John.Plocher at Sun.COM Fri Feb 22 15:57:33 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 15:57:33 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <5c902b9e0802221542m1f996d4g59e4f81b6bbbb8e9@mail.gmail.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <5c902b9e0802221448u7156d3daw4b583ac89231c7fc@mail.gmail.com> <5c902b9e0802221542m1f996d4g59e4f81b6bbbb8e9@mail.gmail.com> Message-ID: <47BF616D.4000000@Sun.Com> Justin Erenkrantz wrote: > Would you be satisfied granting > the rights to the "Built on OpenSolaris" name if the third-party > distribution came with a tool that converted from IPS into whatever > format it deems best-suited for its users? If there was also a way to produce "IPS or whatever" packages for export out into a repository, yes. This would be /exactly/ what "we" would do if "we" ever found a reason to evolve "IPS or whatever" into something else; what's good for the goose had better be good for the gander. -John From justin at erenkrantz.com Fri Feb 22 15:58:49 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Fri, 22 Feb 2008 15:58:49 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <47BF598F.6080702@Sun.Com> References: <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> <47BF598F.6080702@Sun.Com> Message-ID: <5c902b9e0802221558q7991f5e7ic926e69501ea5942@mail.gmail.com> On Fri, Feb 22, 2008 at 3:23 PM, John Plocher wrote: > Because there doesn't seem to be any other reasonable and implementable > way to define compatibility at this level. > > If you wish to claim you are part of the OpenSolaris ecosystem, you need > to /be/ a part of it - whatever that ecosystem is. To make sure we're all on the same page, as I understand what is suggested, this suggested policy means that Nexenta has *no* rights to the OpenSolaris marks - even as "Based on OpenSolaris" because it uses deb as its packaging system. Is that really being inclusive regarding the ecosystem? Mandating IPS in the T&B means that we all agree on one packaging format for all time. We simply don't know if that'll be the case. My concern about enforcing a technical problem in a social document is that it'll be used to block change and innovation. Imagine the day five years from now and Sun decides, "IPS was good, but we need something better." What rules govern the introduction of a new packaging tool into OpenSolaris? According to this, whatever Indiana delivers from Day 1 is the only solution and is immutable. I don't think that's what we really intend. We need to be able to permit people doing exactly what Indiana is doing - developing new packaging formats. When it gets this hard to write the rules, my hunch is that we've probably gone too far. Are we really concerned about an OpenSolaris third-party distribution that doesn't use IPS? If there are users for it and they're happy, who are we to say it must be excommunicated from the 'ecosystem'? In a way, you might end up fracturing the community with such rules - since IPS is not yet production-ready or had extensive real-world feedback, why would we exclude people from the community who want to use pkg or deb or RPM or anything else they might be more comfortable with? Do you consider Nexenta outside the ecosystem or inside it? > Addressable from a package consumer perspective, maybe. But not from > a package producer one - if I use an opensolaris system to build my app, > I should also be able to package it and install it into a repository for > others to use. If your distro doesn't have the packaging tools to let me > do that, I'm no longer able to interact with the OpenSolaris ecosystem. Hosting a package repository is the distribution's responsibility - not ours. We may ultimately decide that it's good to host IPS; but if someone wants to run alien and convert it all to the FOO-format and host it themselves, how is that a devastating problem to the overall ecosystem? > The T&B guidelines don't actually mention IPS; they simply say that > OpenSolaris will be whatever the Indiana project actually produces. > Right now, their prototypes are using IPS, but they are still only > prototypes. All I can say is get involved and cause them to change > if you don't like what is being done. (I don't have any power over > the decision to anoint Indiana; all I can do is work on delivering > a T&B policy that doesn't suck...) The point, as mentioned earlier, how do we ever permit the introduction of IPS++? > I've tried to find a better way of meeting the goal of a compatible > ecosystem; this is the least restrictive way I've been able to come > up with. As it is aparently lacking in the area of packaging, do you > have a better/different metric in mind that we could use instead? I would stress ABI and API compatibility. Folks are capable of understanding that a GNOME package needs GNOME installed somehow. That's the job of the distribution developers to figure out how to best deliver that software. If they make it hard to do, they won't get many users. -- justin From James.Walker at Sun.COM Fri Feb 22 16:28:06 2008 From: James.Walker at Sun.COM (Jim Walker) Date: Fri, 22 Feb 2008 17:28:06 -0700 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF4947.6030709@Sun.Com> References: <47bdae30.2K1xXcJAc8RW6lf+%Joerg.Schilling@fokus.fraunhofer.de> <47bdbd2a.3qu+nYoaaRBkRGtl%Joerg.Schilling@fokus.fraunhofer.de> <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> Message-ID: <47BF6896.7050106@sun.com> John Plocher wrote: > > "It Seams To Me" that the things that are produced by Indiana > really need to be that minimalist core (i.e., enough to > boot into an installer that is network aware in whatever > locale the user needs) AND a set of repositories where > the various packages are to be found. That minimalist core > is then called OpenSolaris. I agree. We must find the current minimum set of packages that defines an OpenSolaris distribution. Certification can proceed once we have that. However, even though Indiana is very minimal, I'm not convinced that it is at the minimum. I think you can use even less code to have a very useful OpenSolaris distribution (ie. embedded applications). I consider the OpenSolaris distribution (Indiana) the "OpenSolaris Reference Distribution" that targets the 80% mark (ie. most OpenSolaris distributions will be like it). But, I think the trademark policy project team needs to dig even deeper and define a more absolute minimum so we hit the 100% mark, so we don't have to change what certification means on day one. Larger distributions are easy once we agree how we handle the small ones. Cheers, Jim From John.Plocher at Sun.COM Fri Feb 22 16:25:31 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 16:25:31 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <5c902b9e0802221558q7991f5e7ic926e69501ea5942@mail.gmail.com> References: <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> <47BF598F.6080702@Sun.Com> <5c902b9e0802221558q7991f5e7ic926e69501ea5942@mail.gmail.com> Message-ID: <47BF67FB.6050303@Sun.Com> Justin Erenkrantz wrote: > To make sure we're all on the same page, as I understand what is > suggested, this suggested policy means that Nexenta has *no* rights to I have not analyzed a recent version, so my comments will probably be less than completely correct :-) > the OpenSolaris marks - even as "Based on OpenSolaris" because it uses > deb as its packaging system. Is that really being inclusive regarding > the ecosystem? The intent of the T&BP is that, if Nexenta can install packages from the various OpenSolaris Repositories (which, given its track record, it will do within 30 seconds of the time the repo manager and IPS code is made available :-), and if a developer on Nexenta can create and export IPS packages into a Repo, then yes, they absolutely should be able to use one of the OpenSolaris marks. > Mandating IPS in the T&B It *IS NOT* mandated there. Rather, the definition is Whatever is done out on OpenSolaris.Org in the Indiana Project [or its successor projects]. As the Indiana project evolves and is replaced by other OS.o efforts that integrate into ON, Admin, Packaging and Install, etc, so will the definition of "OpenSolaris". Right now, this technically means "SVr4 pkgs", since IPS and friends are still prototypes. In fact, since Nexenta includes pkg-get and access to blastwave, [no, I'm not going there... :-)] .... > Do you consider Nexenta outside the ecosystem or inside it? Right now, they are on the edge thru their own choice - there is a lot of harvesting and not a lot of give back, but I'm hopeful that will change. I'd love for them to jump in and contribute their innovation back to the community, and to become a key player in the Indiana project; however, it is their right to chose their own course. > Hosting a package repository is the distribution's responsibility - > not ours. I think that this is where we just have to disagree. Hosting repositories /is/ what makes a distro into an ecosystem. > The point, as mentioned earlier, how do we ever permit the > introduction of IPS++? ARC it (which means addressing compatibility and transition support) and integrate it into our consolidations. Once it is in our consolidations, it (by the definition I am pushing) *becomes* part of OpenSolaris. > I would stress ABI and API compatibility. A starting point, but not sufficient... > That's the job of the distribution developers to figure out how to > best deliver that software. The question that Indiana brings to the table is exactly that: Do we wish to become a distro developer in addition to a kernel (and other things) development project? The answer, "of course", is yes. -John From johnsonnenschein at gmail.com Fri Feb 22 16:30:26 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Fri, 22 Feb 2008 19:30:26 -0500 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF6896.7050106@sun.com> References: <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> Message-ID: <241540330802221630v6a4c74afsed3d9b447f5f232d@mail.gmail.com> On Fri, Feb 22, 2008 at 7:28 PM, Jim Walker wrote: > John Plocher wrote: > > > > "It Seams To Me" that the things that are produced by Indiana > > really need to be that minimalist core (i.e., enough to > > boot into an installer that is network aware in whatever > > locale the user needs) AND a set of repositories where > > the various packages are to be found. That minimalist core > > is then called OpenSolaris. > > I agree. We must find the current minimum set of packages > that defines an OpenSolaris distribution. Certification > can proceed once we have that. > > However, even though Indiana is very minimal, I'm not > convinced that it is at the minimum. I think you can > use even less code to have a very useful OpenSolaris > distribution (ie. embedded applications). > > I consider the OpenSolaris distribution (Indiana) the > "OpenSolaris Reference Distribution" that targets the > 80% mark (ie. most OpenSolaris distributions will be > like it). But, I think the trademark policy project > team needs to dig even deeper and define a more absolute > minimum so we hit the 100% mark, so we don't have to > change what certification means on day one. > > Larger distributions are easy once we agree how we > handle the small ones. +1 I think before we can submit something to Sun for their approval, we should flesh out a minmial compatibility standard -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From justin at erenkrantz.com Fri Feb 22 16:46:19 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Fri, 22 Feb 2008 16:46:19 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <47BF67FB.6050303@Sun.Com> References: <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> <47BF598F.6080702@Sun.Com> <5c902b9e0802221558q7991f5e7ic926e69501ea5942@mail.gmail.com> <47BF67FB.6050303@Sun.Com> Message-ID: <5c902b9e0802221646i29fb072awe4e39efdf4098bc5@mail.gmail.com> On Fri, Feb 22, 2008 at 4:25 PM, John Plocher wrote: > Justin Erenkrantz wrote: > > To make sure we're all on the same page, as I understand what is > > suggested, this suggested policy means that Nexenta has *no* rights to > > I have not analyzed a recent version, so my comments will probably > be less than completely correct :-) > > > > the OpenSolaris marks - even as "Based on OpenSolaris" because it uses > > deb as its packaging system. Is that really being inclusive regarding > > the ecosystem? > > The intent of the T&BP is that, if Nexenta can install packages from > the various OpenSolaris Repositories (which, given its track record, > it will do within 30 seconds of the time the repo manager and IPS code > is made available :-), and if a developer on Nexenta can create and > export IPS packages into a Repo, then yes, they absolutely should > be able to use one of the OpenSolaris marks. Thanks for the clarification - we've perhaps had our wires crossed. If it's okay to have a different packaging system as the core of a third-party distribution - but enforcing that the distribution must have a way to automatically either convert to and from the "OpenSolaris" packaging format (such as an IPS-specific alien or similar tools) into whatever format the distribution chooses, then I'm comfortable with that. Would you be comfortable calling this out in the T&B document as I don't think this has been made very clear so far. -- justin From shivakumar.gn at gmail.com Fri Feb 22 18:51:18 2008 From: shivakumar.gn at gmail.com (S h i v) Date: Sat, 23 Feb 2008 08:21:18 +0530 Subject: [trademark-policy-dev] Vision for the future: Re: Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <47BF0F01.6000806@Sun.Com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> <47BF0F01.6000806@Sun.Com> Message-ID: <319ee2b10802221851w7a5025dbh751e03138d089fea@mail.gmail.com> Mistakenly marked only to me I presume. ---------- Forwarded message ---------- From: John Plocher Date: Fri, Feb 22, 2008 at 11:35 PM Subject: Vision for the future: Re: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft [aarg - this started out as a simple "OK" response and grew into a "what is my vision for the future of the community and the distros it spawns. Sorry about the length - John] S h i v wrote: > Can you please number the issues (even if there is only one currently) > so that referring to it is easier. OK Issue #1: > ISSUE At this point in time, nevada is what we have, and indiana is an > undefined future possibility. How can people today use the OpenSolaris > marks without implying that they are using/doing indiana? > > Given that Indiana will be referred to as OpenSolaris. "People using > OpenSolaris marks without implying they are using Indiana" will remain > a proposition that cannot be satisfied. Agreed. I believe that this is simply a short term artifact of the transition we all are making, and that the dichotomy between Nevada, Indiana and OpenSolaris will go away. My view: Time=yesterday: We have Solaris 10, and its update releases from Sun. This is a mature, enterprise level product. We have SXCE from Sun, a snapshot of ON's bi-weekly build plus a veneer of Solaris 10 special sauce mixed with Nevada special sauce mixed with Indiana special sauce. This is a bleeding edge, OS developer focused snapshot of things yet to come. We have SXDE from Sun, a snapshot of SXCE plus Sun tools and other Sun chosen/provided stuff. This is still a leading edge snapshot, but instead of being aimed at OS developers, it is aimed at ISVs and early adopter users. We have a bunch of loosely connected distros that harvest consolidations off of os.o on a semi-regular basis and combine them with their own distro-specific special sauce and philosophies to produce Schillix, Nexenta, MartUX, Belenix and potentially others. It could be argued that SXCE and SXDE are also in this camp. We have a bunch of projects on OS.o that are prototyping stuff and getting it into the core (ksh93, caiman, ips, and yes, even Indiana) It is interesting to note that Indiana is really simple at this point in time: It is simply ON + Caiman(install) + ips(packaging) + a small amount of special sauce experiments. On a cosmic scale, there is very little substantive difference between SXCE and Indiana - though the perceptive differences are huge. Finally, we have the remnants of what we thought would be the "Solaris version that comes after Solaris 10", which we call Nevada. It maps mostly to the ON consolidation, though there are close ties to the admin, install and sfw consolidations as well. Time=today: Solaris 10, S10u*, SXCE, SXDE and Nevada are unchanged (though the S10 updates continue to be released). Sun is figuring out its roadmap for the future. It is putting its resources into Indiana, and positioning it as the replacement for Nevada, sortof. If yesterday's roadmap was "S10, S10updates, Nevada->S11", then todays view is more like "S10, S10updates, GOSUB(Indiana), f(Nevada,Indiana) -> S11" - where Indiana is intended to modernize and revitalize the more stagnant parts of Solaris, and the definition of f() is left for the future :-) Time=tomorrow (warning, my crystal ball warranty expired yesterday): Solaris 10, and S10u* are unchanged (though the S10 updates continue to be released). Nevada (and the SXCE built from it) start looking more like Indiana as the things Indiana is prototyping start being integrated into Nevada. Sun starts putting its applications and add-on features into repositories that can be accessed from any OpenSolaris distro. SXDE goes away. Indiana (aka OpenSolaris Developer Edition) becomes the replacement distro for ISVs and early adopter users. Because it gets its bits from the OpenSolaris repository, it effectively *is* OpenSolaris. (I added the following bullet to the T&B page to reflect the above assumption: * Potential Metric: The various consolidations that make up the OpenSolaris community will publish their {packages, bits, stuff} in an OpenSolaris Repository, which will be used by Indiana to create the distro called OpenSolaris. -John) Time=ma?ana (not today, Se?or, and not tomorrow, either...) Solaris 10, and S10u* are unchanged (though the S10 updates continue to be released). SXCE goes away. Because all its prototyping is complete and integrated into ON/SFW/Nevada/..., the Indiana project goes away and is replaced with other projects as the community evolves. Nevada remains the place where we build the OpenSolaris kernel and related bits. OpenSolaris has both a set of developer releases (formerly known as SXCEs) and a set of production releases (formerly SXDEs). We all will need to figure out how these map to the release taxonomy, or if the release taxonomy needs to change...) There will be a slew of OpenSolaris-family distros, based on a set of repositories - the ones from OpenSolaris.org, the ones from Sun.Com, the ones from Blastwave, SunFreeware, Debian/Nexenta, Joerg and others, and we will figure out what to call them, how to nurture them and how to make the community be something that they feel comfortable participating in. There will be a slew of OpenSolaris-derived things that are not "general purpose distros" in their own right - appliances, embedded, turnkey, whatever, and we will figure out what to call them, how to nurture them and how to make the community be something that they feel comfortable participating in. It may be that Solaris 11 is created from one of these OpenSolaris production releases, or maybe it *is* the production release mechanism. Who knows, it may not even be called Solaris 11. Someone will figure out what to call it, we will nurture the effort, and how to make the community be something that they feel comfortable participating in. > Usage of logos need to be made a more liberal and meaningful for > community purposes. Unfortunately, graphic artwork is not handled well under trademark law, which is why all the additional legal cruft is there. There is also a HUGE NEED for logo-related-stuff that can be used in places that aren't web pages. I added these thoughts to the Issues. > A community project should seek art-work and submissions so that it > can be approved and made available instead of the message provided at > the link. I think that, while this may be desirable, it is out of scope for this policy document. > Due to some of the reasons above we have not been able to use it in > our UG. Please contact JimG or SaraD to work out those issues - I believe they are willing and able to help. -John From shivakumar.gn at gmail.com Fri Feb 22 19:30:29 2008 From: shivakumar.gn at gmail.com (S h i v) Date: Sat, 23 Feb 2008 09:00:29 +0530 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <47BF21D8.9090205@Sun.Com> <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> Message-ID: <319ee2b10802221930y3c895b8ema2de6aadea90cd33@mail.gmail.com> On Sat, Feb 23, 2008 at 1:05 AM, John Sonnenschein wrote: > Okay, another question then. > > What subset of packages are required to qualify for compatibility? > > Say for example I build a distro that only includes SUNWips, > SUNWos-net ( or whatever ON will be called ) and the rest are pulled > from another repo ( blastwave, for example ). > > And lets say this distro builds on top of a whole slew of weird > packages that are mostly incompatible with anything that indiana > includes ( say... the freebsd userspace ) compiled with a compiler > that makes ABI compatibility an impossibility ( like gcc 2.95 ) > > Is it still " BreakFOO - an OpenSolaris Distribution " ? technically > you can use IPS to pull in all the indiana stuff and make it > compatible, and it includes a subset of packages that indiana does, > but in and of itself it is not. > It is just a matter of time before OpenSolaris is equated to Indiana and not ON. History of the term is then forgotten. When that happens any other community distro even those at OS.o may not want to use the TM term to avoid confusion. With a product to use, world associates the term with the product and the community is assumed to revolve around it. Cut & slice whichever way, the result is a given. Changing the fine-print will not help. I have not been for Indiana using the name, but given the restriction, I do not see why any community distro would want to identify their independent efforts with "OpenSolaris - the distro" by using the term "built on OpenSolaris technology". Hence no point fighting for the term. The current T&M guidelines in the wiki (0.12 version) seems to be good enough within the limitation we have. -Shiv From shivakumar.gn at gmail.com Fri Feb 22 19:53:08 2008 From: shivakumar.gn at gmail.com (S h i v) Date: Sat, 23 Feb 2008 09:23:08 +0530 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> Message-ID: <319ee2b10802221953o52c9b16s1a75bb347c08f97b@mail.gmail.com> On Fri, Feb 22, 2008 at 9:16 PM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 1:33 AM, S h i v wrote: > > ISSUE At this point in time, nevada is what we have, and indiana is an > > undefined future possibility. How can people today use the OpenSolaris > > marks without implying that they are using/doing indiana? > > I disagree that it implies Indiana. Remember that Indiana is just a > prototype and will eventually disappear leaving only what we have > already today. > Please go figure. > > > Usage of logos need to be made a more liberal and meaningful for > > community purposes. > > This needs explanation. > After completing reading my mail you could have revisited this statement :) Anyway, I will initiate a separate thread on advocacy-discuss once these TM discussions are through. > A license of some form is going to be necssary. > > > Also have a look at "Please, No Modifications (thanks!)" section of > > http://www.opensolaris.org/os/about/buttons page. > > A community project should seek art-work and submissions so that it > > can be approved and made available instead of the message provided at > > the link. > > You can discuss this with Advocacy. I believe they have the ability to > get such materials approved by Marketing. > > > > The logos can definitely do with some face-lift and community > > contribution while being in line with the trademark policy. It is good > > to seek submissions here as well. > > Unfortunately, I don't think you understand the need a company has to > protect trademarks. They can't allow anyone to use the trademark in > any way possible and still have it be commercially viable. > Please read my mail in its entirety. Nowhere is doing away with checks & balances that any TM holder needs is suggested. Expounding any further on this topic on this list would be a digression to the current discussion on TM. I will initiate a separate thread on advocacy-discuss once my task list becomes free. Having a look at ubuntu's TM policy might be useful http://www.ubuntu.com/aboutus/trademarkpolicy It is very clear about how people can and cannot use the TM that canonical uses, there are no confusions as to how canonical intends to use and it is no less controlled than the current draft for OS.o on the wiki, but the intent of getting the community involved comes out well in their usage of the wordings. Their processes, infrastructure, etc enable the intent as well. They have mechanisms for community inputs, designs. The control (I wouldn't use "quality control" here) is via the need for getting the approval. This I already suggested in my earlier mail. -Shiv From shivakumar.gn at gmail.com Fri Feb 22 20:55:01 2008 From: shivakumar.gn at gmail.com (S h i v) Date: Sat, 23 Feb 2008 10:25:01 +0530 Subject: [trademark-policy-dev] Brandorr: recompiling "OpenSolaris" from the sources? In-Reply-To: <47BF2928.6070707@Sun.Com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> Message-ID: <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> Hi John, On Sat, Feb 23, 2008 at 1:27 AM, John Plocher wrote: > > > ...and call it xyz OpenSolaris? (This last > > question is for the room.) > > The room does not need to answer - the lawyers already have: > http://www.sun.com/policies/trademarks/ > > Sun trademarks cannot be included in the name of any third party > product, technology, program or service. This includes free and > educational materials, open source distributions, and the titles > of informational web sites. > My assumption is that the link you posted is the current sun's TM policy governing the current trademarks and product portfolio. If this is what opensolaris community needs to abide by going forward, the discussion becomes redundant. People are not allowed to provide educational materials(even non-commercial) for "OpenSolaris" ! Community needs *some term* that it can use, without the fear of being sued, seriously ! I would encourage people to see the "Permitted Use" section of Ubuntu's trademark policy (http://www.ubuntu.com/aboutus/trademarkpolicy) and see what aspects can be applicable and used for OpenSolaris. It covers 3rd party usage as well as usage for community advocacy. -Shiv From John.Plocher at Sun.COM Fri Feb 22 23:03:20 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Fri, 22 Feb 2008 23:03:20 -0800 Subject: [trademark-policy-dev] Packaging deviations... In-Reply-To: <5c902b9e0802221646i29fb072awe4e39efdf4098bc5@mail.gmail.com> References: <47BF1C86.8070704@Sun.Com> <241540330802221107y690badcex4c10e82f1a0e6863@mail.gmail.com> <5c902b9e0802221137q5f124610t626652ec2dd55de8@mail.gmail.com> <47BF2D63.8080107@Sun.Com> <5c902b9e0802221352p44dc6036q383c8e94abc1d44c@mail.gmail.com> <47BF598F.6080702@Sun.Com> <5c902b9e0802221558q7991f5e7ic926e69501ea5942@mail.gmail.com> <47BF67FB.6050303@Sun.Com> <5c902b9e0802221646i29fb072awe4e39efdf4098bc5@mail.gmail.com> Message-ID: <47BFC538.8050206@Sun.Com> Justin Erenkrantz wrote: > Would you be comfortable calling this out in the T&B document as I > don't think this has been made very clear so far. -- justin I put in a placeholder && will flesh it out later. -John From swalker at opensolaris.org Sat Feb 23 00:24:11 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 02:24:11 -0600 Subject: [trademark-policy-dev] Comments and closure needed on the current OpenSolaris Trademark and Branding policy draft In-Reply-To: <319ee2b10802221953o52c9b16s1a75bb347c08f97b@mail.gmail.com> References: <47BDB577.6040203@Sun.Com> <5c902b9e0802211112r41322da4p4e59f337c4084ba4@mail.gmail.com> <47BDE937.5060406@Sun.Com> <32F2C7BB-6CC2-4E61-8944-6D27CA6B0BF1@gmail.com> <47BE51A1.90703@Sun.Com> <47BE5543.9060809@Sun.Com> <319ee2b10802212333h136e9779k7d09d729063c23fa@mail.gmail.com> <319ee2b10802221953o52c9b16s1a75bb347c08f97b@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 9:53 PM, S h i v wrote: > On Fri, Feb 22, 2008 at 9:16 PM, Shawn Walker wrote: > > > On Fri, Feb 22, 2008 at 1:33 AM, S h i v wrote: > > > ISSUE At this point in time, nevada is what we have, and indiana is an > > > undefined future possibility. How can people today use the OpenSolaris > > > marks without implying that they are using/doing indiana? > > > > I disagree that it implies Indiana. Remember that Indiana is just a > > prototype and will eventually disappear leaving only what we have > > already today. > > > > Please go figure. > > > > > > > Usage of logos need to be made a more liberal and meaningful for > > > community purposes. > > > > This needs explanation. > > > > After completing reading my mail you could have revisited this statement :) > Anyway, I will initiate a separate thread on advocacy-discuss once > these TM discussions are through. > > > > A license of some form is going to be necssary. > > > > > Also have a look at "Please, No Modifications (thanks!)" section of > > > http://www.opensolaris.org/os/about/buttons page. > > > A community project should seek art-work and submissions so that it > > > can be approved and made available instead of the message provided at > > > the link. > > > > You can discuss this with Advocacy. I believe they have the ability to > > get such materials approved by Marketing. > > > > > > > The logos can definitely do with some face-lift and community > > > contribution while being in line with the trademark policy. It is good > > > to seek submissions here as well. > > > > Unfortunately, I don't think you understand the need a company has to > > protect trademarks. They can't allow anyone to use the trademark in > > any way possible and still have it be commercially viable. > > > > Please read my mail in its entirety. Nowhere is doing away with checks > & balances that any TM holder needs is suggested. > Expounding any further on this topic on this list would be a > digression to the current discussion on TM. I will initiate a separate > thread on advocacy-discuss once my task list becomes free. > > Having a look at ubuntu's TM policy might be useful > http://www.ubuntu.com/aboutus/trademarkpolicy > It is very clear about how people can and cannot use the TM that > canonical uses, there are no confusions as to how canonical intends to > use and it is no less controlled than the current draft for OS.o on > the wiki, but the intent of getting the community involved comes out > well in their usage of the wordings. Their processes, infrastructure, > etc enable the intent as well. They have mechanisms for community > inputs, designs. > The control (I wouldn't use "quality control" here) is via the need > for getting the approval. This I already suggested in my earlier mail. I don't think that conflicts with Sun's statements so far. You always have the option of producing something and then asking for approval *before* releasing or using that material. I know Sara Dornsife would likely be very open to community contributions for marketing materials. This is definitely something to approach the Advocacy CG with. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Sat Feb 23 00:36:16 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 02:36:16 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF5CA6.6050009@Sun.Com> References: <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5CA6.6050009@Sun.Com> Message-ID: On Fri, Feb 22, 2008 at 5:37 PM, John Plocher wrote: > Shawn Walker wrote: > > what is in core... > > IMO, this is what would distinguish the various distros - how "complete" > they were for the types of things you do. > > If you were the "I want it all there" type of guy, you'd gravitate > towards one of those 3-DVD-Set distros; on the other hand, if you had > a nice fat internet pipe, you might rather use a LIVE-USB-Stick distro, > or even one of those micro-CD found in the latest issue of OpenSolaris for > Dummies :-) > > I agree with Justin - this level of detail is way out of scope , > even though someone will need to address it before we actually have > such a beast. That depends. Are we talking about the right to use the trademark to describe a distribution as "OpenSolaris compatible" or are we simply talking about the right to call a distribution "based on OpenSolaris". If the former, I disagree, if the latter, I can agree. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Sat Feb 23 00:34:03 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 02:34:03 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 5:14 PM, John Sonnenschein wrote: > > On Fri, Feb 22, 2008 at 6:07 PM, Shawn Walker wrote: > > [moved to trademark-policy-dev] > > > > On Fri, Feb 22, 2008 at 5:01 PM, Garrett D'Amore > > wrote: > > > > > > John Sonnenschein wrote: > > > > On Fri, Feb 22, 2008 at 5:37 PM, Shawn Walker wrote: > > > > > > > >> On Fri, Feb 22, 2008 at 4:14 PM, John Plocher wrote: > > > >> > John Sonnenschein wrote: > > > >> > > On Fri, Feb 22, 2008 at 3:12 PM, John Plocher wrote: > > > >> > > > > >> > >> Distribution name: OpenSolaris > > > >> > >> > > > >> > >> * Sun has decided to fund Project Indiana to build a base distro, which it > > > >> > >> will call OpenSolaris. > > > >> > >> * Potential Metric: The various consolidations that make up the OpenSolaris > > > >> > >> community will publish their {packages, bits, stuff} in an OpenSolaris > > > >> > >> Repository, which will be used by Indiana to create the distro called > > > >> > >> OpenSolaris. > > > >> > >> * Would this be a distribution made from exactly the bits in "The OpenSolaris > > > >> > >> repository" (no less but definitely also no more)? > > > >> > >> o This would seem to require that either > > > >> > >> A) Sun would need to not include things not on OS.o in their > > > >> > >> product, or > > > >> > >> B) Other distros could be supersets of the OS.o stuff and still > > > >> > >> call themselves OpenSolaris. > > > >> > >> > > > >> > >> Does this capture your intent? > > > >> > > > > > >> > > It certainly does not capture mine. The sticking point is the last > > > >> > > one, that is, KIndiana is precluded from existing by virtue of the > > > >> > > fact that Indiana proper includes GNOME. Unless Indiana and the metric > > > >> > > for compatibility are built from a minimalist core ( ie, no GUI save > > > >> > > for twm ), I don't see how a superset is workable > > > >> > > > > > >> > > > > >> > > > > >> > Thanks - this actually helps clarify something for me :-) > > > >> > > > > >> > "It Seams To Me" that the things that are produced by Indiana > > > >> > really need to be that minimalist core (i.e., enough to > > > >> > boot into an installer that is network aware in whatever > > > >> > locale the user needs) AND a set of repositories where > > > >> > the various packages are to be found. That minimalist core > > > >> > is then called OpenSolaris. > > > >> > > > > >> > The branding for OpenSolaris would then need to allow for > > > >> > superset distros, so that Sun can create a distro that has > > > >> > GNOME and NetBeans and Java, and you can create a distro that > > > >> > has KDE instead. > > > >> > > > > >> > On a technical level, the only difference between the two distros > > > >> > would seem to be A) the recipe of packages to be installed > > > >> > by default, and B) the set of packages pre-cached on the > > > >> > distribution media (so that the default recipe can be installed > > > >> > without requiring network access...) > > > >> > > > > >> > As Shawn said, because of IPS/Repos, the ecosystem is > > > >> > available for getting new applications AND for satisfying > > > >> > their install-time dependencies. This would mean that any > > > >> > OpenSolaris-core containing distro could ultimately run any > > > >> > Built-for-OpenSolaris application. > > > >> > > > >> Putting on my ISV hat for a moment; I think you have to draw the line somewhere. > > > >> > > > >> That is, as an application developer, I want to be able to distribute > > > >> my applications for any OpenSolaris user without triggering a > > > >> multi-gigabyte download. > > > >> > > > >> That means that the core should not be so minimalist that to use > > > >> almost any basic application that most of a repository needs to be > > > >> installed. > > > >> > > > >> For example, I distribute an Adventure Game Runtime engine for Ubuntu > > > >> users that relies on all of the libraries being fetched from apt-get, > > > >> etc. > > > >> > > > >> Most users were happy because they either: > > > >> > > > >> 1) already had those libraries installed > > > >> > > > >> 2) weren't bothered by extra downloads from the repository > > > >> > > > >> However, some users were bothered at having to download a lot more, > > > >> since the < 1mb download suddenly turned into > 10mb. > > > >> > > > >> My gut tells me that you have to pick a desktop and a set of libraries > > > >> to part of core compatibility and that those should be installed by > > > >> default. > > > >> > > > >> OS X, Windows, etc. are successful, in part, precisely because > > > >> developers can depend upon a rich set of APIs always being available > > > >> on a user's system. > > > >> > > > >> On OS X, for example, you can always depend on the "Core" libraries > > > >> being present for audio, graphics, etc.. > > > >> > > > >> On Windows, you can always count on GDI, DirectDraw, DirectSound, and > > > >> international text services > > > >> > > > >> I guess what I'm getting at is that developers should be able to > > > >> distribute their applications in such a manner that if they stick with > > > >> a "core set" of dependencies, the user won't have to download anything > > > >> additional and they will get a fully functional application. > > > >> > > > >> That is my ultimate personal goal in this compatibility thing. > > > >> > > > >> Does that mean that certain libraries and environments might get > > > >> favoured somewhat over another? You're darn right. But that is a > > > >> benefit to users and ISVs. > > > >> > > > >> Cheers, > > > >> > > > > > > > > And then you have to pick a desktop, and then you have to pick some > > > > RDBMSes and then you pick... blah blah, before you know it we've got a > > > > 3-DVD pile of software as the core distro. > > > > > > > > It is up to the distribution maintainer to choose a rich set of > > > > software to include in order that his distro doesn't feel like "the > > > > one where installing basic software takes a long time". As far as > > > > compatibility in terms of the trademark goes, however, the fact that > > > > they can pull from the same repo is sufficient IMO. > > > > > > > > Once again, +1 to Plocher's initial idea. > > > > > > > > > > I agree that 3 DVDs shouldn't be necessary. But I also think a fairly > > > rich "core" is useful. > > > > > > One of the problems I have with the current trend is sticking > > > *everything* into the WOS. Automake is probably as good a poster child > > > as anything else. ISVs don't need that for software delivery. > > > > > > Or, to pick another example, the fact that we seem to have a bajillion > > > alternatives -- do we really need to ship 3 or 4 SCMs? (Or do they > > > really need to be part of the "compatibility core".) > > > > > > I think it would be useful for someone to look at pruning down what is > > > in the WOS to just the stuff that folks are likely to need to *use* 3rd > > > party software. (Not to *develop* it!) Then where "conflicts" exist > > > (e.g. different MUAs or different font libraries, or whatever!) pick > > > just *one* that gets to be called "core". All the other options then > > > need not be part of the requirements for a "compatible" branding, and it > > > can then be possible to give ISVs good "advice" about what APIs and > > > other foundation software they should be building on top of if they want > > > to retain maximum compatibility with OpenSolaris distributions. > > > > That sums up my view fairly nicely. ISVs and users both will have a > > better experience when there is a rich, core set of APIs that they can > > depend on. > > And that's great and all, Sun OpenSolaris ( indiana ) should totally > ship with all that, and they should be in the repo so all "- An > OpenSolaris Distribution " distros can use it > > We're talking about what the core that the trademark requires though, > so I'm not convinced that forcing certain decisions that impact more > than just one distribution should be taken lightly like that. > > Once again for the folks in the back of the room. If it is in the > repo, then it can be installed with an ISV's software. It may just > take longer to install on fluxDiana - An OpenSolaris Distribution, and > the users of fluxDiana should know that. There needs to be some compromise though. How do you think a user will feel when he's told, "Oh yeah; this is OpenSolaris compatible, but to use some applications you may need to download half a gigabyte or more from some repositories because we don't like some of the libraries." That really isn't conducive to a forward-thinking platform that seeks wide adoption. I think the compromise here needs to be that compatibility has to be based on target audiences. I can agree with the idea of a minimal core for the purposes of being able to carry a tagline, like, "based on OpenSolaris". However, if you want to produce something that is actually useful to ISVs, etc. then you're going to need: * OpenSolaris Desktop Compatible * OpenSolaris Server Compatible * OpenSolaris Embedded Compatible The Desktop one is the one I'm personally the most interested in and that is going to have to make choices I know you won't like. Already I can hear the questions, "Why do we have to make choices?" I'll tell you why. Developers don't want to be able to choose from 30 different graphics and input libraries. They want *one* very well supported, integrated, performing library that does what they need. Developers don't want to have to support several different sound APIs (pulse audio, aRTS, eSD, etc.). They just want to support one Audio API. Not only that, from the perspective of the community, I think it is far more important to focus on quality than quantity. Solaris tends to integrate FOSS software differently than GNU/Linux distributions do. Whereas most GNU/Linux distributions tend to just repackage something and "tack it on" to the system; Solaris tends to modify it to a certain extent to make it feel as though it is actually part of the system (e.g. providing SMF profiles, etc.). Finally, resources, time, and attention are finite. By choosing to focus on several rich, core APIs for a platform, you can increase developer productivity and user confidence in your platform through the resulting focus in resources, stability, etc. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Sat Feb 23 00:38:18 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 02:38:18 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF6896.7050106@sun.com> References: <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> Message-ID: On Fri, Feb 22, 2008 at 6:28 PM, Jim Walker wrote: > John Plocher wrote: > > > > "It Seams To Me" that the things that are produced by Indiana > > really need to be that minimalist core (i.e., enough to > > boot into an installer that is network aware in whatever > > locale the user needs) AND a set of repositories where > > the various packages are to be found. That minimalist core > > is then called OpenSolaris. > > I agree. We must find the current minimum set of packages > that defines an OpenSolaris distribution. Certification > can proceed once we have that. > > However, even though Indiana is very minimal, I'm not > convinced that it is at the minimum. I think you can > use even less code to have a very useful OpenSolaris > distribution (ie. embedded applications). > > I consider the OpenSolaris distribution (Indiana) the > "OpenSolaris Reference Distribution" that targets the > 80% mark (ie. most OpenSolaris distributions will be > like it). But, I think the trademark policy project > team needs to dig even deeper and define a more absolute > minimum so we hit the 100% mark, so we don't have to > change what certification means on day one. > > Larger distributions are easy once we agree how we > handle the small ones. I would agree that it is the minimum; but the question is how useful the minimum will actually be. I don't think it would be fair or right to take the system down to the bare minimum and then give everyone using that minimum the right to have a blanket "OpenSolaris compatible" tagline. Such an indicator would mean little in practice. Now if that tagline were "based on OpenSolaris"; that's a different story. Perhaps then, the easiest to create a compatibility trademark usage guideline for first would be for the server environment. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Sat Feb 23 00:54:47 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 00:54:47 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> Message-ID: On 23-Feb-08, at 12:34 AM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 5:14 PM, John Sonnenschein > wrote: >> >> On Fri, Feb 22, 2008 at 6:07 PM, Shawn Walker > > wrote: >>> [moved to trademark-policy-dev] >>> >>> On Fri, Feb 22, 2008 at 5:01 PM, Garrett D'Amore >>> wrote: >>>> >>>> John Sonnenschein wrote: >>>>> On Fri, Feb 22, 2008 at 5:37 PM, Shawn Walker >>>> > wrote: >>>>> >>>>>> On Fri, Feb 22, 2008 at 4:14 PM, John Plocher >>>>> > wrote: >>>>>>> John Sonnenschein wrote: >>>>>>>> On Fri, Feb 22, 2008 at 3:12 PM, John Plocher >>>>>>> > wrote: >>>>>>> >>>>>>>>> Distribution name: OpenSolaris >>>>>>>>> >>>>>>>>> * Sun has decided to fund Project Indiana to build a >>>>>>>>> base distro, which it >>>>>>>>> will call OpenSolaris. >>>>>>>>> * Potential Metric: The various consolidations that make >>>>>>>>> up the OpenSolaris >>>>>>>>> community will publish their {packages, bits, stuff} >>>>>>>>> in an OpenSolaris >>>>>>>>> Repository, which will be used by Indiana to create >>>>>>>>> the distro called >>>>>>>>> OpenSolaris. >>>>>>>>> * Would this be a distribution made from exactly the >>>>>>>>> bits in "The OpenSolaris >>>>>>>>> repository" (no less but definitely also no more)? >>>>>>>>> o This would seem to require that either >>>>>>>>> A) Sun would need to not include things not on >>>>>>>>> OS.o in their >>>>>>>>> product, or >>>>>>>>> B) Other distros could be supersets of the OS.o >>>>>>>>> stuff and still >>>>>>>>> call themselves OpenSolaris. >>>>>>>>> >>>>>>>>> Does this capture your intent? >>>>>>>> >>>>>>>> It certainly does not capture mine. The sticking point is the >>>>>>>> last >>>>>>>> one, that is, KIndiana is precluded from existing by virtue >>>>>>>> of the >>>>>>>> fact that Indiana proper includes GNOME. Unless Indiana and >>>>>>>> the metric >>>>>>>> for compatibility are built from a minimalist core ( ie, no >>>>>>>> GUI save >>>>>>>> for twm ), I don't see how a superset is workable >>>>>>>> >>>>>>> >>>>>>> >>>>>>> Thanks - this actually helps clarify something for me :-) >>>>>>> >>>>>>> "It Seams To Me" that the things that are produced by Indiana >>>>>>> really need to be that minimalist core (i.e., enough to >>>>>>> boot into an installer that is network aware in whatever >>>>>>> locale the user needs) AND a set of repositories where >>>>>>> the various packages are to be found. That minimalist core >>>>>>> is then called OpenSolaris. >>>>>>> >>>>>>> The branding for OpenSolaris would then need to allow for >>>>>>> superset distros, so that Sun can create a distro that has >>>>>>> GNOME and NetBeans and Java, and you can create a distro that >>>>>>> has KDE instead. >>>>>>> >>>>>>> On a technical level, the only difference between the two >>>>>>> distros >>>>>>> would seem to be A) the recipe of packages to be installed >>>>>>> by default, and B) the set of packages pre-cached on the >>>>>>> distribution media (so that the default recipe can be installed >>>>>>> without requiring network access...) >>>>>>> >>>>>>> As Shawn said, because of IPS/Repos, the ecosystem is >>>>>>> available for getting new applications AND for satisfying >>>>>>> their install-time dependencies. This would mean that any >>>>>>> OpenSolaris-core containing distro could ultimately run any >>>>>>> Built-for-OpenSolaris application. >>>>>> >>>>>> Putting on my ISV hat for a moment; I think you have to draw >>>>>> the line somewhere. >>>>>> >>>>>> That is, as an application developer, I want to be able to >>>>>> distribute >>>>>> my applications for any OpenSolaris user without triggering a >>>>>> multi-gigabyte download. >>>>>> >>>>>> That means that the core should not be so minimalist that to use >>>>>> almost any basic application that most of a repository needs to >>>>>> be >>>>>> installed. >>>>>> >>>>>> For example, I distribute an Adventure Game Runtime engine for >>>>>> Ubuntu >>>>>> users that relies on all of the libraries being fetched from >>>>>> apt-get, >>>>>> etc. >>>>>> >>>>>> Most users were happy because they either: >>>>>> >>>>>> 1) already had those libraries installed >>>>>> >>>>>> 2) weren't bothered by extra downloads from the repository >>>>>> >>>>>> However, some users were bothered at having to download a lot >>>>>> more, >>>>>> since the < 1mb download suddenly turned into > 10mb. >>>>>> >>>>>> My gut tells me that you have to pick a desktop and a set of >>>>>> libraries >>>>>> to part of core compatibility and that those should be >>>>>> installed by >>>>>> default. >>>>>> >>>>>> OS X, Windows, etc. are successful, in part, precisely because >>>>>> developers can depend upon a rich set of APIs always being >>>>>> available >>>>>> on a user's system. >>>>>> >>>>>> On OS X, for example, you can always depend on the "Core" >>>>>> libraries >>>>>> being present for audio, graphics, etc.. >>>>>> >>>>>> On Windows, you can always count on GDI, DirectDraw, >>>>>> DirectSound, and >>>>>> international text services >>>>>> >>>>>> I guess what I'm getting at is that developers should be able to >>>>>> distribute their applications in such a manner that if they >>>>>> stick with >>>>>> a "core set" of dependencies, the user won't have to download >>>>>> anything >>>>>> additional and they will get a fully functional application. >>>>>> >>>>>> That is my ultimate personal goal in this compatibility thing. >>>>>> >>>>>> Does that mean that certain libraries and environments might get >>>>>> favoured somewhat over another? You're darn right. But that is a >>>>>> benefit to users and ISVs. >>>>>> >>>>>> Cheers, >>>>>> >>>>> >>>>> And then you have to pick a desktop, and then you have to pick >>>>> some >>>>> RDBMSes and then you pick... blah blah, before you know it we've >>>>> got a >>>>> 3-DVD pile of software as the core distro. >>>>> >>>>> It is up to the distribution maintainer to choose a rich set of >>>>> software to include in order that his distro doesn't feel like >>>>> "the >>>>> one where installing basic software takes a long time". As far as >>>>> compatibility in terms of the trademark goes, however, the fact >>>>> that >>>>> they can pull from the same repo is sufficient IMO. >>>>> >>>>> Once again, +1 to Plocher's initial idea. >>>>> >>>> >>>> I agree that 3 DVDs shouldn't be necessary. But I also think a >>>> fairly >>>> rich "core" is useful. >>>> >>>> One of the problems I have with the current trend is sticking >>>> *everything* into the WOS. Automake is probably as good a poster >>>> child >>>> as anything else. ISVs don't need that for software delivery. >>>> >>>> Or, to pick another example, the fact that we seem to have a >>>> bajillion >>>> alternatives -- do we really need to ship 3 or 4 SCMs? (Or do they >>>> really need to be part of the "compatibility core".) >>>> >>>> I think it would be useful for someone to look at pruning down >>>> what is >>>> in the WOS to just the stuff that folks are likely to need to >>>> *use* 3rd >>>> party software. (Not to *develop* it!) Then where "conflicts" >>>> exist >>>> (e.g. different MUAs or different font libraries, or whatever!) >>>> pick >>>> just *one* that gets to be called "core". All the other options >>>> then >>>> need not be part of the requirements for a "compatible" branding, >>>> and it >>>> can then be possible to give ISVs good "advice" about what APIs and >>>> other foundation software they should be building on top of if >>>> they want >>>> to retain maximum compatibility with OpenSolaris distributions. >>> >>> That sums up my view fairly nicely. ISVs and users both will have a >>> better experience when there is a rich, core set of APIs that they >>> can >>> depend on. >> >> And that's great and all, Sun OpenSolaris ( indiana ) should totally >> ship with all that, and they should be in the repo so all "- An >> OpenSolaris Distribution " distros can use it >> >> We're talking about what the core that the trademark requires though, >> so I'm not convinced that forcing certain decisions that impact more >> than just one distribution should be taken lightly like that. >> >> Once again for the folks in the back of the room. If it is in the >> repo, then it can be installed with an ISV's software. It may just >> take longer to install on fluxDiana - An OpenSolaris Distribution, >> and >> the users of fluxDiana should know that. > > There needs to be some compromise though. > > How do you think a user will feel when he's told, "Oh yeah; this is > OpenSolaris compatible, but to use some applications you may need to > download half a gigabyte or more from some repositories because we > don't like some of the libraries." > > That really isn't conducive to a forward-thinking platform that seeks > wide adoption. > > I think the compromise here needs to be that compatibility has to be > based on target audiences. > > I can agree with the idea of a minimal core for the purposes of being > able to carry a tagline, like, "based on OpenSolaris". > > However, if you want to produce something that is actually useful to > ISVs, etc. then you're going to need: > > * OpenSolaris Desktop Compatible > * OpenSolaris Server Compatible > * OpenSolaris Embedded Compatible > > The Desktop one is the one I'm personally the most interested in and > that is going to have to make choices I know you won't like. > > Already I can hear the questions, "Why do we have to make choices?" > > I'll tell you why. > > Developers don't want to be able to choose from 30 different graphics > and input libraries. > > They want *one* very well supported, integrated, performing library > that does what they need. > > Developers don't want to have to support several different sound APIs > (pulse audio, aRTS, eSD, etc.). > > They just want to support one Audio API. > > Not only that, from the perspective of the community, I think it is > far more important to focus on quality than quantity. > > Solaris tends to integrate FOSS software differently than GNU/Linux > distributions do. Whereas most GNU/Linux distributions tend to just > repackage something and "tack it on" to the system; Solaris tends to > modify it to a certain extent to make it feel as though it is actually > part of the system (e.g. providing SMF profiles, etc.). > > Finally, resources, time, and attention are finite. By choosing to > focus on several rich, core APIs for a platform, you can increase > developer productivity and user confidence in your platform through > the resulting focus in resources, stability, etc. You are just wrong. The vast majority of distro maintainers will choose sane things to place in their distro by default (because the distro that requires 2Gb of download for any simple install will not grow a large user base ), and there's no reason to preclude others from the canonical name simply because of a couple design choices, or introduce shovelware to the trademark. Trying to derail trademark discussion in order to ensure no distro but the one you want can possibly exist is not a productive course of action. a nucleic core that allows one to build a distribution with any number of packages in a centralized repo should be sufficient. Particularly seeing as how the GNOME team will make GNOME live in opensolaris with as much integration as possible, and similarly with KDE ( I know stefan's been talking about ACL support in dolphin ). The difference between distros would then be the packages pre-cached on the install medium. I believe this to be a fair compromise that takes all parties in to consideration, and we should just leave it at that. Users will decide which distribution is best integrated, and the others will die out. Even in this scenario, the rising tide lifts all boats, because the same packages that KOpenSolaris improves will be available to GOpenSolaris and FluxOpenSolaris by virtue of a centralized repository that is the definition of compatibility. From johnsonnenschein at gmail.com Sat Feb 23 01:00:58 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 01:00:58 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> Message-ID: On 23-Feb-08, at 12:38 AM, Shawn Walker wrote: > On Fri, Feb 22, 2008 at 6:28 PM, Jim Walker > wrote: >> John Plocher wrote: >>> >>> "It Seams To Me" that the things that are produced by Indiana >>> really need to be that minimalist core (i.e., enough to >>> boot into an installer that is network aware in whatever >>> locale the user needs) AND a set of repositories where >>> the various packages are to be found. That minimalist core >>> is then called OpenSolaris. >> >> I agree. We must find the current minimum set of packages >> that defines an OpenSolaris distribution. Certification >> can proceed once we have that. >> >> However, even though Indiana is very minimal, I'm not >> convinced that it is at the minimum. I think you can >> use even less code to have a very useful OpenSolaris >> distribution (ie. embedded applications). >> >> I consider the OpenSolaris distribution (Indiana) the >> "OpenSolaris Reference Distribution" that targets the >> 80% mark (ie. most OpenSolaris distributions will be >> like it). But, I think the trademark policy project >> team needs to dig even deeper and define a more absolute >> minimum so we hit the 100% mark, so we don't have to >> change what certification means on day one. >> >> Larger distributions are easy once we agree how we >> handle the small ones. > > I would agree that it is the minimum; but the question is how useful > the minimum will actually be. > > I don't think it would be fair or right to take the system down to the > bare minimum and then give everyone using that minimum the right to > have a blanket "OpenSolaris compatible" tagline. Such an indicator > would mean little in practice. Normally, this would be true. However, as this minimum also includes a link to a centralized package repository shared by all "certified compatible" distributions, it is a red herring. For example, RealONEPlayer depends on GTK. I've just installed twmOpenSolaris that includes the absolute bare minimum required for compatibility, which is to say, not a whole lot. User goes to install RealOne from an IPS package. IPS sees the dependencies, fetches them from the central repo ( as this is a compatible distribution, the repo is part of that compatibility ). User gets RealOne just like in any other compatible distro. This is the definition of compatibility. Speed of install has absolutely nothing to do with whether a distro is compatible or not. > > Now if that tagline were "based on OpenSolaris"; that's a different > story. > > Perhaps then, the easiest to create a compatibility trademark usage > guideline for first would be for the server environment. I'm not sure how to parse that sentence ... From Joerg.Schilling at fokus.fraunhofer.de Sat Feb 23 03:41:46 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Sat, 23 Feb 2008 12:41:46 +0100 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF6896.7050106@sun.com> References: <47bdae30.2K1xXcJAc8RW6lf+%Joerg.Schilling@fokus.fraunhofer.de> <47bdbd2a.3qu+nYoaaRBkRGtl%Joerg.Schilling@fokus.fraunhofer.de> <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> Message-ID: <47c0067a.LQ3pl9A+TkRoQEat%Joerg.Schilling@fokus.fraunhofer.de> Jim Walker wrote: > I consider the OpenSolaris distribution (Indiana) the > "OpenSolaris Reference Distribution" that targets the Before we did define the intended guaranteed behavior of the reference, you cannot call it a "reference". 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 John.Plocher at Sun.COM Sat Feb 23 10:17:28 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Sat, 23 Feb 2008 10:17:28 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> Message-ID: <47C06338.9080104@Sun.Com> [**** PLEASE **** learn to trim your replies!] Shawn Walker wrote: >>>>> ... 120 lines deleted ..... >> Once again for the folks in the back of the room. If it is in the >> repo, then it can be installed with an ISV's software. It may just >> take longer to install on fluxDiana - An OpenSolaris Distribution, and >> the users of fluxDiana should know that. > > There needs to be some compromise though. Agreed. But IMO not here. Those choices need to be hashed out and articulated in and by the distro community and presented (thru the ARC? thru the OGB? ...) as a set of Best Practices that we encourage distro makers to follow. Why Best Practices and not Policy? Because it is ultimately up to the Distro producer to cater to their users - if they are happy downloading gigabytes, they the distro will be successful, otherwise it won't. -John From John.Plocher at Sun.COM Sat Feb 23 10:24:09 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Sat, 23 Feb 2008 10:24:09 -0800 Subject: [trademark-policy-dev] Do we need to define a OSOLcommon base package set in the T&B policy? In-Reply-To: References: <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> Message-ID: <47C064C9.90707@Sun.Com> John Sonnenschein wrote: > For example, RealONEPlayer depends on GTK. I've just installed > twmOpenSolaris that includes the absolute bare minimum required for > compatibility, which is to say, not a whole lot. User goes to install > RealOne from an IPS package. IPS sees the dependencies, fetches them > from the central repo ( as this is a compatible distribution, the repo > is part of that compatibility ). User gets RealOne just like in any > other compatible distro. This is exactly what Blastwave and Debian and Uubuntu do today. People grumble a bit the first time the system downloads CSWcommon and friends, and then they get over it. If the distroCG figures out the moral equivalent of CSWcommon for OpenSolaris, I'll bet that *every* distro will choose to include it on their media. Even if they guess wrong, the Repo will have everything else, so mismaches and omissions are easily rectified... End of story :-) -John From John.Plocher at Sun.COM Sat Feb 23 10:33:43 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Sat, 23 Feb 2008 10:33:43 -0800 Subject: [trademark-policy-dev] Uubuntu T&B policy? In-Reply-To: <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> Message-ID: <47C06707.9000403@Sun.Com> S h i v wrote: >> Sun trademarks cannot be included in the name of any third party >> product, technology, program or service. This includes free and >> educational materials, open source distributions, and the titles >> of informational web sites. >> > > My assumption is that the link you posted is the current sun's TM > policy governing the current trademarks and product portfolio. > If this is what opensolaris community needs to abide by going forward, > the discussion becomes redundant. This seems exactly the same as the Uubuntu requirement (below), but not as nicely worded. It is also why I am working on this T&B document to ensure that the things that the community wants to be able to do are allowed. Since SHIV mentioned it, what is the general feeling about the UUBUNTU T&B guidelines? What if we simply used them instead of what we have here? (note that I have *NO* idea whether Sun would be happy doing this) -John Uubuntu's draft T&B policy (which is *really* worth reading) says: http://www.ubuntu.com/aboutus/trademarkpolicy > The Trademarks > > Canonical owns a number of trademarks and these include UBUNTU, > KUBUNTU, EDUBUNTU, and XUBUNTU. The trademarks are registered in > both word and logo form. Any mark ending with the letters UBUNTU > or BUNTU is sufficiently similar to one or more of the trademarks > that permission will be needed in order to use it. This policy > encompasses all marks, in word and logo form, collectively referred > to as ?Trademarks?. and > While there may be exceptions, it is very unlikely that we will > approve Trademark use in the following cases: > > *Use of a Trademark in a company name. > *Use of a Trademark in a domain name.... > *The calling of any software or product by the name UBUNTU > (or another related Trademark), unless that software or product > is a substantially unmodified Ubuntu product, or properly labelled > as a "Remix" as described above. > *Use in combination with any other marks or logos. This include use > of a Trademark in a manner that creates a "combined mark," or use > that integrates other wording with the Trademark in a way that the > public may think of the use as a new mark ... or ... conveys an > impression that the two are tied in some way. From johnsonnenschein at gmail.com Sat Feb 23 10:51:25 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 10:51:25 -0800 Subject: [trademark-policy-dev] Uubuntu T&B policy? In-Reply-To: <47C06707.9000403@Sun.Com> References: <3BD9751B-8BF0-423D-9324-AC3F12FEE913@gmail.com> <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> <47C06707.9000403@Sun.Com> Message-ID: the Ubuntu t&b explicitly states that one cannot build a commercial distribution with the name, I would hope that commercial distributions would be something we'd want to explicitly allow. And commercial advertising to a limited degree ( joyent's "we use opensolaris, it rocks" ad campaign ). (this is, assuming Sun would permit commercial usage in this way) the clause about 'minimal and unsubstantial' changes is unclear. I thing explicitly spelling out what you can and cannot change to retain the compatible mark is important ( that is, " can pull IPS packages off the Sun repository and use them unmodified" ) On 23-Feb-08, at 10:33 AM, John Plocher wrote: > S h i v wrote: >>> Sun trademarks cannot be included in the name of any third party >>> product, technology, program or service. This includes free and >>> educational materials, open source distributions, and the titles >>> of informational web sites. >>> >> >> My assumption is that the link you posted is the current sun's TM >> policy governing the current trademarks and product portfolio. >> If this is what opensolaris community needs to abide by going >> forward, >> the discussion becomes redundant. > > This seems exactly the same as the Uubuntu requirement (below), but > not as > nicely worded. It is also why I am working on this T&B document to > ensure > that the things that the community wants to be able to do are allowed. > > Since SHIV mentioned it, what is the general feeling about the > UUBUNTU T&B > guidelines? What if we simply used them instead of what we have here? > (note that I have *NO* idea whether Sun would be happy doing this) > > -John > > Uubuntu's draft T&B policy (which is *really* worth reading) says: > > http://www.ubuntu.com/aboutus/trademarkpolicy > >> The Trademarks >> >> Canonical owns a number of trademarks and these include UBUNTU, >> KUBUNTU, EDUBUNTU, and XUBUNTU. The trademarks are registered in >> both word and logo form. Any mark ending with the letters UBUNTU >> or BUNTU is sufficiently similar to one or more of the trademarks >> that permission will be needed in order to use it. This policy >> encompasses all marks, in word and logo form, collectively referred >> to as ?Trademarks?. > > and > >> While there may be exceptions, it is very unlikely that we will >> approve Trademark use in the following cases: >> >> *Use of a Trademark in a company name. >> *Use of a Trademark in a domain name.... >> *The calling of any software or product by the name UBUNTU >> (or another related Trademark), unless that software or product >> is a substantially unmodified Ubuntu product, or properly >> labelled >> as a "Remix" as described above. >> *Use in combination with any other marks or logos. This include >> use >> of a Trademark in a manner that creates a "combined mark," or use >> that integrates other wording with the Trademark in a way that >> the >> public may think of the use as a new mark ... or ... conveys an >> impression that the two are tied in some way. > _______________________________________________ > trademark-policy-dev mailing list > trademark-policy-dev at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/trademark-policy-dev From swalker at opensolaris.org Sat Feb 23 11:38:41 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 13:38:41 -0600 Subject: [trademark-policy-dev] [ogb-discuss] "Open-SunOS" for OS/Net src base? __/__ Re: Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <319ee2b10802221930y3c895b8ema2de6aadea90cd33@mail.gmail.com> References: <47BEE439.10104@Sun.Com> <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <47BF21D8.9090205@Sun.Com> <241540330802221135x6ac725a9mdc030853baa36f86@mail.gmail.com> <319ee2b10802221930y3c895b8ema2de6aadea90cd33@mail.gmail.com> Message-ID: On Fri, Feb 22, 2008 at 9:30 PM, S h i v wrote: > On Sat, Feb 23, 2008 at 1:05 AM, John Sonnenschein > wrote: > > Okay, another question then. > > > > What subset of packages are required to qualify for compatibility? > > > > Say for example I build a distro that only includes SUNWips, > > SUNWos-net ( or whatever ON will be called ) and the rest are pulled > > from another repo ( blastwave, for example ). > > > > And lets say this distro builds on top of a whole slew of weird > > packages that are mostly incompatible with anything that indiana > > includes ( say... the freebsd userspace ) compiled with a compiler > > that makes ABI compatibility an impossibility ( like gcc 2.95 ) > > > > Is it still " BreakFOO - an OpenSolaris Distribution " ? technically > > you can use IPS to pull in all the indiana stuff and make it > > compatible, and it includes a subset of packages that indiana does, > > but in and of itself it is not. > > > > It is just a matter of time before OpenSolaris is equated to Indiana > and not ON. History of the term is then forgotten. > When that happens any other community distro even those at OS.o may > not want to use the TM term to avoid confusion. With a product to use, > world associates the term with the product and the community is > assumed to revolve around it. > > Cut & slice whichever way, the result is a given. Changing the > fine-print will not help. People act like Indiana is the hugely different thing. It really isn't. It's essentially just ON + IPS + configuration + a lot of hard work + and a few component changes I think it proves that you can have a radical increase in usability and perception without diverging very far from "canon." -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From justin at erenkrantz.com Sat Feb 23 12:22:10 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Sat, 23 Feb 2008 12:22:10 -0800 Subject: [trademark-policy-dev] Uubuntu T&B policy? In-Reply-To: <47C06707.9000403@Sun.Com> References: <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> <47C06707.9000403@Sun.Com> Message-ID: <5c902b9e0802231222y38c2e259h339efc09a42ed841@mail.gmail.com> On Sat, Feb 23, 2008 at 10:33 AM, John Plocher wrote: > Since SHIV mentioned it, what is the general feeling about the UUBUNTU T&B > guidelines? What if we simply used them instead of what we have here? > (note that I have *NO* idea whether Sun would be happy doing this) It is a very well-thought out policy, but in some cases, I think it's a bit stricter than the current draft (commercial distributions) and a bit looser in other places (remixes). Yet, outright adoption of their T&B policy (with s/Ubuntu/OpenSolaris/ and s/Canonical/Sun/ and appropriate attribution) would be a nice way of not reinventing the wheel and a good way of acknowledging positive actions from other projects. So, I would be happy to see that policy adopted outright even if it's not a 100% fit - but a 95% fit for the community. If that 5% bugs us later, the policy can be revised then. -- justin From johnsonnenschein at gmail.com Sat Feb 23 12:36:19 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 12:36:19 -0800 Subject: [trademark-policy-dev] Uubuntu T&B policy? In-Reply-To: <5c902b9e0802231222y38c2e259h339efc09a42ed841@mail.gmail.com> References: <8E48A7E9-DA94-4399-907E-B0A2A51FB4DB@gmail.com> <47BF1377.4090700@Sun.Com> <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> <47C06707.9000403@Sun.Com> <5c902b9e0802231222y38c2e259h339efc09a42ed841@mail.gmail.com> Message-ID: On 23-Feb-08, at 12:22 PM, Justin Erenkrantz wrote: > On Sat, Feb 23, 2008 at 10:33 AM, John Plocher > wrote: >> Since SHIV mentioned it, what is the general feeling about the >> UUBUNTU T&B >> guidelines? What if we simply used them instead of what we have >> here? >> (note that I have *NO* idea whether Sun would be happy doing this) > > It is a very well-thought out policy, but in some cases, I think it's > a bit stricter than the current draft (commercial distributions) and a > bit looser in other places (remixes). Yet, outright adoption of their > T&B policy (with s/Ubuntu/OpenSolaris/ and s/Canonical/Sun/ and > appropriate attribution) would be a nice way of not reinventing the > wheel and a good way of acknowledging positive actions from other > projects. So, I would be happy to see that policy adopted outright > even if it's not a 100% fit - but a 95% fit for the community. If > that 5% bugs us later, the policy can be revised then. -- justin -1 The ubuntu policy is far more restrictive than I think we ought to be, even not taking in to account the ability for commercial distributions. I'm going to dig my heels in about John P.'s new proposal ( ie, nucleus + IPS + repo ) because I think it precisely captures the compatibility we're going for, while still allowing distro maintainers a lot of freedom to innovate without having to sacrifice the ability to use the trademark in a useful way From swalker at opensolaris.org Sat Feb 23 12:38:26 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 14:38:26 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> Message-ID: On Sat, Feb 23, 2008 at 2:54 AM, John Sonnenschein wrote: > On 23-Feb-08, at 12:34 AM, Shawn Walker wrote: > > There needs to be some compromise though. > > > > How do you think a user will feel when he's told, "Oh yeah; this is > > OpenSolaris compatible, but to use some applications you may need to > > download half a gigabyte or more from some repositories because we > > don't like some of the libraries." > > > > That really isn't conducive to a forward-thinking platform that seeks > > wide adoption. > > > > I think the compromise here needs to be that compatibility has to be > > based on target audiences. > > > > I can agree with the idea of a minimal core for the purposes of being > > able to carry a tagline, like, "based on OpenSolaris". > > > > However, if you want to produce something that is actually useful to > > ISVs, etc. then you're going to need: > > > > * OpenSolaris Desktop Compatible > > * OpenSolaris Server Compatible > > * OpenSolaris Embedded Compatible > > > > The Desktop one is the one I'm personally the most interested in and > > that is going to have to make choices I know you won't like. > > > > Already I can hear the questions, "Why do we have to make choices?" > > > > I'll tell you why. > > > > Developers don't want to be able to choose from 30 different graphics > > and input libraries. > > > > They want *one* very well supported, integrated, performing library > > that does what they need. > > > > Developers don't want to have to support several different sound APIs > > (pulse audio, aRTS, eSD, etc.). > > > > They just want to support one Audio API. > > > > Not only that, from the perspective of the community, I think it is > > far more important to focus on quality than quantity. > > > > Solaris tends to integrate FOSS software differently than GNU/Linux > > distributions do. Whereas most GNU/Linux distributions tend to just > > repackage something and "tack it on" to the system; Solaris tends to > > modify it to a certain extent to make it feel as though it is actually > > part of the system (e.g. providing SMF profiles, etc.). > > > > Finally, resources, time, and attention are finite. By choosing to > > focus on several rich, core APIs for a platform, you can increase > > developer productivity and user confidence in your platform through > > the resulting focus in resources, stability, etc. > > You are just wrong. How am I wrong? Prove it please. I've worked at ISVs for several years; and have been one personally. As a result, I'm fairly certain my experience proves otherwise. > The vast majority of distro maintainers will choose sane things to > place in their distro by default (because the distro that requires 2Gb > of download for any simple install will not grow a large user base ), > and there's no reason to preclude others from the canonical name > simply because of a couple design choices, or introduce shovelware to > the trademark. No one has suggested any such thing. You seem to have a different perception though. You also continue to enjoy using derogatory terms to describe software without any justification or explanation as to what that means. Why is that you must use the belittlement of other people's work to support your thoughts on this? > Trying to derail trademark discussion in order to ensure no distro but > the one you want can possibly exist is not a productive course of > action. If I was actually doing that; I would agree. However, I am not despite your assertions to the contrary. > a nucleic core that allows one to build a distribution with any number > of packages in a centralized repo should be sufficient. Particularly > seeing as how the GNOME team will make GNOME live in opensolaris with > as much integration as possible, and similarly with KDE ( I know > stefan's been talking about ACL support in dolphin ). The difference > between distros would then be the packages pre-cached on the install > medium. Again, you seem to assume that I am choosing GNOME or KDE or that I have mentioned any such specific thing. I have not. > I believe this to be a fair compromise that takes all parties in to > consideration, and we should just leave it at that. Users will decide Fair is as far as the beholder believes; in this case neither of us seems to agree on what is fair. > which distribution is best integrated, and the others will die out. > Even in this scenario, the rising tide lifts all boats, because the > same packages that KOpenSolaris improves will be available to > GOpenSolaris and FluxOpenSolaris by virtue of a centralized repository > that is the definition of compatibility. As software designers and engineers; it is our responsibility to make the best choices for users while offering a compromise of flexibility. Apple is routinely criticised for not allowing user choice in their products; yet, they are wildly, commercially successful. I think the strategy speaks for itself. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Sat Feb 23 12:39:56 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 14:39:56 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47C06338.9080104@Sun.Com> References: <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> <47C06338.9080104@Sun.Com> Message-ID: On Sat, Feb 23, 2008 at 12:17 PM, John Plocher wrote: > [**** PLEASE **** learn to trim your replies!] Sorry, my webmail client hides most of the sections from me by default :) > Shawn Walker wrote: > >>>>> ... 120 lines deleted ..... > > >> Once again for the folks in the back of the room. If it is in the > >> repo, then it can be installed with an ISV's software. It may just > >> take longer to install on fluxDiana - An OpenSolaris Distribution, and > >> the users of fluxDiana should know that. > > > > There needs to be some compromise though. > > Agreed. > > But IMO not here. > > Those choices need to be hashed out and articulated in and by the distro > community and presented (thru the ARC? thru the OGB? ...) as a set of > Best Practices that we encourage distro makers to follow. > > Why Best Practices and not Policy? Because it is ultimately up to the Distro > producer to cater to their users - if they are happy downloading gigabytes, > they the distro will be successful, otherwise it won't. I don't agree. How can you allow distributions to claim compatibility when the compatibility is merely defined as "best pratices" ? This just doesn't sit right in my gut at all. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Sat Feb 23 12:44:30 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 14:44:30 -0600 Subject: [trademark-policy-dev] Do we need to define a OSOLcommon base package set in the T&B policy? In-Reply-To: <47C064C9.90707@Sun.Com> References: <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C064C9.90707@Sun.Com> Message-ID: On Sat, Feb 23, 2008 at 12:24 PM, John Plocher wrote: > John Sonnenschein wrote: > > For example, RealONEPlayer depends on GTK. I've just installed > > twmOpenSolaris that includes the absolute bare minimum required for > > compatibility, which is to say, not a whole lot. User goes to install > > RealOne from an IPS package. IPS sees the dependencies, fetches them > > from the central repo ( as this is a compatible distribution, the repo > > is part of that compatibility ). User gets RealOne just like in any > > other compatible distro. > > > This is exactly what Blastwave and Debian and Uubuntu do today. People > grumble a bit the first time the system downloads CSWcommon and friends, > and then they get over it. My problem with that is the assumption that everyone has free, unlimited bandwidth. By ensuring that core platform software is always distributed in certain cases, you ensure that the experience of users remains good. > If the distroCG figures out the moral equivalent of CSWcommon for OpenSolaris, > I'll bet that *every* distro will choose to include it on their media. I dearly hope so. > Even if they guess wrong, the Repo will have everything else, so mismaches > and omissions are easily rectified... I'm not entirely convinced just yet. Fedora tried to take the same approach with the repository bit and they did run into some problems. Namely, with multiple community repositories, people tend to want to recompile certain packages or libraries with different flags. This ends up putting the user in package dependency hell because they can never pick and choose software they want, they have to choose between one entire repository or another. Again, my concern here is compatibility. I'm still not clear on what is being agreed on here. Does compatibility just mean "has a core platform set available in repositories if needed" or does it also ensure a consistent quality and level of user experience for each distribution? These are not easy questions to answer. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Sat Feb 23 12:46:53 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 12:46:53 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> <47C06338.9080104@Sun.Com> Message-ID: <241540330802231246k156b2eb5we8fe06afb2cf0bbe@mail.gmail.com> On Sat, Feb 23, 2008 at 12:39 PM, Shawn Walker wrote: > On Sat, Feb 23, 2008 at 12:17 PM, John Plocher wrote: > > Those choices need to be hashed out and articulated in and by the distro > > community and presented (thru the ARC? thru the OGB? ...) as a set of > > Best Practices that we encourage distro makers to follow. > > > > Why Best Practices and not Policy? Because it is ultimately up to the Distro > > producer to cater to their users - if they are happy downloading gigabytes, > > they the distro will be successful, otherwise it won't. > > I don't agree. How can you allow distributions to claim compatibility > when the compatibility is merely defined as "best pratices" ? > > This just doesn't sit right in my gut at all. Because you continue to ignore the fact that part of compatibility we are talking about here involves the availability of the package repository. ISVreally-cool-software depends on SUNWgnome-libs ... I'm using twmOpenSolaris. Installing ISVreally-cool-software pulls in SUNWgnome-libs. That's compatibility. By definition. If twmOpenSolaris is a pain to use because installing any third party stuff involves a 200MB download, then people won't use it. This trademark policy involves indiana not at all, except to say the repository of software that indiana pulls from is part of what is required for compatibility. Indiana can include whatever packages it needs to be the best most comfortable user experience it can, and that's great. However, if I want to build a distro that's uncomfortable to use for whatever reason, so long as an ISV can distribute their software as an IPS package and expect it to work, that ought to be perfectly fine in terms of trademark policy. Indiana will still carry the canonical " OpenSolaris " name, and there's nothing we can do about that regardless of if we want to or not, and it will continue to be a sort of reference distribution. -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From johnsonnenschein at gmail.com Sat Feb 23 12:51:58 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 12:51:58 -0800 Subject: [trademark-policy-dev] Do we need to define a OSOLcommon base package set in the T&B policy? In-Reply-To: References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C064C9.90707@Sun.Com> Message-ID: <241540330802231251h2ec3bbe5ge5a97a0ee4d28c75@mail.gmail.com> On Sat, Feb 23, 2008 at 12:44 PM, Shawn Walker wrote: > On Sat, Feb 23, 2008 at 12:24 PM, John Plocher wrote: > > John Sonnenschein wrote: > > > For example, RealONEPlayer depends on GTK. I've just installed > > > twmOpenSolaris that includes the absolute bare minimum required for > > > compatibility, which is to say, not a whole lot. User goes to install > > > RealOne from an IPS package. IPS sees the dependencies, fetches them > > > from the central repo ( as this is a compatible distribution, the repo > > > is part of that compatibility ). User gets RealOne just like in any > > > other compatible distro. > > > > > > This is exactly what Blastwave and Debian and Uubuntu do today. People > > grumble a bit the first time the system downloads CSWcommon and friends, > > and then they get over it. > > My problem with that is the assumption that everyone has free, > unlimited bandwidth. By ensuring that core platform software is always > distributed in certain cases, you ensure that the experience of users > remains good. > Trademark policy says nothing of the users. It does not concern them. Trademark policy is an agreement between Sun and distribution maintainers on the use of the name. Poor distributions will simply not gain enough user traction to cause a problem > > If the distroCG figures out the moral equivalent of CSWcommon for OpenSolaris, > > I'll bet that *every* distro will choose to include it on their media. > > I dearly hope so. Me too, but they ought not be /required/ to carry it on the install medium. ( especially if you want embedded distributions to be allowed to carry the trademark... there's no reason whatsoever they ought to carry X11 + GTK + Gnome and the whole bit when all they need is the kernel and a few other things ) > > > Even if they guess wrong, the Repo will have everything else, so mismaches > > and omissions are easily rectified... > > I'm not entirely convinced just yet. Fedora tried to take the same > approach with the repository bit and they did run into some problems. > > Namely, with multiple community repositories, people tend to want to > recompile certain packages or libraries with different flags. This > ends up putting the user in package dependency hell because they can > never pick and choose software they want, they have to choose between > one entire repository or another. So the solution then is to define the trademark in terms of the package repo that is immutable. That is, if weirdnessOpenSolaris wants to completely replace the userland or some other such thing, their repo cannot conflict with the osol repo ( they must install their software to /opt, or somewhere else where it won't cause problems ) > Again, my concern here is compatibility. I'm still not clear on what > is being agreed on here. > > Does compatibility just mean "has a core platform set available in > repositories if needed" or does it also ensure a consistent quality > and level of user experience for each distribution? > > These are not easy questions to answer. sure it is, the consistent user experience isn't compatibility. It may be a good thing to have, but that's not binary compatibility, and therefore is not the domain of the trademark policy. -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From John.Plocher at Sun.COM Sat Feb 23 12:54:56 2008 From: John.Plocher at Sun.COM (John Plocher) Date: Sat, 23 Feb 2008 12:54:56 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> <47C06338.9080104@Sun.Com> Message-ID: <47C08820.6080909@Sun.Com> Shawn Walker wrote: > I don't agree. How can you allow distributions to claim compatibility > when the compatibility is merely defined as "best pratices" ? Compatibility is "can download and run stuff from repos". Not "has distribution media that contains cached versions of some set of packages in the repo" If the Distro CG comes up with a OSOLcommon that it feels SHOULD be a requirement, then I expect it to present it (to ARC, OGB, whatever) as a Policy, where it will get adopted, and probably cause an update to the T&B Policy to reflect the new information. Meanwhile, I think the above definition is sufficient to get us moving in the right direction, with a good expectations that it isn't actually harmful to our goals. -John From johnsonnenschein at gmail.com Sat Feb 23 13:04:04 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 13:04:04 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47C08820.6080909@Sun.Com> References: <241540330802221445q48bcfab1m31e7d3eea5eb4755@mail.gmail.com> <47BF5438.6090102@opensolaris.org> <241540330802221514n398aa942ib074ac0e362f7222@mail.gmail.com> <47C06338.9080104@Sun.Com> <47C08820.6080909@Sun.Com> Message-ID: <241540330802231304x25b0150dn7d08b7b0229a1c20@mail.gmail.com> +1 we can fight over OSOLcommon later. For now, this seems to work sufficiently well to keep all parties on the same page On Sat, Feb 23, 2008 at 12:54 PM, John Plocher wrote: > Shawn Walker wrote: > > I don't agree. How can you allow distributions to claim compatibility > > when the compatibility is merely defined as "best pratices" ? > > > Compatibility is "can download and run stuff from repos". > > Not "has distribution media that contains cached versions of some set of packages in the repo" > > If the Distro CG comes up with a OSOLcommon that it feels SHOULD be > a requirement, then I expect it to present it (to ARC, OGB, whatever) > as a Policy, where it will get adopted, and probably cause an update > to the T&B Policy to reflect the new information. > > Meanwhile, I think the above definition is sufficient to get us moving > in the right direction, with a good expectations that it isn't actually > harmful to our goals. > > -John > > > _______________________________________________ > trademark-policy-dev mailing list > trademark-policy-dev at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/trademark-policy-dev > -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From justin at erenkrantz.com Sat Feb 23 13:36:21 2008 From: justin at erenkrantz.com (Justin Erenkrantz) Date: Sat, 23 Feb 2008 13:36:21 -0800 Subject: [trademark-policy-dev] Uubuntu T&B policy? In-Reply-To: References: <241540330802221029p2506f2ag7f854273b59fd7b5@mail.gmail.com> <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> <47C06707.9000403@Sun.Com> <5c902b9e0802231222y38c2e259h339efc09a42ed841@mail.gmail.com> Message-ID: <5c902b9e0802231336h1b067608v109c70ac40bfcf4@mail.gmail.com> On Sat, Feb 23, 2008 at 12:36 PM, John Sonnenschein wrote: > -1 > > The ubuntu policy is far more restrictive than I think we ought to be, > even not taking in to account the ability for commercial distributions. My initial reaction was similar, but I then looked at the current T&B policy draft and, based on my reading, I don't think Sun would permit commercial usage either for anything that used the "based on OpenSolaris" or similar marks. If Sun wishes to permit such usage, then they can strike those phrases from Ubuntu's policy and replace it with wording that such usage is permitted. But, my understanding is that Sun doesn't wish to permit such usage. I'd be happy to be corrected on this. -- justin From Glynn.Foster at Sun.COM Sat Feb 23 13:48:22 2008 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Sun, 24 Feb 2008 10:48:22 +1300 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47BF6896.7050106@sun.com> References: <47bdae30.2K1xXcJAc8RW6lf+%Joerg.Schilling@fokus.fraunhofer.de> <47bdbd2a.3qu+nYoaaRBkRGtl%Joerg.Schilling@fokus.fraunhofer.de> <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> Message-ID: <47C094A6.8010802@sun.com> Hey, Jim Walker wrote: > John Plocher wrote: >> "It Seams To Me" that the things that are produced by Indiana >> really need to be that minimalist core (i.e., enough to >> boot into an installer that is network aware in whatever >> locale the user needs) AND a set of repositories where >> the various packages are to be found. That minimalist core >> is then called OpenSolaris. > > I agree. We must find the current minimum set of packages > that defines an OpenSolaris distribution. Certification > can proceed once we have that. I actually disagree to some extents. I don't think there has to be a minimum set of packages defining a core. I think the repository is that core. Any derivation based on a selection of packages from that repository (or set of 'official' repositories) should be compatible. All distributions *must* have that repository first on its priority list, and then have the ability to install from separate repositories after. The emphasis and focus should be encouraging people to contribute directly to that repository IMO, and make it a better place for everyone. Glynn From johnsonnenschein at gmail.com Sat Feb 23 13:54:31 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 13:54:31 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47C094A6.8010802@sun.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> Message-ID: <241540330802231354w5698cc5foa7803c83ab57736c@mail.gmail.com> On Sat, Feb 23, 2008 at 1:48 PM, Glynn Foster wrote: > Hey, > > > Jim Walker wrote: > > John Plocher wrote: > >> "It Seams To Me" that the things that are produced by Indiana > >> really need to be that minimalist core (i.e., enough to > >> boot into an installer that is network aware in whatever > >> locale the user needs) AND a set of repositories where > >> the various packages are to be found. That minimalist core > >> is then called OpenSolaris. > > > > I agree. We must find the current minimum set of packages > > that defines an OpenSolaris distribution. Certification > > can proceed once we have that. > > I actually disagree to some extents. I don't think there has to be a minimum set > of packages defining a core. I think the repository is that core. Any derivation > based on a selection of packages from that repository (or set of 'official' > repositories) should be compatible. > > All distributions *must* have that repository first on its priority list, and > then have the ability to install from separate repositories after. The emphasis > and focus should be encouraging people to contribute directly to that repository > IMO, and make it a better place for everyone. > +1 This is precisely what I'm advocating for, and what I assume JohnP means. ( given the allowance for building a distro value-add repositories on top of the offical one, eg indiana + NONUSlibdvdread ). -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From Glynn.Foster at Sun.COM Sat Feb 23 14:19:17 2008 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Sun, 24 Feb 2008 11:19:17 +1300 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47C094A6.8010802@sun.com> References: <47bdae30.2K1xXcJAc8RW6lf+%Joerg.Schilling@fokus.fraunhofer.de> <47bdbd2a.3qu+nYoaaRBkRGtl%Joerg.Schilling@fokus.fraunhofer.de> <47BDF210.4030905@Sun.Com> <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> Message-ID: <47C09BE5.8050302@sun.com> Glynn Foster wrote: > All distributions *must* have that repository first on its priority list, and > then have the ability to install from separate repositories after. The emphasis > and focus should be encouraging people to contribute directly to that repository > IMO, and make it a better place for everyone. Having thought about this a little more (sorry, I know I'm coming late to the game for most), here's roughly what I have in my head - OpenSolaris Distribution Derivations made purely from a single official OpenSolaris repository only ie. this could include a GNOME or KDE based derivation *providing* those packages were in the repository, or an embedded OS for example OpenSolaris Compatible Derivations made from a single official OpenSolaris repository *and* additional 3rd party repositories. The official repository would be first in the search path. An install would become 'OpenSolaris Compatible' the moment an additional IPS repository was added to an original OpenSolaris Distribution install. Built on OpenSolaris Any package that was built on either an OpenSolaris Distribution or OpenSolaris Compatible one. OpenSolaris Inside (or similar?) Any binary based on OpenSolaris source code I think the real use issues are around why people want to create derivations, and what they may want to change. Branding and configuration are 2 that come to mind - in this case, it's likely to be additional packages that provide branding content, and tweak configuration, and end up being 'OpenSolaris Compatible' rather than the 'OpenSolaris Distribution' itself. Glynn From johnsonnenschein at gmail.com Sat Feb 23 14:25:31 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 14:25:31 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47C09BE5.8050302@sun.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> Message-ID: <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> On Sat, Feb 23, 2008 at 2:19 PM, Glynn Foster wrote: > > > Glynn Foster wrote: > > All distributions *must* have that repository first on its priority list, and > > then have the ability to install from separate repositories after. The emphasis > > and focus should be encouraging people to contribute directly to that repository > > IMO, and make it a better place for everyone. > > Having thought about this a little more (sorry, I know I'm coming late to the > game for most), here's roughly what I have in my head - > > > OpenSolaris Distribution > > Derivations made purely from a single official OpenSolaris > repository only ie. this could include a GNOME or KDE based > derivation *providing* those packages were in the repository, > or an embedded OS for example > > OpenSolaris Compatible > > Derivations made from a single official OpenSolaris repository > *and* additional 3rd party repositories. The official repository > would be first in the search path. An install would become > 'OpenSolaris Compatible' the moment an additional IPS repository > was added to an original OpenSolaris Distribution install. Does this include for example, packages that would never and can never exist in the official repository, such as for example libdvdcss, which is not legal for distribution within the US (and therefore sun can't distribute it ) > Built on OpenSolaris > > Any package that was built on either an OpenSolaris Distribution > or OpenSolaris Compatible one. > > OpenSolaris Inside (or similar?) > > Any binary based on OpenSolaris source code > > > I think the real use issues are around why people want to create derivations, > and what they may want to change. Branding and configuration are 2 that come to > mind - in this case, it's likely to be additional packages that provide branding > content, and tweak configuration, and end up being 'OpenSolaris Compatible' > rather than the 'OpenSolaris Distribution' itself. Aside from that one point (and the "opensolaris inside" name ;) I still like my "- An OpenSolaris Technology" tag) I agree with you also the matter of non-compatible non-IPS capable distributions such as nexenta ( " - built with opensolaris technology " seems okay to me... thoughts ? ) -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From Glynn.Foster at Sun.COM Sat Feb 23 14:36:38 2008 From: Glynn.Foster at Sun.COM (Glynn Foster) Date: Sun, 24 Feb 2008 11:36:38 +1300 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> Message-ID: <47C09FF6.7050205@sun.com> Hey, John Sonnenschein wrote: >> OpenSolaris Compatible >> >> Derivations made from a single official OpenSolaris repository >> *and* additional 3rd party repositories. The official repository >> would be first in the search path. An install would become >> 'OpenSolaris Compatible' the moment an additional IPS repository >> was added to an original OpenSolaris Distribution install. > > Does this include for example, packages that would never and can never > exist in the official repository, such as for example libdvdcss, which > is not legal for distribution within the US (and therefore sun can't > distribute it ) Yes - there may never be an official OpenSolaris Distribution that can support non-open formats IMO. Glynn From johnsonnenschein at gmail.com Sat Feb 23 14:45:13 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 14:45:13 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47C09FF6.7050205@sun.com> References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> <47C09FF6.7050205@sun.com> Message-ID: <241540330802231445x70a02ec3mee0103ab990434fc@mail.gmail.com> On Sat, Feb 23, 2008 at 2:36 PM, Glynn Foster wrote: > Hey, > > > John Sonnenschein wrote: > >> OpenSolaris Compatible > >> > >> Derivations made from a single official OpenSolaris repository > >> *and* additional 3rd party repositories. The official repository > >> would be first in the search path. An install would become > >> 'OpenSolaris Compatible' the moment an additional IPS repository > >> was added to an original OpenSolaris Distribution install. > > > > Does this include for example, packages that would never and can never > > exist in the official repository, such as for example libdvdcss, which > > is not legal for distribution within the US (and therefore sun can't > > distribute it ) > > Yes - there may never be an official OpenSolaris Distribution that can support > non-open formats IMO. Okay, as long as that's clarified. In that case, +1 to your penultimate post in it's entirety. -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From Joerg.Schilling at fokus.fraunhofer.de Sat Feb 23 14:45:24 2008 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Sat, 23 Feb 2008 23:45:24 +0100 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> Message-ID: <47c0a204.hgpYSffhVGR9fv/S%Joerg.Schilling@fokus.fraunhofer.de> "John Sonnenschein" wrote: > > OpenSolaris Compatible > > > > Derivations made from a single official OpenSolaris repository > > *and* additional 3rd party repositories. The official repository > > would be first in the search path. An install would become > > 'OpenSolaris Compatible' the moment an additional IPS repository > > was added to an original OpenSolaris Distribution install. > > Does this include for example, packages that would never and can never > exist in the official repository, such as for example libdvdcss, which > is not legal for distribution within the US (and therefore sun can't > distribute it ) It of course can be supported by OpenSolaris as you just need to load/install the related package from a server in Europe. 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 johnsonnenschein at gmail.com Sat Feb 23 14:46:50 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 14:46:50 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47c0a204.hgpYSffhVGR9fv/S%Joerg.Schilling@fokus.fraunhofer.de> References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> <47c0a204.hgpYSffhVGR9fv/S%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <241540330802231446u570147b1p2896a994a4c7f709@mail.gmail.com> On Sat, Feb 23, 2008 at 2:45 PM, Joerg Schilling wrote: > "John Sonnenschein" wrote: > > > > > OpenSolaris Compatible > > > > > > Derivations made from a single official OpenSolaris repository > > > *and* additional 3rd party repositories. The official repository > > > would be first in the search path. An install would become > > > 'OpenSolaris Compatible' the moment an additional IPS repository > > > was added to an original OpenSolaris Distribution install. > > > > Does this include for example, packages that would never and can never > > exist in the official repository, such as for example libdvdcss, which > > is not legal for distribution within the US (and therefore sun can't > > distribute it ) > > It of course can be supported by OpenSolaris as you just need to load/install > the related package from a server in Europe. Indeed, but then to pre-cache the package on a distribution would seem to make it an "- OpenSolaris Compatible " distribution, rather than being allowed to carry the canonical name "OpenSolaris" unadorned -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Sat Feb 23 15:58:54 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 17:58:54 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802231354w5698cc5foa7803c83ab57736c@mail.gmail.com> References: <18366.56532.244094.778528@gargle.gargle.HOWL> <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <241540330802231354w5698cc5foa7803c83ab57736c@mail.gmail.com> Message-ID: On Sat, Feb 23, 2008 at 3:54 PM, John Sonnenschein wrote: > On Sat, Feb 23, 2008 at 1:48 PM, Glynn Foster wrote: > > Hey, > > > > > > Jim Walker wrote: > > > John Plocher wrote: > > >> "It Seams To Me" that the things that are produced by Indiana > > >> really need to be that minimalist core (i.e., enough to > > >> boot into an installer that is network aware in whatever > > >> locale the user needs) AND a set of repositories where > > >> the various packages are to be found. That minimalist core > > >> is then called OpenSolaris. > > > > > > I agree. We must find the current minimum set of packages > > > that defines an OpenSolaris distribution. Certification > > > can proceed once we have that. > > > > I actually disagree to some extents. I don't think there has to be a minimum set > > of packages defining a core. I think the repository is that core. Any derivation > > based on a selection of packages from that repository (or set of 'official' > > repositories) should be compatible. > > > > All distributions *must* have that repository first on its priority list, and > > then have the ability to install from separate repositories after. The emphasis > > and focus should be encouraging people to contribute directly to that repository > > IMO, and make it a better place for everyone. > > > > +1 > > This is precisely what I'm advocating for, and what I assume JohnP > means. ( given the allowance for building a distro value-add > repositories on top of the offical one, eg indiana + NONUSlibdvdread > ). I think I finally understand how I crossed wires earlier. I was under the mistaken impression that this minimum set was going to affect what was expected to be in the repositories and what was required for compatibility. If the minimum set is just the minimum to be able to use the trademark, and then the rest of the repository access is required for compatibility, that would be agreeable to me. I think I'm finally in agreement here. Thanks for your patience. -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Sat Feb 23 15:59:48 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 17:59:48 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <47c0a204.hgpYSffhVGR9fv/S%Joerg.Schilling@fokus.fraunhofer.de> References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> <47c0a204.hgpYSffhVGR9fv/S%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: On Sat, Feb 23, 2008 at 4:45 PM, Joerg Schilling wrote: > "John Sonnenschein" wrote: > > > > > OpenSolaris Compatible > > > > > > Derivations made from a single official OpenSolaris repository > > > *and* additional 3rd party repositories. The official repository > > > would be first in the search path. An install would become > > > 'OpenSolaris Compatible' the moment an additional IPS repository > > > was added to an original OpenSolaris Distribution install. > > > > Does this include for example, packages that would never and can never > > exist in the official repository, such as for example libdvdcss, which > > is not legal for distribution within the US (and therefore sun can't > > distribute it ) > > It of course can be supported by OpenSolaris as you just need to load/install > the related package from a server in Europe. The one problem will be software that doesn't support such a configuration at runtime. Some software requires that you compile it with certain options enabled. I believe such software should be fixed to support runtime loading in those cases so that we don't have to recompile "supported" packages just so such things can be enabled. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Sat Feb 23 16:00:30 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 16:00:30 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BF1138.6060605@Sun.Com> <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <241540330802231354w5698cc5foa7803c83ab57736c@mail.gmail.com> Message-ID: <241540330802231600u7e29f082u54bc4646e3119317@mail.gmail.com> On Sat, Feb 23, 2008 at 3:58 PM, Shawn Walker wrote: > > On Sat, Feb 23, 2008 at 3:54 PM, John Sonnenschein > wrote: > > On Sat, Feb 23, 2008 at 1:48 PM, Glynn Foster wrote: > > > Hey, > > > > > > > > > Jim Walker wrote: > > > > John Plocher wrote: > > > >> "It Seams To Me" that the things that are produced by Indiana > > > >> really need to be that minimalist core (i.e., enough to > > > >> boot into an installer that is network aware in whatever > > > >> locale the user needs) AND a set of repositories where > > > >> the various packages are to be found. That minimalist core > > > >> is then called OpenSolaris. > > > > > > > > I agree. We must find the current minimum set of packages > > > > that defines an OpenSolaris distribution. Certification > > > > can proceed once we have that. > > > > > > I actually disagree to some extents. I don't think there has to be a minimum set > > > of packages defining a core. I think the repository is that core. Any derivation > > > based on a selection of packages from that repository (or set of 'official' > > > repositories) should be compatible. > > > > > > All distributions *must* have that repository first on its priority list, and > > > then have the ability to install from separate repositories after. The emphasis > > > and focus should be encouraging people to contribute directly to that repository > > > IMO, and make it a better place for everyone. > > > > > > > +1 > > > > This is precisely what I'm advocating for, and what I assume JohnP > > means. ( given the allowance for building a distro value-add > > repositories on top of the offical one, eg indiana + NONUSlibdvdread > > ). > > I think I finally understand how I crossed wires earlier. > > I was under the mistaken impression that this minimum set was going to > affect what was expected to be in the repositories and what was > required for compatibility. > > If the minimum set is just the minimum to be able to use the > trademark, and then the rest of the repository access is required for > compatibility, that would be agreeable to me. > > I think I'm finally in agreement here. Thanks for your patience. Great! So it seems like everyone's on board. That's fabulous ( and considering our community dynamic, a once in a lifetime happening ) -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Sat Feb 23 16:07:45 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 18:07:45 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802231600u7e29f082u54bc4646e3119317@mail.gmail.com> References: <47bf2444.iGQpG+R+Hi0N9gDx%Joerg.Schilling@fokus.fraunhofer.de> <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <241540330802231354w5698cc5foa7803c83ab57736c@mail.gmail.com> <241540330802231600u7e29f082u54bc4646e3119317@mail.gmail.com> Message-ID: On Sat, Feb 23, 2008 at 6:00 PM, John Sonnenschein wrote: > > On Sat, Feb 23, 2008 at 3:58 PM, Shawn Walker wrote: > > > > On Sat, Feb 23, 2008 at 3:54 PM, John Sonnenschein > > wrote: > > > On Sat, Feb 23, 2008 at 1:48 PM, Glynn Foster wrote: > > > > Hey, > > > > > > > > > > > > Jim Walker wrote: > > > > > John Plocher wrote: > > > > >> "It Seams To Me" that the things that are produced by Indiana > > > > >> really need to be that minimalist core (i.e., enough to > > > > >> boot into an installer that is network aware in whatever > > > > >> locale the user needs) AND a set of repositories where > > > > >> the various packages are to be found. That minimalist core > > > > >> is then called OpenSolaris. > > > > > > > > > > I agree. We must find the current minimum set of packages > > > > > that defines an OpenSolaris distribution. Certification > > > > > can proceed once we have that. > > > > > > > > I actually disagree to some extents. I don't think there has to be a minimum set > > > > of packages defining a core. I think the repository is that core. Any derivation > > > > based on a selection of packages from that repository (or set of 'official' > > > > repositories) should be compatible. > > > > > > > > All distributions *must* have that repository first on its priority list, and > > > > then have the ability to install from separate repositories after. The emphasis > > > > and focus should be encouraging people to contribute directly to that repository > > > > IMO, and make it a better place for everyone. > > > > > > > > > > +1 > > > > > > This is precisely what I'm advocating for, and what I assume JohnP > > > means. ( given the allowance for building a distro value-add > > > repositories on top of the offical one, eg indiana + NONUSlibdvdread > > > ). > > > > I think I finally understand how I crossed wires earlier. > > > > I was under the mistaken impression that this minimum set was going to > > affect what was expected to be in the repositories and what was > > required for compatibility. > > > > If the minimum set is just the minimum to be able to use the > > trademark, and then the rest of the repository access is required for > > compatibility, that would be agreeable to me. > > > > I think I'm finally in agreement here. Thanks for your patience. > > Great! > > So it seems like everyone's on board. That's fabulous ( and > considering our community dynamic, a once in a lifetime happening ) Indeed; sorry about earlier. I still believe that we need to establish *recommendations* (guidelines) for distributions to help guide them to the best experience for users. There's a lot of knowledge that gets lost (and work duplicated) in building distributions (KDE, XFCE, or GNOME flavours); and I would like to see that more formally documented since Solaris has different requirements for those environments than other operating systems. However, that's something the Distribution CG in cooperation with other CGs can do. I would expect certain guidelines to eventually become requirements for certain compatibility claims, just as any other platform. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From johnsonnenschein at gmail.com Sat Feb 23 16:10:05 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 16:10:05 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> <47c0a204.hgpYSffhVGR9fv/S%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <241540330802231610s21fcb450y212eee5a270c070c@mail.gmail.com> On Sat, Feb 23, 2008 at 3:59 PM, Shawn Walker wrote: > On Sat, Feb 23, 2008 at 4:45 PM, Joerg Schilling > wrote: > > > "John Sonnenschein" wrote: > > > > > > > > OpenSolaris Compatible > > > > > > > > Derivations made from a single official OpenSolaris repository > > > > *and* additional 3rd party repositories. The official repository > > > > would be first in the search path. An install would become > > > > 'OpenSolaris Compatible' the moment an additional IPS repository > > > > was added to an original OpenSolaris Distribution install. > > > > > > Does this include for example, packages that would never and can never > > > exist in the official repository, such as for example libdvdcss, which > > > is not legal for distribution within the US (and therefore sun can't > > > distribute it ) > > > > It of course can be supported by OpenSolaris as you just need to load/install > > the related package from a server in Europe. > > The one problem will be software that doesn't support such a > configuration at runtime. Some software requires that you compile it > with certain options enabled. I believe such software should be fixed > to support runtime loading in those cases so that we don't have to > recompile "supported" packages just so such things can be enabled. While I agree, I think that's a technical implementation detail best left to the package maintainer in charge, rather than the trademark discussion cheers -John -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From johnsonnenschein at gmail.com Sat Feb 23 16:12:47 2008 From: johnsonnenschein at gmail.com (John Sonnenschein) Date: Sat, 23 Feb 2008 16:12:47 -0800 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: References: <47BF2CCA.9020604@Sun.Com> <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <241540330802231354w5698cc5foa7803c83ab57736c@mail.gmail.com> <241540330802231600u7e29f082u54bc4646e3119317@mail.gmail.com> Message-ID: <241540330802231612s721ad8deg91e79ed55e070bd3@mail.gmail.com> On Sat, Feb 23, 2008 at 4:07 PM, Shawn Walker > I still believe that we need to establish *recommendations* > (guidelines) for distributions to help guide them to the best > experience for users. Sure, I encourage it. I'm no usability expert, but it's good to have recommended best practices anyways > > There's a lot of knowledge that gets lost (and work duplicated) in > building distributions (KDE, XFCE, or GNOME flavours); and I would > like to see that more formally documented since Solaris has different > requirements for those environments than other operating systems. > > However, that's something the Distribution CG in cooperation with > other CGs can do. > > I would expect certain guidelines to eventually become requirements > for certain compatibility claims, just as any other platform. I still disagree with you here. I think recommendations are best left as recommendations, to be ignored at the target's peril ( as they affect nobody but those they are intended for ) Regardless, that has no bearing on the current discussion -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From swalker at opensolaris.org Sat Feb 23 16:13:28 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 18:13:28 -0600 Subject: [trademark-policy-dev] Uubuntu T&B policy? In-Reply-To: <5c902b9e0802231336h1b067608v109c70ac40bfcf4@mail.gmail.com> References: <47BF1A33.6060808@Sun.Com> <241540330802221058p48410e05v735c561941bd158@mail.gmail.com> <5b5090780802221127h66e5d55cn6bab8b1e6fabf137@mail.gmail.com> <47BF2928.6070707@Sun.Com> <319ee2b10802222055o45ddeadcn1c75ea2492fd478e@mail.gmail.com> <47C06707.9000403@Sun.Com> <5c902b9e0802231222y38c2e259h339efc09a42ed841@mail.gmail.com> <5c902b9e0802231336h1b067608v109c70ac40bfcf4@mail.gmail.com> Message-ID: On Sat, Feb 23, 2008 at 3:36 PM, Justin Erenkrantz wrote: > On Sat, Feb 23, 2008 at 12:36 PM, John Sonnenschein > wrote: > > -1 > > > > The ubuntu policy is far more restrictive than I think we ought to be, > > even not taking in to account the ability for commercial distributions. > > My initial reaction was similar, but I then looked at the current T&B > policy draft and, based on my reading, I don't think Sun would permit > commercial usage either for anything that used the "based on > OpenSolaris" or similar marks. If Sun wishes to permit such usage, > then they can strike those phrases from Ubuntu's policy and replace it > with wording that such usage is permitted. But, my understanding is > that Sun doesn't wish to permit such usage. > > I'd be happy to be corrected on this. -- justin I haven't heard any such statements yet. That's the purpose of this proposal though. This proposal obviously is going to require give and take with Sun. I'm fairly certain Sun does want to enable commercial distributions, I just don't know to what degree yet. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben From swalker at opensolaris.org Sat Feb 23 16:14:49 2008 From: swalker at opensolaris.org (Shawn Walker) Date: Sat, 23 Feb 2008 18:14:49 -0600 Subject: [trademark-policy-dev] [ogb-discuss] Should the OGB respond to Sun's OpenSolaris name-use decision? In-Reply-To: <241540330802231610s21fcb450y212eee5a270c070c@mail.gmail.com> References: <241540330802221239m7cd7df94xbbca214a4585ed42@mail.gmail.com> <47BF4947.6030709@Sun.Com> <47BF6896.7050106@sun.com> <47C094A6.8010802@sun.com> <47C09BE5.8050302@sun.com> <241540330802231425q533a6a18sb2abfbfbcc9bbeb7@mail.gmail.com> <47c0a204.hgpYSffhVGR9fv/S%Joerg.Schilling@fokus.fraunhofer.de> <241540330802231610s21fcb450y212eee5a270c070c@mail.gmail.com> Message-ID: