[arc-discuss] The relationship between OpenSolaris and ARC

[Joerg Schilling]

Darren Reed <Darren.Reed at Sun.COM> wrote:

> There are some interesting things going on here...
> 1) various people inside OS believe that Joerg should
>    integrate (or request the integration of) his tools
>    himself;
> 2) it would appear that Joerg doesn't want to do that
>    dance himself;

Here seems to be a missunderstanding: I would if I could but I can't.

Correct would be: "can't do the dance alone/himself".

But see also below.


> 3) OS appears to Joerg to move in hostile ways by
>    "deliberately" making changes that conflict with
>    what he's done.
>
> So for the people in (1), consider (2).
>
> I'll add a note that as an open source participant there's
> a mindset that perhaps Sun people miss, that being that
> various people in open source just want to *write* s/w,
> it is up to others to use/integrate it, *if* they want it.
> This is particularly relevant to (2).  Or to put it
> differently, we don't ask the people who write gcc,
> wireshark, etc, to do the legwork and integrate their code
> into OS, so why should we be asking Joerg?  If someone
> asked me, today, to integrate ipfilter into opensolaris
> today, I'd probably say no because it *feels wrong* -
> software should be pulled in by the people that want
> to use it, not pushed in by the authors.  I don't know
> if I can explain it any better/more than that.

There are different possible levels of interaction.

a)	Do everything by your own
	This works if you can do all needed decisions for the
	way of integration also.

b)	Discuss the problems and do cooperated work
	If other people define constraints, this is the most probable
	way to go.

c)	Sit and wait until it has been done by others
	This is the only way to do it if interaction is impossible.


> Note that (2) applies more to individual works, such as
> star/mkisofs, etc, rather than work done for a larger open
> source project.

Could you explain this, what do you call "a larger open source project"?

GNUtar is less source than star (although it compiles to a bigger binary;-)
Many people would call it a larger OSS project because there have been many 
contributors over the years.

mkisofs is a work that has been done by many people but most other people
including the original author did lose interest (most OSS writers stop coding
on their "babys" after ~ 3-5 years) and if I did not take over maintainership, 
the project was dead since 1999. Now of course much code has been replaced or 
rewritten and today more than 50% of the current mkisofs source has been written 
by me.

... if you believe that "a larger open source project" needs a public writable 
SCCS, then even OpenSolaris does not met the conditions.

I would really be interested to know what you call "a larger open source 
project" and why you do so.


> For (3), Joerg has to stop thinking everyone is out to get
> him.  I often think the same because of the shifting sands
> that are the header files and interfaces for networking
> across open source os's but get over it with some (sometimes
> loud) swearing at the monitor or occasionally in email and
> then let it go.

One big problem in the past was that a few people started personal attacks
in a technical discussion.


> All that said, by advice to Joerg is to stop taking decisions
> by Sun's engineers to do whatever personally and just
> workaround whatever problems the OS engineers that work for

I was technical before. If people believe I was personal, this is a 
missinterpreation.

Jörg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) Jörg 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