On 11/10/08 15:27, C. Bergstr?m wrote:
> Hi
>
> I apologize if this comes across as slightly flippant, 

Not at all. You are most welcome to comment. :)

> but I'm not even 
> sure where to even go to get these things resolved/looked at.
>
> 1) The general community attitude towards new members in #opensolaris in 
> irc is bad on a regular/daily basis  There is one individual @Sun in 
> particular who probably needs some mentoring or a corrective talk.  This 
> could possibly could be resolved by a code of conduct policy on a wiki 
> somewhere.  


We have a code of conduct here: http://opensolaris.org/os/about/values/. 
This is somewhat new so some people may not know about it. It's not 
really meant to be a set of rules, per say, but more of a guideline 
about how we should all treat each other (with respect). :)


> I've given it some thought, but haven't had a chance to make 
> a draft. (Normally, /trolls/ are kicked, but currently only @Sun people 
> are given OP privileges to resolve such matters.)  

I believe that #opensolaris was started by non-Sun guys initially, but I 
don't know who runs the channel these days. There have been complaints 
in the past, but I've also heard that lately it's been ok. The only 
thing I can say is to issue a complaint with the leaders. We point to 
the channel from this community site, but we didn't create it and Sun 
doesn't run it.

> I have logs, other 
> community members willing to back this and records if anyone wants to 
> look into this further. (Or even better I can make a list of additional 
> != @Sun community members who are quite active in the channel and 
> responsible.)
>
> 2) dclarke posted back in September [1] about OpenSolaris being more of 
> a community effort.  So taking a look at the onnv-notify bug fixes for 
> November [2] would you say that this is working very well?  

It's a work in progress, no question.

We have a sponsor program for code contributions to the OS/Net 
consolidation, and that has improved a great deal lately in terms of 
addressing the backlog: 
http://opensolaris.org/os/bug_reports/request_sponsor/

Also, keep in mind that contributions come into the entire program via 
the many projects and communities on the site, but we don't collect 
those stats in one place so it's difficult to see. But I always run into 
people who are contributing to various projects that I didn't 
necessarily know about (which is very cool). So, for project- and 
community-based contributions like these, people have to get involved in 
a given community or project and get to know the developers in that area 
since each project would have different needs and different engagement 
methods.

We've started to write up some general ways to contribute:
http://opensolaris.org/os/participate/
http://www.opensolaris.org/os/communities/participation/

And, of course, the OpenSolaris 2008 distribution is designed, in part, 
to eventually enable the community to contribute packages.


> If I had to 
> guess based on loose observation I would say this is probably a 
> combination of three things.
>     a) Technical issues which for various reasons deters people from 
> contributing.
>   

Agree. The bar is high, but people have been contributing all along.

>     b) Lack of a real community to give people != @Sun something to be a 
> part of
>   

We list some pointers to five or so other distros here 
http://opensolaris.org/os/downloads/ that I think that offers a view 
into other projects growing outside opensolaris.org. To me, those are 
pretty big contributions to the OpenSolaris community. The community is 
also growing in other ways, non-code ways. There are people evangelizing 
OpenSolaris around the world in user groups, at conferences, at 
universities, etc. Heck, there are even people creating and managing 
conferences now, which is really cool.

Simon and Bonnie have some data points on contributions too:

http://blogs.sun.com/webmink/entry/opensolaris_open_getting_more_so
http://blogs.sun.com/bjc/entry/cool_stats

There's no question this has taken time, and I'd say that most people 
involved in the core project would like things to move much faster (and 
those people are doing their best to make that happen).


>     c) Most devs I've talked with perceive the policy and procedure for 
> even trivial patches to be too high.  (Whether this is truly the core or 
> most important reason I'm not unsure.)
>   

I think this is a good idea, and the scm-project is looking into it. 
Here's a review of some of the issues that the scm-migration project is 
working on regarding open development: 
http://mail.opensolaris.org/pipermail/scm-migration-dev/2008-October/003323.html

> Glynn Foster has summed up in a single line what I consider to be the 
> most true statement in free open source.
>
> "Get. Use. Learn. Love. Spread. Only then, in my opinion, can we even 
> think about Contribute??  [3]
>   

And I would agree generally. As we make things easier to get and use 
we'll naturally engage many more people and that will substantially 
increase contributions and a cycle developers. However, we can't forget 
that people have been contributing from day one of the OpenSolaris 
project, and they helped boot-strap the project from nothing. In other 
words, it took some time to get enough people, code, and infrastructure 
in place to do some of the things we are doing now. And at this point, 
we have the ability to move much faster, actually. Good stuff.

> (Additional thanks to Ian Murdock for blogging about it [4])
>
> So with this in mind where is the point of failure? [5]
>   

There is no single point of failure. This is a very big project that is 
emerging into the open from the inside of a company. It's also global. 
So, no one sees all parts at any given time. Everyone knows we need more 
infrastructure to facilitate code contributions, and this has been 
openly discussed in many forums. We'll get there. :)

> [5] In fairness I could see part of my argument as partially wrong 
> because I'm sure people @Intel 

Intel is contributing, yes. So is AMD. And there are others.

Jim

-- 
Jim Grisanzio
Sr. Program Manager, OpenSolaris Engineering
http://blogs.sun.com/jimgris/



Reply via email to