[Ironruby-core] "Start spreading the news"

Tomas Matousek Tomas.Matousek at microsoft.com
Sat Aug 7 23:26:47 EDT 2010


Re CI server: cool, let's get one up and running then. Ideally we would have 2 - one running on Windows and other on Linux to make sure that IronRuby works well on both platforms. 
Both should run the same test-suite (irtests script). The harness might need some tweaks for Linux and you're welcome to submit patches. 

Re "Code review emails are just one extra step that's not needed":
Code reviews are absolutely needed! It's not a bureaucracy, it's a quality gateway. Nobody should commit anything without a code review from the code owner. For now, that would be mostly me (core, libraries, csproj files), Jim (test harness, infrastructure) and Jimmy (Silverlight, ironRack). If anybody's interesting in owning some part of IronRuby let know the current owner.

Actually, what do you mean by "canonical" repo? What would make GIT repo canonical? 

Tomas

-----Original Message-----
From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Michael Letterle
Sent: Saturday, August 07, 2010 7:34 PM
To: ironruby-core at rubyforge.org
Subject: Re: [Ironruby-core] "Start spreading the news"

We've had a CI server set up on CJ's hardware for a while, it's actually worked from time to time as well :) http://twitter.com/IronRubyCI

I'm more then willing to get that up and running again, it wasn't a general CI server though, it was mostly for mine and Ivan's changes for compilation under Linux.

And Code review emails are just one extra step that's not needed, it's just bureaucracy added on by the current situation, it's a barrier to entry and a hassle.  Don't misunderstand, the work gone into them IS appreciated, but it's just another *groan*.

And I find it hard to believe that the only reason for an internal canonical repo was lack of external CI.. that's not what's being implied, is it?

On Sat, Aug 7, 2010 at 10:13 PM, Tomas Matousek <Tomas.Matousek at microsoft.com> wrote:
> Is there anybody who decided not to submit a patch based upon the limitations/requirements of the current process and who would contribute otherwise?
> Is anybody willing to run a continuous integration server that guards the repo from erroneous patches? A canonical repo needs to have such a gatekeeper. Our internal SNAP system has provided this functionality for the internal TFS repo.
>
> BTW, "Sync to TFS" was an artifact of lack of automated sync tool. We now have that tool, thanks to Jim, so you should see less of these. Code review emails has been sent for most of the changes that were made to Ruby repo anyways. So if you watched the mailing list you could easily track what's going on. Only comments to changesets that included some internal infrastructure changes weren't sent to the mailing list.
>
> Tomas
>
> -----Original Message-----
> From: ironruby-core-bounces at rubyforge.org 
> [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of William 
> Green
> Sent: Saturday, August 07, 2010 5:36 PM
> To: ironruby-core at rubyforge.org
> Subject: Re: [Ironruby-core] "Start spreading the news"
>
> I agree with Michael here. As long as the canonical source for IronRuby lives in TFS behind the big blue firewall, with only MSFT employees as core committers, there is a problem.
>
> What I propose is that we, as a community, designate the github repo 
> as the canonical one, and that whoever controls it opens up commit 
> access beyond the MSFT core team, perhaps to those that have already 
> had accepted contribution. (this may already be in place, I don't
> know)
>
> Bottom line is that MSFT has decided they no longer desire to invest in the project. If we want IronRuby, we're going to have to make it happen, on our own.
>
> --
> Will Green
> http://hotgazpacho.org/
>
>
>
> On Aug 7, 2010, at 8:00 PM, Michael Letterle <michael.letterle at gmail.com> wrote:
>
>> If Microsoft was simply sponsoring development and putting its stamp 
>> on it, it wouldn't be a problem.. but the fact that MS "owns" it IS a 
>> bottleneck, as been from the start.  "Sync to TFS" commits make me 
>> cringe everytime I see them.
>>
>> On Sat, Aug 7, 2010 at 7:07 PM, Jim Deville <jdeville at microsoft.com> wrote:
>>> I guess I never really sent the mail that I meant to send about my 
>>> progression.
>>>
>>>
>>>
>>> Similar to Jimmy, I've also made the incredibly hard decision to 
>>> move on as well. I've been working on the JavaScript team now for 
>>> about 2 weeks. I've made this decision for many of the same reasons 
>>> as Jimmy, but for various personal reasons, I decided to remain with 
>>> MS instead of leaving. I will still be working with IronRuby, it 
>>> just won't be my primary function anymore.  I also had a wonderful 
>>> time on the team, and I will really miss working directly with them.
>>>
>>>
>>>
>>> I agree that IronRuby is not a big community, but I don't know that 
>>> it is a problem unless we let it be one J. I also feel that MS 
>>> owning the code isn't a problem. It's not like it really benefits 
>>> anyone inside of MS, it benefits us, the users J. It also benefits 
>>> the Ruby community at large if we continue to make IronRuby a great 
>>> product that people want to use in their .NET applications, because 
>>> it shows people Ruby J
>>>
>>>
>>>
>>> I understand that this is sad, and for some, expected L, but I hope 
>>> that people still continue doing the awesome things people have been 
>>> doing J
>>>
>>>
>>>
>>> JD
>>>
>>>
>>>
>>> From: ironruby-core-bounces at rubyforge.org
>>> [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Orion 
>>> Edwards
>>> Sent: Sunday, August 08, 2010 2:43 PM
>>>
>>> To: ironruby-core at rubyforge.org
>>> Subject: [Ironruby-core] "Start spreading the news"
>>>
>>>
>>>
>>> I'm sure most of you have seen this already, but I hadn't seen 
>>> Jimmy's "farewell Microsoft" blog post posted to the list, so here 
>>> it is if anyone hasn't read it
>>>
>>>
>>>
>>> http://blog.jimmy.schementi.com/2010/08/start-spreading-news-future-
>>> o
>>> f-jimmy.html
>>>
>>>
>>>
>>> I'd like to thank Jimmy for his work thus far on IronRuby, and 
>>> certainly wish you the best of luck at your new job and living in 
>>> NYC. From that point of view, it's happy days.
>>>
>>>
>>>
>>> The news about Microsoft's internal handling of IronRuby however 
>>> makes me rather sad, and frankly a bit worried about the future of 
>>> IronRuby in general. If there's now only 2 part-time developers 
>>> working on IR inside Microsoft, and they've been told not to develop 
>>> any new features, then where does that leave us?
>>>
>>>
>>>
>>> While IronRuby is open source (yay), It unfortunately hasn't seem to 
>>> have become quite big enough (unlike JRuby) to be self sustaining.
>>>
>>>
>>>
>>> I personally also perceive somewhat of a problem, in that (again, 
>>> unlike with JRuby) Microsoft "owns" IronRuby. Last I knew, 
>>> committers still had to sign a copyright assignment form to MS, and 
>>> MS controls all the repositories and other assorted stuff. (AFAIK 
>>> Charles has personally controlled all the JRuby repos since day one and still does).
>>>
>>> This provides a LARGE (to me at least) disincentive to contribute to 
>>> IronRuby in the future. Because MS owns IronRuby, I feel like if I 
>>> were to commit code, it would not be for the benefit of a nice 
>>> friendly community-driven open source group, it would be for the 
>>> benefit of Microsoft Corporation.
>>>
>>> I was happy to accept this when MS were putting a lot of work into 
>>> IronRuby, but now that they're not, my feeling is "So Microsoft have 
>>> bailed and left us hanging, why on earth would I want to do work on 
>>> their behalf after they've just done that?"
>>>
>>>
>>>
>>> At any rate, I've not committed any code to IronRuby (I've come 
>>> close several times, but never just had the time) so my opinion is 
>>> largely irrelevant, I just hope that other potential committers with 
>>> more time and skill than I don't end up feeling this way too :-(
>>>
>>> _______________________________________________
>>> Ironruby-core mailing list
>>> Ironruby-core at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/ironruby-core
>>>
>>>
>>
>>
>>
>> --
>> Michael Letterle
>> IronRuby MVP
>> http://blog.prokrams.com
>> _______________________________________________
>> Ironruby-core mailing list
>> Ironruby-core at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/ironruby-core
> _______________________________________________
> Ironruby-core mailing list
> Ironruby-core at rubyforge.org
> http://rubyforge.org/mailman/listinfo/ironruby-core
> _______________________________________________
> Ironruby-core mailing list
> Ironruby-core at rubyforge.org
> http://rubyforge.org/mailman/listinfo/ironruby-core
>



--
Michael Letterle
IronRuby MVP
http://blog.prokrams.com
_______________________________________________
Ironruby-core mailing list
Ironruby-core at rubyforge.org
http://rubyforge.org/mailman/listinfo/ironruby-core



More information about the Ironruby-core mailing list