On Nov 10, 2013, at 6:58 PM, Rodolfo García Peñas <k...@kix.es> wrote:
> Thanks again for your reply.
>
> As you can see in their last mail, about the idea of include more commiters
> ("Bad idea"), things doesn't change too much. For me is unacceptable.
>
> If more people thinks that is good moment to make a fork and create a new
> repo with more than one commiter, more democratic, I will collaborate in the
> new fork.
I see one big problem with that: based on the mailing list activity, you are
going to be the only regular committer to that fork. A few people delivering
patches sporadically count just as much as they count right now (and you are
the only so far to propose a fork at all).
What is really undesirable is that it sounds very much like it’s going to be a
hostile fork. Hostile forks by a single person are really ephemeral, the
history tells us. And a „more democratic” process only means a thing if there
are at least a few people whose really useful contributions were rejected.
Can you please name those people? Can you please also list the really useful
patches that have been rejected beyond hope of ever being accepted?
There was a fork of CNOME called GoneME around GNOME 2.8 (back in 2004). [1]
It’s quite a good illustration of how forks that are started out of annoyance,
die. Mind you, the author managed to make quite a buzz[2] and there were people
willing to help (initially). I fail to see annoyed crowds here, so it’s safe to
predict that a fork would die in, say, a month, after its author’s enthusiasm
vanishes.
What *can* be done, though, is a fork in a sense of a Github fork. That is,
your repository with all your cool ideas goes public, and so does the process
around it. Issues get filed and tracked, people do more forks to test their
ideas on, there are pull requests, which after a few stabilising rounds go
upstream, that is, here. Call it a „friendly fork“ or something like that, so
there is still a single project at the core, and there is NO bloody „Rodolfo
vs. Carlos“, or, for that matter, „Anyone vs. Carlos“, stanza in it.[3]
I can see a good thing in it that so far WindowMaker so far doesn’t have a real
public bug tracker, it should really grow one. Also, it would keep code reviews
consolidated, so you could easily see there are reasons for not accepting a
patch right away.
Also, I hope it gives you a gist of understanding that „democracy“ doesn’t
quite work in software projects, especially after you yourself reject a few
pull requests which don’t add (or subtract) value and get an angry reply from
someone who spent their precious time on it.
I’m sorry for jumping in with such a long-winded reply, having not contributed
a single line of code to WM. It’s just that I care for this project too much to
see it killed.
____________
[1] http://www.akcaagac.com/index_goneme.html
[2]
http://tech.slashdot.org/story/04/07/25/144244/project-goneme-fixes-perceived-gnome-ui-errors
[3] What the actual hell. repo.or.cz allows forking repos. Why not? However, it
doesn’t have issue tracker and a nice UI for pull requests.
