Am 04.05.2010 07:19, schrieb Jonathan Leto:
That being said, I think if everyone who is interested can give three-ish constructive reasons (on parrot-dev and/or blog post) for why Parrot should convert to VCS $X, that would be useful.
My shot: http://perlgeek.de/blog-en/perl-6/distributed-vcs.html
I think these reasons should concentrate on why it is good for the Parrot community in general, and not concentrate on personal reasons. I think case studies are very valuable here as well. We should interview developers from other open source projects who have converted VCS's recently, and see what we can learn from them. Not learning from their mistakes is just plain dumb.
In particular I know of two projects that migrated away from SVN: Rakudo and the eigen2 linear algebra.
Both were fairly unspectacular; the Rakudo migration caused some confusion because Patrick did the migration, and then disappeared for some time (due to travel or family issues); but we managed, and I haven't observed any big mistakes except clear, up-front communication.
Eigen planned to migrate to git, but ended up with mercurial. The reason was simply that somebody volunteered to do the mercurial transition, and nobody stepped up for the conversion to git. There was one day on which the developers were asked not commit anything. And there was a period of about two days when the Eigen wiki was not properly listing contributors, but there was no major fallout either.
I think most core developers agree that Subversion is not serving us well. Everyone wants to be more productive and have fewer hoops to jump through to do amazing things with Parrot. Let's try to make the process of migrating something that brings us together instead of dividing us.
+1 on that. Cheers, Moritz _______________________________________________ http://lists.parrot.org/mailman/listinfo/parrot-dev
