At Tue, 25 May 2010 10:39:46 +0800, Dean Michael Berris wrote:
At the risk of alienating myself, I think there is a faulty assumption here. I see no reason to believe that there is going to be a clear cut single "winner" with DVCS. The systems clearly differentiate themselves. Quality of windows OS support, performance, and learning curve are just a few examples. If the future turns out as I expect, choice of DVCS is going to be similar to a choice of vim or emacs - mac or windows.
True, but then even in the choices available, we're going to have to start somewhere. And I think the largest reach for the most part would be to support git and then if someone does feel strongly about it, then make it work for mercurial, bazaar, subversion, or <insert your favorite version control system here>. Like Dave has already pointed out, Pip already abstracts this plugin subsystem out anyway, so it's just a matter of writing the glue that Pip can use to understand other VCSes -- I don't see how choosing to support git first or thinking that it will win over the others is a bad thing though.
Actually, it turns out I was (sort of) wrong: one of the things you can do with git that you can't with mercurial is important to making this system efficient: getting a list of the branches and tags in a remote repository. With mercurial, you have to clone the whole danged thing. I'm still inclined to tell people who want to use other VCSes to use adaptors like hg-git, git-svn, etc. -- Dave Abrahams Meet me at BoostCon: http://www.boostcon.com BoostPro Computing http://www.boostpro.com