[Bioperl-l] Possible migration to git/github

Chris Fields cjfields1 at gmail.com
Sat Apr 17 16:18:13 UTC 2010


All,

I wanted to get the BioPerl community's general input on a possible Subversion to git/github migration for the BioPerl repository.  The plans haven't been finalized in any way.  We haven't decided, for instance, on whether to make github the main site, a read-only mirror, or a downstream mirror for forking/pulling code from the community (with a central upstream dev repo for releases).  As a preliminary run I will be pushing a test run to github of bioperl-live (all tags/branches) this weekend.  

Thoughts have been circulating among the core developers for some time about a migration at some point, however recent problems with anon access to code has made this a more pressing issue.  The overall impression I am getting from the community at large is a move would be a positive thing and would swing the doors wide open for users to contribute much more easily (via forking). 

Therefore, I would like to outline the positive/negative aspects of a move to git/github, and a proposed path to migration if one should occur.  Remember, this is a project with 15+ years of code that has already undergone a recent migration from CVS->SVN.  Retaining history and branches/tags for all bioperl projects is key.  We also need to think logistically about other issues particularly with github (number of collaborators allowed per account, mapping authors, etc etc).

Personally, I think we should be moving forward with this as soon as possible, but I personally like git/github so I'm a bit biased.  The recently-added support for in-line code review comments and SVN support has particularly swayed me:

http://github.com/blog/626-announcing-svn-support
http://github.com/blog/622-inline-commit-notes

Thoughts?  Comments?  Flames?  Suggestions?

chris



More information about the Bioperl-l mailing list