[Python-Dev] Looking for VCS usage scenarios (original) (raw)

Terry Reedy tjreedy at udel.edu
Wed Nov 5 19:32:41 CET 2008

Paul Moore wrote:

2008/11/5 David Ripton <dripton at ripton.net>:

All timings very approximate:

Time for average user to check out Python sources with bzr: 10 minutes Time for average user to check out Python sources with git or hg: 1 minute Time for average user's trivial patch to be reviewed and committed: 1 year I love DVCS as much as the next guy, but checkout time is so not the bottleneck for this use case. :-) That's a fair point. But it's not the point I was trying to make, which is that I'd want whatever DVCS is chosen to make the initial experience of a casual user / newcomer as easy as possible. Why discourage them in the first 10 minutes (which, BTW, is much faster than my experience with bzr last time I tried the Python repo) when we can make them suffer for a whole year? :-) :-)

It does get to the point that the current bottleneck is code review, to the point that people may not submit patches because it seems nearly useless*. And often, when a patch does get reviewed, the diff is obsolete and needs to be redone versus the changed 'current' trunk. I presume that patches as branches would alleviate this last part.

So I think easier review should be a prime consideration for infrastructure improvement. If I go to the tracker now and click on a 'patch', I get a sometime easy, usually difficult, and sometimes impossible to read diff. With a wide-screen monitor, I would like a side-by-side display with differences marked, as with Guido's code review tool and other such displays I have seen here and there.

*The current quick review and implementation of doc suggestions and patches, on the other hand, has encouraged more submissions from me and, I believe, others.

More information about the Python-Dev mailing list