On Thu, Aug 14, 2008 at 4:25 PM, Robert Godfrey <[EMAIL PROTECTED]> wrote:

> I don't think review then commit is going to work personally.  For a
> start you now have a three way exchange before a commit can be made
> (put on JIRA... get reveiwed... wait for OK... commit)...  and the
> latency introduced will generate a much greater chance of patches
> failing due to conflicts...   While I agree a git based source control

I would hope that this would be relatively rare, and while it does
introduce some latency it does also ensure that the patch has been
reviewed before it goes in.

Potential patch conflicts are less of a problem from my PoV than
having a ton of Jiras and commits that haven't gotten past the "ready
for review" stage because people are inherently lazy and it's easier
to not do it.

> system would make this feasible... we are not in such a world.  The
> majority continue to use subversion and that is the system Apache is
> providing us.

I'm not suggesting that we mandate it, but we could all switch to
using git over svn. It was mostly just an idle comment ;)

- Aidan
-- 
Apache Qpid - World Domination through Advanced Message Queueing
http://cwiki.apache.org/qpid
"Nine-tenths of wisdom consists in being wise in time." - Theodore Roosevelt

Reply via email to