[2990] in BarnOwl Developers
Thoughts on our current code review policy
daemon@ATHENA.MIT.EDU (Nelson Elhage)
Wed Mar 23 21:30:07 2011
Date: Wed, 23 Mar 2011 21:29:50 -0400
From: Nelson Elhage <nelhage@nelhage.com>
To: barnowl-dev@mit.edu
I'm curious what people about how our current code review policy is
working out. I'm personally getting the feeling that it's not working
great; Code reviews sit around forever, and we have tons of unmerged
cleanup commits and branches that often get lost waiting on review and
then get forgotten about.
BarnOwl also doesn't have a large pool of developers that know the
code well, and so I'm not convinced anyone can effectively review
changes for e.g. subtle interactions with other pieces of the code,
which is one of the things that code review would potentially provide
us. And so largely code review ends up just nitpicking on style and
other local issues, that, while not irrelevant, also just don't matter
all that much.
My personal feeling at this point is that barnowl would be better
served by a model where it's easy to get things into master (or into
some other "blessed" bound-for-master branch), and we do testing and
review from there, potentially in batches periodically / prior to
releases / whatever, than the current model, which mostly seems to
result in code sitting in peoples' INBOXes forever.
Does anyone else have thoughts, or agree or disagree?
