On 2013-07-11 2:14 PM, Jet Villegas wrote:
I may have a skewed view of things, but I personally have not had problems
getting prompt code reviews. I also don't have a problem talking to my
reviewers about what I'm hacking on. I'm hesitant to throw a bunch of
technology at this problem, if it's a social issue. Code reviews are a
conversation and more code has to come with more conversations.
For what it's worth, there's some data on the subject available (inside
the firewall only, unfortunately) here:
https://metrics.mozilla.com/bugzilla-analysis/ReviewIntensity.html#sampleInterval=day&sampleMax=2013-07-15&sampleMin=2013-01-15&requestee=
You can drill into that data on a per-project or per-team basis, but in
aggregate you can see that in the last six months more than two-thirds
of reviews already happen in less than a day, and it tails off pretty
sharply after that.
That's not to say there's not room for improvement there; the ~2200
reviews that take two or three days, and ~1700 that take four to six, vs
the ~12500 one-days, still seems like a lot, and I agree with Jet that
some communication - particularly with new contributors or first-timers
- would go a long way to relieving the stop-energy problem even if you
can't get to the review in the next eight hours.
There's a blip at the right hand side of the graph there, but a casual
inspection of the "over 10 days" pile seems to indicate that those are
mostly hard problems, not neglect.
- mhoye
_______________________________________________
dev-platform mailing list
dev-platform@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-platform
