+1 on gerrit. For example very grateful that mice put his changes on vm snapshots on reviewboard.
I really disagree with code commit bring a privilege of committers. --Alex David Nalley <[email protected]> wrote: On Sun, Feb 3, 2013 at 1:50 PM, Chip Childers <[email protected]> wrote: > On Sun, Feb 3, 2013 at 2:23 AM, Marcus Sorensen <[email protected]> wrote: >> Are you thinking there should be a regular moratorium or something similar >> just before the cut, so that the quality of the features as a whole can be >> evaluated, or are you just concerned that the last minute features didn't >> get proper review? I think that as long as there's a time-based release >> were going to have features rushed, we either need to be OK with it and >> allow for time and ability to fix it afterward, or have some very stringent >> quality control prior to merge. We can maybe start with the former and work >> toward the latter. > > That's exactly right. We're either going to do time-based release > schedules, or we're not. And if we are doing time-based releases we > need to be concerned about the quality of commits coming into master. > Anyone want to get gerrit up and running? I'd be more than happy if > we had a strict review process for ALL commits coming into master, > regardless of commit access permissions. I've had some folks express interesting in setting up Gerrit for code reviews. That said - it's not like we can't review code that is already committed - we get commit mails after all. Gerrit just helps automate some of that. --David
