On Aug 12, 2013, at 9:32 AM, Lloyd Hilaiel wrote: > On Aug 12, 2013, at 4:27 PM, Johnathan Nightingale <[email protected]> > wrote: > >> On Aug 9, 2013, at 9:11 PM, Mark Finkle wrote: >> >>> My only strong opinions are: >>> >>> 1. Using bugzilla as the one source of truth for bugs. Even b2g had to do >>> it. >>> 2. ELM is the place where the code ends up for nightly builds. How it gets >>> there, I don't care. But we have a test infra that works with the hg repos >>> and we need it to run. >> >> A +1 to both of these, but particularly the first - it has been our >> experience over and over that when we move away from bugzilla as the source >> of truth, it bites us in numerous and unpleasant ways. At this point, it's >> the nearest thing you'll encounter to a project-wide edict, but it's >> absolute law when it comes to work that impacts Firefox desktop, Android, or >> OS. > > I'm hearing the same thing from everyone. > > decision: > 1. As far as the client engineering team - "ELM is the place where code ends > up" - it doesn't matter how it gets there.
I agree. > 2. Anything that has cross-team implications (like say, gavin needs to review > a patch from lloyd), goes in bugzilla. That is a narrower framing than I'd proposed, which suggests you may have exceptions in mind - any examples? The test for "should it be in bugzilla" is broader than "does it have cross team implications?" or "does it require review?"; it's more like "is it work that others in the project will want to track?" A lot of teams use bugzilla metadata to track and audit their work, and it's also useful for after-the-fact spelunking. J --- Johnathan Nightingale VP Firefox Engineering @johnath
_______________________________________________ Sync-dev mailing list [email protected] https://mail.mozilla.org/listinfo/sync-dev
