----- Original Message ----- > Le 26/03/2014 12:59, Ryan VanderMeulen a écrit : > > On 3/26/2014 6:03 AM, Julien Wajsberg wrote: > >> Le 25/03/2014 07:32, Kevin Hu a écrit : > >> In v1.3, Fabrice approved all patches that entered the v1.3 branch, > >> blockers or not blockers. Now I see that for v1.4 we don't do the same, > >> and we rather do like previous releases: blockers have automatic > >> approval. > >> > >> Do we have a rationale for this change? Do we know whether Fabrice's > >> hard work was useful (or not)? > >> > >> Thanks > >> > > > > My understanding is that the intent is that v1.4 checkins will > > eventually be more restricted once we get further along the process. > > Sort of like the Gecko train model where Aurora has fewer uplift > > restrictions than Beta, but the bar gets higher as we get closer to > > release. If that makes sense.
This is the intention. I don't have the exact process worked out with the relman team yet in terms of milestones but I do think that every blocker should undergo risk evaluation (at some point in the process) before the associated patches land. > And I would welcome such a policy :) I just want to see reasonings > behind changes, especially that we change this for each and every release! The role Fabrice played in 1.3 is very similar to the role that release management should play and will play in 1.4 and beyond. To be clear, the release management team will assess each uplift request and solicit input from engineering, QA, l10n, etc. when required to make an informed approval decision. My intention is to establish a consistent policy that will remain intact release to release. I welcome feedback and you can hold me accountable for this. Lawrence _______________________________________________ dev-b2g mailing list [email protected] https://lists.mozilla.org/listinfo/dev-b2g
