On 4/20/16 11:53 AM, Armen Zambrano G. wrote: > Would it make more sense to have a relbranch instead of using ESR?
Oh lordy, no! It's hard enough diverting engineering work to supporting a single ESR 9 months after the fork. Why would we do two of them? How would a relbranch differ from ESR? > IIRC ESRs are stable for a period but when we uplift we uplift > everything new. > For this XP relbranch we would only take security patches. That's what we do with an ESR: security and stability patches only. In practice not many stability patches once it's in decent shape after the first couple releases--only if one of the security patches breaks something. We don't uplift things to ESR, ESRs reach their end-of-life and then people migrate to the next ESR. > It would serve the purpose of keeping our users secure where they're but > saving some energy in making new features also XP compatible. > > Setting an N months EOL expectation (plus another criteria[s]) might be > wise rather than leaving it open ended. Putting XP people on an existing ESR does both those things. It's the second that's problematic. When the ESR we've diverted the XP folks to dies, what then? If we still have a significant chunk of XP users they will have nowhere to go. We could try to extend the life of that ESR, but that could result in more work back-porting fixes and building/testing/shipping extra releases than simply working around XP issues until ESR 52 next year. -Dan Veditz _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
