Am 10.10.2017 um 21:19 schrieb Andrea Pescetti: > Jim Jagielski wrote: >> I feel bad leaving RM at this stage... we are so close, and I don't want >> us to lose momentum. But I also don't want there to be drama > > Yes, let's avoid drama, really. > > But I'll still send this "modest proposal", based on experience: > > 1) We just go ahead and release 4.1.4 as scheduled (i.e., we release > RC4), at some point within the next 7 days. The "synchronize labels" > bug becomes a "Known Issue" in the release notes. > > 2) Starting today, we incorporate fixes on the AOO415 branch and aim > at releasing it in, say, early November; AOO415 is only meant to fix > 4.1.4 regressions and we try to be reasonably strict about this (i.e., > we try to avoid the never-ending "just another small fix" syndrome). > > 3) We still wait until, say, end of October before producing RCs for > 4.1.5 since this gives us the time to receive reports of other > possible bugs/regressions to be fixed in 4.1.5. > > Can this work? This covers the case where someone finds a bug the day > after release; this way we have a window for fixing it in a short term. > > My suggestion to keep 4.1.4 was simply pragmatic: those who never > updated the metadata might believe that 4.1.4->4.1.5 is only a > one-line change, while the change affects dozens of files and hundreds > of XML files related to updates and living on another domain. Not > something that can be done in 10 minutes. Deleting the tag was bad, > but we never relied on tags: 4.1.3 was tagged several months after it > was released. To us a release is/was simply the distributed source > package, not a tag in SVN. But this is a side discussion. > > Now the question is, is it OK for us to go ahead, release 4.1.4 with > this "known issue" and commit to fixing this and other possible > regressions in 4.1.5 next month? Technically, the Release Manager (and > I still consider Jim to be in charge, of course!) has the last word. > But I'd say everybody is welcome to comment.
I think this is a reasonable way... Another (minor) regression is the Polish dictionary (shame on me!). Already fixed in trunk and committed to 4.1.5 by Jim. I may have found another regression, but I need some time to test it and put it in words... And then it needs some time fixing it. So +1 for the timeline. Matthias > Regards, > Andrea. > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org > For additional commands, e-mail: dev-h...@openoffice.apache.org >
smime.p7s
Description: S/MIME Cryptographic Signature
