Andre Poenitz wrote: > On Fri, Oct 11, 2002 at 03:35:23PM +0900, Rob Lahaye wrote: > >>Would it be really such a bad idea to branch off 1.4.0 immediately at the >>moment 1.3.0 goes into feature freeze? > > > Probably a bad idea as people like me would have even less incentive to > contribute to the stabilization of the frozen branch. > > Another option that _might_ work slightly better is to branch the > releases not from CVS head of some arbitrary day but from 'best CVS within > the last few month'. So if many people are happy with 'early August' but > not with 'September 20', we could use 'early August' as start for the > release...
Not a bad idea; a graph with "date" on the X-axis and "stability & happiness" on the Y-axis, would help to find the maximum we need. However,...:). Such a graph would actually be VERY interesting. Does LyX grow like an exponential with a cosine fluctuation, or does it more look like the NASDAC stock exchange...? --- Another thing that keeps popping up in my humble mind: is the present 1.3.0 as it is now, that much worse than the present 1.2.x release; in stability terms, that is. I don't think so, which means you could simply release what is there as 1.3.0 at this very moment; if you can do it for 1.2.x, you surely can do the same for 1.3.0. Yes there are bugs, and there always will be bugs. But they cannot be that serious. I've been using 1.3.0 from almost its start for three papers already; no serious problems at all; I'm not a power user, though. If you release now, you/we can focus on the bugs that the real users are going to complain about. Fix them then, and release 1.3.1 etc. Just a thought. Rob.
