Am 16/12/2013 17:21, schrieb Makarius: > On Mon, 16 Dec 2013, René Neumann wrote: > >> That is, have Isabelle-yyyy-x and Isabelle/Tools-yyyy-x.z with the >> invariant that for every z Isabelle-yyyy-x and Isabelle/Tools-yyyy-x.z >> work well together. >> >> Again: This is just a quick idea, and I have no insight into how feasible the >> differentiation is, especially on the very low levels. > > This would basically mean a second line of development, where problems of > published versions are sorted out incrementally later on. In principle > Isabelle2013-2 over Isabelle2013-1 was that already. The Coq guys also do > that > routinely, with 8.4, 8.4pl1, 8.4pl2, and 8.4pl3 coming out this week (they > have > 5 times more people working on that). > > We did not do this in the past, because it is an extra effort. I am already > feeling mostly alone on just the main line of Isabelle release, with more and > more people just following "the" repository version and not really taking > notice > of the main thing: the published stable releases.
I suspect the mailing list gives a skewed picture because the most vocal people on it are the power users who tend to follow the development closely. But having a stable release is absolutely essential and your work on it is highly appreciated. I hope/assume you are aware of this. (I know that this is not exactly what you were talking about.) Tobias > Going from Isabelle2013 to Isabelle2013-1/Isabelle2013-2 I've spent about 3 > months doing some actual work, and then 3 months trying to get everything > released. With a second line, this ratio would become even worse. > > > Makarius > > > _______________________________________________ > isabelle-dev mailing list > [email protected] > https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev > _______________________________________________ isabelle-dev mailing list [email protected] https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev
