On Thu, 2006-03-30 at 22:13 -0800, Paul Eggert wrote: > Ralf Wildenhues <Ralf.Wildenhues@xxxxxx> writes: > > > While I don't mind doing two alpha releases, wouldn't it be more > > sensible to postpone the alpha release until all those pending changes > > are applied? > > No, sorry, by "pending changes" I meant "all the changes that we've > made since 2.59". > > Our main problem will be to get people to actually try the first alpha. For people to try it, you'd have to release it - This hasn't happen so far and we all are facing (exaggeration) the same "ole' stagnating bugged autoconf-crap", similar as it had been during the autoconf-2.13 days. Some people (at least me) are eagerly waiting for autoconf (and even more important to me: automake) to release a new release ... So instead of continuing to wait, I'd recommend to "just release a prerelease with all known bugs" and wait for what's going to happen. If something serious should show up, release a new prerelease ASAP, if nothing serious happens within a (fairly short) prerelease trial time, release "that damned thing". If something seriously should show after the release, fix it and release another release, ASAP. In a nutshell: I recommend a "release early, release often release-policy", instead of continuing to wait for "the day of perfection to happen" :) Ralf _______________________________________________ Autoconf mailing list Autoconf@xxxxxxx http://lists.gnu.org/mailman/listinfo/autoconf
