In fairness, nobody ever "promises", you just make the best guess you can under the circumstances. The problem here, I think, was that there never was a formal announcement that 7.5 was becoming something completely different; I think the change of plans just sort of oozed out into the development community.
If you look at the 7.5 branch activity, the largest number of changes (albeit the smallest ones) happened between rc3 and rc4, which makes sense for a stable release but not for 7.5 as it was originally planned.
So, in short words, the 7.5 release being worked on now is what was originally called 7.6, and was later called 7.5.1. I'm not 100% sure of this but it seems pretty likely.
I see… that's a different impression than I had and explains pretty much. I thought it was meant as "7.5 will be released as planned, but instead of 7.6 being a stable follower of 7.5, this will be 7.5.1 etc. while 7.6 can be branched from master with all the new stuff right after the first 7.5 release".
Originally Posted by bridgman
The problem here, I think, was that there never was a formal announcement that 7.5 was becoming something completely different; I think the change of plans just sort of oozed out into the development community.
The mailing lists archives on sourceforge seem to be somewhat broken. It is hard to get accurate information (or at least was, maybe it has been fixed now). So: yes, such an announcement would have been helpful.
Last edited by Mr. Anderson; 07-14-2009 at 12:56 PM.
Given that everyone now uses a CM tool which handles branches well (git) and the broad availability of packaged builds off development code (edgers) it's not clear that the traditional "unstable releases of Mesa" are actually needed any more.
I don't think the problem is broken archives, but rather that the developers understood what was going on and so no announcement was felt to be required. The mailing lists are aimed at developers, not users; AFAIK the official public schedule is still "when it's ready".
Anything else is usually either (a) original target dates discussed in the context of making content/schedule decisions, or (b) informal "today my guess is..." comments made by folks involved in development, where those comments take on a life of their own and eventually get (wrongly) regarded as plans or commitments.