Given all the time and energy you put into this Remy, I can completely understand.
On the flip side, the overall community would be best served by getting as solid of a point release as possible out sooner rather than later.
If the current 5.0.22 tag is the best you/we have time/energy for, then that seems better than 5.0.19.
I don't know how much better than 5.0.19 this build is. Given that the mood seems to be on the "woops, I did it again" patch, a new tag immidiately seems worthless.
If the plan was to release furiously every couple of weeks or something than I believe everyone's release criteria would be simply "better than the last public release".
Given that you don't expect to tag for "weeks", which I would assume means more like 4-5 than 2 (and would completely understand and concur with this decision as no one can keep up with releases happening every week or two anyway and this sponges up too much time doing release management), then there is a desire to try to get as many of the fixes in process into each public release -- rather than having to refer the masses to CVS for weeks...
One (real) release per month seems to be acceptable.
If you need a particular patch, use the CVS code, it's really easy to build. No one seems to be able to manage it, though.
Rémy
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
