<html><body><div>On Aug 31, 2012, at 12:38 AM, Jonathan M Davis <jmdavisProg@gmx.com> wrote:<br><br><div><blockquote type="cite"><div class="msg-quote"><div class="_stretch"> It hasn't even been a month since the last release and activity has been low <br> enough that I really don't see any point in doing another release right now - <br> especially since the beta stuff _isn't_ properly automated yet, so it seems to <br> take a fair bit of Walter's time to deal with it. That and we need to iron out <br> how we're going to deal with branching and the beta, since it definitely looks <br> like we should be moving to a model where we branch for betas. If we actually <br> take the time to sort that out for the next beta like we should, then that <br> means that the beta will take that much more work (though it will mean less <br> work for future betas - particularly if the release process is properly <br> automated like Andrei keeps asking for). So, it seems completely counter-<br> productive to do a beta right now.<br> </div></div></blockquote><span> </span><br>Does it? If we want to change the release process isn't it better to have as few changes as possible. We could have made a dry-release (i.e. without changes) just after the 2.060 release just to test the new process.<br><br>I think it would be better to figure out how we want the release process to look like as soon as possible and not wait until we have enough changes for a proper release.<br><br>The first thing we should have done after the 2.060 release was to figure out a new release process. Then when we're actual want to make a real release we already have the new process figured out and it won't be a problem.<br><br>A release with the new process should be made as soon as possible.<br><br>--<br>/Jacob Carlborg<br></div></div></body></html>