I think we're leaning slightly towards being overly cautious. Here's little unofficial note about when to push stuff.
(these guidelines may change if/when we have a GOP policy question(s) in the next few weeks; this email should not be seen as detracting from any future policy discussions) Developers have push access because we trust them, both to not be malicious, and also to be reasonable. You should push a patch if, and only if: you think that nobody will complain about it. That's it. More details: 1. I will complain extremely bitterly if it breaks make. 2. I will complain nastily if it breaks a regtest comparison. 3. I will complain if you push something I wanted to comment on. 4. notwithstanding the above, I will never complain about a patch that has gone through an official patch countdown. We're still sorting out a backlog in the patches -- and that will be much worse right after the fixx.py run on Aug 01 -- but by the end of August, we should have the countdown such that any patch without problems will be officially accepted within 96 hours. I think that's an acceptable amount of worst-case delay for a perfect patches. If you want to avoid the countdown, then it really hinges on point 3. That's most difficult to judge, but you have git push access because you've been around for a while. You should have a fairly good idea of what I like to comment on, what Trevor likes to comment on, Neil, Han-Wen, Joe, etc etc. To pick a completely hypothetical example: if you're working on an obscure aspect of the parser, which virtually nobody understands or cares about, and you're totally certain that your patch will cause no problems and nobody wants to talk about it... then go ahead and just push. If you happened to be incorrect about thinking that nobody would want to talk about your patch, and somebody complains... well, just remember that the next time you're wondering if you should just push vs. waiting for the countdown. It's not the end of the world. You could even view it as a game, or like making a bet! :) Cheers, - Graham _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
