On Thu, Jul 18, 2013 at 4:21 AM, Sherif Ramadan <theanomaly...@gmail.com>wrote:
> On Wed, Jul 17, 2013 at 6:47 PM, Lior Kaplan <lio...@zend.com> wrote: > >> What do you think about closing older PR ( > 28 days) ? >> > > First off, thanks for all the hard work. PRs aren't getting as much > attention as they should, and I'd like to say that I certainly haven't been > helping as much as I should with them. What I am noticing though, is that a > lot of these newer PRs, or the ones that are getting more attention, seem > to be fairly cosmetic changes. I mean, spelling mistakes are all great > fixes, and no disrespect to any contribution small or big, but there are > certainly a lot of PRs which people have spent a great deal of time working > on, clearly, that seem to be getting ignored. If we start closing these > strictly because of how long they've been open it would be such a > discouragement to everyone who contributed. I sincerely hope this doesn't > happen. I'd much rather see people spending more time on reviewing some of > the older PRs and seeing if they're worthy of merging or not based on > substance rather than just dismiss them based on how long they've been open. > > I've tried to get in contact with some of the original authors of at least > a couple of PRs in the past weeks to see if they were interested in working > on an RFC to get their PRs merged as they necessitated some more > discussion, but so far have come up empty. So I'll try to help out as much > as I can with the ones that can get immediate attention. > I'm glad to see that question got everyone's attension (: Indeed we started with the easy ones... there isn't any reason simple patches won't get merged very quickly. We have some PR which the disscussion about them is stuck, the question is who and when do we cut it off and send the author to rework his patch. After some feedback was given, if the author isn't responsive, and no one else wants to handle the patch instead, we can reject and ask them to come back when ready. Also, we have more than a few PR which the patch is OK, but are stuck due to a missing test. This is fair enough, but should also have some rule of thumb for these cases. Seeing a long list of PR waiting for a year isn't much encouraging, I would prefer to see people quickly either have their changes accepted, sent to improve the PR or completly rejected. Don't forget the PR can always be resent and the work isn't lost. Kaplan
