Hi, in the light of current events one problem comes up again: how do we deal with enhancement requests reported in Bugzilla?
Currently we count 1'193 open GIMP bugs, of which 607 are enhancement requests. We're drowning in them. In 2015 Jim DeLaHunt pointed out a discrepancy between the website and the documentation, see https://mail.gnome.org/archives/gimp-developer-list/2015-February/msg00003.html Unfortunately the topic wasn't discussed, but still persists and is in a current case time consuming and going blue. In the past I already closed bugs as INVALID until they are agreed upon on the mailing list, but it wasn't considered the best solution. Actually we put the bug on hold, but Bugzilla has no bug state for this. How about this: 1. On enhancement requests we check for Bugzilla duplicates and former mailing list discussions. On duplicates and formerly disagreed requests we close the bug with a suitable reason as DUPLICATE or WONTFIX. 2. If the request is not a duplicate we answer with an friendly stock answer and ask to bring the topic to the developer mailing list. If it's a pure GUI topic, then to the gimp-gui mailing list. Until the topic is discussed with a clear result, we set the bug into NEEDINFO state. 3. On agreement we set the bug into state NEW, otherwise close it as WONTFIX. If the reporter refuses to post the request to the mailing list, then either we post it there if we think it's useful, otherwise we close it as INVALID. If no further discussion is ongoing we close the bugs after an agreed period of time, e.g. 3 weeks since reporting. The goals are getting useful feedback, while keeping the bug base small. It would be nice if we found a final solution now and made it clear at the website and in the documentation. Greetings Sven _______________________________________________ gimp-developer-list mailing list List address: gimp-developer-list@gnome.org List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list List archives: https://mail.gnome.org/archives/gimp-developer-list
