Joerg Heinicke wrote: > > But it's obvious we have a more or less bad bug handling. Most of the > bugs are not even ASSIGNED, so nobody can really see if it's only a bug > report of a user or a confirmed bug. Why don't we accept or reject a bug > within a month at the latest. > And the patches? Why aren't they applied or rejected within ... hmmm ... > let's say a half year at the latest? Of course it's not important that > anybody owns the bug, but they should not be ignored. At the moment 40 > bugs are older than a half year (bug number 15.000 and lower). > > Furthermore the informations on a bug like severity are not use as > possible and as they should. We have 4 blocker and 5 critical bugs - > shouldn't this prevent us from releasing new versions? > I think this is plain simple - as long as noone of the committers has time for it and has fun in doing so, nothing will happen. This is open source development and this means, things are not done because someone "forces" you to do them. Work is done because someone has fun in doing so.
I confess that our bug list could be shorter but there are successful (Apache) projects with much longer bug lists. It's correct that we should at least check the severity of a bug, but even this is in most cases a personal perception. E.g. if someone enters a bug with severity blocker, I look at the bug and say "No, it's only a minor bug" and change it, what do you think might happen then? I guess 70% then would say, "No, it's a blocker" and change again. Hmm.. But the good news is: we decreased our bug list from approx 130 to approx 100 in three weeks! So, if everyone helps in continuing this work we could be bug free in two months :) Carsten
