On 23 December 2011, at 14:38, Michael Mol wrote: > ... > Devs don't seem to like bug reports involving mixes of stable and unstable. I > was mildly scolded for filing a bug report against xemacs' failing to build > with libpng15, when xemacs was stable, and libpng15 wasn't, yet. I'm trying > to being slightly more conservative about what I unmask.
The devs are just sometimes grumpy and, as a group, inconsistent. Assuming there isn't already an existing bug report for that issue (just makes sure you check *thoroughly* - duplicate bug submissions is a pet peeve of mine) it's perfectly legitimate to report it. If you don't report it, then there'll just be a spate of users doing so when libpng15 is stabilised. It may very well be that libpng15 hasn't been stabilised for this very reason, but in that case there should be an existing bug so you don't waste time reporting the problem. I report bugs all the time saying "please mark this ~x86 package as stable, works for me on an otherwise mostly stable system" and I never get any complaints. The sum of your experience and mine is that devs will complain about this only when it suits them. Gentoo devs are over-worked volunteers, so we must accept such idiosyncrasies. But please don't stop fling useful bug reports. Stroller. (Searching shows that your report *was* found valuable, and that your bug was set as a blocker to bug 354479 until bug 385207 was created. This is *exactly* how it should have been handled, and if you hadn't reported the issue then someone else would have had to find out the hard way. The dev's chiding was a poor choice of words - please continue filing such useful bug reports.)
