Sebastiaan Couwenberg, on Tue 13 Sep 2016 12:26:29 +0200, wrote: > The bug are not reproducible in unstable where we build our packages, we > don't build our packages in testing.
But users do build their package in testing. That's one of the whole points of free software: being able to rebuild the software. Saying the users "well, it did build fine at that point of time in unstable" is useless for them: the release is Stretch, not the state of unstable at some point of development in unstable. Now, if we do know that what is in unstable will get into testing, and then it'll become buildable in testing, it is fine. But saying "we build our package in unstable" is not an argument. Samuel
