Is this a feature, or is this a bug?
This one can cause a lot of hand-wringing and a lot of heated tempers, although I'm really never sure why. It gets into the "well, it SHOULD do this, so of course it's a bug and we need to fix it really fast" versus "we didn't screw up and this is something new to us so it's a new feature" argument. This is the realm where people get a bit defensive, particularly when you're facing a cluster of problems that are causing pain.
In truth, it doesn't really matter. Both stories and bugs need to be prioritized, tracked, implemented/resolved, tested, and shipped. However, to forestall meetings where this is argued over, I tend to publicly apply a pretty simple rule:
If it's something we've never tried to do before, then it's a feature. If it's something we've tried to do and have messed up or missed an edge case, then it's a bug.
It's not perfect, and there's still a lot of grey, but it sounds definitive. Usually it's enough to drop it in one bucket or another and move on. Anyone got any better ideas for distinguishing feature from bug?