Again, this will never be absolutely black and white because it can never be black and white. So we decide how much "noise" is allowed in the signal cloud and then develop a process for managing it.

I daresay that if we developed a robust, clearly-defined process followed by corresponding site design, that easily 85% of submissions would hit the right target first. So the question is, HOW do we handle it?

I still think the path for which to be used can be highly contextual. A given participant's need, background, encounters, etc will largely determine whether they believe they are submitting an enhancement or bug, and I'll bet your typical developer leans toward bug and typical end user leans toward enhancement... at least that's been my experience in device testing. I believe Brainstorm should be the default destination though.

What would really be nice is a wizard for lesser experienced participants, one that asks a handful of questions and guides them toward the better venue based on filtering. No need to slap me, Quim, I realize that's unlikely... just brainstorming.