IDK about this implementation ...

OVER-Denouncing ought to be tracked, too, for a user's trustworthiness profile.