This talk covers the theories and functions dealing with validating input and sanitizing output. It starts at the very basic, and wind up with some discussion of the trickier parts of how different kinds of data should be filtered.