Sat Apr 20 12:48:59 EDT 2013

Stilson's Root Locus Adventure

I can see how it gives some insight.  Building linear and circular by
doctoring root locus equations does look like an interesting approach.
However, the subject itself seems rather unstructured, i.e. a lot of
ad-hoc stuff, and additionally, the idea of "constant-Q" controls seem
quite unnatural for discrete time filters.