Published on *Computer Science Department at Princeton University* (http://www.cs.princeton.edu)

Theoretical work in geometric complexity is often justified by its relevance to key problems of computer graphics, most notable the problems of hidden line and hidden surface removal. We consider a geometric structure - the convex drum - both in the context of a theoretical algorithm for polyhedral intersection

and in a practical context giving an algorithm for computing and decomposing unions of polygons. This is used as a model of situations where theoretical ideas can have relevance to actual implementations.