In this technical summary the theory, methods and results from two papers about conveyor-based part feeding are discussed. The papers summarized are: Computing fence designs for orienting parts from Robert-Paul Berretty, Ken Goldberg, Mark H. Overmars and A. Frank van der Stappen [BGOvdS98] and Parts feeding on a conveyor with a one joint robot written by Srinivas Akella, Wesley H. Huang, Kevin M. Lynch and Matthew T. Mason [AHLM00]. The articles under discussion are about sensorless part orienting on a conveyor belt. Part feeding (Figure 1) on a conveyor is a frequent action in manufacturing processes. This orienting often happens before the assembly, as the assembly in general requires the parts to be in a common orientation. Often this orienting of the parts is done by using sensors to read the orientation of the part and overactuated robot systems are used to orient the part. Both articles discuss sensorless orientation of the parts using underactuated systems. The only input to the algorithms presented in the articles is a description of the part. One algorithm outputs an open-loop (meaning that there is no feedback) sequence of motions and the other algorithm outputs fence angles. The goal of both articles is to find a method to reorient the parts into a single common goal orientation, no matter the given initial orientation of the parts.