This article discusses best practices for evaluating open source software for use in library projects, based on the authors' experience evaluating electronic publishing solutions. First, it presents a brief review of the literature, emphasizing the need to evaluate open source solutions carefully in order to minimize Total Cost of Ownership. Next, it describes the process used to conduct a comparative evaluation between two open source electronic publishing systems, highlighting aspects of the methodology not described elsewhere. Finally, it concludes with a discussion of reporting the results of an evaluation to decision-makers.