Within the computer-science community, submitted conference papers are typically evaluated by means of rating, in two respects: First, individual reviewers are asked to provide their evaluations of papers by assigning a rating to each paper’s overall quality. Second, program committees collectively rate each paper as being either worthy or unworthy of acceptance, according to the aggregate judgment of the committee members. This paper proposes an alternative approach to these two processes, based on rankings rather than ratings.