Social sites are extremely popular among users but user interactions in most sites revolve around relatively simple tasks, such as uploading resources, tagging and poking friends. We believe that social sites can go beyond simple interactions among individuals and offer valuable services to well-defined, closed, communities (e.g., an academic, corporate or scientific community). In this paper, we present an example of a closed-community social system, CourseRank, an educational and social site where Stanford students can explore course offerings and plan their academic program. We perform an analysis of 12 months worth of CourseRank data including user contributed information, such as ratings and comments, as well as information extracted from the user logs, and we analyze several aspects of user interactions and user-contributed content in the site, such as activity levels, user behavior and user content quality. Our findings provide useful insights with respect to the potential of closed-community social sites.