Online communities are creating a growing legacy of texts in online bulletin board postings, chat, blogs, etc . These texts record conversation, knowledge exchange, and variation in focus as groups grow, mature, and decline; they represent a rich history of group interaction and an opportunity to explore the purpose and development of online communities. However, the quantity of data created by these communities is vast, and to address their processes in a timely manner requires automated processes. This raises questions about how to conduct automated analyses, and what can we gain from them: Can we gain an idea of community interests, priorities, and operation from automated examinations of texts of postings and patterns of posting behavior? Can we mine stored texts to discover patterns of language and interaction that characterize a community?