The explosion of digital documents on the internet and in the workplace has led to an increasing need for computer systems that help us not only manage the documents but also manage our understanding of these documents and their relationships. This minitrack focuses on how one gains an understanding of a digital document and how that information is communicated. It encompasses retrieval and text analysis methods, including summarization, categorization and genre theory and detection. This outline for a minitrack has given us the opportunity to select for presentation nine fascinating papers in the general area of how you understand the contents of digital documents, and how you communicate and classify them. We have divided this minitrack into three sections, The Next Generation, Filtering, and Navigation and Composition.The Next Generation In the Next Generation session, Bogaraev, Bellamy and Swart present a discussion of the challenges hand-held devices present to useful delivery on news stories in a constrained environment, and explore the concept of a genre consisting of a summary for hand-held devices. They also explore useful metaphors for navigation in this environment. Cooper, Viswanathan and Kazi describe a research project in identifying important content in a series of recorded customer telephone calls, using a combination of speech recognition, text mining and novel user interface development. They also propose new criteria for measuring the accuracy of a speech transcript. Coden and Brown describe a solution to the problem of finding collateral data to a live television broadcast in real time using a combination of speech and other mining technologies, and show a compelling user interface for the presentation of the discovered data and collateral data. They also describe distance-ranking algorithms for filtering repetitive data.Filtering In the Filtering session, Kruschwitz describes the need for intelligent search engines that are adaptable to new domains. They identify not only keywords and metadata but make use of emphasized sections as well. Shepherd, Watters, Duffy and Kaushik explore using adaptive user profiles to rank web pages and attempt to discover whether such a system can learn effectively about a user's preferences. They compare this to genre identification as an alternative tool for personalization. Waern and Rudstrom also investigate user profiles, with the intention of discovering whether users can understand and improve machine-generated profiles.Navigation and Composition In this third session, we learn about genre and navigation of digital documents. Roussinov and Crowston report on genre based navigation of the web they identify five major genre groups that users could find helpful in a genre-based search and report on search studies based on this model. Schneider and Smoliar discuss description and narrative in hypervideo and how these may best be represented to the user in the context of a compelling hyperlinked video “trip report.” Finally, Lamirel, Ducloy and Kammoun describe their work in providing powerful navigation techniques on the images and text describing art works in the BIBAN multimedia database of the Art Nouveau collection of the Muse'e de l'Ecole' de Nancy.