Overview

Expanding on the notion of personal information management (PIM), the Community Information Management (CIM) project is exploring system support for loosely structured, semitrustful communities. Consider, for example, the parents of a youth soccer team that need to share team rosters, phone numbers, practice schedules, action photos, party plans, and other information that may be contributed and updated by various community members. A person may belong to many distinct communities, which change over time with the person’s changing activities, interests, and location. With the emergence of large capacity portable devices, such as cell phones and music/video players with gigabytes of storage, users will want to carry much of their daily information with them so that it is readily available and yet share select items freely within their communities. Such sharing may take place not only via “live” network services but also during interpersonal encounters where direct device-to-device communication is possible. Shared documents and their associated communities should be managed in a seamless way while maintaining accessibility, consistency, and privacy. Addressing the needs of community information management requires research in a number of areas including storage and communication as well as identity, trust, replication, provenance, and contextual awareness.

Research Issues

Fundamental questions to be addressed through this research include the following:

• Communities: How are communities formed? Who keeps track of their membership? How do users discover communities and join them?
• Naming: Do documents and other information items have global, human-sensible names or simply unique IDs? Do communities have global names? If so, what is the structure of names and how are they managed?
• Storage: Where is information stored? Who makes decisions about the placement of information? Do users want personal copies of all information that pertains to them or are they willing to let some of it reside on shared servers?
• Replication: What is the role of replication? What information is replicated to facilitate sharing, to avoid centralized trusted repositories, to increase availability, or to provide reliable backup? What protocols are needed for peer-to-peer, partial replication? What consistency is desired for replicated information? How do community members directly exchange new and updated items?
• Security: How is information protected? Who is allowed to join which communities? How does privacy and preservation affect where information is stored? Who is trusted to update info? How is information recovered if trust is misplaced?
• Schemas: How is information structured? Must the community information management platform be aware of data types and tailor its behavior accordingly? Should the system enforce constraints on information items?
• Lifetime and versioning: For how long is information retained? Do documents and other items automatically disappear from a community when they are no longer relevant? How do people access previous versions of updated information items?
• Context: Does a user’s context, such as his location or meeting schedule, affect what information is visible to the user and how information is managed on the user’s behalf?
• Notification: How do community members monitor the systems’ behavior? Can participants register to be informed of new information items or other conditions of interest?

Interns

• Dan Peek, University of Michigan, Summer 2006
• Meg Walraed-Sullivan, U. C. San Diego, Summer 2006 and 2007
• Prince Mahajan, University of Texas, Summer 2008
• Kaushik Veeraraghavan, University of Michigan, Fall 2008

Overview

• People
• Publications