Increasing users ’ control over privacy

P rivacy lobbyists and advocates have named privacy surrounding health information, mobility, and genetic information as some of the most significant issues of our time. Financial and behavioral privacy are also at stake. Even as governments around the world offer services requiring that our personal data be moved online and prepare to upgrade systems to share data among multiple agencies, privacy advocates are working double-time to encourage these governments to build privacy into their systems from the ground up. In an ideal world, there would be no conflict among privacy and e-commerce, high-quality healthcare, and the tracking of personal movement. But in reality, we haven’t achieved such balance, although multiple stakeholders in government, business, and the community aspire to this end. In this article, we present the Personal Context Agent Networking architecture (Pecan) for e-privacy, which we developed to support users in making informed privacy-related decisions in the presence of uncertainty and across a variety of situations and interacting entities. Pecan is currently in prototype stage, and we’re working on some user interface details—specifically, adding natural language capability. We plan to release the experimental prototype in December 2005. We provide a holistic picture of online privacy, identifying privacy’s position in trust and the applicability of a layered model in understanding the complex privacy domain. We apply the model to build a semantic-aware privacy Web ontology, associated external privacy Web services, a client-side user architecture for privacy control, and requirements of the computer–user interface to support user-controlled e-privacy.