Towards a Hippocratic log file architecture

Abstract

The World Wide Web (WWW) is fast becoming the central location for goods, services and information. The very factors that make the Internet such a powerful medium, combine to make the Internet a treasure trove of personal information regarding individual Web users. Users' movements and information are logged as they navigate the Web, often without their knowledge and definitely without their explicit consent. This has lead to internet users voicing concerns over the loss and violation of privacy. Inspired by the Hippocratic Oath, Agrawal et al. [2002] introduced the concept of Hippocratic database systems. These systems are responsible for the privacy of data they manage, and must comply with ten defined principles. In a previous paper, the authors investigated the feasibility of applying the ten principles of Hippocratic databases to the management of log files. The focus of this paper is to expand on the originally proposed Hippocratic log file architecture by introducing a layered view of the architecture. In examining this layered view, the major processes that would be involved in the implementation of Hippocratic log files will be discussed.