Automating the Internet: Agents as User Surrogates

The explosion of the World Wide Web as a global information network brings with it a number of related challenges for automation. First, nontechnical users should be able to benefit from the information available on the Web without being overwhelmed by technical detail. Second, users should be freed from mundane and repetitive browsing tasks. Third, and most critical, information from the Web should be available in the format and combination that best fit the user's cask, regardless of the pages on which the information was originally found. The article looks at these issues for Internet automation in the context of new software agent technologies that act as user surrogates for carrying out routine Web activity. Such surrogates enable automation of all interactions with HTML pages and forms-not merely the retrieval of specific URLs-and also the flexible integration of Web information into customized reports and other applications.