In early November of 1965, at the height of the cold war, 30 million people living in the most densely populated region of the United States experienced a cascading power failure which blacked out almost the entire Northeast in just under fifteen minutes. Twelve years later, in the summer of 1977, the New York metropolitan region experienced another massive power failure whose cause and duration were at least partly attributable to the preventive measures adopted in the wake of the previous blackout. The 1977 failure, and the devastating rioting it engendered, in turn prompted a dramatic increase in industry regulation which set the stage for today's move toward deregulation. This dynamic of system momentum, unanticipated failure, social response, path-dependent "solution," and subsequent unanticipated failure is the subject of the Blackout site, a web-based research project.

To explore the ways in which this large technological system structured social experience, and was in turn restructured by societal pressures when it failed, the Blackout History Project has identified three target populations for study: 1) the broad public of electricity consumers who lived through the blackouts, 2) the electric utility employees who managed the power grid during both failures, and 3) the public officials and utility officers who sought to manage the crises and prevent them from recurring. The web offers a unique opportunity to establish channels of communication with widely dispersed and socially discrete target populations. The increasing popularity of the web makes it easy for potential respondents to interact with the site, regardless of geographic location. Multiple layers of archival and authored content, structured by a lattice of hypertext-based forums, allow potential respondents and site visitors to navigate according to their divergent interests, gradually branching out to explore unfamiliar topics available at the click of a mouse. Interactive mechanisms -- including a survey form, chat room, newsgroup discussion threads, source material submission form, and context-sensitive comment forms -- are currently under development which will elicit respondent feedback and integrate it into the site with recursive specificity, hopefully producing a history that is literally made by its participants.

Traditional methodologies, such as archival research and face-to-face interviewing, have proven essential to the Blackout History Project's research design. Respondents require some means by which they can adapt to new technologies; established approaches presumably offer a comfortable "on-ramp" to less familiar ways of communicating. Just how essential more conventional methods are to the success of this new approach should be evident by the time of the conference, and this point will serve to illuminate the question of whether the web actually improves our ability to explore the history of technology. This paper will present a two-tiered analysis -- of the blackouts as historical event, and of the Blackout site as historical method -- to explore the ways in which history and technology intertwine.