Democratic technologies: Openness, decentralization, and the success of information systems.

Abstract

There have been many failed and forgotten information systems, resulting in losses of millions of dollars. Some of these systems were created before the Internet heyday in the mid-1990s, some after. Researchers write about these systems variably as centralized and decentralized, or open and closed. With regards to failed systems, writers are often very hopeful about the system's future even after it fails. This leaves two issues that have not been dealt with adequately in the literature. One, why have some information systems failed, even though the systems were backed by experienced and financially strong companies? Two, why is the academic literature divided on the open and closed or centralized and decentralized nomenclature? How are these two seemingly similar approaches related, as they clearly must be? This dissertation seeks to answer these two related questions. First, it asks how openness and decentralization are related by analyzing the available literature. Said literature is diverse, ranging across disciplines, and completely disconnected. By connecting the literature, the concepts of centralization and decentralization are found to be an important subset of the open and closed framework. Once the way for a theoretical approach has been made clear, the dissertation then uses case studies to analyze how the openness of a system affects its success. Four cases are used, ranging from open to closed, and from success to failure. The four are the website Slashdot (open, success), Napster (open, failed), CompuServe (closed, success), and videotex (closed, failed). The study focuses on what parts of a system are open or closed, and how this overall affects the survival of the system. In the end, using a socio-technical approach to systems, two areas, actors and the institutions, are found to be of primary importance. Actors and institutions are not technological, but are social, actors on the user end of the system and institutions on the managing end. It is the democratic balance between these two groups that determines the success of information systems.