Eric Pacuit

Topics in Social Software: Information in Strategic Situations

    Social software is an emerging interdisciplinary field devoted to the design and analysis of social procedures. This new field has recently gained the attention of a wide range of research communities, including computer scientists, game theorists and philosophers. The main idea behind social software is that constructing and verifying social procedures should be pursued as systematically as computer software is pursued by computer scientists. The logical systems developed in this thesis are intended to facilitate such an analysis.

    Although the analogy between computer software and social software is strong, there are some important differences. For example, two issues which are important for an analysis of social procedures but less crucial for computer software are the exchange of (and occasional hiding of) information, and the provision of incentives. Concurrency theory, cryptography and distributed computing have all addressed the first issue. However, many of the underlying assumptions in these fields make applying these results to social procedures unrealistic. The second issue has been more or less the province of game theory. But game theory tends to study the area in rather simple terms lacking the sophisticated tools of computer science such as modularization or data types. The objective of this thesis is to develop formal frameworks which may be used to verify social procedures, with special attention paid to naturally modeling the flow of information in a social situation.