Law-Governed Coordination and Access Control in Distributed Systems

Computing increasingly involves interactions between autonomous and heterogeneous agents distributed throughout a network, agents that have been designed separately, by different people, with little, if any, knowledge of each other. For such disparate agents to work safely and harmoniously together, they must be able to trust each other to observe certain unifying principles and protocols. Unfortunately, in such an open system one cannot expect universal voluntary compliance with any given principle. If required, such a principle must be imposed on all components of a given system. A flexible, scalable, and fault tolerant mechanism for imposing such principles and protocols over open groups of distributed agents is being developed under this project. This mechanism, implemented as a toolkit called Moses, is based on the concept of law-governed interaction (LGI), under which all messages exchanged between the members of a given group of agents are governed by an explicit, declarative, and locally enforced set of rules, called the law of this group. Some of the the potential applications of LGI are being explored. ***