Autonomous decentralized community concept and architecture for a complex adaptive information system

The global information service in the Internet is a heterogeneous and rapidly evolving environment. Constantly, new information services are added, others are modified, removed or in fault, making it more and more intractable to maintain a coherent image of the information environment. Moreover, users' interests and demands for information services are rapidly changing. In this paper we propose the concept of an Autonomous Community Information System (ACIS) to meet the users' requirements and guarantees the evolution and continuity of the information systems. It allows individual end-users (community members) to communicate directly with one another and share information without relying on any centralized authorities to organize the network. Moreover, it does not load up any single node excessively. For an efficient communication among the community members, we propose an autonomous decentralized community construction technique. It makes hotspots in the community network very improbably and satisfies the fairness by distributing the network traffic evenly among the community members. In addition, it is highly scalable because the complexity of this technique at each node grows logarithmically with the number of the community members.