Towards the Design of Robust Peer-to- Peer Communities

Peer-to-peer (P2P) sharing of resources, and technologies for facilitating resource sharing have witnessed tremendous advances in the recent past. A.1; these technologies become commonplace, emphasis must be placed on the survivability of such communities in the face of non-cooperative peers (freeriders, malicious users). While incentive-based approaches provide possible solutions, similar problems in ecological populations are solved by complex social interactions that have evolved over the ages. Evolutionary biology has addressed these problems and numerous models of cooperation between selfish organisms have been proposed to explain how factors such as altruism, guilt, and the sense of justice have evolved in spite of harsh life-or-death conditions. These studies provide blueprints for essential computational techniques in support of stable, scalable, robust, and highly cooperative P2P communities. In this paper, we present a range of stable models of social interaction, their relevance to P2P communities, the associated computational bottlenecks in the context of P2P networks, and motivate the need for the next generation of structured and unstructured resource sharing networks.