Over the last half century, the Internet has moved from a medium for resource sharing to one centered on the access and distribution of content. This research effort leverages insights from social science research to characterize shared interest in content distributing services and explore the implications of these findings on the design of existing and future services. Its agenda is predicated on the observation that a thorough understanding of shared interest in content - what content people want to share, when, and where - is key to the design, implementation and management of content distribution services, dictating these services' architectures and dynamic organizations, determining performance experienced by users and the impact they have on the underlying network. The effort leverages the view of users in a popular content distribution system to explore these and other aspects of content distribution and serve as a testbed for gaining social scientific insights in the evolution and dynamics of shared interest and content distribution networks.
Broader Impact: Research findings from this effort will be made available as publications, datasets and software. When applicable, the software will be integrated into existing systems such as BitTorrent, building on our previous successes. The impact of the proposed work will naturally extend beyond existing services and networks to inform the design of future content distribution services and new content-centric Internet architectures through close interaction with the Internet Engineering Task Force and collaborating research labs. In addition to the promising research questions and practical benefits, the project will have a direct impact on education in communication, network and distributed systems.
