I am very pleased to serve as guest editor of this special issue of Sigecom Exchanges on "Economics of Peer-to-Peer Systems." In this issue, we have a nice mix of research papers and position papers surveying the progress made thus far in this area and mapping out a vision for where the field should be going.
The first paper, entitled "Markets Are Dead, Long Live Markets," looks at why economic approaches to resource allocation in computer systems have thus far not become mainstream and makes an interesting case for why mechanism design and system design should be integrated.
The second paper, entitled "Incentives for Content Availability in Memory-less Peer-to-Peer File Sharing Systems," examines the issue of content availability in peer-to-peer file sharing systems. Previous work has shown that a fixed contribution scheme in terms of the number of files shared per unity of time can result in maximal social welfare, but does not provide mechanisms to enforce this. This paper presents a fixed contribution scheme which does not require the use of system memory, but instead relies only on the amount of time peers are contributing resources to ensure they contribute sufficiently.
The third paper, "A Non-Manipulable Trust System Based on EigenTrust," proposes a strategy proof mechanism that provides incentives for peers to share files in a peer-to-peer file-sharing system and is non-manipulable by selfish interests of peers. The proposed mechanism approximates trust scores based on the well-known EigenTrust scheme.
The fourth paper is entitled "Drawing Crowds and Bit Welfare." Though BitTorrent has become a successful mechanism for sharing files, its greedy, Tit-for-Tat strategy still depends somewhat on "donations" by hosts with complete file copies. This paper demonstrates two counterintuitive schemes in which additionally selfish behavior on the
part of these hosts can improve aggregate system performance.
The fifth paper, entitled "Overcoming Free-Riding Behavior in Peer-to-Peer Systems," surveys recent research that focuses on the design of distributed systems consisting of rational participants. In particular, the paper describes work related to the problem of free-riding in peer-to-peer systems and list the remaining open questions that must be explored to eliminate this problem
The final paper provides a summary report for the Sixth ACM Conference on Electronic Commerce.