Landon Cox: The Economics of Peer-to-Peer Networks
Computer scientists are drawn to the many uses of networks
Friday, October 21, 2005
Durham, N.C. -- Imagine snapping a photograph at a party with your cell phone
and getting the names of the people in the picture automatically
inserted into the resulting image file.
To make such a scenario possible, there’s the technical challenge of getting the cell phones of the people being photographed to transmit identifying radio frequencies to the cell phone taking the picture and the software to interpret the signals. There’s also the economic challenge of figuring out what incentives people would need to agree to participate in such a network of sharing digital identities in exchange for automatic photo captions.
Those are the kind of technical and economic issues that Duke professor Landon Cox, explores in his research of peer-to-peer computer networks.
Cox has returned to Duke this year as an assistant professor of computer science after having received his bachelor’s of science in computer science and mathematics from Duke six years ago. He is due to receive his Ph.D. in computer science from the University of Michigan this December.
Peer-to-peer networks, Cox explains, are interconnected groups of computers that perform a function – such as distributing files, routing phone calls or transmitting wireless connections -- without a central server controlling the system. Instead, in peer-to-peer networks, each individual computer interacts with its neighbors according to its own parameters. The most well-known peer-to-peer networks are file-sharing programs, like Napster and Kazaa, used to swap music files; peer-to-peer-based internet phone services, like Skype, are becoming increasingly popular.
Cox is drawn to peer-to-peer networks because of their promise of providing cost-effective, fault-tolerant, efficient, convenient solutions to computing activities like backing up files, transporting data to specific physical locations and sharing personal information in ways that benefit all involved. But Cox is also intrigued by the human dimension of the networks.
“Peer-to-peer all rests on this assumption that people will provide excess resources for everybody else,” he said. “So, the really interesting thing is what happens if people are actively trying to subvert this cooperation and trying to get away with getting something for nothing -- how do we deal with it?
“It’s very easy to model these sharing networks as economies or markets,” he said. “And that’s where the economic tools really become interesting and useful.”
Cox said he is interested in designing these peer-to-peer markets and designing the mechanisms that encourage certain behaviors. “Fundamentally I like to build things; I’m very much an engineer at heart.”
