P2P as R&D
by Dale Dougherty
Right now, anyway, it's too early to be picking winners (unless it's really your job to gaze into a crystal ball). What's going on in the P2P space is research and development into the shape of the Internet. This research
is coming from a variety of sectors: well-funded, commercial companies; seat-of-the-pants startups; well-known universities; and distributed, open source projects some of which are no more than proposals on SourceForge.
So, at O'Reilly P2P Conference, the kind of question to ask is -- what is being learned? How do they see the current state of the Internet? Many are here saying that the Internet is broken, that even the process of sharing a document via a Web site can involve getting permission from an ISP, paying to register a domain name and other such bureaucratic steps. Clay Shirky pointed this out in his keynote, explaining how Napster by making it easier to share files creates "usability at the network layer."
One area that P2P focuses today is discovering what Shirky calls "resources at the edge of the network." These resources are devices, people, and programs. These resources are under-connected, and under-utilized.
P2P seeks to connect these resources without necessarily establishing a central control point. However, some centralization seems to be necessary in most of the P2P systems, and this is an interesting area of research. Clay Shirky said that we should ask of any P2P system if it is "decentralized enough."
For now, it's good to have a lot of different approaches because each is an investigation into solving important problems. Moreover, each is a kind of an assault on the established way of thinking. And for some of us, it is a little bit odd to think of the Internet or the Web as the establishment; but it is (and it isn't). Many of the P2P efforts are valuable critiques on the capabilities of the Internet, and its emergence as a mass medium.
Of course, all the P2P players are hoping that their research leads them to establish a new way of doing things. At that point, the winner will be obvious and the business model will be crystal clear.