Napster Inc. may be on shaky ground after a US appeals court Monday said the music file swapping service probably violates recording industry copyrights. But peer-to-peer computing can mean much more than simply exchanging a few music files over the Web, and the nascent computing model probably has its best days ahead of it, according to analysts and industry executives.
In simple terms, P2P (peer-to-peer) describes a model of computing in which computers talk to each other over the Web in order to share information and computing resources. Applications based on the model can include file sharing, such as in Napster's service, harnessing idle processing power from thousands of PCs to solve complex problems, or tapping unused storage in PCs and servers to house large databases.
Napster was crucial in helping P2P companies like Groove Networks Inc. of Beverly, Massachusetts, achieve recognition, said Richard Eckel, a company spokesman.
Groove has developed a P2P platform combining software and services that allows small groups inside companies to work together or with other companies. In essence, the company provides an expanded VPN (virtual private network) where companies work across firewalls and boundaries to share information and resources.
Napster brought to light the power of the emerging P2P computing model and helped inform the public about its potential, Groove's Eckel said.
"Napster really helped educate the market," he said. "The down side is that P2P is defined by Napster. We are entering the next phase of P2P."
Indeed, P2P will be the focus of the technology's first industry conference in San Francisco Thursday, where products on display are likely to shift the focus away from its file-sharing roots. P2P products were also on show at Demo 2001, a high-tech trade show this week in Phoenix, Arizona, which also played host to a panel discussion on the topic."Peer-to-peer is more than Napster -- it's not an application, it's an infrastructure," said Chris Shipley, an industry analyst and executive producer of the Demo conference. "Distributed computing is really the thing that's fundamentally going to allow us to scale the Internet."
The emerging model faces some daunting challenges, however, not the least of which is finding sustainable business models for delivering P2P services, attendees at Demo said. On the flip side, consumers may want some payback for the use of their computers in P2P tasks. If a large pharmaceutical company makes billions of U.S. dollars by harnessing the processing power from PCs to design a new drug, for example, those consumers aren't going to be happy if all they receive in return is a free mouse pad, noted Shipley.
"We have to look at the issue of how consumers get rewarded," echoed Michael Tanne, CEO of XDegrees Inc. of Mountain View, California, which offers infrastructure services for building distributed computing applications.
Improving the performance of the Internet could be reward enough, other Demo panelists said. PC users could become part of a distributed caching network in which Web pages are stored in an unused portion of their hard drive. Such a system could boost Web performance for everyone, because users accessing the page for the first time could download it from their neighbor's PC instead of from a server halfway around the world.
Sharing computer resources doesn't have to involve consumer PCs, either, others said. Large corporations could benefit by making better use of resources within their own organizations. VPNs and firewalls used today to keep intruders at bay could form a kind of natural boundary within which a company's P2P network could operate, attendees at Demo said.
"Part of the issue that's being worked out now is how to deal with the firewall," said Neil Iscoe, chief executive officer of eCertain, which makes security software for P2P networks.
Other issues to be addressed include matters of user authentication, the level of access any given user has to another PC, and writing software applications that can take advantage of a distributed computing infrastructure -- a potentially daunting task.
While the path ahead may not be clear, one thing many observers seem to agree on is that Napster's fate is largely irrelevant to the future of P2P computing. The company has done trailblazing for P2P, and the industry can do the rest from here.
Even Jordan Ritter, a cofounder of Napster, agreed with that. He left Napster to become vice president of Round 1 Private Capital Marketplace Inc., which offers a Web-based workflow automation system for the finance industry.
"Napster could go away completely and it would make no difference to the future of peer-to-peer," Ritter said.