Peer-to-Peer Technology (P2P)

views updated

PEER-TO-PEER TECHNOLOGY (P2P)

Broadly speaking, peer-to-peer (P2P) technology employs a network to put individuals in direct contact with each other in some form. A simple telephone call to a friend, then, could be considered a form of P2P. In contemporary parlance, however, P2P refers almost exclusively to computer-based systems of sharing information directly with others via the Internet. Peer-to-peer technology has its roots in locally-based hardware arrangements in which each individual system shares certain identical features and capabilities. In the Internet Age, however, P2P refers usually to specific applications rather than hardware arrangements.

The peer-to-peer philosophy and network architecture eliminate the direct top-down relationship between clients and their servers in favor of linking each connected individual as a peer, eschewing the centrality of traditional networks by placing the focal points at each individual computer. Users in a P2P network can pool their resources, sharing each other's files, storage systems, and applications, thereby paving the way for extensive collaboration and efficient information sharing.

Perhaps the most well known example of peer-to-peer technology in the late 1990s and early 2000s was the Napster music file-sharing system, consisting of a distributed network of millions of personal computers, whereby users could search for and download songs directly over the Internet for free. While Napster itself generated a flood of controversy due to allegations of copyright infringement and record companies' fears that such technology would undermine their position in the music business, the technology itself raised many eyebrows in corporate America.

Another major thorn in the side of copyright owners in the early 2000s was Freenet. Created by Ian Clarke, Freenet went Napster one better in that there was no central network for authorities to focus on. Most P2P applications in the early 2000s, including Napster, still required a central server or directory. Freenet consisted of encrypted, anonymous files in the manner of Napster, and was designed ostensibly to protect individual privacy on the Internet by ensuring open access to all files and absolute anonymity, while simultaneously shielding the distribution of shared files from scrutiny. Copyright holders screamed that the system would destroy the integrity of their copyrights by making their work freely and anonymously available to any and all Internet users. But the ability to bypass the need for central networks and the maintenance and administration that accompanies them was particularly attractive to companies seeking to modernize their corporate organizations for the Internet Age. In spring 2001, Intel gave Clarke and his startup firm, Uprizer, $4 million to develop a decentralized distributed networking system for commercial applications.

In July 2000, Napster turned itself into a secure membership-based service with the help of the media conglomerate Bertelsmann. In so doing, Napster created a scheme for copy protection that limits what can be done with the downloaded files. For example, only the PC used in the download is able to play the songs downloaded through Napster. While such moves by what had once been an extremely threatening P2P service were a breath of fresh air to copyright holders, these kinds of restrictions embedded into P2P technology worried some observers interested in the future of P2P applications, insisting that they will inhibit P2P's potential. Others, however, see the field as diverse enough to incorporate a number of eventual niches.

Investment in P2P business applications was pouring in by the early 2000s. Business Week, in a special industrial/technology edition in the spring of 2001, reported that roughly $300 billion in venture capital had been funneled into about 150 P2P startups. P2P companies generate their revenue by selling software packages that consumers can then use to communicate-and collaborate—directly with one another, bypassing central servers. Peers in these distributed networks, according to Business Week, bypass central computing points altogether, so power devolves to the distributed, individual computer nodes, in what proponents herald as a more democratic form of computer networking.

In a sense, P2P brings to life the distribution of power for which the Internet was originally designed, in that it allows computers to communicate directly with each other without going through any intermediary channels. However, Business Week noted that the difference between early Internet designs and the P2P of the early 2000s lay in the vastly superior power of contemporary PCs. Napster, for instance, instead of storing music files on a central server, simply provided means for users to access other members' hard drives where the music files were actually stored. In short, P2P put the largely untapped potential of today's powerful PCs to work. Most computers—particularly those used at home but also in office settings—harbor much more power than is ever actually used.

Security, of course, is a major concern for businesses looking to implement P2P networks. Since P2P allows users direct access to others' hard drives, the protection of corporate data gave pause to executives who saw P2P as the ultimate way to generate efficiency and encourage employee collaboration. The different layers of access that often accompany corporate intranets and extranets would necessarily have to be extended to the P2P systems to afford access to company computers. The idea of installing yet another hierarchy of access controls is unappealing to systems managers who already have several of those to maintain.

Another obvious security concern was the heightened need to safeguard against malicious or careless P2P users uploading viruses directly onto others' computers. In 2001 the first P2P virus appeared as a requested media file on the Gnutella file-sharing system, which infected the users' computers upon download. Corporate systems operators were especially wary of such possibilities to compromise data or cripple crucial systems in a corporate network. And, of course, the larger and more integrated the P2P network, including overlapping and intermingling P2P systems, the more likely that such viruses could potentially spin out of control.