Related LABS tests:
Best Torrent clients
Best private filesharing applications
This feature story appears in the April 2006 issue of PC Authority magazine.
Few technologies have had a more sordid past than peer-to-peer networking. The Napster revolution in 1999 created an unstoppable crusade from users that wanted to share files, and copyright holders trying to prevent it. But like a multi-headed hydra, with each court battle that shut down one network, another would spring up, and we now have more choice than ever to share files. Copyright infringement may be its main use at the moment, but it's an ignorant assumption that piracy is its foundation. People want to share information, and they want to use the technologies available to them to do it.
And they will. Strings of lawsuits later, and P2P is showing no signs of decline. It currently accounts for more Internet traffic than any other medium, and businesses are trying to harness its power to distribute content faster, cheaper, and wider. Piracy is still rampant, but peer-to-peer is now entering a new phase of maturity, where private, secure networks are helping groups connect together, and technologies like BitTorrent have revolutionised the way files are distributed.
P2P now offers more ways for people to connect than ever before. While Napster dealt a healthy blow, and Kazaa made a slight detour, the future of P2P is brighter than ever. This feature looks beyond the teething problems and examines the future of P2P networking, the latest networks, and how you can make the most out of the original bad boy of Internet.
Power to the people
At its core, the Web has always been peer-to-peer. The fact that any individual with an Internet connection, a PC and HTTP server software can throw up a site contributed significantly to the exponential expansion of the Web, and the Internet as a whole. But while the early web featured interconnected, distributed nodes (web servers), it was still based on a traditional client/server system, rather than the P2P networks around today.
The first mainstream P2P services used a combination of interconnected nodes (or peers) and centralised tracking servers. These servers aren't involved in the file transfer, rather they act as 'matching' services to connect one node to another. The most well known network to use this strategy was Napster. Napster was predominantly a music sharing network, which hosted a server to track users and files. Even though the actual process of file swapping was happening between the nodes, Napster's involvement led to its eventual demise where it was ordered to pay several millions of dollars to copyright holders. Only recently, after being acquired by Roxio, has it relaunched into a legitimate business.
But far from stymieing the use of P2P networks, it initiated increasing interest in filesharing, driving the creation and use of decentralised networks which do not need a centralised server to function. When a node connects to a decentralised P2P network, it will often connect to a small number of nodes that are (hopefully) online. Each node that it connects to is also connected to its own list of other nodes, and so on until, in theory, each node can form a link to every other node.
To search for a file, a request is sent out to only those nodes that have a direct connection, and the request is subsequently 'passed on' to other nodes on the network. By not having a central point of failure, this can keep the network always up and running. However, searching for files can be slow as the user waits for all nodes on the network to return the results.
All networks around now use either centralised or decentralised formats, and often a combination of both.