Wireless Peer to Peer
By Kevin Werbach, Mon Nov 27 00:00:00 GMT 2000
Wireless data and peer-to-peer (P2P) networking,
perhaps the two hottest technology trends of the past year, are linked.
P2P applications allow for direct connections between users at the edge
of the network, instead of requiring data to flow through central
servers. Although today's wireless networks don't operate in
P2P fashion, it's possible to imagine a future in which some of
them do. As wireless devices become more powerful and data-centric, P2P
architectures will make more and more sense.
What is Peer-to-Peer?
P2P technologies have been closely associated with Napster, the music-sharing service that has been downloaded nearly forty million times since its launch in fall 1999. Napster allows users to search and download music in MP3 format directly from other users' hard drives over the Internet, with no restriction on sharing copyrighted material. Due to its stunning growth, and its threat to the music industry, Napster has generated tremendous attention in the media. The Recording Industry of America Association (RIAA) filed suit against Napster alleging that it encouraged copyright violations. Despite this, Napster recently signed a far-reaching partnership with Bertelsmann to create a secure version for licensed content.
Napster is a P2P application because users transfer files between one another, rather than going to a central Website. This eliminates the need for data centers with expensive server farms and high-bandwidth connections. It also makes it harder to control what kinds of files users transfer. There's nothing about digital music downloads that requires a P2P architecture, though, and nothing about P2P that limits it to this application.
In reality, P2P is a form of decentralized networking architecture that has been the subject of computer-science research projects and special-purpose distributed computing applications for years. Only now, though, are companies exploring the potential of P2P for general-purpose end-user services. For example, Groove Networks, led by Lotus Notes creator Ray Ozzie, has created a P2P groupware tool that facilitates ad hoc collaboration among teams of workers. Other startups are using P2P architectures for functions as diverse as search, content delivery, scientific research and instant messaging.
P2P unlocks the latest processing power at the edge of the network. That, more than anything else, is the heart of the P2P value proposition. Today's "client" PCs have a tremendous amount of processing power and storage, most of which goes un-used most of the time. In a client-server configuration, common on the Internet today, users' PCs are often treated as little more than dumb terminals passively displaying Web pages that are assembled at the center of the network.
P2P architectures use those PCs to offload tasks from the core. With millions of Internet users, the combined power of distributed edge processors far exceeds what any central server can achieve. P2P approaches also take advantage of another kind of intelligence at the edge of the network: people. If I find someone whose favorite 10 songs match my own, it's a good bet I'll be interested in some other music they like.
P2P in communications
How does all this relate to wireless?
The communications world has also traditionally looked at devices at the edge of the network as nothing more than dumb appendages. Central switches control how information is routed and the services available to users. Carriers determine the protocols and applications their networks will support, and control how those applications are marketed to users.
The Internet, which uses decentralized routing to move data around, challenges this traditional communications model. The Net has no central control points. Its "carriers," the Internet service providers, don't decide what services and applications are available on the network. As a result, innovators such as Yahoo!, Amazon.com and eBay were able to set up shop and attract customers without going through any gatekeepers.
"It was always in the interests of the phone companies for the Internet not to succeed," says Marc Andreessen, developer of the Mosaic and Netscape Navigator Web browsers and now chairman of Silicon Valley startup LoudCloud. Andreessen, who is excited about P2P and is personally investing in P2P startups, sees a strong parallel between the Nethead/Bellhead battle in communications and the current debate over P2P services such as Napster.
Today, wireless handsets have more processing power than the average PC did a few years ago. Data and voice are converging through devices such as Handspring's forthcoming VisorPhone plug-in for its Palm-based personal digital assistant. Storage and display capacity on end-user devices are also growing, as is the bandwidth available through wireless data networks. For the same reason it makes sense to tap the latent power of PCs at the edge of the network, there is value in thinking of wireless handsets as something more than glorified microphones.
The success of NTT DoCoMo's iMode wireless data service shows the power of allowing anyone to create applications and provide them to users. In contrast to most US and European mobile operators, who tightly control standards and services on their wireless data offerings, DoCoMo created an easy-to-use authoring standard and opened up its platform. Tellme, a leading voice portal service that allows people to access automated information over the phone using speech recognition, has a similar program that allows virtually anyone to develop voice-based applications and make them available.
Though DoCoMo's and Tellme's services aren't P2P, strictly speaking, they point the way to a future in which phone-based data services and users interested in those services connect directly. The network becomes simply the carrier of information; the intelligence and processing happen between the service provider on one end of the network and the user at the other end.
Peering into the future
All the exciting P2P applications being developed for PCs may also start showing up on wireless devices. With several manufacturers set to introduce handsets this Christmas with built-in MP3 music players, a Napster-like P2P file-sharing service is a natural starting point. One of the biggest P2P services in the PC world, instant messaging, is already a big hit on mobile phones in Europe through short message service (SMS). P2P groupware platforms such as Groove will no doubt support wireless devices as one way to send information into their collaborative spaces.
But that's only the beginning. Internet-connected wireless devices move with their users rather than being tied down to a specific location. Peer-to-peer reaches a whole new level when you add the concept of physical proximity between nodes. For basic voice communication, which requires little bandwidth but requires global visibility, it may still make sense to route calls up to a base station and into the existing telephone switching matrix. On the other hand, when the services involved are local and data-oriented, direct connections can be much more efficient.
Think of a wireless notification service that alerts you to interesting places or businesses when you're close by, or that pulls information directly from your contact-management application as you begin a conversation with a sales prospect. These kinds of services will become much more feasible with short-range personal-area networking technologies such as Bluetooth, which are slated to go into widespread deployment next year.
As Internet connectivity becomes much more widespread, we could eventually see wireless networks themselves go P2P. Despite promises that third-generation wireless (3G) will soon allow mobile phones to support megabit data rates, wide-area wireless networks will remain bandwidth-constrained for some time. The technical challenges and investment required to deploy high-speed data connections are substantial, especially since wireless operators generally don't see a revenue model to recover these costs.
Here's an alternate scenario. In urban areas in the US and elsewhere, there is bandwidth everywhere. Office buildings have high-speed local-area networks and homes increasingly have broadband connections using digital subscriber line (DSL) or cable modems. With Bluetooth or a wireless local area network technology such as IEEE 802.11, a mobile device could tap into the bandwidth of the landline network in the vicinity. When the user walked away from the wired network or got into a car, the device could automatically "down-shift" to the slower wireless network until another local connection was available. It's easiest to conceive of this model for data services, though it could work for voice as well.
Will something like this really be deployed? It's hard to say. Much depends on how quickly robust wireless data networks and technologies such as Bluetooth are rolled out. One thing we've learned from the growth of peer-to-peer services - and the Internet itself - is that it's difficult to predict when new technologies will take off, and it's even more difficult to predict the killer applications. Hardly anyone saw Napster or instant messaging coming, yet they are two of the fastest-growing computer applications ever created.
All that's clear is that there are pieces out there waiting to be put together somehow. In the case of peer-to-peer on desktop PCs, it was the untapped processing and storage capacity of machines at the edge of the network combined with the Internet a means to link those machines together. It's difficult to predict how the story will play out in the wireless world, though some of the conditions are starting to look similar. It may be location-based services, or online games, or social applications that push wireless P2P over the edge in the way Napster did for PCs. We'll just have to wait and see....
Kevin Werbach is the Editor of Release 1.0, an influential monthly report that covers the converging worlds of technology, communications and the Internet. He also co-organizes the annual PC Forum and High Tech Forum conferences for technology industry executives.
Kevin is known worldwide as a leading thinker on topics such as the future of e-business, network architecture, convergence and technology policy. An active participant in online communities for over fifteen years, he is particularly interested in the complex ways that new technologies intersect with markets and society.