Intel, still intent on providing developers with some
basic foundation technology for peer-to-peer networking, has delivered its second set of software
development tools for Microsoft users. It hopes the Peer-to-Peer Accelerator Kit for Microsoft .NET
will spur the development community into building P2P Web services products.
The move makes it clear what side Intel is supporting in the Web services battle, and in what direction it expects P2P technology to develop. It doesn't look as if the company will support any non-Microsoft Web services efforts, either in the current product line or future ones.
Context Intel was one of the earliest of the large
companies to put weight behind the peer-to-peer movement, and it uses the technology internally.
Interest in P2P developed rapidly last year in reaction to the huge success of Web-based
file-sharing applications such as Napster. But most of that interest died away again almost as
rapidly, although activity in a few key areas -- such as grid computing -- remains strong. P2P
itself is not a product category in its own right, but an enabling technology for other
Intel released its P2P trusted library software earlier this year, addressing some of the security concerns that continue to surround P2P. It claims not to be trying to establish standards, but says that base-level technology is required if P2P capabilities are to be added to existing applications or developed from the ground up. Its motive is that more PCs, servers and networking equipment (such as routers and network interface cards) will be sold if P2P takes off.
Technology Why the sole focus on Microsoft's .NET?
Intel's belief is that the platform is poised to take off as the next popular development
environment, and says that the .NET framework offers a distributed programming model that is easy
enough to use -- something that's vital for the development of PTP applications.
About 80% of all P2P software will be associated with file sharing, observers believe. One area Intel is concentrating on is "remoting," which supports interaction and file copying between applications hosted on different peers, and also deals with such issues as availability, location independence and encryption. For instance, the Intel toolkit supports the ability to send files even if one of the peers is offline.
Intel has also developed technology to help deal with firewalls, using a new permanent naming scheme that devices would always carry, regardless of their location with regard to the firewall. Other tools cover security and certification, search, content distribution, collaboration and authentication. The kit includes source code, application demos and documentation. The toolkit will be made available for free at Microsoft's .NET news Web site, www.gotdotnet.com, in early December, although that date may change depending on the rollout of Microsoft's technology.
Competition Intel's approach with its P2P tools is
to offer them on a take-it-or-leave-it basis. Plenty of others -- including Groove Networks,
Liquidnet, NextPage, Consilient, SmartPeer, Toadnode, Xdegrees and OpenCola -- are offering P2P
development tools for specific market areas as diverse as content delivery, supply chain
management, securities trading and business-to-business exchanges.
As an alternative to .NET, Sun Microsystems is pushing its JXTA foundation technology, which it has also put into the public domain. Sun claimed in September that it had seen more than 100,000 downloads for JXTA after five months, and that the JXTA.org community had about 6,000 members. Like Intel, Sun hopes its involvement in JXTA will drive sales of its hardware, but also give a boost to its Java development language.
Conclusion Indications to date suggest that the appeal is minimal. Most deployments at customer sites thus far have been limited to small parts of an organization, as users attempt to evaluate the usefulness of the technology and its maturity. Intel's approach -- addressing the largest development base out there and offering free tools -- is probably the right one at this stage.
Expanding P2P architecture beyond its current niche will be challenging, however, because problems around such issues as security, robustness and scalability persist. IT departments are suspicious of information sharing, and on the whole, favor the opposite approach -- information lockdown. And that creates a major problem when trying to promote P2P in mainstream business applications. Any meaningful deployments aren't likely to happen until 2004, if at all.
the451 (www.the451.com) is an analyst firm that provides timely, detailed and independent analysis of news in technology, communications and media. To evaluate the service click here.