Information is power, and now a set of emerging technologies that enables more connectivity between computer networks could mean more information -- and more power -- for computer users everywhere.
Advocates say we're on the verge of the absolute democratisation of information: a digital utopia, they claim, where your Internet connection brings you all the knowledge you need by connecting you to all the resources that carry it. And thanks to some new technologies it could all take place away from the watchful eye of any central authority.
The central technology at work here is peer-to-peer networking. It may sound like something that takes place in the House of Lords, but it's not at all old-fashioned.
A peer-to-peer network is one in which each workstation has equivalent capabilities and responsibilities. This differs from the traditional client/server and Web architectures, in which some computers are dedicated to serving the others. When you request a file over the Internet it is typically retrieved from a public Web server. Peer-to-peer networking means that file-transfers can take place directly without relying on server computers; and if one part of the network is shut down, the rest will keep running.
The Internet could conceivably become one giant computer through peer-to-peer networking, with important consequences that go beyond file trading and into the realm of copyright, intellectual property and personal freedom.
Peer-to-peer infrastructures aren't the sole preserve of copyright-infringing anarchists however.
Groove, a new product from the creator of Lotus Notes, is new Internet communications software that allows people to make direct connections for real-time interaction. Groove offers a computing platform that can be used for building peer-to-peer computing business applications, which allow the enterprise to connect instantly with partners and customers in a secure and immediate interactive environment.
Groove's open-architecture platform fully supports standard data formats, protocols, APIs, and development tools, so applications can be developed with a minimum of time and investment.
It is a peer-to-peer application that resides on a local machine.
The Jini technology infrastructure, from Sun Microsystems, is a system architecture that joins machines on which Java programming language objects are running. Jini technology enables such objects to work together as though they were on a single, very powerful computer.
"Jini abstracts the physical connectivity from the job at hand. Services such as applications register themselves on the network and announce their functionality to the other elements. Since it's Java-based it can be widely employed," Simon O'Gorman, Consultant Systems Engineer, Horizon Open Systems Ltd, told ElectricNews.Net.
When the 'use-anywhere', Java language first appeared there was talk of a paradigm shift. Many expected Java to dominate software development completely within a year or two. But paradigm shifts take time: there are legacy systems that cannot be abandoned or upgraded and education and training is needed to provide qualified developers to ensure the technology's practicability and availability. Mundane concerns like limited bandwidth and technological readiness will curb the chances of a futuristic peer-to-peer society coming anytime soon.
Many believe that business adoption is necessary for the peer-to-peer concept to be accepted by the mainstream masses. It is probably only when business fully embraces the peer-to-peer concept that it will develop sufficiently to overcome today's technological barriers.
It sounds like the perfect open computer architecture, but there are inherent disadvantages.
Because peer-to-peer networking replaces reasonably efficient client-server transactions with many-to-many packet flurries, network efficiency is drastically reduced. Further, a fully decentralised model means that there's no central location to build a business from.
Then there's the question of trust: do we really want to be interconnected? Some information is confidential and there are simply some people you don't want to communicate with for one reason or another. The prospect of absolute connectivity is probably more appealing than the reality.
MAKING LIGHT WORK
Distributed computing is another factor behind the rapidly changing face of modern computing and informational structures.
This is a form of information processing in which work is performed by separate computers linked through a communications network. Fully distributed computing has separate computers working individually on small fragments of a larger task.
Intel uses the processing power that is left redundant on various machines in various locations when employees have gone home for the night. Employees submit projects to the system and several hundred workstations perform the tasks overnight based on priority.
Intel estimates that it has saved USD500m in computer purchase and support costs over the past decade by employing a distributed strategy.
Despite some doubts about its viability, many believe the technology hurdles to distributed computing may be on the verge of disappearing.
Improved network capacity, bandwidth and high-speed Internet connections in homes and offices has made computer connections more efficient and VPN software has made it secure. Since most of the processing is done on machines and not on the network (unlike peer-to-peer) distributed computing is less affected by technological limitations.
The distributed computing model will probably find acceptance primarily within the confines of a single organisations rather than as an Internet-wide phenomenon using thousands of unsecured computers. A closed environment, such as an educational institution or research lab, allows for greater control over the variables than a mass-market consumer deployment.
Again, it is the vast differences between computer platforms, software versions and numerous other irregularities, that make widespread use of distributed computing impractical for now.
Distributing processing works if multiple computers performing a complex task can accomplish as much as a single supercomputer. That is still a matter for debate amongst the experts, but in the meantime new distributed networks keep appearing.
The potential for file-transfer, dissemination of knowledge, communication, copyright-infringement and hacking are massive via these new structures and technologies. In terms of where it will eventually lead us, it seems that most experts agree that technological constraints coupled with security concerns will prevent a digital utopia from coming soon. But all agree that the times they are a changin'.
The battle between Napster and the giants of the music industry is testimony to the fears of the established order and the potential dominance of the new. And recent events indicate the most likely future for Napsteresque technologies. Napster allows music enthusiasts to search and browse for MP3 files from the computers of other Napster users who are on-line and exchange songs for free in MP3 format. Napster calls it "sharing music on a person-to-person, non-commercial basis."
The music industry couldn't agree. In July, US District Judge Marilyn Patel in California placed an injunction on the Napster service, pending a trial in which the service was to be charged with copyright infringement.
Now the US Ninth Circuit Court of Appeals is deciding whether to shut Napster down even before the trial begins later this year.
"It's clear that a better music-industry strategy would be not to ban Napster's technology, but to make it their own," Jerome Kuptz, a Gnutella protocol contributor, claimed in a recent edition of Wired magazine. Exactly that happened early this month, when German media giant Bertelsmann announced a "strategic alliance" (and an investment of USD100m) with Napster. Bertelsmann subsidiary BMG was one of four US music companies involved in the lawsuit to have Napster shut down. Bertelsmann has withdrawn its suit.
As to the future: "We're going to find a huge upsurge in awareness of peer-to-peer technology," Donal McCarthy, Vice-President, Multichannel E-business, Ebeon, told ElectricNews.Net.
"We're going to find two key types of peer-to-peer application over the next while, one that uses a central server to facilitate the sharing of information (Napster, for example) and one that doesn't have that central server and is truly peer to peer (Groove, for example). I think that we're in the first generation of this technology. We are also seeing the development of business models within the business-to-business arena too. I think that Jini and Groove will have a big part to play here. Peer-to-peer knowledge-sharing solutions will become more popular as business-to-business relationships become centered on the internal expertise of each company."
Emmet Cole is at
|