To: DMaA who wrote (12582 ) 2/10/1998 1:29:00 PM From: Moonray Read Replies (1) | Respond to of 22053
Next Internet hopes to cut through tangled Web BOSTON (Reuters) - Buried near the end of President Clinton's State of the Union address was a pitch for funding the Next Generation Internet (NGI), an initiative most Americans still know little about. NGI advocates hope it will transform today's slow and sometimes unreliable Internet into a high-speed, multimedia superhighway within the next three to five years. Many of the Internet's most glamorous promises such as worldwide video teleconferencing, virtual classrooms and diagnosing disease from half a globe away have been impossible on any sizable scale because the traffic signaling and routing systems are not sophisticated enough. For example, the Internet's computers cannot distinguish between a video teleconference, in which voice and picture have to arrive at exactly the same moment without delay, and e-mail, which probably could move along in a slower lane without inconvenience. ''Today, the Internet is like a one-lane highway with unlimited access,'' said Douglas Van Houweling, the head of Internet 2, a consortium of universities, the National Science Foundation and technology companies at the heart of NGI. The members of Internet 2, each of whom invests at least $500,000 a year and often much more, hope the NGI will solve this and related problems of handling huge volumes of information at speeds 100 to 1,000 times faster than the current Internet. They are designing the standards and protocols companies will use to create products that can fulfill the Internet's promise. EVENTUAL EUROPEAN LINK POSSIBLE Similar work also is being sponsored by federal agencies with strong research interests such as the Defense and Energy departments, NASA and the National Institutes of Health. Eventually, Internet 2 may be linked with a similar European network called the TEN-34 Consortium based in Cambridge, England. Participants hope to follow the pattern of the current Internet, which began as a private network among scientists and eventually exploded into today's World Wide Web. In development since 1996 but formally organized only late last year, the Internet 2 consortium is now laying the foundation. Sometime in the next few months they will begin testing new applications, most focused on educational needs. Internet 2 will serve as both a laboratory for the new technology and a separate, high-capacity Internet for its members. Within five years, parts of the new technology will be transferred to the so-called commodity Internet, the one almost everyone uses now, van Houweling said. ''The idea is to design things that can be used in an off-the-shelf way in real time using Internet protocols,'' University of Washington Professor Ronald Johnson said. ''This is not just a high-end, pie-in-the-sky, super-performance problem. It's a meat-and-potatoes, garden variety issue, which will determine people's interactions with health care, education and professional opportunities.'' PEOPLE SHOULD SEE REAL DIFFERENCES IN FIVE YEARS Individuals should notice real differences in their interactions on the World Wide Web in the next five years, van Houweling said. ''It will be snappy, quick, just as if you have the material on your local disk. People won't think twice about having a video phone conversation on the Internet.'' For a very few, this technology already is a reality. ''The question is, can millions of people do it simultaneously? That's the main problem,'' said Bill Hawe, vice president of architecture for Bay Networks, a California networking company and Internet 2 member. To accomplish that, scientists must figure out how to make the Internet's computers say much more when they talk to each other. The computers must be able to recognize different types of data that require different levels of service, assign each piece of data to the right category and get it to its destination on time and in sync with related data. And the computers must do this with vastly greater amounts of information than they process now, at higher speeds, with a reliability and security that companies can guarantee to consumers. This means developing more complex routing and switching systems and expanding their inherent capacity. ''Now, you're lucky to make it work at one billion bits per second,'' Washington University's Johnson said. ''We need to push the state of the art to deal with hundreds of billions, then trillions of bits per second. That's where the future is taking us.'' o~~~ O