By Dr. Philip Baczewski, Associate Director of Academic Computing
If you liked the original
If you had to pinpoint a trend in popular film for the latter part of this century the fondness for producing sequels would certainly be noticeable. Its not that the sequel was invented in the last 25 years. Its just that it has definitely been overused. Weve seen six Star Trek (original series) films, a contemporary sequel to Gone with the Wind, a handful of Lethal Weapon spin-offs and over 500 Rocky films. Some sequels are an improvement. For example, discriminating geeks everywhere will agree that Star Trek II was much better than the first movie of the same series (popular wisdom is that all the even numbered Star Trek films are better than the odd numbered ones). On the other hand, many, if not most sequels do not measure up to their original models. Who will remotely remember Speed II and who can justify the making of a sequel to Caddyshack.?
Computer technology sequels are even more frequent than film sequels, except with computers we call them upgrades. Sometimes its hard to tell if the sequel is an improvement over the original. The usual elements are there, but they supposedly act better. The difference with technology is that its ongoing march makes upgrades, if not necessary, for the most part desirable. There can be no argument that computer technology is more ubiquitous and usable than it was 20 years ago. If you doubt that improvement, lets play a round of pick up and reorder the punch cards some time. The latest addition to the computer technology arena is the Internet, and the demands placed upon it are quickly outpacing the protocols and data transport schemes upon which it was developed. This can only mean one thing. Its time for Internet2!
The next generation Internet?
Internet2 exists. Well, it almost exists. The Internet2 project is a cooperative effort by a number of U.S. universities and research centers which hopes to promote the development of new networking and application technologies in support of the needs of the academic community (soon to include UNT see "Internet 2 at UNT" from "Campus Computing News" in the January issue and "Campus Computing News" from this issue of Benchmarks Online). The first baby steps of this new network will be taken by the end of this month when the high-speed network project named Abilene begins to route traffic across thousands of miles of fiber-optic cable. Access to Internet2 will be limited to academic and research institutions. To understand the rationale in this regard, you need to study some of your Internet history.
It all began as a research project to see if there was a way to preserve a working communications network in the event of a nuclear war. Arpanet was a U.S. Department of Defense project which eventually linked academic and research institutions involved in military research. Many of the Internet communication methods (or protocols) that are still in common use on todays Internet were developed back in the early Arpanet days of the 1970s. At the same time, the development of computer communications networks such as BITNET and CSNet encouraged academic interest in wide area networking, linking more and more scholars around the U.S. and the globe.
The Arpanet was a patchwork of connections between institutions, and in the late 1980s, the National Science Foundation funded the NSFNet, a high-speed (for then) network backbone to link the many academic and research institutions which had become members of Arpanet as well as additional institutions interested in this new technology. NSFNet further spurred the development of new protocols and applications. In fact, the National Center for Supercomputer Applications (NCSA) came up with a program called Mosaic, which ran on systems employing a graphical user interface (at the time, UNIX X Windows systems, Macintoshes, and a bit later on a fledgling PC-based program called Windows 3.0). Mosaic was the first widely used World Wide Web browser, and was the basis for further developments including a little program now called NetscapeŽ (just think -- if it hadnt been for that government research funding, youd probably be reading this article using your favorite Gopher client).
By the early 1990s, NSFNet had begun to route some commercial traffic, some commercial Internet service providers had begun to appear, and it was apparent that this Internet thing was a must see. In 1993, NSFNet got out of the business of being the primary carrier of Internet traffic, and the era of the expanding commercial Internet had begun. For many of us in the academic community, by this time the Internet was a familiar tool. The "gee whiz" factor had long worn off and wed seen many of the amazing things that are done inside todays Internet browser windows as peoples research projects posted on the Internet as shareware or freeware to try out. It was a very active academic and research community which laid the foundation for the Internet we know today.
Its been about 6 years since the Internet went commercial, and its already almost obsolete. I heard a recent statistic that 50% of all U.S. households owned personal computers and the growth in home computing is almost directly attributable to the Internet. (An aside I find it particularly ironic that Ken Olsen, founder of the Digital Equipment Corporation was attributed as saying that there was no need for anyone to have a computer in their home, and now that company he founded is owned by Compaq, a company that makes personal computers.) As more people use the Internet, more addresses are needed and more "bandwidth" is needed to support the increased amount of information moving over the network (think of bandwidth as a pipe; the larger the pipe, the more that can flow through; in other words, the larger the pipe, the faster your teapot fills with water; the greater the bandwidth, the faster those graphics can be transmitted to your browser screen). Its not just the number of people using the Internet that is straining its ability to function, but it is also the format of information which is being transmitted - digital video, for example geometrically increases the amount of information transmitted.
Internet2 limits access to academia in order to spur the development of technology which will solve some of the problems happening on the Internet. In fact, one of the requirements for joining is a commitment to contribute to the development of new protocols and networking technologies. Abilenes corporate sponsors, Cisco, Nortel, and Qwest Communications International, have donated equipment and resources in the hopes of benefiting from these new developments. For example, network multicasting (kind of like airwave broadcasting) can cut down on consumed bandwidth by making one video stream available to any computer "listening" on the network. Quality of Service protocols may help ensure an uninterrupted data stream for those applications that require it (you really dont want that patented Internet pause in the middle of a remotely-guided surgical procedure). Internet2 will also help develop the next version of the Internet Protocol (the IP in TCP/IP), among other things, geometrically increasing the number of addresses which can be supported on the network.
The Early Reviews
While Internet2 has been in the formation stage for at least a year or more, the start of operations of the Abilene network backbone makes this project even more of a reality. Having a live and technologically advanced network on which to test some theories will go a long way towards developing new Internet technologies. It is too soon, however, to tell what impact this project will have on the future Internet. It may not be possible (or desirable) to recreate the tumultuous and sometimes unorganized environment which spurred the development of the Internet software we commonly use today. It is certain that some process will need to affect change in this area of technology in order for the Internet to remain a growing, vibrant, and useful resource. Still, if Internet2 isnt a hit, I guess we can always wait for BITNET IV: THE REAWAKENING.
* For a trip down memory lane see "The End of the Original Internet" that appeared in the March/April 1995 issue of Benchmarks NewsJournal.. Also, The Internet Society's "A Brief History of the Internet" is fascinating reading and "Hobbes' Internet Timeline v4.0" helps put it all in perspective.
Comments, Questions? Send them to Philip Baczewski.