Cerf's Up Cerf's Up
|
Vinton G. Cerf
|
|
|
|
|
Dear Internauts:
September 2 is the 35th anniversary of the installation of the first node of the ARPANET at UCLA in the laboratory of Professor Leonard Kleinrock. Kleinrock was the principal investigator on the ARPANET Network Measurement Center project. Many of us involved in the development of the Internet were present when this first ARPANET node, called an Interface Message Processor or IMP, was installed. Developed by Bolt Beranek and Newman, the IMP was a packet switch (as distinct from traditional telephony circuit switches).
The ARPANET was the first wide-area demonstration of packet switching as a computer communications technology. When the first node was installed, Steve Crocker, was there - he headed the project to develop the Network Control Protocol (a forerunner of the Internet's Transmission Control Protocol). Jon Postel was by then the editor of the Request for Comments (RFC) series of documents and eventually became the Internet Assigned Numbers Authority. Mike Wingfield was there - he had developed the hardware interface between UCLA's Scientific Data Systems Sigma-7 computer and the IMP. I think Charley Kline was there (he was one of the programmers who worked on the operating system of the Sigma-7 and later did some early testing of the computer/IMP interface and inter-IMP communication. I was there too.
Little did we know that this marked a more significant beginning than simply the first installation of the first node of the ARPANET. That was an important development in its own right. Soon SRI International, UC Santa Barbara and the University of Utah were part of the ARPANET. Bob Kahn and his colleague, David Walden, soon visited UCLA to "kick the tires" and try out the software I had developed to assist the Network Measurement Center to do its work. Kleinrock wanted to compare his mathematical queueing models of the networks predicted behavior with actual measurements.
Bob Kahn wanted to force the network into extreme conditions where anomalies might be uncovered. He found several ways to lock the network up by injecting high levels of traffic into the system from the Sigma-7. It was during this early work that Bob and I got to know one another and this relationship led to our later work in 1973 that led to the definition of the Internet architecture and its associated protocols.
2004 marks the 30th anniversary of publication of the first paper on the Internet. It was published in the IEEE Transactions on Communications, May 1974 issue and titled "A Protocol for Packet Network Intercommunication" by Vinton Cerf and Robert Kahn. Little did we know thirty years ago that this research effort would spawn countless initiatives by individuals and organizations, ultimately world wide, to explore the applications made possible by a network of hundreds of millions of programmable devices, soon to be billions.
Comparing the networking world of 1974 to the networks of 2004 is sobering and exhilarating at the same time. The fastest backbone available to the nascent Internet over the ARPANET backbone was operating at 50,000 bits per second. That is NOT a typo. Of course, today, one finds backbone speeds in the tens of gigabits per second with literally terabits per second carried in aggregate over a single optical fiber.
The applications running on the ARPANET were simply email, remote terminal access and file transfers. While these applications continue to be used, they have been joined by the World Wide Web, GRID computing, Instant Messaging, endless varieties of Peer-to-Peer transfers including the song-swapping Napster and its derivatives and successors, voice over IP, all kinds of collaboration and conferencing tools and even an extension of the Internet to operate across the solar system (the InterPlaNetary Network or IPN - see www.ipnsig.org)! The latter has morphed further into an examination of delay and disruption tolerant networking (DTN - see www.dtnrg.org).
It should be no surprise that the experiments in delivering Internet capability over radio links of the mid-late 1970s have re-emerged in the 21st Century in the form of IP-enabled mobile phones, wireless local and metropolitan area networks (e.g. IEEE 802.11, 802.16, 802.20) and satellite systems such as DirecTV. Phased arrays that were once classified military systems technology are showing up in commercial offerings for efficient fixed wireless Internet access. Code Division Multiple Access technology, part of the Packet Radio experiments of the 1970s has emerged as a key ingredient in the mobile 3G environment, delivering hundreds of kilobits up to a megabit per second in burst capacity.
This past year has seen substantial debate on a variety of issues highlighted by the continuing use of the Internet. The exchange of MP3-coded audio raised an uproar in the music media camp and led to numerous lawsuits claiming copyright infringement. One can easily imagine a similar scenario in the case of digital video. However, Apple's iTUNE experiment has reportedly delivered on the order of 25 million songs online since inception. At 99 cents per song, this is a good deal for consumers. Electronic books have had an up and down year with Barnes and Noble closing down its ebookstore. My personal experience is mixed. I like reading books on my laptop since I am on the thing most of the day anyway to do email and surf the Web. But I soon discovered that an ebook is not like a paper book. You can't easily loan it out and if you change computers, beware, it's not very easy to move the copies of the ebooks to the new platform. Until it is as easy to move copies around as it is to move paper copies, ebooks may be a small niche growing smaller. That's a sad outcome in my opinion.
There is renewed interest in delivering video online - Disney has introduced its MovieBeam service using over-the-air data transfers to keep a 160 GB disk drive full of up to 100 movies of 1.5 GB each. Ten movies a week are downloaded into each subscriber's hard drive. Whether the cable companies are prepared to cannibalize their own pay per view business with a similar service remains to be seen.
Radio Frequency ID devices (RFID) are another hot topic with WalMart and Procter & Gamble pushing hard for their suppliers to deliver good with embedded chips. These chips would make it easier to do inventory control and check out at retail stores. Any passive object could be labeled with an RFID - and the resulting combination would allow tracking for inventory control, maintenance record keeping, shelf-life surveillance, and a host of other applications. It should be obvious that such devices could be a great benefit in health care to assure the right medication gets to the right patient and that the medication is still potent.
At the end of 2003, a major conference was conducted in Geneva under the rubric World Summit on the Information Society (WSIS). It was heavily attended by representatives from around the world. The most controversial topics included intellectual property protection, funding for developing country application of information technology, and the governance of the Internet.
The Internet Corporation for Assigned Names and Numbers, ICANN, was a topic of considerable debate. Some countries thought that the domain name system, Internet address system and related unique identifiers associated with the Internet protocol suite should be moved from the responsibility of ICANN to the International Telecommunications Union or other UN body. In my opinion, a great deal of this discussion missed the mark entirely. There are an enormous number of matters that could be said to fall until the general heading of Internet governance. A small part of that is the oversight of the domain name system and the Internet address allocation process. By far, the larger aspect of Internet Governance involves such things as intellectual property protection, fraud, law enforcement, electronic commerce rules and conventions, consumer protection, and a host of other matters that are far beyond ICANN's remit. ICANN has been structured to promote a bottom-up development of policy associated with domain names and Internet addresses. The matter of protocol parameters is essentially the responsibility of the Internet Engineering Task Force. IANA simply records their assignments according to standardization actions taken by the Internet Engineering Steering Group. On occasion, some of these parameters may be assigned by the Internet Architecture Board.
After all the debates leading up to and during the actual Geneva WSIS event, many of the more thorny problems were remanded to the care of a Task Force to be set up by the UN and reporting to its Secretary-General. These matters will be revisited during the 2003-2005 period with a new summit to take place in Tunis in December 2005. I sincerely hope that the Task Force will take steps to broaden the matters under consideration for Internet governance to go well beyond the limited scope of ICANN. And I hope that the Task Force will also recognize that ICANN plays a key role in stabilizing a small but key part of the Internet Technology. ICANN should be permitted to do its job and, assuming it is successful, even the small remaining link between the US Department of Commerce and ICANN will devolve to be no different from all the other relationships that ICANN maintains with governments around the world through its Government Advisory Committee and its interactions with the "country code Top Level Domain operators" (ccTLD operators).
At Cerf's UP we will try to keep you current on developments in the Internet, networking technologies and related regulatory and policy issues affecting the Internet's evolution. We hope to make Cerf's UP a useful point of reference for issues related to the development of the Internet. And of course, you will also find many personal perspectives on the Internet expressed in presentations, interviews, personal observations, technical writing, prose and poetry.
Cerf's UP is evolving to keep pace with the Internet. I am very interested in your perspective so feedback is welcome. Please feel free to email me your questions and comments about Cerf's UP.
Now, I invite you to explore Cerf's UP! See you on the Net!
Vint
| 
