FRANCAIS

ABOUT THE
INTERNET SOCIETY
ISOC Mission Statement
Membership

CONFERENCE OVERVIEW
Working Party
RealVideo Broadcast
Mbone Broadcast
At-a-Glance
Program
Conference and Program
   Committees
Geneva and Palexpo
Call for Papers
Plenary Speakers
Evening Events
Internet Access Room
BOF Meetings
Pre-Post Tours
Chapter Activities
Internet Related Meetings
Reports From The   Conference

PRE-CONFERENCE EVENTS
K-12 Workshop
Developing Countries Networking Symposium
Technical Tutorials
Network Training Workshops

SPONSORSHIP + EXHIBITION
Invitation to Sponsors
INET'98 Sponsors
Previous INET Sponsors
Sponsor Benefits
Exhibition Hall

MEDIA/PRESS
Press Releases
Media Accreditation and Form
Official INET'98 Publications

REGISTRATION, HOUSING, TRAVEL
Registration Information and Form
Hotel Information and Form
Tour Information and Form
Airline Travel

HELP PROMOTE INET'98

Organizations/Companies Displaying the INET'98 Logo

FREQUENTLY ASKED QUESTIONS

Reports

These reports were written by a team of local volunteers: Angela Merino, Assina Bounis, Celia Boyer, Eric Bianchi, Irčne Butor, Julian Albert Kilker and Melisa Makzume. The reports summarise information for people not able to attend the sessions. Their comprehensiveness and accuracy are not guaranteed. For more information, please contact the presenters directly. Their e-mail addresses are available at http://www.isoc.org/inet98/program.shtml

Track #1: New Applications

Panel: Unexpected Outcomes: Perspectives on the Development of the Internet

By Paul Gillingwater and Julian Kilker, 23 July 1998

"What were you supposed to be doing when you did what you did?" This question was posed by the Chairman of the Panel, Ken Klingenstein of the University of Colorado, to an illustrious group of early Internet innovators, inviting them to recall the elements of serendipity that could be used to develop a "Theory of Surprise."

Vinton G. Cerf of MCI, current Chair of the Internet Society and occasional television actor (he had a walk-on part in one episode of the Gene Roddenberry science fiction television series "Earth: Final Conflict") regaled the audience with a series of anecdotes, both technical and personal, of his "unexpected" experiences during the early days of the development of the ARPAnet, the first generation of today's Internet. He first became involved with computers as a senior in high school, when (with the permission of UCLA) he and a fellow student made their way into the computer room via a window at the weekends to use a Bendix G15 with paper tape I/O. In 1965, he was due to go to Los Alamos, but ended up staying behind to run Quicktran (a Fortran time-sharing system) on an IBM 7044. Studying the assembly code of the Quicktran system made him realize the gaps in his education, so he went to UCLA to study operating systems in more depth. While there, he became interested in the use of computers to measure the operations of other computers, and worked on the wittily-named

"Snooper Computer." Unfortunately, this name caused a certain amount of anxiety once the press got hold of it, and questions were asked in Congress before the team wisely changed the name to an obscure acronym that didn't imply that the computers were being used for privacy invasion.

These and similar experiences of being in the right place at the right time led Vint to develop two fundamental propositions in relation to the Theory of Surprise. One, "Don't Plan on Anything, Because That Will Be Wrong." And two, "When You Have a Choice Between Risky and Non-Risky Things, Always Take the Risky Things, Because They're a Lot More Fun!" The growth of the Internet has not surprised Vint (although the WWW did grow much faster than he expected), but he remembered thinking in the early 1980s that businesses would understand the benefits and efficiencies of e-mail much earlier than they did. In summary, citing his wife's recent experience with a cochlea implant to restore her hearing, Vint predicted that there will be a lot more bio-electronics (bionics), not only for corrective purposes, but also to provide enhancements of human abilities.

The second speaker, Brian Carpenter of IBM, described the epiphany he experienced in 1974 when he built his first network (as a side-effect of a particle physics experiment at CERN), and realized that "Wow! Computers can talk to each other!" It only took 20 years after this for politicians, journalists and entrepreneurs to discover the Internet.

Brian amused the audience with his failed predictions. In 1985, he said that OSI (the CCITT specification for open systems interconnection) would be successful. Wrong. In 1989, he predicted that at least X.400 (the OSI e-mail component) would be successful. Wrong again. And in 1994, he predicted the death of transaction processing. That's three out of three!

Based on these experiences, he concluded that technology always beats policy, because it often gives users what they need. He emphasized that standards do matter, because they freeze change long enough for someone to implement workable solutions, but that he preferred the IETF (Internet Engineering Task Force) approach to standards development, because it does not freeze things too solidly, allowing for innovation as needs change.

Finally, Brian observed that all outcomes are unexpected by at least somebody. Tiny innovations can end up having a major impact (for example,Tim Berners-Lee's realization that hypertext could be implemented over TCP/IP sockets led to the WWW), and that if you wanted to play it safe, you had to change your plan every day.

The third speaker, Doug Van Houweling, also shared with the audience some of his failed predictions. In 1975, he stood up in front of an educational conference and said that future education budgets would be 95% on people and only 5% on technology. Wrong. As a number of economists pointed out to him later, computers are substitutable for people. At a more personal level, he described his fascination with a friend's romance and eventual marriage being supported by online communication. The lesson is that the networking is not just about technology, but about people and technology, and that the two have to be considered together.

Doug also noted that something as apparently chaotic as the Internet, where no single authority has responsibility, could only have developed in the educational sector, where no one is in charge. (In response to a later comment, he agreed that University Computer Centers resisted the introduction of the Internet and should be excluded from this generalization.) In summary, the path to rapid change was to avoid strict hierarchy, decentralize, and delegate decision making to the people directly involved. Unexpected outcomes of the future will be centered around human and societal interaction via the Internet, with communities of interest and affection being a more viable model than hierarchies.

The next speaker, Clifford Lynch, reported on his work in 1979 to help build MELVYL, a joint on-line catalog of the entire holdings of the nine libraries of the University of California system. In the late 1980s this system was connected to the Internet, and one of the unexpected consequences was that most users of the system were not only out of state, but some were even from other countries! This led to some interesting conflicts with the policies of funding bodies, who had to consider how their money was being used to benefit a much wider community than that they originally intended.

A similar situation occurred with the finalization of the Z39.50 searching protocol, which saw content owners worrying because their competitors could provide the front-end interface to their valuable data, thereby depriving them of control over what the customer was seeing. A similar struggle has recently been occurring on the Internet, with practices such as framing and caching of other peoples content. A further interesting result of the way the Internet has grown is the way that intermediary networks without business relationships to the two end parties become critical channels for reliable information delivery, without any formal contracts being in place.

As for his prediction, Clifford suggested that there will be much greater availability of valuable information in the fields of investing and medicine. He warned that there is a risk to the "social archive", because journals, periodicals and other material normally preserved for years in printed form are now being produced exclusively on the Web, leading to an increasing loss of information. Print media survives relatively easily through a species of benign neglect, while digital media survives only through heroic efforts.

The session was brought to a conclusion by Mr Klingenstein, who summarized his theory of unexpected outcomes by an anecdote relating to the invention of the SNMP (Simple Network Management Protocol.) This occurred in a bar one evening, and achieved the subsequently-admired elegance of using only five commands through the simple requirement that it had to fit on a single cocktail napkin. Sometimes, the most surprising ideas are the simplest to express, yet can have a profound impact on our lives.