A Malasyian Journal:

Changing the world, quietly

(18 October 1998)

Copyright 1998, D. Crocker, Brandenburg Consulting
A series of notes on living and working in Malaysia, during Jackie's Fulbright Fellowship to Universiti Putra Malaysia, near Kuala Lumpur. Copies may be freely distributed, but must retain this preamble. Previous notes are located at <http://www.bbiw.net/amj>. To subscribe send me a note. /Dave
<mailto: dcrocker@bbiw.net>

More than anything these notes I am writing concern lessons in perspective. Not all of them focus on experiences in Malaysia, directly, but all are affected by it. I've just had an unexpected and upsetting experience and hope you will not mind my exploring it with you:

Jackie and I are visiting Sarawak this weekend. It is the southern of the two Malaysian states on the island of Borneo. A little over one year ago, we visited the northern state, Sabah, and I was confronted with the reality of an undeniably changed world. In the U.S., we still think of Borneo in terms of head hunters in the jungles. Indeed, Jackie works with a professor from the Iban tribe in Borneo and he acknowledges that his great-grandfather did hunt heads, as did all of the warriors in those days. A warrior could not get married unless he had some heads to show as proof of his bravery.

However what we see now and saw a year ago, are modern towns with the usual conveniences. Most astonishing to me, during our trip to Sabah last year, was that the conveniences included a "cybercafe" for Internet access. This fact of global access, reaching all the way to the "wilds" of Borneo, brought home to me just how profound the effect of the Internet is. I was reminded of that fact again, here in Sarawak, when I received news of the death of one of the Internet's true pioneers, Jon Postel.

Few of us get to participate in activities that really do change the world. Fewer still can be counted as principal contributors. For the Internet, a number of people have been put forward as pioneers, some deserving of the label and some not. All of the ones being touted enjoy the limelight. Jon was a notable exception. He only reached the public eye recently and he never sought or enjoyed it. For twenty-five years, he worked to help the community rather than to garner recognition. Most of his effort was in doing operational "scut" work, things that no one else was interested in but that needed doing. So he administered the technical publications series, he administered assignment of registration values for technical protocols, he administered assignment of Internet addresses and Internet names, and he administered operation of the servers that map between names and addresses.

There is no glory in doing administration and operations. Quite the opposite. People notice when it is done badly but rarely offer praise when it is done well. People in administrative positions often become petty bureaucrats. Since there is so little reward in the job, they artificially make it a base of power to increase their sense of self-importance. So it has confused some who heard Jon referred to as the Internet numbers "czar".

They did not realize that the community imparted the title to Jon out of affection and deep appreciation for his having brought order to essential infrastructure services. In particular the community used that term in full knowledge that Jon took his position as a trust, rather than as an opportunity for personal power. We always knew that his views came from legitimate beliefs and we never had to worry that he was somehow considering political or personal advantage. We might not agree with him, but we always knew he was driven first by a concern that the right thing be done.

All this might give you the wrong idea about Jon. I was not a close friend, so I cannot claim to have known him well, only long. But he was entirely human. I certainly knew him well enough to find him a pain to deal with, sometimes, just like anyone else.

To qualify for responsibility over an infrastructure service, one must be conservative. Every change is a danger to the stability of operation, so every change must be resisted. Jon suited that requirement far better than some of us would have liked. In response to most suggestions for change, Jon's first response was "no". It took me many years to learn to put an idea before him and then walk away, rather than to press the arguments in its favor. If I pressed, he would entrench against. If, instead, I walked away, he would always think the issues through carefully and respond constructively. For those of us who believe that at any moment we know Ultimate Truth, it is frustrating to have to deal with someone who approaches things more carefully. Frustrating, but very helpful.

Jon was part of the student mafia that formed the original Computer Science department at UCLA. He went to Van Nuys high school, in the San Fernando Valley of Los Angeles, with my brother Steve, Vint Cerf, and a number of others who formed that first team of students in this new field, at UCLA. It is easy to acknowledge the professors who create an academic department, but it is also easy to forget the role of the first students. In these heady days of the sixties, this crew happened into the beginnings of a research project investigating shared access for long-distance data communication, designed to be robust against failure. They were inventing the Arpanet, which became the Internet. What they did not realize was that they were also inventing a culture.

I was hired onto that project in 1972, just in time for the first public demonstration of the Arpanet in Washington, D.C. The technology had been under development and testing for 3 years and it was starting to move into an operational phase, although continuing experiments with network behavior would often crash the entire, international system. There were a number of teams involved around the country. Officially the one at UCLA was the "Network Measurement Center" since the principal investigator was a leader in queuing theory and one of the research goals in creating the Arpanet testbed was to measure the actual behavior against what the queuing theory work had predicted. Jon, Vint, my brother Steve, and others did participate in that work, but they served a role which I believe was more important in the long run: They led efforts to develop uses for the net, and they created the foundation for a community approach to that development.

I had dropped out of college and this was my first full-time job. My brother had introduced me to computers ten years earlier, but I had limited experience and no formal training. This was not a particularly good background for someone joining a high-powered research project funded by the high-flying Advanced Research Projects Agency. Yet these folks never acted condescending or dismissive. Quite the contrary they were always open to any effort to help. It was the perfect opportunity for real learning and contribution and I watched it repeated with many others who joined the team over the next four years, at UCLA and elsewhere on the Net.

Jon had the dubious privilege of getting me as an office mate. One day I noticed a think-piece that had been distributed by a graduate student at the University of Hawaii. It complained about poor performance over the satellite link to the Arpanet, and suggested a particular approach to solving it. I told Jon that it sounded pretty reasonable to me and might be worth developing as an optional enhancement to the Telnet terminal access protocol. Jon concurred. I had never done any technical work but said I'd be interested in trying to write the specification, if he would help me. He agreed. He mentored the process perfectly, always praising my newest version and then observing a number of fatal flaws. His style was so clear and direct that I was convinced he knew exactly how the protocol should be done but was humoring me through the learning process. I had no understanding of the general ignorance about building network protocols, at that stage of the industry.

Eventually, the specification stabilized and we published it. A few people decided to implement but it soon died away, in spite of his publishing a revision a bit later. After a few years I asked Jon about the reason it failed and he said that it apparently had a fatal flaw which caused client and server machines to lose synchronization with each other. Almost no one knows of this protocol today, but I consider it a superb example of the real "decision" process of the Internet community. Someone suggested an idea. Some others fleshed it out. Still more people tested it. No one complained about authority or scope of responsibility, or following a particular process. No one worried about egos or power. The focus was on the problem and its possible solution. The problem was serious enough and the idea appealing enough to get some people interested in exploring it. The idea failed, but it failed on its merits.

In the last two years, Jon found himself painfully in the public eye. Some of his work had suddenly become quite interesting outside of the core Internet technical community, primarily because a decision at the US National Science Foundation made some of the activities under him worth a lot of money. This started an astonishing sequence of geo-politics and public platform-seeking by many people who had no experience with Internet development, administrations or operations. The money begat power, the power begat politics and the politics begat publicity seekers. Through all of it, Jon focused only and exactly on the underlying work. If he had a failing, it was in refusing to engage in the politics and, perhaps, in failing to institute some changes in his organizational structure sooner. Unfortunately these failings led to his being pilloried by some, with the press all-to-ready to report the dramatic language, but generally unwilling or unable to provide thoughtful analysis.

I recently asked Jon whether he was able to get any real work done, now, or whether he was entirely consumed by the politics which surrounded the changes to his group. He admitted that he had not been able to do any other work for nearly a year. I wonder how I would feel if I spent 25 years offering a community his kind of public service, only to find myself attacked so ruthlessly.

He was given some awards over the last year. Perhaps in response to the attacks, the professional community finally acknowledged his contribution formally. In spite of this praise, it must have been a serious blow to Jon, who has always been so modest and so well-intentioned, to be treated to such attacks. In 1991 he had heart operation and early this month he went into the hospital to have another. It cannot have helped his state of mind to be under exactly the sort of public pressure that he had always avoided. What effect did that pressure have on his ability to recover?

Vint Cerf is again Chair of the Internet Society's Board of Trustees and he has already pledged that there will be a Jonathan B. Postel Service Award, given to those who have contributed to the Internet community. Vint's announcement came just as I was deciding that we needed some sort of continuing acknowledgement of Jon's role in developing not just Internet technology, but Internet culture. I think the service award is exactly the right formal monument.

However I also hope that those engaged in the effort to evolve the organization that Jon built over the last ten years will give him a living, and more practical, monument. I hope that they will emulate his commitment to the community and his focus on constructive, pragmatic evolution, eschewing personalities and politics, and emphasizing community benefit. I hope that as the various factions continue the debate for the evolution of his work, each participant asks themselves carefully and honestly whether their contribution is worthy of Jon.


Dave Crocker

Brandenburg Consulting
Fax: +1(408)246 8253

Tel: +60 (19) 3299 445
Post Office Box 296, U.P.M.
Serdang, Selangor 43400 MALAYSIA

Tel: +1 (408) 246 8253
675 Spruce Dr., Sunnyvale, CA 94086 USA