Exploring the Internet: A Technical Travelogue/Prologue
Prologue
The Document Liberation Front
It was August 1991, and I was sitting in a cafe in Geneva, Switzerland, having lunch with Tony Rutkowski, a man who was a senior official in the International Telecommunication Union (ITU) but whom nobody could seem to pin an exact title on. While many called him the General Counsel of the ITU (for reasons that will become apparent), Tony prefered to refer to himself as "Counsellor to the Secretary-General," a title with enough ambiguity to serve his purposes.
Tony and I were celebrating a minor palace coup, a year in the making and the result of a bizarre partnership. How all this came to be can be traced to my propensity to flame.
For several years. I've been complaining about the high cost of international standards documents published by the ITU and other international standards groups. After all, I write professional reference books for a living and these documents are my raw materials.
At first the flames started as simple whining, but soon they grew to full-fledged diatribes in publications running the gamut of technical sophistication from Data Communications to Connexions. I got lots of quotes from illustrious people like Jon Postel, Editor of the Internet Request for Comments (RFCs), who explained that keeping international standards away from "random people" was hurting their acceptance. The gestation time for software based on Open Systems Interconnection (OSI) serves as ample illustration of Postel's point.
The fabled power of the press produced an effect about as pronounced as, say, an editorial in favor of truth and beauty. An occasional nod from a passing reader, but no cries of outrage.
As a sanity check, I posted a note on the Internet asking if anybody else thought that the inaccessibility of international standards was hindering technical progress. This hit a nerve. In fact, the resulting firestorm surprised even survivors of historically notable dialogues such as "should pornography be distributed on the net?"
The Internet has a wide community of users, but a convenient taxonomy splits them into flamers and firefighters. Flaming is easy to explain. You send a message and bingo, you've made 1,000 people all hit the delete key in their mail readers. The power to make so many people twitch is seductive and some people occasionally succumb. In fact, some succumb all too often.
The firefighters have a different motivation. Their happiness is inversely proportional to the size of their electronic inbox. Not everybody was terribly happy when their inboxes started filling with "me too," "I agree," and "another minor yet trivial point" variants on the first wave of reaction.
Dan Lynch, the founder of Interop Company, is a firefighter. He didn't want to see this situation continue forever, so he sent me a copy of a memo written by Tony Rutkowski at the ITU. Turns out Tony was doing the same sort of complaining I was, but was directing his comments internally to the huge ITU bureaucracy.
Tony and I exchanged a few e-mail messages and found we were in violent agreement: standards have to be widely available or the standards are irrelevant. We started trying to figure out what could be done about it.
Turns out that we agreed on another point. We both feel that the ITU has a fairly dubious basis for asserting copyright protection on standards, since a standard spends months (often years) wandering around the public domain as a working-group document before it ever becomes blessed as an official standard and is "published."
You can't make a speech to a crowded room and then, one week later, tell people that your speech contained valuable trade secrets. You can't publish a newspaper and then unpublish it. Once you give something away, it stays given away.
The lack of valid copyright protection is one argument for making standards more widely available, but there is a much more important one. If you believe, like Tony and I fervently did, that the ITU does valuable work, then the standards are an essential public good. Standards are laws and laws must be known to be observed.
Failing to make the standards widely available was, in the long run, going to make the work of the ITU irrelevant. Other standards-making bodies were promulgating standards, and despite its official United Nations treaty status, the ITU had to compete with other groups for the attention of implementors and users.
I sent Tony a note and suggested that since we both agreed that the ITU really had an indeterminate legal basis for asserting copyright, I would simply take all 19,000 pages of the Blue Book, scan it, run it through OCR software, and post it on the Internet for distribution by anonymous FTP, a service that allows a site to give the public access to file archives using the File Transfer Protocol to any user who logs in as "anonymous." (The Blue Book is the massive set of 1988 standards developed by the ITU's Consultative Committee on International Telephone and Telegraph to govern the operation of everything from telephone signalling systems to packet data networks to ISDN to high-speed moderns.)
Now, there are a couple of points worth keeping in mind. Tony was a senior lawyer for a powerful United Nations group that made lots of money selling these documents. While Tony certainly sympathized with my goals, I wasn't quite sure how he was going to react to this form of standards terrorism. Putting a lawyer on notice that you plan to relabel his corporate assets with a $0 price tag is kind of like putting Honda stickers on the motorcycles parked outside a Harley bar.
The other point is that the Blue Book is over 19,000 pages long. Scanners and OCR are great technology, but it would take an awfully long time to carry out such a project on my little home PC. In fact, in a world of finite resources, you might even call my plan an idle threat.
Idle threat or no, this message caught the ITU's attention. One year later, in June of 1991, after endless messages back and forth to Tony, I had received a fax from the Secretary-General asking me to post all ITU standards on the Internet on an "experimental" basis. I promptly booked a plane ticket to Geneva for August to pry the data out of the cold, clammy hands of the bureaucracy. Next thing I knew, I was in Geneva, ready to go to work.
❒
"Mr. Rutkowski's office." She scribbled some notes on a pad.
"I'll give him the message," Julie said, hanging up the phone. Something wasn't right with this picture. As she walked out, I inquired why she didn't just have the phone roll over to her desk if Tony doesn't pick it up.
"We have nothing that sophisticated here at the ITU," she responded with a smile and a strong dose of sarcasm.
A few minutes later, Tony came back and we started talking while Julie ran some errands. Her phone rang and Tony sprinted out to the outer office to take the call.
The ITU was using a very old donated Siemens PBX. This PBX was actually quite advanced for its time, but its time had long passed. This was a feature-rich PBX, but most of the features were so difficult to use that people didn't use them. Tony brought in his AT&T phone from home just so he could have a few stored numbers.
Until recently, in fact, the entire ITU had one fax machine for over 900 employees. A single fax machine was more the result of bureaucratic desire to limit outside communications than a cost factor. The fax machine was in the Secretary-General's office. This meant that sending a fax could require approval right up to the Secretary-General. To get the Secretary-General's approval, you had to go through the Deputy Secretary-General, which would require recursive approval right on down the food chain. Needless to say, to communicate through this medium you would have to be pretty strongly motivated.
What is really funny about all this is that ITU, as part of its enabling treaty, has the right to free telecommunications throughout the world. This perk is meant to allow easy communication to the 160 member countries, and even goes so far as to allow the ITU to set up terminal clusters at off-site meetings with X.25-based links back to Geneva.
I'm not quite sure what I expected at the ITU, but I kind of figured a modicum of computer literacy. Nothing fancy, mind you, but at least things like basic phone tricks, fax machines, and maybe even electronic mail. After all, this is the home of X.400, the message handling system to end all message handling systems. Was the ITU using X.400, though? Not a chance! They had Digital's proprietary Message Router and the infamous All-in-(N)One as a user interface. X.400 connectivity was achieved by a poorly functioning gateway.
This lack of computer literacy at the ITU was what made Tony Rutkowski such an odd duck. When he came to Geneva he saw that none of the international bureaucracies had Internet hookups. So, Tony did what any self respecting technocrat would do and called up the nearest Internet provider, in this case the CERN physics laboratory, and asked for a guest account. Several years later, Tony was still the only ITU employee with access to the Internet.
Tony had taken a rather strange career path to his position at the ITU (a position he left at the end of 1991). After studying engineering at the Florida Institute of Technology, he became quite active in civil rights. Eventually, he landed himself a seat on the city council, getting further involved in policy and politics.
From there, through a roundabout road, his interest in telecommunications drew him to Washington, D.C., where he went to work for the chief scientist at the FCC. The chief scientist was also an Internet old boy, and the FCC was promptly installed as part of the Internet. While working at the FCC by day, Tony put himself through law school at night at American University.
Eventually, Tony went up to Boston to become the publisher of Telecommunications. In Boston, he alleviated the drudgery of punching out a magazine by working as a Research Associate at MIT's Media Laboratory.
Altogether, this mix of lawyer, regulator, and engineer was just-the combination needed when Pekka Taijanne took over the helm of the ITU. A physicist by training, Tarjanne was determined to bring the ITU into the telecommunications age and asked Tony to work as his personal assistant.
❒
My first face-to-face meeting with Tony, after countless e-mail messages, was in August when I took the train in from Paris and went to the twelfth floor of the ITU tower. As I got off the elevator, a short man with a beard darted in after me, posted a piece of paper in the elevator and jumped back out again just as the doors slammed shut.
This was Tony. He was trying to get into every elevator to post a notice of my lecture the next day. The lecture was news to me, so I read the notice. Along with the obligatory hype, the notice explained that I would be lecturing about a host of topics ranging from the future of networking in the civilized world to how more bandwidth would lead to world peace.
After catching the last elevator, Tony and I wandered across the street to the International Organization for Standardization (ISO). It is ironic that the acronym used by this global standards organization is a dyslexic variant of its real name.
ISO is a rather strange group. Unlike the ITU, ISO is a private organization. Like ANSI in the U.S., ISO's lack of real legitimacy to carry out its quasi-governmental functions makes the organization a bit insular and very conservative. Located in a squat, drab building, ISO occupies the dimly lit corridors of the third and fourth floors. The halls are almost always empty, but occasionally one can see a functionary dart out of one office and flit into another.
Tony and I wandered around the halls to find people to invite to my lecture. The purpose of our visit was to make sure everybody at ISO knew about the ITU experiment. I was a bit worried that putting ISO on notice would be considered confrontational, but Tony assured me that there was "nothing wrong with pushing forward the state-of-the-art a bit."
My lecture, by the standards of public bureaucracies, was a stunning success. Aside from the obligatory ITU contingent trying to figure out what their management was up to, there was a large crowd of functionaries from diverse UN contingencies interested in the Internet. Only a few people fell asleep, and a few even asked questions. This was described by several people as a "dynamic" performance.
I described how the standards would be posted on the network, how different access protocols would be used, and how making standards widely available had been so successful for the TCP/IP protocols. I explained how, once the data was digital, we could all start using advanced services, write better code and how, when all was said and done, we would enter a state of standards equilibrium, a nirvana of documentation.
Unbeknown to us at the time, the audience included an operative from ISO, sent over to keep tabs on the radical goings-on across the street. When my own spies informed me of this the next morning, I placed a call to Dr. Zakharov, head of the ISO computer group. I explained who I was and said that, at the request of ITU management, I would be happy to come over and give them a courtesy briefing on our little project.
"That sounds mighty exalted. Come over Thursday morning."
❒
Two days later, I donned my suit and dashed through the rain to the other side of Rue Varembe. I bypassed the obligatory visit to ISO reception and went straight to the third floor and back through a warren of offices, where I was ushered into the inner sanctum of Dr. Zakharov.
Walking in, I saw an older man behind the desk wearing the sort of safari suit that the British favor. "You are Malamud?" he asked. "I have asked McKenzie to join us" he said, beckoning to a young man wearing a shirt and tie who nodded at me with a glum gaze.
"McKenzie is our database expert. I have asked him to be here because what we have here is a database issue, not a network issue."
"Let me first say that I know all about this Internet of yours," he continued. "I used the Internet even before it was the Internet. In fact, before it was the ARPANET. This Internet of yours may be fine for researchers, but here at ISO we have real clients to serve, so enough about this Internet."
I was still standing. A pause introduced itself, so it seemed an appropriate moment for a few formalities.
"Glad to meet you," I said, holding out my hand.
McKenzie continued to glance at the floor and Dr. Zakharov grunted in what I took to be an amiable fashion so I gathered my courage and made a little speech about how my efforts with the ITU were experimental, provisional, something we could learn from, temporary, and any other non-threatening platitudes I could think of.
My little speech contained a fatal flaw. I had used the pronoun "we." Now, "we" is not totally inappropriate as this effort was based on the efforts of several interested parties, but I must admit that the corporate offices of Me, Inc. do not actually teem with corporate retainers.
"We?" Zakharov pounced. "Who is 'we'?" he demanded. "How many members are on your team, anyway?"
Like many corporate types, my lack of a formal institutional affiliation did not sit well with him. I sidestepped his question and explained that, in my view, the question of publishing standards was, rhetoric notwithstanding, just not that hard. After all, there were hundreds of ways to get data on-line, ranging from Wordstar to bitmaps to SGML (an ISO standard).
To say that ISO viewed my little project as a threat was an understatement. Zakharov let slip that the Secretary-General of ISO had sent a letter to his counterpart at the ITU protesting the experiment. Zakharov made it quite clear that he viewed my approach as simplistic, misguided, and naive.
I gently tried to make the point that there were positive aspects to this project that should not be overlooked, such as increasing the public awareness of the vital work that groups like ISO were undertaking by making primary resource documents available to a wider audience.
Once again, my diplomatic efforts seemed less than successful. "People don't need to read the standards," Dr. Zakharov snapped. Indeed, he felt that ISO standards were basically unreadable and that normal people shouldn't even bother to try. What was apparently needed was some guru to write a book, and that these second or third-generation digested versions of standards were more appropriate for the general public. Turns out that Zakharov had himself written such a book ("a monograph, actually") on some long-forgotten CAD standard.
I'm all in favor of promoting book sales, but there is still no substitute for the original documents. A programmer working on an implementation should be required to RTFM ("Read the F***ing Manual" or, in the staid company of standards potatoes, "Read the FTAM Manual"). There is no excuse for second-hand knowledge of something as vital as international standards.
So far, we had spent 45 minutes of a one-hour meeting about databases, and the subject of databases still had not come up. It didn't look like it was going to. McKenzie still hadn't said anything and didn't appear to be about to burst into action.
As I paused to gather my forces, Zakharov jumped in. "Another thing, young man. Do the people in charge of the Internet know about your efforts?"
He seemed convinced that when some appropriate official learned that I would be distributing documents—"with graphics, no less"—on the network, my renegade activities would be quickly stopped.
(Let's put this into perspective. The NSFNET backbone transfers well over 14 billion packets per month. Even if my server was wildly popular, the backbone could handle it. Just to be sure, I was going to send the big computer companies archive tapes so they didn't need to FTP the entire file store into their internal networks.)
I told him that indeed my activities were supported by appropriate officials. The chairman of the Internet Activities Board (IAB), for example, had offered to be one of the anonymous FTP sites for the standards. I didn't tell Zakharov that the incoming chair of the IAB, who had also chaired several ISO committees, had on his home computer many ISO documents and had half-seriously offered them to be posted along with the Blue Book.
"It's not that we're against the Internet," Zakharov explained. "We just want to make sure that all the issues are properly considered. In fact, we are using the TCP/IP protocols here at ISO." ISO had been unable to find enough OSI software to keep their internal network going, so they used a combination of Novell and TCP/IP.
My allotted hour was nearing the end, so I once again began mumbling conciliatory phrases about experiments, cooperation, and similar appropriate fin-de-meeting noises.
"So what do you want from us?" Dr. Zakharov cut in, ever the diplomat.
I pulled out my extra large shovel and explained that since ISO obviously had concerns about the experiment, I was simply here to explain my efforts. "By the way," I continued, "ISO is welcome to join in our little experiment. I have plenty of disk space left."
Zakharov grunted. McKenzie sat.
Zakharov rose. McKenzie rose.
Our little meeting was over.
Back at the ITU, the machinery had swung into action and, with a few false starts, my tapes were being cut.
With a week to kill before my plane left to go back to the States, I decided to try and get in a little fun. Sitting in Tony's office using Telnet to read my mail in Boulder, I looked next to his mouse pad and saw a disorderly stack of business cards. On top was the card of one Jan Gruntorad, head of data communications at the Czech Technical University in Prague.
With the opening of Eastern Europe, this seemed an appropriate omen. I picked up the phone and gave him a call. On Sunday morning, after an all-night train ride, I arrived in Prague. In addition to the usually high ratio of tourists to infrastructure, I had timed my visit to coincide with an international congress of 100,000 Jehovah's Witnesses. Every hotel was full. Each clerk in an endless parade of hotels grunted the same explanation: "We have Jehovah Congress today."
"They've always caused me problems, too" I cracked to one surly reservations clerk, attempting to inject a little levity into my fruitless tour of hotels. The clerk stared at me, giving me the same blank look she would give a moldy potato. My sense of humor was evidently not appreciated.
Most hotels in Prague are privately run, but Cedok, the state tourist agency, still books most of the rooms. In one hotel, for example, 80 percent of the rooms were controlled by Cedok. Cedok, which advertises itself as "the biggest travel agency in Czechoslovakia" (until recently the only travel agency, so it had a long head start), doesn't believe in reservations. When I asked to make a reservation for the next day, they told me to come back the next day and they would see what they could do.
After 8 different hotels, I found a room. No wonder. The Prague Palace, at U.S. $230 per night (payable in foreign exchange) would be no bargain in London or New York. In Prague, it might be considered extortionate.
I had the day off, so I slipped into the mob of Jehovah's Witnesses and set off to see the sights. For lunch, I stopped at one of the overcrowded, spartan restaurants and ordered the house special, a pizza topped with ham, eggs, and peas. The beverage menu also featured "Bois Avocat," a French drink which could be translated either as the blood of a lawyer or some form of avocado nectar. I had a beer.
With relief, I finished my sightseeing and we it back to my room. The number of tourists in Prague was staggering, but they didn't seem to have anything to do. With few restaurants, most milled around in the streets. The Jehovah's Witnesses smiled a lot. It was as if Walt Disney had taken over the country, but had named Kafka managing director.
The next day, I went over to the Czech Technical University. Jan Gruntorad ushered me into his office wearing a Hawaiian shirt, having come in from his holiday a day early to see me. His office was piled high with papers, computer boxes, a huge 3270 terminal, and books. An ancient TV was on top of one cabinet, and on the other were dozens of empty bottles of various kinds. In other words, my kind of office.
For many years, Czechoslovakia had suffered a technology embargo due to COCOM regulations and, like other Eastern bloc countries, concentrated on reverse-engineering IBM mainframes. Jan took me into their computer room and showed me a Russian mainframe clone, together with a Bulgarian-made front end processor (FEP). The MVS-like operating system had been written locally and it had taken two years to integrate it all into a semi-working system.
The Bulgarian FEP and the Russian mainframe were right out of an episode of "Lost in Space." Full of knobs and mechanical registers, it reminded me of early computers like the ENIAC, which can be seen at the Smithonian's American History Museum in Washington, D.C. In a corner were a pile of removable disk packs, each the size of a keg of beer.
Forcing designers to reverse-engineer systems and to serve as super-systems integrators had produced one positive effect. Many of the engineers at the school were quite sophisticated. Unfortunately, however, the operating system had never quite run right and the users never had enough confidence to invest their time learning to do useful work.
In 1989, with the revolution, things changed dramatically. Czechoslovakia was able to put in a 9,600 bps leased line to Linz in Austria and join the European Academic Research Network (EARN), the European BITNET. In Bratislava, another school started using UUCP links over dial-up lines to Vienna and became part of EUnet.
An important breakthrough for Gruntorad came in 1990, when IBM donated a 3090 mainframe as part of the IBM Academic Initiative. Starting with Czechoslovakia, but later including other countries such as Hungary and Poland, plunking supercomputers down in key locations gave an important boost to Gruntorad' s networking efforts (you can't build a shopping mall without upgrading the roads around it).
EARN had expanded to include seven other facilities in Prague. Efforts were underway to extend the national backbone to Bratslava and other cities. From zero users in 1989, the system grew to over 1,500 users by the summer of 1991.
Gruntorad is much more than just the BITNET manager, however. He was leading the effort to install a national networking infrastructure. He wanted to see UUCP, NJE (for BITNET), and TCP/IP protocols all sharing an infrastructure of leased lines.
Leased lines, as in many European countries, are quite expensive in Czechoslovakia. A 9,600 bps line from Prague to Brno, for example, costs 36,000 crowns per month (U.S. $1,242/ month). With monthly salaries averaging 3,000 crowns, a leased line becomes a major line item in the budget. The PTT didn't have the infrastructure in place to properly support digital lines, so a 64 kbps line would be provided by ganging up six 9,600 bps line.
Although I managed to time my departure with what appeared to be the majority of the Jehovah's Witnesses, some other deity must have been on my side. Not having bought a ticket allowed me, through some mysterious bureaucratic rule, to bypass the hundreds of people waiting to check in and report to the excess baggage window.
The burly representative from Czechoslovakian State Airlines who sold me my Swissair ticket even smiled when I suggested she should feel free to close the airport after I had gone—a sure solution to overcrowding. They even allowed me to change my crowns back into dollars without the obligatory official receipt proving that I hadn't used the black market.
Zürich was a real contrast to Prague. I settled into an expensive hotel with small rooms and crossed the street to a beer house, where I feasted on three kinds of sausages, each accompanied by a large frothy mug of local beer.
The next day, I went to meet Urs Eppenberger, a technical manager at SWITCH who explained how the Swiss research network was being established. SWITCH, which stands for Swiss Telecommunication System for Higher Education and Research, was established six years ago by the federal government and the cantons of Switzerland to start and maintain a university research network and a national supercomputer center.
Urs works in the SWITCH network office, which consists of eight engineers, an administrator, and lots of computers. SWITCH is a network infrastructure with services like e-mail. The infrastructure consists simply of leased lines and Cisco boxes installed at 11 sites around the country, plus links into the X.25 Telepac network. The interface between the campus and SWITCH is clearly defined at the Ethernet plug. SWITCH maintains its own system, including configuring the routers.
Originally, the network was going to be religiously OSI, using services such as X.400, FTAM, VTP, and any other OSI services available. The system quickly switched over to a more pragmatic approach, supporting TCP/IP, X.400 over OSI, and even DECnet.
The network is very Swiss. It is carefully planned and carefully implemented. Impeccably engineered, not too adventurous, and focusing on offering a reliable production service, SWITCH is an orderly addition to the carefully manicured Swiss landscape.
I took the train back through the rolling hills of the Bernese Oberland and Fribourg to Geneva. There, I went to see Brian Carpenter, head of networking for CERN, the European physics laboratory.
CERN is one of the major centers for the study of high-energy physics in the world. The laboratory has over 5,000 visiting scientists, plus another 3,000 permanent staff members. CERN is funded by 17 European countries, including a substantial East European membership. The laboratory is mission oriented, a place to do serious, world-class research. It has, as have many other physics laboratories such as Fermilab in Chicago, been on the forefront of networking as a way of allowing physicists to exchange data.
A typical CERN experiment in high-energy physics might involve 400 scientific staff from 20 to 40 different institutions. For example, one of the experiments on the Large Electron-Positron (LEP), the new 27-km accelerator, involves physicists from seven CERN members working alongside scientists from China, India, Israel, Hungary, the U.S., and Russia.
Because of CERN's constituency, it has become one of the global Internet hubs. Much of the connectivity to Eastern Europe funnels through CERN. The CERN mail gateway routes messages for hundreds of networks.
Internally, the CERN network runs the usual motley assortment of technologies, including gigabit Ultra networks, FDDI, and over 3,500 stations on 60 different Ethernets. Over 400 gigabytes per month are transferred on the FDDI backbone alone. Over 1,000 stations on a 4 Mbps token ring are used to control the accelerator.
WAN links at CERN are equally impressive. Over 150 gigabytes per month are transferred out of the laboratory over wide area networks. The laboratory has a 384 kbps link to MIT, a Tl line to Cornell, an El line to Bologna, and an 8 Mbps feed to the CHEOPS satellite. Francois Flückiger, deputy head of networking, estimated that of the total 15 Mbps of international bandwidth devoted to research networking, 11 Mbps terminate at CERN.
I went back to Paris to accomplish the more serious business of eating and drinking until my plane left to take me back to the U.S. In August, Paris transforms itself. The French all leave on their annual four- to six-week holidays. The absence of Frenchmen contrasted sharply with the surge in the American presence.
It seemed like the entire population of New Jersey had come to see the Eiffel Tower. They had brought with them a large portion of Sony's Camcorder output for the past few years. This called for guerilla tourism on my part, avoiding the Champs Élysées and concentrating instead on seedy cafes on the Left Bank. Before leaving, however, there was one bit of business I wanted to accomplish.
On my last day in Paris, I called Gerard Poirot and arranged to visit his office, located in the shadows of what used to be the Bastille. Poirot is an evangelist for videotex. One of the original marketing managers for the French Minitel system, he is active in efforts to spread this underrated technology into the U.S.
When you get a telephone, France Telecom gives you the choice of a phonebook or a computer terminal. As a result, over 3 million people have, free of charge, one of the squat brown Minitel terminals in their home. Of these 3 million terminals, France Telecom estimates 80 percent are actually used. Even if the number is inflated, this was still, until very recently, the largest single network in the world.
The core of Minitel is directory assistance, however. Poirot had a formidable memory, rattling over a dozen phone numbers off the tip of his tongue. When his memory failed, however, he reached over to his Minitel terminal and tapped the two digits to connect him to directory assistance. He typed in the last name and the city, and the fax, voice, and audiotext numbers all appeared on the screen. (Audiotext is the French word for phone-based services like stock reports or voice mail.) Gerard picked the menu item for the voice line, and the number was automatically dialed for him.
Minitel is much more than just directory assistance. The usual bulletin board services—pornography, dating services, IQ tests, and sports results—were all present, but with a few additional surprises.
For example, Nynex had the New York Yellow Pages up on Minitel. Prohibited from offering electronic yellow pages services in the U.S. by Judge Greene's restraining order, Nynex had no such restraint in France and was evidently testing their U.S. system for future deployment. Gerard and I looked up the names of all Bangladeshi cafes in Manhattan and jotted down their locations for reference on subsequent trips to the Big Apple.
Most access to Minitel was based on the French X.25 network, Transpac. With over 300 X.25 switches connected at speeds of up to 2.048 (El), Transpac and Minitel have a symbiotic relationship. A service like Minitel does no good if there is no access path to it. Likewise, an X.25 network is a road to nowhere without useful services on it.
Minitel was a case study in providing useful information services. By the beginning of 1990, Minitel had over 12,000 service providers. Even intra-company transactions, incuding banking, insurance, and order processing, occurred on the network. The directory service alone received well over 10 million calls per day and train reservations received over 1.5 million calls per day.
Leaving Gerard Poirot, I went to catch my plane back to the U.S. Through some fast talking, I managed to get my no-refund, no-change, economy ticket upgraded to a first class seat. It's amazing what you can get, if you just ask.
I sat munching my Beluga caviar and drinking my Dom Perignon champagne and decided that travel wasn't so bad if you just get the right seat. By the time the single malt scotch, the vintage port, and the freshly-peeled kiwi fruits had come and gone, I even became optimistic about the weeks ahead converting ITU data. Drink can certainly cloud your judgment.
At the New York airport on the way back to Denver, I sat in the bar. Over in the corner was a motley assortment of people that could only be a rock and roll band on tour. Leather pants, nose rings, orange hair, and a half-dozen drinks in front of each person were just a few of my clues.
At the check-in counter, the band members stood around making the agents nervous. Evidently, they had booked their seats late and were not pleased with their assignments of middle seats. As I was in the same boat, I stood with this crew harassing the ticket agent. I surreptitiously flashed my frequent flier card and she took my ticket and gave me an aisle seat.
Getting on the plane, I saw that I was next to one of the disgruntled musicians, this one adorned with shoulder length hair and various martial adornments on his leather jacket. He squeezed over me and sat down.
"You have to be with a rock and roll band," I ventured.
He grunted, presumably in confirmation.
When we got into the air, he called the flight attendent over and ordered two double screwdrivers ("hold the orange juice") and began methodically going at them. I drank a beer and started reading The Chinese Screen.
To my great surprise, he leaned over and asked "Is that Somerset Maugham?" I said it was, and we sat there discussing late nineteenth century English literature.
His name was Würzel and he was in Motörhead, the quintessential British loud rock band. They were on tour with Judas Priest and Alice Cooper in a heavy metal extravaganza. Würzel's job was lead guitarist.
"Basically, I try and play as loud as I can," he said, explaining his job.
"Well, somebody's got to do it," I replied. This comment pleased Würzel so much he waved to the flight attendent to bring two more double vodkas for himself and a beer for me, simultaneously repeating my new rationale for his existence to his mates scattered a few rows up. The flight attendent was treating this crew quite gingerly lest they mistake the plane for one of their hotel rooms, and promptly brought the drinks.
Before landing, Würzel handed me his business card and invited me to the big show at some location he couldn't remember. Alas, I was still reading my mail by that evening and let this opportunity for cultural enrichment pass me by.
Although I got back in late August, it was into the second week of September before my tapes from the ITU finally arrived, leaving us about 20 days to convert the tapes before the hooptedoodle began.
At INTEROP 91 Fall, on October 11, a live video link to Geneva had been arranged. Pekka Tarjanne and Tony Rutkowski would make the big announcement about the newly freed Blue Book to a packed and hushed audience in San Jose.
My job was to get the standards converted and on the network before the announcement so there would be no turning back. Until we actually let the documents loose on the Internet, there was always the possibility that the bureaucrats would somehow gain the upper hand and stop the experiment.
A week spent trolling the halls of the ITU had produced documentation on about half of the proprietary, in-house text formatting system they had developed many years ago on a Siemens mainframe. The computer division had given me nine magnetic tapes, containing the Blue Book in all three languages. Despite repeated tries, we had gotten nowhere on trying to also get the CCIR radio recommendations and reports, even though the Secretary-General and the head of the CCIR had both appeared anxious to see this data online.
Along with the magnetic tapes, we had half a dozen TK50 and TK70 cartridges from the ITU VAXen. The cartridges contained PC files that were stored on a VAX using DEC's PC networking products. We had two types of files, one of which was known to be totally useless.
The useless batch was several hundred megabytes of AUTOCAD drawings, furnished by the draftsmen who did the CCITT illustrations. Diagrams for the Blue Book were done in AUTOCAD, then manually assembled into the output from the proprietary text formatting system. The draftsmen were very helpful and quickly said I could have any data I needed.
On my way out of meeting with the draftsmen, however, one of them starting asking some questions about scanners and babbling on about TIFF files. I was puzzled. Why should I care about scanners and TIFF files when I had the diagrams in the original formats?
Turned out that AUTOCAD was indeed used for the diagrams, with the exception of any text in the illustrations. The textless diagrams were sent over to the typing pool, where people typed on little pieces of paper ribbon and pasted the itsy-bitsy fragments onto the illustrations. Come publication time, the whole process would be repeated, substituting typeset ribbons for typed ribbons. A nice production technique, but the AUTOCAD files were useless.
The rationale for this bizarre document production technique was that each diagram needed text in each of the three official languages that the ITU published. While AUTOCAD (and typing) was still being used, the ITU was slowly moving over to another tool, MicroGrafix Designer. There, using the magical concept of layers, they were proudly doing "integrated text and graphics."
The second batch of DOS files looked more promising. Modern documents, such as the new X.800 recommendations, were being produced in Microsoft Word for Windows. My second batch of tapes had all the files that were available in the Word for Windows format, the new ITU publishing standard.
To do the conversion, I was quite lucky to be working with Sun Microsystems on a research grant. They had sent over two large servers for a research program I was participating in and graciously agreed to allow us to use one server to post standards on the network. Without their help, we wouldn't have had the resources to do anything.
❒
The ITU had documented the format we could expect the tapes to be in. Each file had a header written in the EBCDIC character set. The file itself used a character set seemingly invented by the ITU, known by the bizarre name of Zentec. The only problem was that the header format wasn't EBCDIC and the structure the ITU had told us would be on the tape wasn't present.
Using Captain Crunch Decoder Rings, we finally figured out a collating table for the mystery header character set and managed to hack the files off the tape. There were large amounts of data at the beginning and end of files which seemed useless and was simply deleted. We crossed our fingers that the deleted information would not be needed later and indeed, it wasn't.
Next, we had to tackle TPS. This text formatting language was as complicated as any one could imagine. Developed without the desire for clarity and simplicity I had come to expect from the UNIX operating system and its tools, I was lost with the Byzantine, undocumented TPS.
The solution was to take several physical volumes of the Blue Book and compare the text to hexadecimal dumps of the files. I then went to the Trident Café and spent a week drinking coffee trying to make sense of the data I had, flipping between the four files that might be used on any given page of text trying to map events in the one-dimensional HexWorld to two-dimensional events in the paper output.
Inbetween trips to the coffee house, I was trying to take care of diagrams and the PC files. Diagrams were simple: I sat down every morning for a few hours and scanned in diagrams. The diagrams were saved as TIFF and EPS files, then uploaded to our Sun server.
The PC files were all unloaded onto a VAX, then moved over to the Sun, then downloaded at 9,600 bps to my home network. There, the files were loaded into Word for Windows, and then exported as Rich Text Format, the Microsoft proprietary standard for open document interchange. The RTF files were then converted to Word Perfect and ASCII, and all four file formats were sent back up to the Sun.
All told, it wasn't unusual to be downloading and then uploading 10 to 20 megabytes per day, all using a 9,600 bps modern. Still, this was the easy part. It was TPS that almost killed us.
Finally, after pages and pages of PERL code, we had the beginnings of a conversion program. We had tried to use the software developed at the ITU to convert from TPS into RTF, but the code had been worse than useless.
The day before leaving for INTEROP, I was still working desperately away on the conversion program. Tables and equations were still not coming out the way I had wanted them to, but finally, it came time to start hand editing. Any tables that couldn't convert properly were thrown out. Same with equations.
At 2 A.M., with the SuperShuttle coming at 6 A.M., it was finally time to pack for INTEROP (and the six weeks of travel that would immediately follow INTEROP). The data wasn't perfect, by any means, but the Blue Book was on the Internet ready to be distributed. The Bruno project was ready to roll.
As I packed, I reflected on what had been an awful 20 days, programming like a madman to do the conversion. It was amazing how the ITU had been doing a conversion, with lots of people available, but had reportedly estimated that it would take a total of 10 years, and roughly U.S. $3.2 million, to complete the job. By my calculations, if it would have taken 40 days to make the conversion perfect, my time would have been worth, on the ITU scale of reality, U.S. $10,000 per hour. My only hope was that I might be able to use this data to justify an increase in my consulting rates.
As I was going to sleep, I tried to figure out how bad a case of bureausclerosis one would have to have to in order to turn a 20 day (40 to make it perfect) conversion effort into 10 years of agony. Over the next few months, as I visited Geneva again and saw the experiment run into the famed international bureaucracy, I was to learn how it could easily take 10 years, or, more likely, never get done at all.