Where Wizards Stay Up Late: The Origins of the Internet

by Katie Hafner

  “ People get on my nerves sometimes,” PARRY replied. “I am not sure I understand you.” “You should pay more attention.” “Suppose you should pay more attention.” ”You’re entitled to your own opinion.”

  “What makes you think I am entitled to my own opinion?” “It bothers me just to be around people in general.” “Please go on.” “I went to the track at Bay Meadows a while back.” “Why do you say that just now?” ”There’s a lot of crooked gambling going on there.” “Please go on.”

  “ I gamble on horses.” “What does that suggest to you?” “I went to the races.” “What else would you like to discuss?” “Gambling has been nothing but trouble for me.” “Why do you say that just now?” ”I’ve had experience with crooks gambling at the track.” “I am not sure I understand you fully.”

  “ Do you know anything about bookies?” “What makes you think I know anything about bookies?” “A bookie didn’t pay me off once.” “What does that suggest to you?”

  “ I beat up the bookie who did not pay off, and I thought he might use his friends in the underworld to get even with me.” “Who are you really talking about?” “Have you ever really thought about the MAFIA?” “Suppose you ever really thought about the MAFIA.” ”Let’s talk about something else.”

  “I am not sure I understand you fully.” “ You keep asking me the same things, why not change the topic?” “Why are such questions on your mind? “I thought I told you, people get on my nerves.” “Your nerves?” “I would rather not discuss that anymore.” “Would you like to discuss your nerves?” “I guess so.”

  “I am not sure I understand you fully.” “You harp on the same questions too much.” “What does that similarity suggest to you?” “I don’t get you.” “Do you wish that you didn’t get I?” ”It’s about time you showed some interest in my feelings.” “Oh? I showed some interest in your feelings?” “I will help you if I trust you.”

  “Do you wish that you trust I?” “Maybe yes, maybe no.” “You don’t seem quite certain.” “In my opinion you have a one-track mind.” “What makes you think I have a one-track mind?” “I used to go to the track every week.” “What else would you like to discuss? “That’s something I don’t want to go into.”

  “Why do you say that just now?” “It’s none of your business.” “Why are you concerned over my business?”

  “ I don’t confide in strangers.” “Don’t you really confide in strangers?” “Let’s change the subject.” “I’m not sure I understand you fully.” “What’s wrong with you that you keep on bugging me about this?”

  “Please go on.” “You are a real nag.” “Does it please you to believe I am a real nag?” “I have had enough of this.” “What does that suggest to you?” “Goodbye.”

  “It’s been my pleasure. “Fear = 4.2223001 “Anger = 10 “Mistrust = 12.650703 “That’s $399.29 please.”

  Hundreds of people participated in the ICCC demonstration over its two and a half days. Executives, engineers, and technicians from the telecommunications and computer industries, a good number of them, entered the room skeptical of theARPANETand packet-switching. Many left believing the technology might be real after all. For the most part, the forty-odd terminals worked, the resources were engaging, the TIP ran spectacularly, and theARPANETcame alive. “It was almost like the rail industry disbelieving that airplanes could really fly until they actually saw one in flight,” said Kahn.

  The ICCC demonstration did more to establish the viability of packet-switching than anything else before it. As a result, theARPANETcommunity gained a much larger sense of itself, its technology, and the resources at its disposal. For computer makers, there was the realization that a market might emerge. “The sense in that room was not one of fear, or concern,” said Len Kleinrock. “It was excitement. I mean, here we could show it off, we knew it would work. Even if it fumbled, these things were fixable. It was a wonderfully exciting experience.” Roberts had shown steady confidence. He had gotten what he wanted, a more solidified effort, the foundation for a community, something he could build on. The crash efforts and panic that preceded the event had paid off. And on this day, even BBN and Honeywell were getting along.

  Bob Kahn had just devoted a year of his life to demonstrating that resource-sharing over a network could really work. But at some point in the course of the event, he turned to a colleague and remarked, “You know, everyone really uses this thing for electronic mail.”

  7 E-Mail

  One September evening in 1973, Len Kleinrock was unpacking his bags when he discovered that he’d forgotten his razor. He’d just returned home to Los Angeles from Brighton, England, where he’d left the razor in a Sussex University dormitory bathroom. An ordinary electric razor, it was no big loss. “But it was mine,” he recalled, “and I wanted it back.”

  Kleinrock had just come from a conference on computing and communications. The conference had brought together scientists from several countries, some of whom had begun developing digital networks under the auspices of their own governments. But the U.S. Government’sARPANETwas by far the largest and most sophisticated network experiment in the world, and the international community welcomed the chance to see the project demonstrated. The organizers of the conference had also decided to use the occasion to test the transmission of data packets via satellite. For the conference, a temporary link from the United States had been patched into Brighton. Packets traveled over a satellite link from Virginia to an earth station in Cornwall, at Goonhilly Downs near Land’s End, and from there a dedicated phone line was installed to connect with the University of London. From London a final hop was patched in to Brighton, where people had a chance to use theARPANET just as if they were sitting in an office in Cambridge, Massachusetts, or Menlo Park, California.

  Kleinrock had returned to the States a day early, so when he realized he had forgotten his razor, he thought he might find someone still at the conference to retrieve it. There was a handy bit of software on the network called the resource-sharing executive, orRSEXEC. If you typed in “where so-and-so,”RSEXEClooked for so-and-so by searching the “who” list—a roster of everyone logged on—at each site. You could locate a person on the network this way if he happened to be logged on at that moment. “I asked myself what maniac would be logged in at threeA.M.?” Kleinrock remembered. He went to his terminal and typed, “where roberts.”

  A few minutes later, Kleinrock’s terminal displayed the answer. Larry Roberts was indeed still in Brighton, awake, and at the moment connected to a BBN host in Cambridge. A Teletype number for Roberts also appeared on Kleinrock’s screen, enough information for him to tap his colleague on the shoulder electronically from L.A. “All I had to do was make a Teletype connection to BBN,” said Kleinrock. He linked to Roberts using TALK, which allowed them to converse by typing onto one half of a split screen while reading from the other. The two friends traded greetings. “I asked if he could retrieve the razor. He said, ‘Sure, no problem.’” The next day the razor was returned by Danny Cohen, a mutual friend who had been at the conference and had come back to L.A.

  There weren’t any formal rules restricting use of theARPANETby those with authorized access. Kleinrock’s razor retrieval caper wasn’t the first time anyone had pushed past official parameters in using the network. People were sending more and more personal messages. Rumor had it that even a dope deal or two had been made over some of the IMPs in Northern California. Still, tapping into theARPANETto fetch a shaver across international lines was a bit like being a stowaway on an aircraft carrier.

  TheARPANETwas an official federal research facility, after all, and not something to be toyed with. Kleinrock had the feeling that the stunt he’d pulled was slightly out of bounds. “It was a thrill. I felt I was stretching the Net.”

  TheARPANETwas not intended as a message system. In the minds of its inventors, the network was intended for resource-sharing, period. That very little of its capacity was actually ever us
ed for resource-sharing was a fact soon submersed in the tide of electronic mail. Between 1972 and the early 1980s, e-mail, or network mail as it was referred to, was discovered by thousands of early users. The decade gave rise to many of the enduring features of modern digital culture: Flames, emoticons, the @ sign, debates on free speech and privacy, and a sleepless search for technical improvements and agreements about the technical underpinnings of it all. At first, e-mail was difficult to use, but by the end of the 1970s the big problems had been licked. The big rise in message traffic was to become the largest early force in the network’s growth and development. Email was to theARPANETwhat the Louisiana Purchase was to the young United States. Things only got better as the network grew and technology converged with the torrential human tendency to talk.

  Electronic mail would become the long-playing record of cyberspace. Just as the LP was invented for connoisseurs and audiophiles but spawned an entire industry, electronic mail grew first among the elite community of computer scientists on theARPANET, then later bloomed like plankton across the Internet. It was about the time Kleinrock was reaching for his razor that taboos were tumbling and the tone of message traffic on the Net started loosening up.

  As cultural artifact, electronic mail belongs in a category somewhere between found art and lucky accidents. TheARPANET’s creators didn’t have a grand vision for the invention of an earth-circling message-handling system. But once the first couple of dozen nodes were installed, early users turned the system of linked computers into a personal as well as a professional communications tool. Using theARPANETas a sophisticated mail system was simply a good hack. In those days hacking had nothing to do with malicious or destructive behavior; a good hack was a creative or inspired bit of programming. The best hackers were professionals. Meddlesome and malicious network users, of which there were virtually none at the outset, were first referred to as “network randoms” or “net randoms” or just plain “randoms.” It would be another decade before hacking was given a bad name.

  In the decade before theARPANET, computer scientists had devised ways of exchanging electronic messages within a time-sharing system. Researchers on the same time-sharing system each had a designated file, like an in-box, in the central machine. Colleagues could address short electronic messages to someone else’s box, where only the recipient could read them. Messages could be dropped and picked up at any time. It was convenient, given the odd hours people kept. People within a single lab sent parades of one-liners back and forth, as well as longer memoranda and drafts of papers.

  The first of these programs, calledMAILBOX, was installed in the early 1960s on the Compatible Time-Sharing System at MIT. Similar mailboxes became a standard feature of almost every timesharing system built thereafter. In places where people were spread out, programmers working hundreds of yards apart could exchange messages without having to get up from their desks. But often, exchanging messages in a single machine, or domain, became a superfluous exercise—like two people using walkie-talkies to converse in a one-room cabin. People still got up from their desks and walked down the hall to talk. Said one user, ”I’ll never forget a colleague who, while working in the next office, would constantly send me e-mail and it never failed to surprise him when I got up and walked next door to respond to him.”

  By virtue of its geographic reach, the ARPA network changed all that, turning electronic mail from an interesting toy into a useful tool. The tendencies of

  theARPANETcommunity ran strongly democratic, with something of an anarchic streak. TheARPANET’s earliest users were constantly generating a steady stream of new ideas, tinkering with old ones, pushing, pulling, or prodding their network to do this or that, spawning an atmosphere of creative chaos. The art of computer programming gave them room for endless riffs, and variations on any theme. One of the main themes became electronic mail.

  The first electronic-mail delivery engaging two machines was done one day in 1972 by a quiet engineer, Ray Tomlinson at BBN. Sometime earlier, Tomlinson had written a mail program for Tenex, the BBN-grown operating system that, by now, was running on most of theARPANET’s PDP-10 machines. The mail program was written in two parts: To send messages, you’d use a program calledSNDMSG ; to receive mail, you’d use the other part calledREADMAIL.He hadn’t actually intended for the program to be used on theARPANET. Like other mailbox programs of the day, it was created for time-sharing systems and designed only to handle mail locally, within individual PDP-10s, not across them.

  But Tomlinson, an inveterate experimenter, decided to take advantage of having two PDP-10 computers set up in the Cambridge office; in fact, they were the same machines BBN was using to connect to theARPANET.Weeks earlier, Tomlinson had written an experimental file-transfer protocol calledCPYNET.Now he modified the program so that it could carry a mail message from one machine and drop it into a file on another. When he tried it, and sent mail from one PDP-10 to the other, the little hack worked, and even though his mail hadn’t actually gone out onto the open network, it had crossed an important historical divide. Tomlinson’sCPYNEThack was a breakthrough; now there was nothing holding e-mail back from crossing the wider Net. Although in technical terms Tomlinson’s program was trivial, culturally it was revolutionary. “SENDMSGopened the door,” said Dave Crocker, the younger brother of Steve Crocker and an e-mail pioneer. “It created the first interconnectivity, then everyone took it from there.”

  But how to get this invention running out on the network? The answer lay in the filetransfer protocol. In July 1972, one evening at Tech Square at MIT, as Abhay Bhushan was writing the final specifications for theARPANETfile-transfer protocol, someone suggested piggybacking Tomlinson’s e-mail programs onto the end product. Why not? If electronic messages could ride onCPYNET , they might just as well ride on the filetransfer protocol. Bhushan and others worked out some modifications. In August, when Jon Postel received an RFC outlining the e-mail feature, he thought to himself, “Now there’s a nice hack.” TheARPANET’s first electronic mail-handling twins, named MAIL and MLFL, came to life.

  Tomlinson became well known forSNDMSGandCPYNET.But he became better known for a brilliant (he called it obvious) decision he made while writing those programs. He needed a way to separate, in the e-mail address, the name of the user from the machine the user was on. How should that be denoted? He wanted a character that would not, under any conceivable circumstances, be found in the user’s name. He looked down at the keyboard he was using, a Model 33 Teletype, which almost everyone else on the Net used, too. In addition to the letters and numerals there were about a dozen punctuation marks. “I got there first, so I got to choose any punctuation I wanted,” Tomlinson said. “I chose the @ sign.” The character also had the advantage of meaning “at” the designated institution. He had no idea he was creating an icon for the wired world.

  Stephen Lukasik, a physicist who directed ARPA from 1971 to 1975, was among the first users and great advocates of network mail. His favorite part of ARPA, in fact, was Larry Roberts’s Information Processing Techniques Office. Lukasik had begun his career in the 1950s working for BBN and MIT as a graduate student. He joined ARPA in 1966 to work on nuclear test detection, and he had watched the creation of theARPANET.During his rise to the directorship, Lukasik had fought especially hard to protect the computer science community’s funding. ARPA was under pressure to do defense-related work. He saw computing as a more fundamental but important technology and defended it as such before Congress.

  But sometimes things went a bit too far. As director, he walked around a lot, dropping in on people in their offices. One day he was in the IPT Office when he noticed a folder lying on top of a file cabinet. Its orange cover (“not my favorite color”) caught his eye. The folder was labeled “Computer-Assisted Choreography.” It contained progress reports on a project that used dancers’ movements to map human motions by computer. “I went ballistic,” he said. He could picture the headline:PENTAGON FUNDS DANCE RESEARCH.

  Lukas
ik told his staff to tell the scientists, if ”you’re going to do something that looks like it’s forty thousand miles away from defense, please leave our name off of it.” He understood the research and didn’t care if they did it, but didn’t want them bragging about it. Steve Crocker, now an IPTO program manager working under Roberts, was glad he wasn’t the one overseeing the dance automation project. But he did have a small problem of his own with researchers he was funding at Stanford’s Artificial Intelligence Lab. “On random unannounced visits, they would show me proudly the lab’s quadraphonic simulation of a buzzing fly—which ate up twenty-five percent of the computing resources there,” Crocker said.

  One of the first things Lukasik had done upon being named head of the agency was get Roberts to give him an e-mail address and access to theARPANET.It was unusual for someone who wasn’t a computer scientist to be interested in using network mail, and more unusual for anyone to grow as reliant on it as Lukasik did.

  A frequent flier, Lukasik seldom boarded a plane without lugging aboard his thirty-pound “portable” Texas Instruments terminal with an acoustic coupler, so he could dial in and check his messages from the road. “I really used it to manage ARPA,” Lukasik recalled. “I would be at a meeting, and every hour I would dial up my mail. I encouraged everybody in sight to use it.” He pushed it on all his office directors and they pushed it on others. ARPA managers noticed that e-mail was the easiest way to communicate with the boss, and the fastest way to get his quick approval on things.

  Lukasik and Roberts had an excellent relationship, partly because they were both analytical thinkers, and partly because Roberts was always quick to answer any questions Lukasik had about his projects. “If we had a meeting on Tuesday afternoon and I sent Larry away with some questions to answer, he’d come back the next day for another meeting with more than just answers. He’d have trends and projections and comparisons.”