Just for Fun : The Story of an Accidental Revolutionary

by Linus Benedict Torvalds

  For the first two weeks, I didn’t even bother bringing over my computer. Not counting my army service, those two weeks were the longest span of time that I had been away from a computer since I had been eleven years old and sitting on my grandfather’s lap. Not to dwell on this, but it still holds the record for being my biggest stretch—as a civilian—without a CPU. Somehow. I managed (again, the details aren’t interesting). My mother, the few times I saw her then, would mutter something about “a triumph of Mother Nature.” I think my sister and father were just stunned.

  Soon, Tove went out and got a cat to keep Randi company. Then we settled into a nice pattern of spending evenings alone or with friends, waking up at 5 A.M. so she could get to her job and I could go to the university early, before anyone would be there to disturb me, and read my Linux email.

  King of the BALL

  I

  The birth of version 1.0 meant something new for Linux: the need for public relations. I would have been just as happy to introduce the new version to the world pretty much the way I had introduced previous versions. I would write something on the newsgroup like “Version 1.0 is out. Deal with it.” (Okay, not in those exact words.)

  A lot of other people thought it was much more of a big deal. They wanted version 1.0 for marketing purposes. There were all these budding commercial companies that had started to sell Linux. To them, version 1.0 was important for psychological, not technical, reasons. I couldn’t disagree. The fact is, it looks bad when you try and sell version 0.96 of an operating system.

  I wanted it out because it was a milestone for me, and because it meant I could stop fixing bugs for a while and go back to development. The companies and the Linux community wanted to foist it onto the public in a major way.

  We needed a public relations strategy. I wasn’t going to personally champion the effort. I wasn’t interested in putting out press releases or making statements. Others thought it should be done that way, so others volunteered to pick up the torch. This was pretty much how Linux itself was done—and somehow, it all actually worked.

  Lars was one of the driving forces behind making that first official release a real event. He and a few others thought the university would be the most appropriate place to make the announcement. It made sense. My bedroom was too small. And it would have set a wrong precedent to host the announcement at a commercial site. So Lars volunteered to coordinate the event with the university. The computer sciences department at the University of Helsinki was small enough that he could just talk to the head of the department.

  The University of Helsinki was more than happy to offer up the main auditorium of the computer sciences department for the introduction of Linux Version 1.0. And why not? How often does a university have anything worthy of television news?

  I did agree to give a talk. But it bore none of the horror of my Ede experience. Okay, some things were harder, now that I think about it.

  Like having my dad in the audience. And the fact that it was on Finnish TV. It was the first time I ever got the chance to see myself on television. Both of my parents were in the audience (but I’m fairly certain they weren’t sitting together). Tove was there, too. It was the first time my dad met Tove, so for me it was more than just the announcement of Version 1.0. Since I was involved in the last-minute speech preparations, like making sure my slides were okay, I wasn’t there when they actually met. That apparently happened when they were walking into the auditorium. Maybe I caught it out of the corner of my eye.

  In that talk, and in virtually every other one over the next several years, I spoke not so much about the technology but about Open Source. It was nice. It changed some opinions about Linux inside the computer science department. Before that, Linux was something the computer science department was proud of, and mildly encouraged. But after the announcement, people within the department started taking Linux more seriously. After all, they had seen it on the news.

  Over the years, some have suggested that the university was trying to take credit for Linux. That wasn’t the case. The department had always been very supportive. They even gave me a job that enabled me to work on Linux on their time. And that was in the early days, so nobody was saying, “‘Let’s push this because some day it will be world famous.” But at the same time they were pleased to be a big part of the announcement. It provided great public relations. I know there are now more Swedish-speaking students in the computer sciences department, which had always been overshadowed by the Polytechnic University.

  Success envy is considered a Finnish cultural characteristic. And as Linux became better known in the world at large, I got a lot of questions about whether I had problems with people at the university being envious of me. The opposite turned out to be true; they were very supportive. Early on they started getting rid of X terminals and installing PCs with Linux instead.

  The announcement launched Linux into the above-the-radar zone in Finland, and it started generating publicity elsewhere, too. A lot of the early headlines came about because some journalist had stumbled over Linux and got excited about it. From a business standpoint, Version 1.0 was never very challenging to any of the big players. Linux was getting the market that Minix and Coherent had. But there was little attention outside that community, which was fine. It was far more attention than I had expected initially.

  Regardless, journalists, mostly from trade publications, started knocking at my door—literally. It didn’t make Tove happy to wake up on a Saturday morning to find a Japanese reporter bearing gifts—usually watches, as they probably heard somewhere that I have a thing for them—and wanting to do an interview. It made her even less happy when I would invite them in. (It was a pattern I would repeat for years, until we made our new house a Journalist-Free Zone. In my least considerate moments, I would even forget to tell Tove that I had invited a journalist to our house for an interview—and I would forget, too. The reporter would show up and Tove would have to entertain him or her until I made it home.) Then there were the fan websites that started popping up, such as the one based in France that primarily consists of a much-updated gallery of embarrassing photos of me. Like the one of me from a Spektrum meeting: I’m shirtless, drinking a beer, looking studly.

  Not.

  And it wasn’t only journalists or Linux hacker types who were showing an interest. Suddenly, people with big expense accounts wanted to talk to me about their technology. Unix had long been seen as an operating system with vast potential, mostly because of its power and multitasking capabilities. So corporations that were interested in Unix started keeping an eye on Linux. One of those was the networking company Novell, which had started a skunkworks project based on Linux. It was a Unix desktop they evolved called Looking Glass. It looked nice, but it was up against a wall: It lacked the standard of the time, which was the Common Desktop Environment.

  In August 1994, they said they would pay me to visit them in Orem, Utah, to talk to them about their desktop. Novell was offering me my big chance to see America, so I told them I would accept if they would pay for me to visit another U.S. city. Even as an unworldly Finn I kind of suspected that Orem—and even Salt Lake City—weren’t quite representative of the rest of the country. They suggested Washington, D.C., but I didn’t want to go there. I figured it would be just like any other capital. They suggested New York, but I thought it would be more interesting to go to California.

  Inside Novell’s headquarters, it was hard to determine just how serious people were about the project. (In the end, they ended up being not very serious at all; they eventually killed the project, and the nine people involved started up Caldera.) But I was getting my first taste of the United States, where I somehow figured I would live at some point in my life. Novell’s commitment to Linux notwithstanding, the United States seemed to be the center of the growing technology universe.

  My visit to the United States was a bit of a jolt. The first thing that struck me was how new everything was, compared to Europe. The Mormon
Church had had its 150th anniversary a few years before my visit, so they had cleaned up the main temple. It was shining white. Coming from Europe, where all the churches are old and have the patina of time, I could only think of one thing when I saw the white temple: Disneyland. It looked like a fairy-tale castle, not a church. And in Orem I made the mistake of checking out the hotel’s sauna. It was one of those porta-potty saunas, literally made of plastic, and it was barely hotter inside than outside. I came away from it thinking they can’t do saunas in the United States and feeling a little homesick.

  And I started learning the ropes. Just as visitors to Finland learn quickly not to start up conversations with random strangers in bars, I learned that in Utah—and, I later learned, the rest of America—you cannot rationally discuss the subject of abortion or rifles. There’s a 50-percent chance that you’ll get somebody who’s very emotional about those issues, and it’s easy to get into a big fight about something that shouldn’t be fought about. People don’t get hung up about those issues in Europe. The reason people get so defensive about their own positions in the United States is that they’ve heard the other position so much. There are probably more rifles per capita in Finland than in many other places, but they’re mostly used for hunting. It’s not a big thing.

  Another thing I quickly learned during my first days on U.S. soil: Root beer sucks.[2]

  After Utah, I flew to San Francisco and really, really liked it. I spent so much time walking around the city that I developed a major case of sunburn and had to remain indoors for an entire day.

  I remember walking across the Golden Gate Bridge, looking up at the Marin Headlands, excited about the possibility of hiking in those hills as soon as I crossed the bridge. But by the time I made it to the Marin side, I had lost all interest in walking anymore. I could never have predicted that six years later, almost to the day, I would be sitting at the crest of those windy Headlands, looking out on the Pacific Ocean, San Francisco Bay, the bridge, the fog, and San Francisco itself, explaining all this to David’s tape recorder.

  It would take only a year for me to return to the United States. I came back to speak at DECUS, (Digital’s User Group) in New Orleans. There were only forty people at the meeting, so it wasn’t a terrible ordeal. Best of all, that’s when I met Maddog, a.k.a. Jon Hall. He was a technical marketing person for Digital Unix and an old-time Unix user. He was responsible for sending me over for the talk. Maddog, who is known for his chest-length beard and his absurd sense of humor (not to mention his propensity to snore), heads Linux International, an organization that works to support Linux and Linux users. He’s also a godfather of my daughter Patricia.

  Another legacy of that New Orleans meeting: Maddog arranged for Digital to lend me an Alpha. That’s how Linux got ported to something other than a PC. Before that time, people had ported Linux to other architectures. There was a port to a 68K, the Motorola 68000 machines used by Atari and Amiga. But in those cases Linux didn’t work on both platforms at the same time. To make that version of scaling work, you rip out the pieces that don’t work on the new platform and you write new pieces. But the Alpha was the first real port of Linux. Basically the same sources worked on both the PC and Alpha. You add an abstraction layer so that the same code gets compiled in two different ways to work on two different architectures. It’s still the same code, but it ends up working on different architectures.

  When we released Version 1.2 in March 1995, the kernel had grown to include 250,000 lines of code, the new magazine Linux Journal claimed a 10,000-reader circulation, and Linux was capable of running on Intel, Digital, and Sun SPARC processors. That was a big step.

  II

  It’s 1995 and there is a host of growing commercial versions, and Linux companies are attracting a strong following. The university has elevated me from a teaching assistant to a research assistant, which means more money and less time teaching. I’m slowly—very slowly—completing the coursework for my master’s degree, which is about porting Linux to different architectures. Tove has introduced me to squash, and we have a weekly game; we’re fairly evenly matched.

  From this bliss, a problem emerges. It turns out that an opportunistic fellow from Boston has registered the trademark for the word Linux. Not only that, but he sent email to the Linux Journal and a few other Linux companies asking for 5 percent of their revenues as a “thank you” for the trademark.

  When I heard about this, I felt a twinge of déjà vu. The guy’s name sounded familiar. I checked my email archives and saw that maybe a year and a half earlier he had sent me an unsolicited email in which he first asked me if I believed in God, and then said he had a tremendous business opportunity for me. This was before spamming became a global obsession, the innocent years before anyone thought to pollute the Internet with offers of get-rich-quick schemes. No, I hadn’t bothered responding to the fellow’s email, but because it was so unusual for its time, I saved it.

  So we had a little crisis on our hands. We were hackers. Nobody thought about checking the trademark register.

  The guy wasn’t a professional trademark squatter. He apparently just did it this one time. Trademarks come in different categories, and he registered his in the computer category. You have to submit evidence with your trademark application, so he gave the trademark office a disk on which he claimed to have a program called Linux.

  There was some panic. Everybody in the Linux community knew we would contest the trademark. The problem was, we didn’t have an organization for putting up a good fight. There wasn’t even enough money to hire a lawyer. None of the companies felt comfortable about laying down the required amount, which was $15,000. (Today they go through that in a month’s worth of Mountain Dew.) But at the time, it was a considerable amount of money for a single company. So Linux Journal and some other companies decided to pump money into Linux International, which would fight the trademark. Linux International had been started in Australia by a person named Patrick D’Cruze who migrated to the United States in 1994 to help promote Linux worldwide. The year of the trademark dispute was the year when Maddog became its executive director. Everybody trusted him, and still does.

  I was in Finland, trying to beat Tove in squash or to beat Avuton in snooker, and I had no interest in getting involved in this. I just wanted the entire nightmare to go away. My preference at the time was to just get rid of the trademark, to get it declared invalid because of prior use in the industry. We had enough paperwork to show that Linux had, in fact, a history of prior use. The trouble was, our lawyer convinced us that it would be a wasted effort, that we should not even try to get Linux declared a public domain instead of a trademark. The only way for it to really be in the public domain, he explained, was for it to become generic. And Linux at the time wasn’t that generic. The trademark office probably wouldn’t even consider it to be generic today. We could lose the battle, he said. Or if we invalidated the old trademark, somebody could possibly come along and trademark it anew.

  The solution he suggested was to transfer the trademark to somebody else. My vote went to Linux International, but there was a lot of opposition to that. Linux International was young and unproven. People were worried about Linux International being taken over by commercial interests. (It hasn’t happened, I might add.) There was also strong concern about who would eventually take over for Maddog if he were to step down.

  So all eyes looked to me. The lawyer suggested that the legal arguments would be easier if the Linux trademark were to be transferred to me because I was the original user of the word. That’s the strategy we took. We reached an out-of-court settlement because that seemed like the easiest and cheapest thing to do. Like most out-of-court settlements, the details can’t be discussed. Not that I even know them. I was happily uninterested in it all.

  When I went back to check my original email from the guy, I realized that it had nothing to do with patents. It was obvious that he just wanted to talk to me. Maybe he tried to contact me to get me to pay him.
Or maybe, if I had shown myself to be a true believer and soul brother in his faith, he would have just given me the trademark. I don’t know.

  I accept the fact that some people are not morally all there. But what was more irritating at the time was the fact that the entire trademark system put this onus on me, who had done nothing wrong, to go out and fight the guy.

  As a result of the messy squirmish, I hold the Linux trademark. What that means is, when companies like VA Linux file for an IPO, their prospectus has to mention the fact that the company doesn’t even own the trademark for half of its name. (In that particular case, the company was involved in the legal process of getting approval from me to use the word Linux.) But I’ve gotten accustomed to that sort of thing.

  The trademark episode was just an unexpected growing pain for Linux. And a distraction. But no sooner was it settled then another surfaced: An engineer at Intel’s research lab in Portland, Oregon, said his company was using Linux in its exploration of new architectures. He asked me if I wanted to move there for a six-month internship.

  Tove and I had spoken in vague terms about possibly living in the United States. She knew how much I had enjoyed my few visits there, root beer notwithstanding. We agreed that the opportunities—not to mention the climate—were better in America. (By the way, I am totally convinced that the U.S. system of motivating employees is far more realistic, and produces better results, than the European model. In Finland if a worker is much better than his colleagues, you give him just a little more money and keep it very quiet. In America, you give him a lot more money—and it works.) The internship seemed like a great way of testing the waters, or, since it was in the Pacific Northwest, the rainwaters, and we agreed that I should pursue the opportunity. But I was ambivalent. I felt a bit uneasy about leaving school without having finished my master’s. Something inside me, possibly the memory of my professor grandfather, didn’t take to the notion of being a dropout. In the end, my feelings didn’t even matter. The engineer’s manager decided that it would be difficult for me to obtain the required six-month work permit from the U.S. Immigration and Naturalization Service for the internship.