Just for Fun : The Story of an Accidental Revolutionary

by Linus Benedict Torvalds

  So we stayed in Helsinki. By the time Finland’s legendary revelers were toasting the arrival of New Year’s Day, 1996, I was inching toward the master’s degree finish line. I only needed to complete one small course to finish the credit requirements. And I also had to write my thesis. Ironically, it would be the first time I got any academic credit for my work on Linux, which had consumed most of my life throughout my college years.

  The year 1996 brought with it a wake-up call. In egalitarian Finland, you get a state-mandated seniority raise after you’ve been at a job for three years. When I saw my first pay stub reflecting the new salary, I felt a jolt: I had been working at the university long enough to be given seniority. Would I spend my entire working career there? Was I destined to become my grandfather? Remember my description of him from earlier in this book: bald, overweight, and not smelling like anything. I started checking myself in the mirror with some regularity. My hairline was creeping back a couple of millimeters. Extra kilos seemed to have made their way to my once-skinny torso. I was twenty-six and for the first time in my life I was feeling old. I had been at the university for going on seven years; I knew I could graduate fairly quickly if I got my act together.

  My ten-year-old daughter Kaley thinks it’s the apex of super-stardom to have someone buy you a penguin. We are sitting around a campfire on a clear Sierra night when Linus explains that a Linux user group in Bristol, England, bought him a penguin. Kaley cannot imagine that he hasn’t bothered to visit the creature. Then he elaborates: Actually, they didn’t buy him a penguin, but instead sponsored one in his name. And the sponsorship lasts for a year, he thinks.

  The Torvalds family is trying to understand the concept behind ’smores. Somebody looks up from his roasted marshmallow and makes the mistake of asking how the penguin came about as the planetary symbol of Linux in the first place.

  “The penguin was my idea,” says Tove. “Linus was trying to find a symbol for Linux because people were asking, ‘Shouldn’t there be a symbol?’ He was thinking of things he’d seen. The Linux companies had their own symbols. One of the companies had a pink triangle as its symbol. But I knew that was the international symbol for gays, so I told him that had already been taken. He said he would like to have something nice, something sympathetic.

  I thought about penguins. Linus had been bitten by a fairy penguin at a zoo in Australia. He likes to pet things. He’s always poking at stuff like rattlesnakes. Those penguins at the zoo were about one foot high, and he just reached into the cage to pet one of them. He kind of played with his fingers as if they were fish. The penguin came at him, bit him, noticed he was not a fish. He got bitten by a penguin but he liked it anyway. I got the feeling he was sold on penguins after that. He wanted to see penguins wherever it was possible.

  “So when he started looking for a symbol I said, ‘Why don’t you have a penguin because you fell in love with those penguins?’ He said, ‘Okay, I’II think about it.’”

  Here’s where Linus, sitting maybe three bodies away from Tove, shakes his head.

  “No, it was not her idea,” he says. “She’s wrong.”

  This was a departure. Linus and Tove don’t make a habit of disagreeing. Tove is astounding in her ability to deftly handle the responsibilities of the girls and the household—and a famous husband—fending off journalists with her karate skills. Linus seems downright cheerful about chipping in by occasionally folding laundry or doing his morning chore of making the cappuccino. Even during the stress of a ten-hour car trip with the on-again-off-again needs of a pair of young kids, Linus and Tove handle it all smoothly: Think of the marital equivalent of a well-crafted Scandinavian sofa bed.

  We found the kink.

  The story, according to Linus, is that while Tove may in fact have vaguely mentioned penguins at some early stage, it was in a conversation with two high-ranking Linux types that the icy creatures were first seriously considered as the operating system’s official mascot.

  Tove has her take on this version. “He thought it wasn’t a good idea after all, because it was my idea. He went on thinking about a possible symbol. Then we were in Boston with Maddog and Henry Hall. They started talking about the symbol. I said to them, ‘What about a penguin. Do you think it’s nice?’ They said yes. I think that made Linus think it might be a good idea after all.”

  “Henry Hall said he knew an artist who could draw it for him, but that never happened. The next thing I knew, Linus had asked on the Internet if there were people who wanted to send in pictures of penguins.” He chose a version by Larry Ewing, a graphic artist who works at the Institute for Scientific Computing at Texas A&M University.

  But this wasn’t to be just any penguin. Above all, Linus wanted one that looked happy, as if it had just polished off a pitcher of beer and then had the best sex of its life. Even beyond that stipulation, he wanted one that was distinct. Hence, while all other penguins have black beaks and feet, those features are orange on the Linux mascot, making it look almost like a penguin whose father was a duck. As if Daffy Duck got a little kwazy on a cruise to Antarctica and had a wild one-night-stand with some native fowl.

  III

  News of my decision to work for Transmeta Corporation was greeted with the same reaction in the Linux community as was the news that Tove and I had figured out how to conceive a child and were expecting one at the end of 1996.

  When word leaked out in the spring that Tove was pregnant, the more vocal among Linux user newsgroup participants wanted to know how I planned on balancing the demands of Linux maintenance with those of a family. A few months later, when it became known that I would (finally) be leaving the University of Helsinki to work for the secretive Transmeta in Silicon Valley, the big, worldly debate centered on whether I could possibly keep true to my open source philosophy in a dreaded commercial environment, as opposed to a neutral academic institution. Transmeta was partly funded by Microsoft cofounder Paul Allen, folks declared in protest; some claimed it must be an elaborate scam for taking control of Linux.

  I’m not saying those aren’t valid concerns for loyal members of the Linux community, but it’s just… gimme a break! The fact is, neither the birth of Patricia in December 1996 (and Daniela sixteen months later and Celeste forty-eight months later) nor my job at Transmeta, which began in February 1997, has caused the downfall of Linux. I felt all along that if anything were to negatively affect my work with Linux, I would have taken the obvious necessary step of turning it all over to somebody I trust.

  But I’m getting ahead of myself.

  In the spring of 1996, just as the weather was breaking, I finished the last of my required coursework for my master’s. It was about this time that I heard from Peter Anvin, the Linux community member who three years earlier had organized the online collection that helped me pay off my first PC. Like everybody else who prowled the Linux newsgroups, he knew I would soon be graduating. He had been working at Transmeta for about a year, and approached his boss to explain that he knew this guy in Finland who might be good for the company. He came to meet with me briefly while he was visiting his mother in Sweden. He talked up Transmeta, which was pretty hard to do since it was in stealth mode and he couldn’t tell me much of anything. The rumor among programmers was only that it was involved in developing “programmable chips.” At the very least it was great to meet Peter in person.

  A week after he returned to California, he sent me an email asking when I could come. This was hugely different from the experience I had had with Intel a year earlier, when an engineer wanted to hire me for an internship but it never happened because of the paperwork issues.

  It would be fun just to get a trip to California, I thought.

  This was the first job interview of my life. I didn’t have a CV. I didn’t know what Transmeta was doing; it was in a strange land.

  I was more worried about the implications of moving to the United States than about getting the job itself, so I didn’t even think of my meetings as being inter
views. What mattered more, it seemed, was to learn what these guys were up to. It was a fairly strange interview situation.

  After the first day, I went back to my hotel across the street from Transmeta’s office-park headquarters. In my jet-lagged state I thought it was all interesting, but I also thought that the folks at Transmeta were crazy. At that point the company didn’t have any silicon at all. No hardware. Everything was done with a simulator, and the demonstration of the simulation booting into Windows 3.11 and running solitaire did little to convince me that anything was going to happen. After that first day I wondered if it wasn’t all a waste. I distinctly remember thinking: Maybe this isn’t going to turn out—either as a technological innovation for Transmeta or a job for me.

  I literally slept on it. Actually, it wasn’t much of a sleep. I lay in bed thinking about Transmeta’s plans. Then I started fantasizing about having a backyard palm tree. Then I started ruminating over what I had seen on the simulator. It was a memorable, fitful night, but nothing like the frostbitten anxiety of Ede.

  By morning, I was somewhat excited. By the end of the second day, I was very excited. That’s when the stress began.

  Before accepting Transmeta’s offer, I talked it over with various people. When word got out that I was considering the job, I received a number of other offers. In Finland I got an offer from Tele, which was using Linux in some capacity. Through Maddog I got an offer at Digital. (No offense, but Boston in winter isn’t a whole lot better than Helsinki in winter. Okay, maybe it is.) I talked to some of the Red Hat people. They offered me a job and said they would pay better than Transmeta was paying, even though they had no idea what the proposed salary was because I hadn’t even discussed money with the company. The Red Hat crowd said they would even top Transmeta’s stock options, whatever they would be. But I wasn’t interested in working for any particular Linux company—even one that was fortunate enough to be located in the middle of North Carolina.

  In the end, I got five job offers without ever formally looking for a job. Transmeta’s was the most exciting, by far.

  I said yes. It felt weird. The next thing I did was tell the university I would be leaving. That’s when the stress really began. For me, it was a giant step that meant there was no turning back. We were having a new child, moving to a new country, and I was leaving the safe nest of the University of Helsinki—but first I had to write my thesis. In retrospect, I guess getting all those changes over at once was a good idea. But it was madness.

  There was no formal announcement (why should there be?).

  Just word circulating on the Internet, and the aforementioned debate about whether I would be able to remain true to Linux and free software in the evil corporate environment, and between the changing of diapers. Back then, people had this view of Linux as something that was mainly developed by university students, not settled-down people. I guess it was understandable that they would be nervous.

  I wrote my thesis over a long weekend and turned it in minutes before taking Tove to the hospital to deliver Patricia, who was born forty hours later. That was December 5th, 1996. Being a father seemed like the most natural thing in the world.

  The next several weeks we were busy with Patricia and constantly worrying about obtaining the approvals for our U.S. visa paperwork, which was taking forever. We figured it would help if we were married, so sometime in January—I always have to ask Tove the date—we went to a government office to be officially wed. We had three guests: Tove’s parents and my mother. (My dad was in Moscow). It was a strange time. At some point we shipped most of our belongings to the United States without knowing when we would be able to fly out. To say good-bye to all of our friends, we hosted a housecooling party, the reverse of a housewarming. Twenty people crammed into the small, recently emptied one-bedroom apartment. In good Finnish tradition, everybody got drunk.

  Our visas finally arrived and on February 1 7, 1997, we boarded a morning flight to San Francisco. I remember the temperature in Helsinki: 0 degrees Fahrenheit. I remember Tove’s family at the airport, crying when we said good-bye—they’re very close. I don’t remember if my family was there or not. They must have been. Or maybe not.

  We landed in the United States and made our way through customs carrying a baby and two cats. Peter Anvin was there to greet us as we rented a car for the drive down to Santa Clara, to the apartment complex we had chosen during an apartment-hunting trip we had taken a few months earlier. It all felt surreal, particularly the 70-degree difference in climate from Finland.

  Our belongings wouldn’t arrive for another two months. We spent the first night sleeping on an air mattress we had brought with us. The next day we went out to buy a real bed. Until our furniture made it to California, Patricia had to sleep in her carriage. It was something that really annoyed Tove, although David points out that it is sort of cyclical, referring to the first three months of my life that were spent in a laundry basket. We didn’t do much cooking (we still don’t) and didn’t know where to go for dinner. We ate most of our meals either at the food court of a local shopping mall or at a fast-food place. I remember telling Tove we had to find some new places to eat.

  With the move and getting accustomed to the new job at Transmeta, I didn’t have a lot of time to devote to Linux during those first couple of months. The new job occupied much of my time and my after-work hours were spent with Tove and Patricia, trying to get to know the new area. It was a fairly busy time. We had absolutely no money. I had this great salary, but everything went toward getting furniture. Buying our cars was a hassle because we had no way of establishing a credit history. We even endured hassles proving we were capable of paying for telephone service.

  My computer was on a ship that was inching its way around the Horn of Africa. It was the first period of time when I was quiet on the Web, and my absence worried a lot of people. It was like, Okay, now he’s working for a commercial company….

  Many people asked outright: Does this mean Linux will die off as a free system? I explained that under my agreement with Transmeta I could continue doing Linux. And that I wasn’t going to go away. (I couldn’t think of a way to say that I was just catching my breath.)

  Life in Transmetaland.

  One of the problems with explaining to people how the move to the States and into the commercial world wasn’t going to change me was the fact that Transmeta was just about the most secretive company around. There was only one rule concerning what you could talk about, and that rule was very simple: “Say nothing.” Which just made Linux people sometimes wonder what kind of strange cult I had joined, and whether I was ever coming back. I couldn’t even tell my mother what I was up to—not that she would have been interested.

  What I was doing at Transmeta wasn’t all that strange. The first thing I actually ended up doing was fixing some of the Linux problems that Transmeta had. The company was using a lot of multiprocessor Linux machines. I had never personally worked on the Linux SMP (Symmetric MultiProcessing) issues, and it turned out that many things didn’t really work the way they were supposed to. I took this as a personal affront, and had to fix it, of course.

  But my real work was actually being part of the Transmeta softball team.

  Oh, I mean software team. We didn’t play all that much softball. None of the Silicon Valley leagues would let us join unless we agreed to tell them what we were up to.

  I don’t know how familiar people are with Transmeta. As I’m typing this, we’re actually in our silent period before the (please, God, buy our stock) IPO, and we’re no longer secret, although we’re back in our stealth mode due to SEC rules about initial public offerings. Let’s hope that by the time this book is out, everybody and his dog has heard about Transmeta and bought (subliminal message: STOCK) one or more of our CPUs. Because that’s what Transmeta does—CPU’s. Hardware.

  But Transmeta does more than just hardware. Which is just as well because, quite frankly, I wouldn’t know a transistor from a diode even if one k
icked me in the head. What Transmeta does is simple hardware that relies on clever software to make a simple CPU look like much more than it really is—like a standard Intel-compatible X86, in fact. And with the hardware being made smaller and simpler, the CPU ends up having fewer transistors, which in turn makes it use less power, which as everybody realizes will become increasingly more important in a mobile world. This clever software is why Transmeta has a rather large software team, and why I was there.

  This all fit me quite well. A non-Linux company that did something that was technically quite interesting (understatement of the year—I still don’t know of another company that has ever seriously even tried to do what Transmeta does). And it was in an area that I knew intimately: low-level programming of the quite esoteric 80x86 family of CPU’s. As you undoubtedly recall, it was the act of getting to know that CPU in the first personal computer I owned that had started the whole Linux project in the first place.

  The fact that Transmeta wasn’t a Linux company was also important to me. Don’t get me wrong: I loved fixing Linux problems at Transmeta, and I’ve been involved in several internal projects about Linux. (In fact, these days it’s probably impossible to find a serious technology company without such projects.) But at Transmeta, Linux was secondary—which was just what I wanted. I could continue to do Linux, but I didn’t feel I would have to make the technical compromises that would favor the company’s goals over Linux itself. I could continue to think of Linux as a hobby in which only technology mattered, and nothing else held sway over my decisions.