(FORTUNE Magazine) – They're just a couple of slightly geeky middle-aged guys sitting on a patio overlooking Seattle's Lake Washington on a balmy Sunday afternoon. Jet skis and speedboats drone like cicadas as the old pals reminisce, chuckling over escapades from their adolescence. The skinny one, who at 39 is starting to look a lot like Notre Dame football coach Lou Holtz, tells what it was like to incorporate their first company while still students at a prestigious Seattle private school. The burly bearded one, who is the son of librarians and has a demeanor to match, marvels at how much pizza they ate. They may have been teenage computer prodigies, but flamboyant and rebellious they were not.

They're so nonchalant as they sip their Cokes that it's hard to believe, even though you know exactly who they are--that this pair had the will and drive to create more wealth than any business partners in the history of American capitalism, or that they are, today, the undisputed masters of the digital universe.

But then, this isn't the way Microsoft founders Bill Gates and Paul Allen usually spend their Sunday afternoons. More likely, Gates would be hunched in front of a screen, scrutinizing the latest computer code from one of Microsoft's dozens of development teams, laboring to finish another chapter of his book, The Road Ahead (due out this fall), or making notes for a public appearance somewhere in the world to promote personal computing, Windows-style. (Besides, Gates' idea of relaxation is to race his wife, fellow Microsoftie Melinda French, at identical jigsaw puzzles.)

Allen, 42, who left the hurly-burly of Microsoft in 1983 after an illness, is a little more laid back. But not much: He juggles investments in more than two dozen high-tech companies and runs a pro basketball team and four charitable foundations. If he had a couple of free hours, he might crank up his Stratocaster to jam with his rock band,The Threads, in his recording studio in downtown Seattle.

It was for a special reason that the buddy billionaires--at last count, Gates' 25% stake in Microsoft was worth $13.4 billion and Allen's 9.6% was worth $5.3 billion--invited FORTUNE to sit in on a three-hour chat. This October, Microsoft will celebrate the 20th anniversary of its founding, and the world's most notorious college dropouts wanted to mark the occasion with a wide-ranging bull session.

The meeting at Paul Allen's rambling estate was a one-of-a-kind encounter between fellows so fixated on the future that they'd really rather not look in the rear-view mirror, even for a minute. Once they got rolling, they described both the inside history of a company that has become an American entrepreneurial icon, and the symbiotic relationship of two brilliant friends with similar professional interests and dramatically different personalities, whose diverging paths are already the stuff of business legend.

GETTING TO KNOW YOU GATES: Our friendship started after the mothers' club paid to put a computer terminal in the school in 1968. The notion was that, of course, the teachers would figure out this computer thing and then teach it to the students. But that didn't happen. It was the other way around. There was a group of students who kind of went nuts ALLEN: The teletype room was full of rolled-up paper-tape programs and manuals and everything else. Between classes, or whenever any of us hard-core computer types had a spare period, we would congregate there.

GATES: It was easy to spend a lot of money on computer time. One of the smarter teachers accidentally created an infinite loop and spent something like $60 in one shot. He was so shocked that he didn't want to go near the thing after that. We ran through the mothers' club money pretty fast.

ALLEN: When that happened, the computer service would bill the school, and then the school would charge us or our parents. It wasn't always easy to tell your parents how much you had racked up, so we paid it ourselves when we could. GATES: At the end of the year, Paul massively overdrew his checking account. On graduation day he was with his mother, and the school officials came up and said before he could get his diploma they would have to pay $200 for all this excess computer time.

We were always scrounging for free computer time. One year a student's mother arranged for us to go downtown to a new, commercial computer center. We didn't have to pay for the time as long as we could find bugs in their system and report them. They brought us in like monkeys, but it was a godsend.

ALLEN: At the end of every school day, a bunch of us would take our little leather satchel briefcases and ride the bus downtown to the computer center. Bill and I were the guys that stayed the latest, and afterward we'd go eat pizza at this hippie place across the street. GATES: Some of the professional programmers who worked at the center hated us and would shut us out. But there were guys who actually liked us. One was this super-nice guy who would give us the manuals. Another guy was kind of quiet, but if we bugged him, he would answer questions. Paul and I always wanted to understand machine code [the fundamental level of programming on a machine]. We were confused about the difference between various operations. ALLEN: That's because they'd parcel out the manuals and only give us one at a time. It was like a game with them. They'd give us the manual for the assembler and not the manual for the operating system or the manual for the instruction set. Then they'd wait to see how long it took us to figure out that we only had part of the picture. And then we'd come back and ask for the next manual. GATES: Paul and I got bonded together trying to figure out that machine. ALLEN: The computer center was having financial problems, though. It's where we first learned about bankruptcy. Bill and I were working on our programs one day when somebody came to repossess the furniture. They took the chairs out from underneath us. GATES: Yeah, and we were still typing away, trying to save our programs on these little tapes before all the lights were turned off. BUSINESS BASICS GATES: The event that started everything for us business-wise was when Paul found an article in 1971 in an electronics magazine, on page 73 or something, about Intel's 4004 chip, which was the world's first microprocessor. Paul comes up and says, "Whoa," and explains that this microprocessor thing's only going to get better and better. Sure enough, a year later Intel came out with the 8008 and it was lots better. ALLEN: That's when it hit us how Moore's Law really worked--that each generation of microprocessor chip was basically twice as fast as the previous one and that they got cheaper too. So Bill and I went out and bought our own 8008 for 360 bucks. The chip was all wrapped up in aluminum foil, and we were almost afraid to touch it.

GATES: We thought we could use the 8008 as the heart of a special computer to do traffic-volume-count analysis. We were going to make the machines and sell them to traffic departments. So we set up our first company, which we called Traf-O-Data. ALLEN: Even though Traf-O-Data wasn't a roaring success, it was seminal in preparing us to make Microsoft's first product a couple of years later. We taught ourselves to simulate how microprocessors work using DEC computers, so we could develop software even before our machine was built.

We were always interested in business. We'd go over to Bill's house and read FORTUNE and other business magazines his parents got. Bill would say, "Gee, what does it mean to be No. 1 in the FORTUNE 500? Can you imagine what a company that size does? What do all those statistics really mean?"

We talked about being entrepreneurs. Obviously it was on a smaller scale because we were kids. Microprocessors were instantly attractive to us because you could build something for a fraction of the cost of conventional electronics. That's essentially what we did with the Traf-O-Data computer--only it was too narrow and challenging an area to try to build a service business in.

GATES: I remember, from the very beginning, we wondered, "What would it mean for DEC once microcomputers were powerful and cheap enough? What would it mean for IBM?" To us it seemed that they were screwed. We thought maybe they'd even be screwed tomorrow. We were saying, "God, how come these guys aren't stunned? How come they're not just amazed and scared?" By the time we got to Albuquerque to start Microsoft in 1975, the notion was fairly clear to us that computers were going to be a big, big personal tool. ALLEN: I remember having pizza at Shakey's in Vancouver, Washington, in 1973, and talking about the fact that eventually everyone is going to be online and have access to newspapers and stuff and wouldn't people be willing to pay for information on a computer terminal. GATES: Yeah, we were also fascinated by dedicated word processors from Wang, because we believed that general-purpose machines could do that just as well. That's why, when it came time to design the keyboard for the IBM PC, we put the funny Wang character set into the machine--you know, smiley faces and boxes and triangles and stuff. We were thinking we'd like to do a clone of Wang word-processing software someday. Most personal computers back then didn't even have upper- and lower-case characters. BOOT UP AND DROP OUT

In summer 1973 the friends landed their first real jobs, helping TRW in Vancouver, Washington, use minicomputers to manage and distribute power from hydroelectric dams. Allen, who was getting bored at Washington State University, wanted to drop everything and start a company, but Gates' parents pushed Bill, who had just graduated from high school, to enroll at Harvard.

A year later Gates persuaded Allen to move to the Boston area to take a programming job with Honeywell. That winter Allen ran across a magazine article that would change their lives and, ultimately, just about everybody else's: a cover story in Popular Electronics describing the MITS Altair 8800, "World's First Minicomputer Kit to Rival Commercial Models."

Within weeks Gates and Allen convinced each other this was their big chance. The Altair was practically useless to hobbyists without an easy programming language. In designing the Traf-O-Data machine, the two had written a version of BASIC--a popular and compact language that, with a few tweaks, might work on the Altair. In weeks they cobbled together a version for the MITS machine, finishing it up on the plane, and by April had persuaded MITS to sell it. MITS offered Allen a job and provided the two with office space in its ramshackle headquarters in an Albuquerque strip mall. Soon after, Microsoft was born. GATES: When we signed that first contract with MITS, we referred to ourselves as "Paul Allen and Bill Gates doing business as Micro-Soft.'' I don't remember why we spelled it with a hyphen and a capital "S." We put a credit line in the source code of our first product that said, "Micro-Soft BASIC: Bill Gates wrote a lot of stuff; Paul Allen wrote some other stuff.'' We never officially incorporated until 1981. ALLEN: We had talked about a lot of different names back in Boston, and at some point I said, "Well, the totally obvious name would be Microsoft." GATES: We also had mentioned names like Outcorporated Inc. and Unlimited Ltd., but we were, you know, joking around. We talked a lot about whether we should call it Allen & Gates, but decided that was not a good idea. ALLEN: Yeah. Because companies like DEC and IBM weren't named after personalities, they would have a longevity and identity way beyond the founders... GATES: ...and it seemed like a law firm or like a consulting company to call it Allen & Gates. So we picked Microsoft even before we had a company to name. ALLEN: The building where we first worked in Albuquerque was pretty low rent.

GATES: Next door was a vacuum-cleaner place, then a massage parlor. To get to our offices, you had to walk past the vacuum-cleaner guy. We stayed in this motel down the road called the Sand and Sage. We're talking real sage, not some hypothetical thing. Every morning all the cars in the parking lot had all this sagebrush and tumbleweed that blew underneath them. ALLEN: Our management style was a little loose in the beginning. We both took part in every decision, and it's hard to remember who did what. If there was a difference between our roles, I was probably the one always pushing a little bit in terms of new technology and new products, and Bill was more interested in doing negotiations and contracts and business deals. GATES: We learned a lot of things as we went: Okay, we have to hire people; so what do we do? Okay, we're going to rent space; how do we do that? Okay, we're going to do contracts with people now; I'd get advice from my dad [a prominent Seattle attorney].

Paul and I would talk through every decision, for six or eight hours sometimes. We didn't have many major disagreements, but there was one tiny source of tension: I would always be calling Paul in the morning to tell him it was time to come work on this stuff. He slept even later than me. ALLEN: We used to work really late into the night. You used to sleep under your desk. GATES: Yeah, life for us was working and maybe going to a movie and then working some more. Sometimes customers would come in, and we were so tired we'd fall asleep in front of them. Or at an internal meeting I'd lie down on the floor, because I like to do that to brainstorm. And then I'd just fall asleep. OVERCOMMITMENTS 'R' US GATES: Our basic business strategy was to charge a price so low that microcomputer makers couldn't do the software internally for that cheap. One of the bigger early contracts was Texas Instruments, where we bid $99,000 to provide programming languages for a home computer they were planning. We picked that price because we were too shy to make a bid in the six figures. Afterward we realized they would have paid a lot more, and we thought, "I guess this is what the big shots do: They bid big numbers." ALLEN: We would almost always over estimate our competitors' ability to compete. GATES: Or we'd assume that they were going to execute competently. ALLEN: There were so many times when I'd find an article about a competing product and we'd get depressed for three days. Then we'd find out it wasn't as good as ours ... GATES: ...or didn't even exist. We were in such a flaky business. Our biggest problem was that we talked about so many different ideas, but we never really added up all the people we'd need. When we started selling to Japanese companies, we were so overpromised it was ridiculous. Ricoh licensed every language we had and paid us $180,000 up-front, which was incredible. And then they came back a few months later and said, "Okay, we have another budget this year, so we'd like to buy some more." Well, we'd already sold them everything we had. So we asked, "What can we develop for you?" We agreed to do our first database, a word processor, and a couple of other things. Then we had to figure out how we were going to deliver them, and had to go buy some of them from somebody else. ALLEN: We were so late that Ricoh finally flew a guy over whose whole job was to sit in our office day and night until we delivered. He couldn't help out at all; he was just there to sit and explain to us that his career was going to be over if we didn't get this work done soon. I don't know if that was really true or not... GATES: It could get scary. In our very first contract with MITS, we set them up to sell our BASIC to their customers, rather than us selling to computer buyers directly. We thought it was a good deal because they agreed to make "best efforts" to sell it. But later they decided not to sell to anybody at all because there were so many illegal, free copies of our BASIC floating around, so why try to charge people for it? That really made us mad because we thought it encouraged piracy. We eventually went into arbitration to determine if they were in compliance with the contract. In the meantime we were totally out of money...

ALLEN: ...because MITS was withholding payments from us while the arbitration was going on.

GATES: They were trying to starve us to death. We couldn't even pay our lawyer. They tried to get us to settle, and we almost did, it was that bad. The arbitrator took nine months to issue his damn opinion. But when it was all over, the arbitrator ripped them apart for what they had done. ALLEN: That case really, really scared us. If we had lost, we would have had to start over. Bill would call up his dad for advice. We were on pins and needles the whole time.

GATES: But, you know, through it all, we never borrowed money. I always felt like if we had to, we could have. But we never did. THE DEAL OF THE DECADE By the end of 1978, a wave of impressive personal computers, including Apple Computer's landmark Apple II, had left MITS in the dust. There was no longer any reason for Microsoft-now a dozen employees strong--to hang around Albuquerque. On January 1, 1979, Gates and Allen transplanted the business to the lush and soggy Seattle suburb of Bellevue, where it took root. Within a year the head count had swelled to more than 35 people, and Gates and Allen hired their first professional manager--Gates' old Harvard buddy, a Procter & Gamble marketing man named Steve Ballmer. (They lured Ballmer with a stake in Microsoft. He now owns 5%, worth $2.7 billion.)

Ballmer's arrival marked a turning point. "We knew we had to systematize things," Gates recalls. "Once we had more than 30 people, it was impossible for Paul and me to review or be involved with every line of code we produced." To this day, based on that early insight, Microsoft often limits work teams to 35 people.

But 1980 will be remembered for another series of events. It was the year IBM came to call, looking for programming languages for its secret PC project; also the year Paul Allen negotiated the purchase of an obscure operating system called Q-DOS from a nearby company, Seattle Computer. In a transaction that could well be called the deal of the decade, Gates and Allen would turn around and license Q-DOS to IBM, thus opening the way for Microsoft's domination of the PC software industry. GATES: When IBM came to us in 1980, we thought they were just talking about buying our BASIC for their new PC project. The two guys who came said, "We're just the planning guys, and most of the things we plan don't get done, so don't get too excited." But then we had this really cool discussion about where the technology was going, and how personal computers were about to make a great leap forward. Then they said they'd also like to buy our FORTRAN and COBOL languages, and maybe even more from us. ALLEN: It was like they were bringing up a menu and checking off "all of the above." GATES: So we had this meeting between Paul, Steve Ballmer, me, and Kay Nishi, our partner in Japan. It seemed just like Ricoh all over again. We had told IBM, "Okay, you can have everything we make," even though we hadn't even made it yet. This was one of those marathon meetings.

The question was: Should we also commit to doing an operating system for IBM? It was unusual because I was the one being a tiny bit conservative. Nishi was actually the one saying, "Got to do it, got to do it." ALLEN: Nishi was egging us on because he knew I had been dealing with Seattle Computer and that I thought we could probably make their Q-DOS work for IBM. GATES: We also knew Digital Research [maker of CP/M, the then-dominant operating system for personal computers] wasn't taking IBM seriously enough. So we decided, Why not? It got kind of tense, because there was about a 48-hour period after Ballmer and I had officially offered to license Q-DOS to IBM that we didn't really own it. Paul hadn't yet closed the deal with Seattle Computer, and I was really giving him a really hard time. ALLEN: We were afraid they were going to find out the reason we wanted to buy it was because IBM was our primary customer. If they found that out, the price for Q-DOS would go way up. [Microsoft succeeded in buying the software for only $50,000.] GATES: See, the contract with IBM called for us to do all this work on the design of the machine and all this software. We didn't get paid that much-the total was something like $186,000--but we knew there were going to be clones of the IBM PC. We structured that original contract to allow them. It was a key point in our negotiations. ALLEN: We already had seen the clone phenomenon in the MITS Altair days. Other companies made machines that succeeded because they were very similar to the Altair. For us it had been easy to modify our software so it worked on those machines too. GATES: It seemed like a really great opportunity. IBM was talking about having Sears and ComputerLand sell their machines. They'd been talking with Intel about getting a really good price on the microprocessor. IBM was so hard core about it, we knew their PC might even... ALLEN: ...leapfrog the Apple II, which was the coolest computer at the time. GATES: We worked day and night with this IBM machine in a back room. There were a lot of ups and downs. The one tiff Paul and I had was when he wanted to go see a space shuttle launch and I didn't, because we were late. Still, these guys went to the launch and I was just... ALLEN: It was the first one, Bill. And we flew back the same day. We weren't gone even 36 hours. GATES: Anyway, by the end, Paul and I decided every stupid little thing about the PC: the keyboard layout, how the cassette port worked, how the sound port worked, how the graphics port worked. Between us and a few IBM engineers, we did all that stuff. We felt very intimately involved.

The weirdest thing of all, though, was when we asked to come to the big official launch of the PC in New York, IBM denied us. About four days later we got this form letter that IBM probably sent to every vendor, even the guy who had the capacitors in the machine. It said something like "Dear vendor, thank you for your help, blah, blah, blah." They eventually apologized to us for that. ALLEN'S ORDEAL It's hard to believe now, but the acceptance of Microsoft's DOS as the operating system for the IBM PC wasn't a foregone conclusion when the machine was introduced in 1981. IBM also offered a version of the CP/M operating system. Produced by Digital Research of Monterey, California, CP/M had the advantage of being established on other brands of personal computer. Or was that an advantage?

Initially, since IBM was Microsoft's only operating system customer, Gates and Allen went all-out to make DOS the operating system of choice, urging other software companies to write applications for DOS first and otherwise promoting and cajoling. Within a year the battle was over, DOS ruled the U.S. market for the IBM PC, and the first of what would become a tsunami of PC clones were trickling into the marketplace.

Gates and Allen turned their attention to the nascent PC market in Europe, hoping to repeat the feat. Then the unthinkable happened. ALLEN: We were on a trip promoting some Microsoft products in Europe in 1982, and I had this little bump on my neck. I'd had little bumps before that weren't anything. But I kept feeling stranger and stranger. Then one day in Paris, I just felt really bad and decided I had to go back to the States.

The doctor here in Seattle put me in the hospital that night. At first they told me that it looked like I had lymphoma--an often incurable cancer--and they were really glum. But after the biopsy the doctors came in smiling and said, "You've got Hodgkin's disease and you're going to be fine," which was a little hard to believe, since they also said I was going to need about 22 months of radiation therapy.

During my treatment I tried to continue working at Microsoft. Not full-time, but I'd drop in at meetings and so forth. It was hard because cancer therapy takes a lot out of you. But it was more than that. To be 30 years old and have that kind of shock--to face your mortality--really makes you feel like you should do some of the things that you haven't done. With Hodgkin's disease or any cancer like it, there's basically a two-year window: If you can pass that period without a relapse, then it's probably not going to come back. GATES: It was terrible when Paul got sick. Nobody talked about work or anything; it was just a matter of wondering, Was Paul going to be okay? We had always worked so closely together--I mean, you know, planned stuff together, strategized together. That's why it's always hard when people ask who did what. So it was a real change for me and a huge disappointment that Paul wasn't there. It was really sad to go by his office, because all the memos and magazines would be stacked in there. ALLEN: I took that time to step away from Microsoft and be closer to my family, and do some traveling and other things I'd always wanted to do. After that two-year period, well, I just didn't want to go back to work. I went to Bill and said, "I want to just do something different." I know Bill wished I hadn't decided that. GATES: It was great that Paul got better, and we wanted him back more than anything. But there was just no part-time way to come back to Microsoft. If you were going to be there, you were really going to work hard. We all knew that. It's still that way. ALLEN: After that experience I was just in a different place. WHAT'S NEXT? "Get a life." That's the derisive expression often hurled at computer jocks and other obsessive workers. It's also a pretty good summation of what Paul Allen has tried to do since recovering from Hodgkin's disease--even as his old pal continues to eat, drink, and sleep Microsoft.

Having a $6 billion nest egg helps. Such un imaginable wealth may not buy happiness, but it has allowed Allen to amass collections of authentic Roman statuary and oil paintings by French impressionist masters; to build a massive estate on the shore of Seattle's Lake Washington complete with three houses, a full-size gym, an indoor pool and waterslide, and more garages than most people have kitchen cabinets.

He owns the NBA's Portland Trail Blazers and is building them a $262 million sports and entertainment complex called the Rose Quarter. He owns a jet, so he can take friends to a game and have them back in Seattle by bedtime.

He's a good guy too. His charities have dispensed $6 million so far this year to support libraries, hospitals, AIDS programs, prostate cancer research, and cultural programs including the Oregon Shakespeare Festival. An avid guitarist, he is the main benefactor of the Experience Music Project, a planned museum dedicated to the late Seattle rock great Jimi Hendrix and other Pacific Northwest musicians.

Then there's business. Allen was an early investor in America Online. He owns 80% of TicketMaster, the national computerized ticketing service; 6% of an animal-supply retailer called Petsmart; a piece of a children's educational-products retail chain called Zany Brainy. With Gates he has invested in Darwin Molecular, a genetics research startup.

Not surprisingly, though, the bulk of Allen's investments involve software. He has positions in nearly 20 high-tech companies engaged in multimedia CD-ROMs and software tools, satellite broadcasting, online financial and sports-data services, TV programming, wireless paging and telephony, and electronic commerce. This year he doled out $500 million for an 18% stake in Hollywood's newest studio, DreamWorks.

And, of course, Allen remains a director of Microsoft. Although he claims not to pay much attention to day-to-day affairs, you can tell his buddy Bill still listens when Allen talks about what lies over the technological horizon and the strategies behind his investments. ALLEN: For years now, I've been interested in the information superhighway or whatever you want to call it. The approach I've chosen is to start companies or make strategic investments in companies I think are positioned to take advantage of that huge opportunity. I try to add value as an investor by building synergy between those companies.

Some have no immediate commercial potential, but others are solid businesses in conventional markets: TicketMaster is a consumer service business that sold 55 million tickets last year, and there are ways to broaden the service and deliver it on the Internet. Bill and I are investors in biotechnology too, and I have other investments for fun like Petsmart and the sports team stuff. But most of them are software related. The incredible thing about the software business is, you can create something compelling and exciting out of thin air. GATES: Paul has gone his own way, but we still see this stuff the same way, and we debate about it and share ideas quite a bit. The thing that's totally amazing is that looking at the next 20 years, we both believe there's actually more opportunity to have a broad impact than in the past 20 years. ALLEN: Our industry is still driven by incredible improvements in semiconductor technology that now are removing constraints on communications. You have the very real prospect of continuous high-bandwidth, two-way digital communications to people's homes and places of business coming in five years. God, what can you do with that? A lot of the constraints on our imaginations have gone away. It's like, wow, there's this fertile field; what kind of seeds am I going to plant? What are they going to grow into? GATES: The Internet is the seed corn of a lot of things that are going to happen, and there are so many parallels to when Paul and I were involved in the beginnings of the PC. We said back then, "Don't DEC and IBM know they're in deep trouble?" Here we are, staring at the same kind of situation.

In the computer industry, there's never been a company that's led the way in two successive eras. So really, what Microsoft as a company, or Paul and I as individuals, are trying to do is to defy history and actually take our leadership from the PC era into this new communications era. The odds are against us, and that's what makes it so much fun and so challenging. ALLEN: The Internet is exciting to me as an investor because the barrier to entry is very low. Anybody can put a product out, so success depends more than anything on the quality of the product. GATES: For Microsoft, the Internet is an incredible new platform. Just like cheap microprocessors, cheap communications drives the demand for more great software.

How do we keep up with it all? You have to remember it's all relative--you don't really, in an absolute sense, have to keep up with all the new technologies; you just have to keep up with them better than everybody else.

Some of Paul's companies are classic start ups that are getting in very early and are focused on specific content areas. Eventually you won't think of "the Internet business." You'll think of it more like news, weather, sports, but even that taxonomy isn't clear. ALLEN: When you take sports or music and start adding video and audio to a computer database service on the Internet, and you deliver it with enough bandwidth, and make it interactive, what's the difference between that and an interactive television channel?

GATES: None. ALLEN: All these things are being served up on the Internet and will be, in effect, a click away.

HAVE YOU EVER BEEN EXPERIENCED? So much for reminiscing; so much for looking forward. What was it, after all? To what quality, trait, or circumstance do these young billionaires--already historical figures--attribute their success? Was it greed, brains, obsession, paranoia, luck? Are Gates and Allen humbled by the enormousness of their success? Do they ever pause to gloat about the revolution they helped incite? GATES: There's one thing I have to ask if you remember, Paul. Back in Vancouver, you once said that if you made a million dollars someday, that maybe that would be enough money for anybody. Do you remember talking about that at the apartment? ALLEN: Sounds vaguely familiar. GATES: Really, though, all along we were never motivated in business just because there was something we wanted to buy. Paul and I paid ourselves $36,000 a year until we incorporated, and then we took this huge boost to $60,000. It was enough because there was nothing material we really wanted. ALLEN: That's not quite true. When I showed up in Albuquerque, I had to take an advance from MITS because I had no money to pay for a hotel room. The plane ticket took all my money. And when you got there, we used to go window shopping for fast cars. GATES: You're right. I did buy a used Porsche 911 in Albuquerque; that was my one big desire. Paul bought a big-screen TV, which I think was a seminal event, because watching all those basketball games on it may have gotten to you ... ALLEN: We never really thought about counting our pennies in terms of "this is all the money we could squirrel away for ourselves." Success to us was just plowing it right back in and building the business more. At any time there were lots of software markets we weren't doing that well in--products like word processors and spreadsheets that we've only lately come to lead. There was always some part of the business where we felt like we were the underdog. GATES: The outside perception and inside perception of Microsoft are so different. The view of Microsoft inside Microsoft is always kind of an underdog thing. Today the worry would be: Oh, God, what can we do in the world of the Internet?

In the early years that underdog, almost paranoid attitude was a matter of survival. ALLEN: Yeah. We were always worried that IBM would put a thousand guys on word processors or spreadsheets or BASIC or something like that. GATES: Even though if you look back and see that our sales and profits grew by basically 50% a year for all these years, what I really remember is worrying all the time. If you ask about a specific year, I'd tell you, oh, that was the awful year we had to get Multiplan [a financial spreadsheet] out and establish it, or that was the terrible year we brought out the Microsoft mouse and it didn't sell so we had a warehouse full of them, or it was the miserable year we hired a guy to be president who didn't work out. ALLEN: We were always worried. Sure, we could see the upside, but we could always see the downside. We're both pretty good at being devil's advocates. If Bill would start getting too optimistic, I would say, "Well, wait a moment, aren't you worried about A, B, and C?" I kind of wonder if that doesn't go back to our experiences at that computer center in Seattle where we had the chairs repossessed from under us. Or because Traf-O-Data never succeeded like we thought it would. GATES: We've seen so many failures. This is an industry scattered with failure. But despite that, I still believe in our vision-a computer on every desk and in every home, all running Microsoft software.

And I still believe in the importance of software. We said 20 years ago that eventually people will pay more for their software than they do for their hardware. That isn't yet true. ALLEN: It will be true, though. GATES: I promise it will. We just have to raise our prices ellipse No, I'm just kidding.

But if you had asked me at any point how big Microsoft could be, Paul and I once thought we could write all the software in the world with 100 people. If you had told us that someday we would have more than 5,000 people writing software, we would have just shaken our heads.

I remember coming to work in the morning when our sales were $20 million, and I'd tell myself: "This means we have to sell $100,000 of software before I go home."And I'd think: "How can I possibly do that?" So if in the morning I closed a big deal, I'd think, "Wow, today I did it." I don't do that so much now because Microsoft is more of an institution. The numbers are so big now, it's crazy. Just to keep the thing running at breakeven we have to sell $15 million a day. ALLEN: Remember that day back in 1980 or so, when we figured we'd shipped a million copies of BASIC? We really felt a sense of accomplishment. And we were marveling that, wow, a million people are using our code to do God-knows-what-number of interesting things. That was such a gratifying thing to realize, that you have been able to affect other people's lives in a positive way. GATES: Then we did something kind of funny. Paul and I were in such a different world that we had always stayed away from the normal mainframe software industry. They had a big trade organization--it doesn't exist anymore--and an annual trade show where they gave out a special award if you sold 10,000 copies of a program. Just for the heck of it, we filled out one of these forms and said we shipped a million copies. They didn't even have a category for selling that many. But they sent us this fancy plaque anyway, saying yes, indeed, Microsoft had shipped 10,000 copies of BASIC. It was our first big award. ALLEN: Generally, we don't look in the rear-view mirror much. GATES: Right. Because it's a waste of time, basically. The irony of sitting down like this on our 20th anniversary is that we hardly ever do this. On those rare moments when we do look back, we kind of go, "Whoa." We've been climbing a steep mountain here, and you know, there's still lots there ahead of us. It really is pretty amazing. ALLEN: Geez, it was only 20 years ago that we were sitting there, eating pizza, talking about what we thought a PC would be like, and what we could do. GATES: Still, we're only about halfway to achieving our original dream of a computer in every home and on every desk. I believe that 20 years from now, before we're too old, the industry will have fulfilled that promise. ALLEN: I would hope so before then. GATES: Yeah, but who knows whether we'll be able to maintain our leadership role in it until then.

Reporter Associate Henry Goldblatt