Thinking up business possibilities, in fact, was Bezos' job at D.E. Shaw, an unusual firm that prides itself on hiring some of the smartest people in the world and then figuring out what kind of work they might profitably do. David Shaw, a former professor of computer science at Columbia University, had been wooed to Wall Street by Morgan Stanley, where he specialized in the arcane field of quantitative analysis--using computers to spot trends in the market. He formed his own company in 1988, initially to carry on that kind of work, but with so much brainpower around the office, it seemed a shame to waste it all on Wall Street. It made sense to pursue other businesses too. During much of his four years at Shaw, Bezos "was sort of an entrepreneurial odd-jobs kind of a person," Shaw recalled recently.

Bezos had graduated from Princeton University, majoring in electrical engineering and computer science. The field was unplanned: he had chosen Princeton for its legendary physics department. Shortly after arriving, however, he discovered that he wasn't the smartest guy in the world after all. He felt outclassed by the physics jocks and gravitated to comp-sci.

His first job out of school was at Fitel, a start-up that was building a network to handle international financial trades. He spent about two years there, worked about the same amount of time at Bankers Trust, then got an interview at Shaw.

Actually, it was one of Shaw's partners who interviewed Bezos first and urged the boss to meet him, saying, "He's going to make someone a lot of money someday." Shaw agreed, understanding that Bezos was unusual not only for his balanced intellect--he could handle complex logic as well as articulate his thinking--but also for the overall package: smart, creative, personable, precisely the kind of person they wanted. Over time, Bezos became a specialist in researching business opportunities in insurance, software and then the booming Internet.

But how to take advantage of that online explosion? The Net had been, until 1994, a largely commerce-free zone. It was created by the Defense Department to keep its network of computers communicating in case of nuclear attack. The system then evolved into a network over which university and government researchers could exchange messages and data across most computer platforms.

The government decided to get out of the Internet business and allow private companies to step in and develop it. Bezos recalls, "I'm sitting there thinking we can be a complete first mover in e-commerce." He researched mail-order companies, figuring that things that sold well by mail would do well online. He made a list of the Top 20 mail-order products and looked for where he could create "the most value for customers." Value, in his equation, would be something customers craved: selection, say, or convenience or low prices. "Unless you could create something with a huge value proposition for the customer, it would be easier for them to do it the old way," he reasoned. And the best way to do that was "to do something that simply cannot be done any other way."

And that's what ultimately led to books. There weren't any huge mail-order book catalogs simply because a good catalog would contain thousands, if not millions of listings. The catalog would need to be as big as a phone book--too expensive to mail. That, of course, made it perfect for the Internet, which is the ideal container for limitless information.

Bezos needed to learn the book business fast. Fate was his handmaiden: the American Booksellers Association's annual convention was set for the very next day in Los Angeles. He flew out and spent the weekend roaming the aisles and taking a crash course in the business. Everything he learned encouraged him. The two big wholesalers for books were Ingram and Baker & Taylor. "So I went to their booths and told them I was thinking of doing this." Books, it turns out, are among the most highly databased items on the planet. The wholesalers even had CD-ROMs listing them. It seemed to Bezos as if all the stuff "had been meticulously organized so it could be put online."

Bezos realized he desperately wanted to start his own online bookstore. First he talked it over with MacKenzie. She too had graduated from Princeton, but six years after him; they met at Shaw, where she worked as a researcher. An English-literature major at the university, she had been novelist Toni Morrison's assistant and now had begun a novel of her own. MacKenzie was all for the adventure.

Next Bezos went to Shaw, who said he was sorry to lose such a talented executive but fully understood Bezos' desire to strike out on his own. He cautioned him to make sure, however, that this was what he truly wanted to do. Bezos decided to spend the next two days recalculating the risks.

In his typically analytic way, Bezos cast his decision in what he calls the "regret-minimization framework." He imagined that he was 80 years old and looking back at his life. And suddenly everything became clear to him. When he was 80, he'd never regret having missed out on a six-figure Christmas bonus; he wouldn't even regret having tried to build an online business and failed. "In fact, I'd have been proud of that, proud of myself for having taken that risk and tried to participate in that thing called the Internet that I thought was going to be such a big deal. It was like the wild, wild West, a new frontier. And I knew that if I didn't try this, I would regret it. And that would be inescapable."

Bezos figured that the average Net start-up had a 1 in 10 chance of success; he gave himself a 30% chance. "That's actually a very liberating expectation, expecting to fail," he says. That's exactly what he told his first investors--family and friends: "I think there's a 70% chance you're going to lose all your money, so don't invest unless you can afford to lose it."

"When he called and said he wanted to sell books on the Internet, we said, 'The Internet? What's that?'" remembers Mike Bezos, who initially questioned his son's sanity when he heard him say he was quitting his cushy job to start a company that would probably fail. But this was Jeff, after all, and his parents trusted him and believed in him every moment of his life. In the end, "we talked about it for two minutes," says Jackie Bezos. They ponied up $300,000, a huge chunk of the money they had saved for retirement. "We didn't invest in Amazon," says his mother, "we invested in Jeff." The ROJ--return on Jeff--was substantial. Today, as 6% owners of the company, they're billionaires.