Riding the Wild, Perilous Waters of Amazon.com

You think it's easy navigating a $20 billion company that has never made a dime?

I have seen the future, and it's a little exaggerated. It's the Frederick's of Hollywood padded panty No. 50404, which promises to add a dollop of bump to any behind. I stumbled upon this fetchingly inflated undergarment the same way -- if you believe what you hear -- I will find almost everything in the glorious, not-too-distant future. I found it on Amazon.com.

Despite the fact that the Web site has been hemorrhaging ever larger amounts of cash per quarter since it went up, Jeff Bezos, Amazon's brilliant, charming, hyper and misleadingly goofy mastermind, has managed to keep his three-and-half-year-old company the "it" stock of the late millennium by regularly upping the ante with ever more sweeping intimations about the scale and scope of e-commerce. First it was books, then music and video, then gifts. Now, with his just-opened "Shop the Web" service, which offers links to scores of other online retailers, the menu has been widened enough to include disingenuous lingerie.

But no matter what new foray or feature Bezos is trumpeting, he always keeps our filmy gaze locked on the future, reminding us in a way that somehow manages to be both winningly modest and titillatingly grandiose that this is the merest tip of the iceberg. "We probably know as much about e-commerce as anyone, but it's 2 percent of what we'll know five years from now," he says, and even with his latest bold move, the purchase of a controlling 46 percent stake in Drugstore.com, he points out that this is likely to be only the first of many investments in compatible Internet businesses. While his beleaguered physical rivals are mired in the present, where they have to attend to sticky details like making money, the specifics of Amazon.com's ultimate form remain forever elusive, a lovely shimmering at the edge of the horizon. In this way Amazon.com truly is a virtual company, existing only in the imagination.

And Bezos, a high-school valedictorian who won a Florida State science fair for his project on the effect of zero gravity on the housefly, hasn't pulled this off only with his formidable technical or quantitative skills, but also with his political genius for telling each of his divergent constituencies exactly what they want to hear. To his employees he holds out the rare chance to create from the ground up an entirely new medium that will reshape the world for the better. To his online visitors he offers a kinder and gentler form of commerce, in which a community of customers, armed with the right information, help one another make the purchase that is perfect for each one of them. And to Wall Street he presents a lean, mean virtual juggernaut that, unburdened by such anachronisms as sales clerks, operating hours and inventory, will run circles around any brick-and-mortar dinosaur, whether it be Barnes & Noble or those Arkansas boys from the Wal-Mart Stores.

So far a startling number of people are not just biting but also swallowing. Amazon employees have thrown themselves into their yokes like bovine missionaries. Their customers, 6.2 million and counting, have been so free and easy with their encrypted credit-card numbers that sales have grown an average of more than 50 percent a quarter and ended 1998 at a billion-dollar-a-year pace. At one point early this year, Wall Street gave the company a valuation north of $30 billion, which was more than Texaco's, and made the 35-year-old Bezos, who with his family owns just over half of the shares, one of the wealthiest fellows on either side of the Amazon.

But Bezos may have spun his company a bit too well, if such a thing is possible. The outrageous valuation has brought enormous scrutiny and a certain impatience about the lack of profit. From mid-January to mid-February, the stock lost half its value. And as Bezos struggles to move the company into the black with the whole world watching, it may become clear that the various versions of his vision are not just different but also conflicting. If the promised future ever arrives, we may see that the grand vision for Amazon.com isn't much more than an electronic mall, that as at any big retailer most of the jobs are lousy and pay badly and that Amazon.com's only real goal in regard to its customers, just like that of any other business, is to separate them from their money.

When Tom Clancy puts out a new book, we feel it in our bones in here," says Cedric Ross, a yellow-vested supervisor, as he walks me and my ever-present chaperone, Kay Dangaard, through Amazon.com's distribution center. For the 31-year-old Ross, one of the company's few African-American employees, who is often given the assignment of conducting reporter tours of the warehouse, Amazon.com is his second glimpse of the big time. A few years ago, he was a member of a hard-core Seattle rock band named IMIJ -- as in Hendrix's first name spelled backward -- which was signed to a label, but by the time the band completed its first album, the record company had lost interest in Northwest grunge and moved on in search of the next big thing.

Ross takes me from where the books, CD's and videos arrive from the distributors to where they are packed in individual parcels on a conveyor-belt bundler to where they get picked up by the huge postal and Federal Express trucks and sent off on their journeys around the world. The operation is impressively bustling and efficient, if not that much different from any other big warehouse, with its cement floors and pallets and forklifts.

But on this gray, drizzling morning of Dec. 16, as on any other for that matter, the crucial Amazon.com story isn't unfolding here, but 3,000 miles away, on Wall Street. There, a resounding endorsement of the company by Henry Blodget, then a relatively unknown analyst at C.I.B.C. Oppenheimer, has set the stock price racing upward. Within a couple of hours of the start of trading, the price of one share of Amazon.com stock, which a year and a half earlier when the company went public was worth $9, had opened at $243 and was now over $300. By the end of the day, a second swing in the wrong direction would eat up all but $16 of the huge gain. A little less than a month later, on Jan. 11, soon after a 3-for-1 split, the stock takes off on a similar roller coaster, opening at $158.875 and closing at $160.25, but in between hitting what remains its post-split high of $199.125.

To be exciting, a stock needs a good story, and the Internet and Amazon.com, both growing faster than Otto the overfed goldfish, are good stories, with the potential for only a few happy endings. But contributing to both the high valuation and the violent swings is the feverish activity of a growing group of inexperienced online investors called "day traders," whose strategy consists of little more than buying a stock as it's going up and selling it as quickly as possible the instant it starts to go down. Buttressing this folly is the assumption that by vigilantly staring at their screens, they will always be able to sell in time if things turn really ugly. No wonder Alan Greenspan, chairman of the Federal Reserve Board, has compared Internet investing to playing the lottery.

This is not to say that all the investor enthusiasm for Amazon.com and the Internet can be dismissed as an expression of hype and greed. Last year Amazon.com posted book sales of $610 million, an increase of 313 percent over 1997 sales of $147.8 million, making it in just three and a half years the country's third largest bookseller (virtual or physical), behind Barnes & Noble ($2.7 billion) and Borders (2.3 billion). In only its second quarter of operation, Amazon.com's music division became the largest online music seller, with $33.1 million in sales. At the same time, e-commerce, after a holiday season that dwarfed even the most optimistic estimates, increased by 230 percent to $5 billion. Although that is still less than 1 percent of total North American retail sales of $2.6 trillion, such rapid and seemingly limitless growth poses a serious threat to physical stores, which is why Barnesandnoble.com, bolstered by a $100 million matching pledge from its partner, Bertelsmann, is so committed to winning back some of that business. So far, however, online profits have been minimal at best because of the extreme cost of attracting customers.

Part of the great sell of the Internet has always been its promise to level the playing field, so that Bezos and five comrades working in a garage just outside of Seattle could create a site that looked in its own way as substantial and bona fide as a Barnes & Noble superstore. A month ago Lyle Bowlin, of Cedar Falls, Iowa, working by himself and in his spare time, was able to open a site offering almost the same vast selection as Amazon.com.

But it is far less democratizing than it appears. David E. Shaw, C.E.O. of D.E. Shaw, the hedge fund where Bezos worked before starting Amazon.com, explains: "While it is true that if all you want to do is to put up something for sale, the barriers of entry are extremely low on the Internet. If you actually want to sell a lot of that stuff, they're quite high and getting higher all the time."

You can offer the greatest selection and service on the Net, but without an enormous expenditure for marketing and advertising, no one will come because no one will know you are there. That's why Amazon.com spent approximately $50 million on advertising last year and doled out at least another $50 million to AOL and Yahoo! and MSN to funnel visitors in its direction. On average, online retailers spent $26 on marketing and advertising per sale in 1998, while their physical counterparts spent only $2.50. Until Bezos and other Internet retailers can find a way to attract and keep customers without such an enormous outlay, they are all going to have a hard time making any money.

Nevertheless, despite the ample accomplishments and serious, unresolved challenges, it is the peculiar fate of Amazon.com that both are completely overshadowed by the sensational valuation and volatility of its stock. For all its all-nighters and tattooed punks humping books in the distribution center and golden retrievers wandering the halls in the corporate office, Amazon.com is a $20 billion, 2,100-employee company built on the thin membrane of a bubble, and this brings a manic precariousness to the place that no amount of profitless growth can diminish.

Working at Amazon right now has to be a little like living in Pompeii 2,000 years ago. Much of the culture seems designed to deny the enormous distraction of the looming ticker and to somehow maintain the confidence that the fate of the company still lies wholly in its own hands. The sign in Dangaard's office bravely reads, "We are building an important and lasting company," and employees are beseeched not to concern themselves with the stock price, to keep their heads down and to continue focusing obsessively on satisfying the needs of their customers. But there is a slightly hysterical quality to these exhortations.

"At our last all-hands meeting," Ross says, "I told them that everybody should wake up every morning in terror of not satisfying our customers, because they're the ones we have a relationship with. They're the ones who are sending us money." As a policy the company does not comment on the stock price, and when I ask Ross about the perquisite of his stock options or if he had heard anything about the wild increase that morning, he declines to comment, and even acts almost offended.

This forced and incongruous indifference to money is another conspicuous part of the culture. Although almost every senior employee I question about the day's stock surge is very aware of what has happened, the average worker is expected to act as if it's of little consequence compared to the sweeping mission of Amazon.com, which, however nebulous, is always presented as far loftier than selling a fat pile of books, CD's, videos and blood pressure pills. Bezos never serves up e-commerce as naked capitalism. It's "helping people find and discover the things they want," "helping folks make better purchase decisions" and so on and so on. When it comes to seducing his employees, he offers not just a low paying job with a handful of stock options, but also a life's work.

The company's motto is "Work Hard, Have Fun and Make History," and cynicism is not an option. "We hold ourselves to a very high standard in terms of customer experience," Bezos says, "and we're growing faster than almost any other company has ever grown. So that takes a huge effort, and of course we wouldn't have it any other way, because we're trying to change the world and maybe improve it in a small way, and maybe even a little more than a small way, and that's not supposed to be easy. It's supposed to be hard."

The soft-spoken executive editor, Rick Ayre, who oversees the design and content of the Web pages, indicates that a predisposition to buy into this earnest culture is a job requirement. "I'm looking for people who want to care. I'm looking for people who want things to do that matter, a group of people who want a life with meaning, who want a career with meaning, and we offer that to people because it's a company that provides everyone with a voice and a responsibility for their actions."

Richard Howard, who was fired at the very end of a four-week trial for an entry-level "Tier 1" job in customer service, and later wrote a story for Seattle Weekly titled "How I Escaped From Amazon.com," refers to a tone of "quasi-religiosity" and an "unspoken taboo against any speech or expression (including gallows humor) that betrays your lack of commitment to the long-term success of the enterprise." He also reported that several times in his short tenure, he received unsolicited pep talks from longer hires who came over just to let him know that Amazon.com was the best company they'd ever worked for and that it had been a "life-changing experience." And at one point when I'm interviewing Ayre about the possible dilution of the brand as the company expands beyond books and music to lingerie and deodorant, he looks at me and says, "I can tell you're not a believer."

Maintaining this degree of zealotry is going to be difficult, because the overwhelming majority of the company's employees have tedious, low-paying jobs either in the distribution center or in customer service. Sometime this summer, all the various departments of the company except for the warehouse will move into the former Pacific Medical Center, but in the meantime they're scattered in four locations around town. Later on the same day as my visit to the distribution center, Dangaard, a willowy grandmother from New Zealand, takes me to a Fourth Avenue building, where an entire block-length floor is devoted to Customer Service.

After passing an armed guard, we enter a football-field-size room with eerily subdued lighting whose row upon row of tightly spaced cubicles seem to recede infinitely into the distance. But just as unsettling as the oversize scale and numbers, estimated by Howard at 750, is the hushed intensity as employees talk quietly into headsets, walking novice customers through the process, or tap out e-mails informing existing customers that a certain book they might be interested in has just arrived. No one is flirting or shooting the breeze, and according to Howard this is because they are racing to fulfill a weekly goal of calls and e-mails, all accessible to supervisors making sure that no one deviates unduly from the codified manual of responses called "the blurb index." Jane Slade, head of this division, says that "Seattle being the slacker capital of the world," the employees in her department, who are paid about the same as clerks in book or record stores, tend to be "ridiculously overqualified, with many having multiple degrees."

Despite C.E.O. admonishments not to dwell on the value of the stock price, it's hard to see what else would keep these Tier 1 and 2 employees returning to this vast electronic sweatshop, and the options are a notable part of the job description in the local want ads. When the stock was up over 70 points for a couple of hours that morning, the paper value of even the lowliest workers, with their 100 stock options, had gone up $7,000 that day, and when the stock topped off three and a half weeks later, just after the split, any worker who'd been there a year had a portfolio that was at least briefly worth $120,000. Even in the last few weeks, when the stock has settled at around 100, their portfolios are a not-too-shabby $60,000 each.

In fact, the more you look at an Internet phenomenon like Amazon.com, the more central the stock price seems to the whole enterprise. It's as if the stock price is the business model. Not only does it help bolster the cultlike commitment of the employees needed to provide the slavish service that is Amazon's competitive advantage, but it allows the company to painlessly make hundreds of millions of dollars in acquisitions, paying with stock or cash raised through bond offerings. Amazon.com spent $280 million to buy two companies, including a software provider and an Internet address service, in 1998. The astounding stock price allows the company to deepen its pockets to survive the price wars or whatever other competitive nastiness is surely coming. It is also an enormous branding and P.R. engine, and it's central to the long-term strategy. By the same token, it seems that if the stock were to crater, it would all unravel in a heartbeat.

Shawn Haynes is an earnest, fair-haired young executive who ran the mile for Yale and worked for I.B.M. before taking over the Associates Program, through which the owners of more than 200,000 other Web sites earn a small commission for steering visitors Amazon.com's way. He confides that when he was first recruited by the company, he had deep reservations. "The barriers to entry seemed too low, and I thought there was nothing to prevent competitors from copying our site," says the M.B.A. "But then I spent some time reading through the e-mails from customers, and I was just blown away by how strongly they felt about the company. I decided that there was something going on here that was truly unique."

The warm and gooey place Amazon.com occupies in the hearts and minds of so many of its customers is the great equity of the franchise, and that precious beachhead has been won by deeds and words. Some of that affection is just the natural and deserved spoils for getting there first. On that fateful morning in July 1995, when Bezos, still working out of a 250-square-foot office beside a Tile Barn, hit the enter key on his Sun Sparcstation computer and hung out his shingle in cyberspace, no other mainstream retailer had made the commitment to the Internet. "Barnes & Noble isn't doing this because they wanted to," Bezos points out. "They're doing it because of us. That's just a fact."

Amazon.com did more than show up early. Before going online, Bezos and five co-workers had put in a year of prepping and testing, and it showed. The site was bright and crisp, with a somewhat stark, uncluttered integrity, a virtual iteration of Hemingway's "clean well-lighted place," and although most early users were highly computer literate, the site was so easily navigated that none of that fluency was required. And taking a cue from Apple, they'd strived to humanize the interaction. "Although technology is providing you with the massive gains and services only technology could provide -- search engines, etc.," Ayre says, "you shouldn't be aware of the technology."

And from the minute the site went up, Bezos and his rapidly growing crew of programmers and designers and editors kept tweaking it and making it even simpler and quicker and more "frictionless" while gradually folding in industry firsts, like one-click shopping and one-click gift sending and wrapping. They made it so effortless that almost all it took (other than money) to order a book for yourself or send one as a gift was the inclination to do so. In fact it became necessary for them to post a flashing little reminder on screen informing customers that the transaction was over. "Thank you for your one-click order!" it read. ("Yes, it was that easy.")

At the same time, Amazon's early inordinate responsiveness to customers gave them a heady sense of ownership in this bold new experiment. When an elderly woman e-mailed to say that she loved the service but had to wait for her nephew to come over and break through the packaging, Bezos had it redesigned in a way that was no less protective of the goods, but required much less violence to open. At other times, like when they hunted down used and out-of-print books, their enterprising efforts on behalf of their customers were closer to what one might hope for from the concierge of a five-star hotel than from the clerk of a bookstore. "We may be the most customer-obsessed company to ever occupy planet Earth," says Bezos, who at one point was the company's only customer-service rep. Of course all this proactive groveling is expensive and part of the reason Amazon.com has been spending more than it makes, but according to David Shaw, this was all part of the master plan. "Making money on books was almost irrelevant, compared with establishing Amazon.com as the most trusted brand in this new space."

But at least as important to loyalists as the technological innovations, the reasonable prices and the follow-through was the alternative communal spirit. It begins with the inspired name that Bezos came upon after briefly considering the Dutch-based "aard" (too obscure) and the Disney-esque "cadabra" (too much like "cadaver"). Amazon suggests great size, of course, and a compendious database -- not that all of us knew the big river carries more water than any other in the world. But it's also overflowing with so much third-world, underdog, eco-conscious goodwill that every click on Amazon.com feels like a vote for the rain forest. And even as its market valuation soared past that of Barnes & Noble and Borders combined, the company has clung tenaciously to that image of a principled David pluckily battling its much more powerful and supposedly less principled rivals.

In fact, in many ways, they purloined the tone of those beloved independent bookstores that everyone accuses Barnes & Nobles and Borders of all but wiping off the face of the earth. They posted the same kinds of lists of in-house favorites like "What We're Reading" and "Amazon.com Recommends" that you first saw on the crowded, small round tables at those independents. As a company it consciously tried to give customers the impression, which had actually been the case with those small bookstores, that this was much more a labor of love than a pursuit of wealth. As Ayre tells me: "If you spend a lot of time on the site, I hope you get a sense of the quirky, independent, literate voice, and that behind it all you're interacting with people, and that it's people who care about these things, not people who are trying to sell you these things. My mantra has always been 'the perfect context for a purchase decision.' "

This co-op feel is further reinforced with "reader reviews," in which the site posts customers' reactions to books, both negative and positive, in the hope they might be able to steer you toward a great new title or warn you off a dud. Among the pragmatic reasons Bezos chose to make books his first online product are that there are far more of them in and out of print than any physical bookstore could ever line up on its shelves and that, unlike his second choice, music, the industry was not dominated by a cartel of half a dozen mega-distributors. But beyond that, books are so uncrass, or as Bezos says, "They creak in that nice kind of way," and they let him spin retail like EST. "We've all had books that have changed our lives," Bezos says, "and part of what we're doing at Amazon.com is trying to help you find and discover that next one." In a more cynical climate, "reader reviews" might have been seen as a transparently self-serving way of putting customers to work to boost sales, and customers might notice that they've never come across a listing with more negative than positive reviews, certainly not for a book with any chance of being a best seller. But for Bezos and his customers alike this was seen instead as a shining example of one of the Internet's greatest offerings, which Bezos calls "community" and defines as "neighbors helping neighbors." "You know you live in a community if you can knock on a neighbor's door and borrow a cup of sugar," Bezos adds. "If you can't do that then you don't really live in a community."

That's why the recent revelation by Doreen Carvajal, a Times reporter, that since June the company has been selling spots to book publishers in its posted lists of favorites, like "New and Notable" and "Destined for Greatness," along with special author profiles and extra e-mail support for up to $10,000 a title, was seen by his customer-neighbors as such a betrayal. It turns out they didn't live in a community after all, not even a virtual one. And of course the only publishers who could come up with the money to pay for this preferred treatment were the Goliaths like Random House and Simon & Schuster and Penguin Putnam.

After a flood of e-mail outrage, Bezos announced the next day that the company would be revising the program so that in the future all paid spots would be clearly flagged to let customers know in what instances the glowing praise and special placement had been bought. He offered to accept any returned book, regardless of its condition, that customers felt they had been duped into buying. But even as Bezos rushes to try to limit the karmic damage, Amazon's first P.R. catastrophe is drawing attention to the extremely tight corner the company has backed itself into, where it is likely to find itself caught again and again: between the best interests of customers and the bottom line.

Though the Internet's commercial possibilities have been explored for years by a small number of computer scientists, it wasn't until Marc Andreessen, the 22-year-old Netscape founder, helped invent a means of navigating this vast network, called the Mosaic browser, that it became the focus of this latest stage in the high-tech gold rush. Today, it's not the good coffee that has got Seattle jacked up but stock options, as one little start-up after another races to execute its own way of using the unique attributes of the Internet to siphon off cash from the world's credit-card holders.

"At first the quest was to write the Great American Novel," says Michael Schrage of the M.I.T. Media Lab, "and that was replaced by the Great American Screenplay. Now everyone wants to write the Great American Business Model." In the current entrepreneurial frenzy, business models are like high-concept movie treatments, and the big West Coast venture capital firms like Kleiner Perkins Caufield & Byers, which was an early investor in Amazon.com and whose most celebrated partner L. John Doerr sits on Amazon.com's board, are like studios.

But the problem with most Internet business models is that they are attempts to get around the fact that it is not yet clear how to make money from the business. For example, Buy.com sells books and other goods at cost with the hope of getting enough traffic to be able to earn a profit selling advertising. As for Amazon.com's business model, beyond the basic idea of selling books on an ambitious scale over the Internet, there is nothing novel or ingenious. As Bezos says, "It's not the ideas we're particularly proud of around here, it's the execution." For all the attractive aspects of Amazon's plan, including the advantages of a central distribution model and those of a scale business (in which costs do not rise proportionately with sales) and an excellent cash-flow situation because it gets paid by its customers long before the company has to pay suppliers, none of those aspects are unique or ownable by Amazon.com. Even the billion-dollar-a-year pace is apparently not enough to outrun the costs of advertising, technological upgrades and overhead connected to expansion. All Amazon really has is a special relationship with customers, and with the increasing pressure to finally justify its valuation and show a profit, that's the most valuable thing it has to sell.

Amazon has already come up with two lucrative ways of exploiting this hard-won goodwill, and no doubt is busy hatching others. One is selling to publishers and others preferred access to customers. Another way is by earning a commission for directing that army of trusting customers to other retailers, which could be the key to the whole enterprise. This is the idea of "Shop the Web," although as the range of products expands willy-nilly essentially to anything on the planet that can be sold for a profit, it's hard to see how the company is remaining faithful to its mission stated on the site to "use the Internet to offer products that educate, inform and inspire." Then again, despite presenting the face of an underdog, Amazon.com has always had the agenda of a would-be monopolist. "The idea," says Shaw, "was always that someone would be allowed to make a profit as an intermediary. The key question is, Who will get to be that middleman?"

But whatever minor crisis of conscience Bezos and his army of M.B.A.'s may have suffered before barreling down those first two paths is nothing compared with the problem that is coming regarding the vast stockpile of intimate details Amazon.com has collected on its millions of customers and users. According to Ayre, Amazon breaks down its traffic into "visitors," "users" and "customers," with a "visitor" being anyone who checks out the site, a "user" anyone who gives some information about themselves. Although the company applies a great deal of energy and skill into "converting" visitors up the line, even if it gets only information out of you, it's potentially as good as money.

This record of customers' choices of reading material, which, unless the user declines, Bezos reserves the right to sell, is of such value to marketers because nothing is more nakedly and efficiently revealing. If you're an overweight, power-lusting, cross-dressing father of twins who can't putt, it will all pop right up on the screen; armed with those nuggets, even the most ham-fisted marketer will have more than enough to make you jump and shout and rat out your own mother.

About a half-dozen years ago, Mark Bezos, having just graduated from Texas Christian University at Fort Worth, came into town looking for a job in advertising, and his half-brother, older by six years and then still at D. E. Shaw, put him up on his couch. Soon after his arrival, the two went out and bought a new Nintendo game called Ultima. "It was one of those role-playing games," says Mark, "that takes forever to play because before you can advance to the next level you have to build up an arsenal of weapons and cash." For the next week, Mark would spend his days performing the "menial task of killing minor monsters" but hold off on "the good battles and real evil monsters and the decisions on how to spend the money." When Jeff walked in the door at 10, the first thing he would ask was, "How far have you gotten?" Then they'd hunker down for the next six hours and go to war.

"I know it sounds incredibly dorky," Mark says now about their nightly graveyard shift, "but actually it was a special time for both of us." The two became so obsessed with their quest that Mark extended his stay, not until he had a job, but until Ultima had been routed.

Since throwing the switch in July 1995, Bezos has maneuvered and promoted his company with such cunning and cheek that it's as if he's still on that couch navigating the evil forces of the galaxy with a greasy slice in one hand and a clammy joystick in the other. He showed his perfect command of entrepreneurial story structure even earlier with accounts of how after stumbling on the startling fact that the Internet was growing 2,300 percent a year, he impulsively chucked his big job on Wall Street and headed out West to make his mark on the tech frontier, his young bride, MacKenzie (maiden name Tuttle, Hotchkiss, Princeton), driving their old Chevy Trail Blazer while Bezos furiously typed out the business plan on his laptop and scared up "angels" on his cellular as if he were riffing the whole thing on the fly.

It turns out, however, that he flew to Texas and drove from there. David Shaw takes pains to point out that one of the main things he was paying Bezos to do in his last couple of years with the company was to investigate the commercial possibilities of the Internet, and that many of the important details for his new venture were worked out on his dime under his supervision. Shaw says that when Bezos asked if he could run with this idea himself, Shaw obliged, but made it clear to Bezos that he might at some point go into competition with him.

Nevertheless, the road trip plucked just the right chord, as did his self-depiction as a nerd, which turns out to be just slightly fudged. Born to teen-age parents, Bezos was reared in large part by his maternal grandparents, and from the ages of 4 to 16 spent every summer at their hot, dusty ranch in Cotulla, Tex., learning to brand, vaccinate and castrate cattle and picking up mechanical savvy like rebuilding a John Deere grader. But he knows that social ineptitude and tunnel vision is what we're looking for in our Internet entrepreneurs. It inspires confidence to know that rather than piddling away their young adulthood, they were poring over equations and stealing a march on the future.

And although Bezos emphasizes his obsession with the customer and plays down his concern with rivals, the way he toyed with Barnes & Noble was so instructive that any established brand getting bushwhacked by an Internet upstart is now referred to as "getting Amazoned." He started with a bold print ad campaign that hailed Amazon.com as "Earth's Biggest Bookstore" and used splashy graphs to support the claim that Amazon.com offered a much larger selection than the country's biggest book retailer. Then he got mobile billboards to cruise around Barnes & Noble superstores posing the question, "Can't find that book you wanted?" This was a bit misleading, considering that Amazon.com at that point carried no books at all and was simply ordering them for customers, just as Barnes & Noble or any other bookstore would if they didn't happen to have it in stock.

More important, he held up Barnes & Noble as the embodiment of the doomed past with all its supposed assets reinterpreted as liabilities, and held up Amazon.com as the shining alternative. He did it so well that Amazon.com has almost come to stand for the future itself and any new and improved way of doing things. That Bezos took a particular delight in these gambits can be surmised by the fact that he often arranged to interview prospective hires in the generous physical comfort of a Barnes & Noble's superstore.

The constant taunting egged Barnes & Noble into quickly putting up its own site, and when Barnesandnoble.com failed at first to cut into Amazon.com's big lead, it reinforced the message Wall Street needed to hear: it's not just hype; cyberspace is a whole new animal; none of the old retail skills apply. Then Barnesandnoble.com began to work out the kinks and gained back some market share, running ads with headlines like "Who's Fooling Who?" and its own graphs showing they offer almost twice as many titles. By then, Amazon.com had opened its music store, replaced "Earth's Biggest Bookstore" with "Books, Music and More" and left Barnes & Noble wrapping its fingers around the neck of a phantom. Bezos says, "A year from now I don't think Barnes & Noble will even think of us as a rival."

Even before this, Amazon.com brazenly poached a Wal-Mart vice president, Rick Dalzell, and a dozen other top Wal-mart wonks. In addition to prompting a suit (still unsettled), it signaled that Amazon.com was not content with books and music and other high-end niceties. It wanted a piece of everything. Before Christmas it added a gift section to its menu, where customers could buy toys and games and Barbies and Sonys, and started testing "Shop the Web." And at the very end of last month, after having raised $1.25 billion in a junk-bond offering in late January, Amazon.com bought its share of Drugstore.com, a company that aims to sell everything from breath mints to Viagra, and went after the $150 billion U.S. pharmacy market, almost six times larger than that of books.

"There are a lot of differences between books and drugstores," Bezos says, "but there are a lot of similarities too. Customers want selection, convenience, price and information." He says Amazon.com will share management tips with the company and encourage Amazon's visitors to try the new pharmacy, which will itself be competing with at least half a dozen Internet start-ups, including Planet Rx and Soma.com, which started selling medicine online Jan. 15. Bezos declined to say if he has investments in other e-commerce companies, but says, "We may do more investments like this."

It's this kind of almost routine doubling down every few months of his already massive bet and nervy expansion of the scale of his ambitions that keeps the public and Wall Street on the edge of their seats, makes Bezos a moving target to his rivals and buys him more time to turn a profit. And as he encourages us to imbue his company with the same limitless potential that as a rule we only vouchsafe for ourselves, we have to keep revising upward and outward our best guess of just what Bezos is really after. Does he want to be the reincarnation of Sam Walton? Does he want to be Michael Eisner? Or does he want to be Richard Branson, who made Virgin stand for so much and so little that it can be slapped on a $5,500 seat to London and a can of cola?

Bezos, who once told his younger brother that he "refuses to evaluate either-or equations," most likely wants to be all of these and then some. Indeed, his every gesture, posture and feature, from his standard-issue wardrobe of khakis and brown rubber-soled shoes to his homemade desk fashioned out of a door, seems calculated to belie the incendiary fact that Bezos is one of the world's most ascendant capitalists, with apparently boundless ambitions.

Lately, this enormous reaching has provoked as much skepticism as enthusiasm on Wall Street. Around the time the stock tumbled, in mid-January, Barton Biggs, a global strategist for Morgan Stanley Dean Witter, speaking about Internet stocks in general, told Bloomberg News: "I promise you that, like all bubbles, this bubble will come to a very, very bad end. The only trouble is that none of us know when." And Robert J. Barbera, chief economist at Hoenig & Company, has said that while investors may have dreamed of Amazon killing Wal-Mart, that company has other ideas. "Do you think that Wal-Mart will roll over and die? Don't you think there will be a Wal-Mart.com?" In fact, Wal-Mart has put up a big site with a guarantee in case of credit-card fraud and a flat $3 charge for all shipping costs.

There is a growing suspicion that in the end it could still be the brick-and-mortar dinosaurs who reap the benefits from the Internet trailblazers. Jose Rasco, also at Hoenig & Company, draws comparisons to ill-fated innovators like Atari, the video-game pioneer later bested by Nintendo, and Commodore, which put out one of the earliest versions of the personal computer, and even to Apple, which has had to stage a huge comeback recently just to secure its status as a niche player. He also sees it as a bad sign that Amazon.com, which at one point was championing the advantages of maintaining as little inventory as possible, has recently bought its own distributor in Reno, revealing its need for the same kind of economies of scale as its old-fashioned competitors. "Now they're not exactly a virtual company or a real one," says Rasco, "but in a limbo somewhere between."

A more hopeful possible scenario is that Amazon.com will get bought. James McQuivey of Forrester Research says, "A company like Disney could afford to spend $20 billion for Amazon," and with that possibility he has pushed the company back out onto the horizon, where he can imagine a perfect partnership of information, entertainment and customers.

But by constantly blowing out the dimensions and focus of the company, Bezos has a way of making even the most recent projections by his skeptics and boosters seem out of date and beside the point. As Bezos, video-game style, keeps taking on ever larger and more ornery opponents while reminding himself that "stress is an orthogonal dimension," he seems to be reveling in the game like a man with all kinds of cards up his sleeve. As Graciela Chichilnisky, a Columbia University economics professor who employed Bezos in her arcane financial services start-up right after he graduated from Princeton, puts it: "In the knowledge sector the key issues are rate of innovation and depth of penetration. He will anticipate the changes. I bet on Jeff Bezos' brain."

In fact, Bezos' claim that by the end of the year, Barnes & Noble would no longer consider him a rival almost seems true. According to Nicole Vanderbilt, an Internet analyst for Jupiter Communications, Amazon.com's real rivals now aren't the lumbering store-based retailers but the other heavily trafficked Internet portals and providers, from AOL to Yahoo!, and now, after its acquisition of USA Networks, Lycos. "As impressive as Amazon's 6.2 million customers are," she says, "AOL has the credit-card numbers of 16.2 million customers."

As he widens his scope, Bezos continues to put more and more of his emphasis on building the Amazon brand. "Brands to a certain degree are like quick-drying cement," he has said. "When they're young, they're stretchable and pliant, but over time they become more and more associated with a particular thing and harder to stretch."

Bezos has often spoken about how the Internet, where prices and services can be compared with a couple clicks, "fundamentally shifts the balance of power from the merchant to the customer." Yet the race by all the big Internet players to get huge fast is based on the opposite assumption: that customers will align themselves early on with one of a couple of names and stay put out of sloth and habit. Bob Pittman, formerly of MTV and now of AOL, has expressed this cynical notion. "Coca-Cola doesn't win the taste test and Microsoft is not the best operating system, but in America brands win." But while Coke and Microsoft started with products and then wrapped them in brands, Amazon.com, which sells other people's products, is almost entirely a creation of branding.

Every morning, even before her husband leaves their modest loft and heads to his little office in his sensible Honda, MacKenzie Bezos is at her laptop. While an undergraduate at Princeton, she served as Toni Morrison's research assistant on "Jazz," and for most of the last five years has been researching and writing her own novel, whose protagonist works for the state of California in Sacramento. According to Bezos, she has just finished a first draft.

While her husband has found a way of selling millions of books a year, largely by paring down the process of commerce to the otherwise imperceptible exertion of one fingertip, MacKenzie sits at her desk rewriting sentence after sentence on a single book that may never see print. "It's the hardest job in the world," says Bezos. "Nobody is pushing you. Nobody is even counting on you to do it. And it's lonely. You have to be incredibly self-disciplined." And while Bezos has had to persuade a couple thousand of employees, tens of thousands of shareholders and millions of visitors, users and customers to come along with him for the ride, MacKenzie has just had to persuade herself. But regardless of whose undertaking ends up being more successful or more worthwhile, both are storytellers, and both are spinning epics out of thin air.

Peter de Jonge, co-author of the novel "Miracle on the 17th Green," writes frequently about sports, business and culture.