The lunch plates were cleared long ago, and the waitress gazes vacantly out over an otherwise empty dining room. But George Gilder, his legs propped on a nearby chair, seems rooted in place, not quite ready to leave. We're lingering at a restaurant down the street from his office in Great Barrington, a hamlet set along a rural highway that winds through the southern tip of the Berkshires in western Massachusetts. Here, one of the tech world's more famous — and controversial — prophets is contemplating how he could have been so right over the past half—dozen years and yet seen everything turn out so terribly wrong. A look of anguish clouds his face.
"I knew that it was going to crash, I really did," Gilder says, looking out a window on to Main Street. Since 1996, he has published the Gilder Technology Report, a monthly newsletter that in its heyday was arguably the most influential tout sheet on Wall Street. He glances my way and notices my arched eyebrows. I had plowed through several years' worth of issues, and while I read page after page of praise for a lengthy list of seemingly promising telecommunications companies, I saw nary a hint of warning in anticipation of the Nasdaq's March 2000 tumble and the financial tumult that followed. He adds quickly, "I told people in early 2000 they should sell half their shares in these companies." Then he says, in a tone of self—rebuke: "I didn't say it often. I didn't put it in a newsletter."
He made the recommendation to sell, he admits, only within the limited confines of the Telecosm Lounge, his online salon for newsletter subscribers. He fumbles for words, starting one sentence, then another, before growing uncharacteristically silent and staring off into the distance.
For a moment, he seems to be imagining what might have been if things had turned out differently. Gilder, 62, is the author of a dozen books; he has been shouted down by feminists on The Dick Cavett Show in the 1970s, faced off with Dan Rather over trickle—down economics on 60 Minutes in the 1980s, and debated the future of technology with the likes of Andy Grove and Bob Metcalfe in the 1990s. Now many of his partisans are calling for the tar and feathers. He starts another sentence, and again cuts himself off. Suddenly he squares his body, turns to me, and expels a slight, disbelieving laugh. "When you're up there surfing," he says, "the beach looks beautiful. You never think about what the sand in your face might feel like until after you've crashed.
For a short stretch during the late 1990s, Gilder's newsletter made him a very wealthy man. Anyone taking a cursory look at it might wonder why. Every issue is densely freighted with talk of lambdas, petahertz, and erbium—doped fiber amplifiers. The eighth and final page, however, explains how so geeky a publication attained, at its zenith, an annual subscription base of $20 million. It's on the back page that Gilder lists the stocks he has dubbed "telecosmic" — companies that have most faithfully and fully embraced the "ascendant" telecom technologies in which he believes so wholly and deeply. "For a few years in row there, I was the best stock picker in the world," Gilder says ruefully. "But last year you could say" — here, for emphasis, he repeats each word as a sentence unto itself — "I. Was. The. Worst." Most of the companies listed have lost at least 90 percent of their value over the past two years, if they're even in business anymore.
None exemplifies Gilder's rise and fall more than Global Crossing, which filed for bankruptcy — the fourth—largest ever — in January. Even in a portfolio of flops, the scope and depth of this particular debacle stands out. "It will change the world economy," Gilder wrote a few years ago about the company. After reading its master plan, which called for the laying of fiber—optic cables across the world's oceans and between its great cities, Gilder proclaimed that for 10 years he had been searching for a business this audacious and awe—inspiring. He declared Global Crossing his favorite stock, and staked his financial future on it. While he avoided investing in practically every company he wrote about because of the potential for charges of conflict of interest, this was a notable exception. "Global Crossing going bankrupt?" Gilder asks, a look of disbelief on his face. "I would've been willing to bet my house against it." In effect he did. Just a few years ago, he was the toast of Wall Street and commanded as much as $100,000 per speech. Now, he confesses, he's broke and has a lien against his home.
During a period when blind optimism got the better of so many, no one was more blithely optimistic about our wired future than Gilder. Beginning in the mid—'90s, he advanced the argument that the businesses which most aggressively embrace fiber optics, wireless communication, and other telecommunications breakthroughs would soar in the meteoric fashion of an Intel. It was Gilder, as much as anyone, who helped trigger the hundreds of billions of dollars invested to create competing fiber networks. Then everything imploded, and company after company went under. The telecom sector proved to be an even greater financial debacle than the dotcoms. Yet he's still convinced he was dead—on right in most of his prognostications.
And the damn of it may be that Gilder has a point.
In addition to being famously optimistic, Gilder is also a contrarian. In the mid—1990s —†while the rest of the world was grousing about the slowness with which images and large packets moved over the network, and some very smart people fretted that the Internet would collapse under its own weight — Gilder was already talking about the coming age of network abundance. And being Gilder, he didn't stop there. He vividly imagined a "new epoch of spirit and faith" in which all of us would live in the "majestic cumulative power, truth, and transcendence of contemporary science and wealth." He also coined the termtelecosm to describe the merging of newer technologies, especially fiber optics, with existing telecommunications systems.
Gilder first spied the revolutionary potential of fiber optics at the start of the 1990s, when he shared a conference podium with Will Hicks, one of the field's luminaries. Hicks had predicted that fiber—optic cable, if it were made thin enough, could transmit bolts of light like photons shooting out of a ray gun. Gilder recalls the moment as one of the rare times he encountered someone even more Panglossian than himself. "I had always taken it for granted," he would later write, "that in any assemblage of pundits, I would be the most cornucopian — the most hyperbolically assured that silicon could save the world."
The predictions Gilder has made in the intervening decade suggest that he vowed to never again permit anyone else to convey a vision of the world more exuberant than his own. In 1996 he foresaw that, because of broadband's potential to deliver online learning, within five years "the most deprived ghetto child in the most benighted project will gain educational opportunities exceeding those of today's suburban preppy." It was a preposterous assertion, and hardly the only one that seems absurd in the harsh fluorescent light of the morning. He also claimed that the Web would soon bring on the quick death of both the US Postal Service and television. But none of this rendered Gilder's optimism any less contagious given the light—headed exhilaration of the times.
Yet to dismiss Gilder as just another poster boy for the reckless optimism of the late '90s would be a mistake, for the technical analysis undergirding even his more utopian flights of fancy was prescient. Forget terms like megabits,gigabits, or even terabits when describing the flow of data over the Internet. Soon enough, we'll be measuring things in petabits, or 1 quadrillion (1,000 trillion) bits, because of fiber optics — information traveling via photons flying over strands of glass fiber. The only question is whether we'll see this day as quickly as Gilder imagines. He asserts that by 2004, networks of glass superhighways will deliver 8 petabits per second over optical cables. "I listen very closely to what George says, and then automatically add five years," says Google CEO Eric Schmidt, who first encountered Gilder in the early 1990s when Schmidt was Sun's CTO.
The potential of fiber optics is indeed staggering, though not entirely without precedent, and therein lies Gilder's greatest contribution to the field. He opened minds to the technology by drawing on his understanding of past innovation; Moore's law accurately anticipated that the density of transistors on a computer chip would double every 18 months. Gilder decided to make a prediction of his own, and in 1998 he unveiled an axiom he dubbed the law of the telecosm: The world's total supply of bandwidth will double roughly every four months — or more than four times faster than the rate of advances in computer horsepower.