"Not your father's library, is it?" I remark, as Keller winds down to a stop.

"No," says Keller, with the authority and self-possession of a high-ranking military officer. "And it hasn't been for quite some time."

Keller is exhibit A of that quintessential blend of academic and entrepreneurial pursuits that suffuses Stanford life. Keller isn't just Stanford's head librarian; he's also the publisher of Stanford's HighWire Press -- a commercially self-supporting online publishing house for scientific journals. HighWire, Keller informs me, isn't just about providing access to scholarly information -- it's also about encouraging a "market correction" that will alleviate the current crisis in scholarly journal publishing. Too many journals cost too much these days for cash-strapped libraries to be able to acquire more than a tiny portion of what's available. HighWire offers one potential solution to the problem.

When Keller starts talking about HighWire, he comes off as a dead ringer for a start-up CEO pitching his company to potential investors. I may have come down to Stanford to learn about the Silicon Valley archive project, but I won't be able to leave the campus without hearing plenty about HighWire -- not to mention the ongoing $50 million renovation of Stanford's main library into a high-tech wonderland.

Keller is smooth and engaging -- as eager to talk about medieval monks copying ancient tomes or the mysteries of cellular signal transduction as he is to discuss the role of the Stanford library in the digital era -- but every now and then his patter falls into a slightly formulaic delivery. It's a sure sign of a spiel delivered many times before. And why not? More money always needs to be raised, whether you are a librarian or an entrepreneur. As one of Keller's associates tells me later, the library's renovation hasn't yet been completely paid for: "There are still plenty of naming opportunities available."

Keller takes me to lunch in a plush private room at the Stanford Faculty Club with some of his top lieutenants and Paul Saffo, a director at the Institute for the Future, a technology forecasting think tank. The conversation ranges all over the place -- at one point Saffo and Keller are trading tips on how to get the most from a new-model Nokia cellular phone -- but the flow always returns to the valley's most cherished icon, the start-up entrepreneur.

As well it should. After all, one can argue that Stanford President David Starr Jordan kicked off the unofficial beginning of Silicon Valley way back in 1909, when he bankrolled (with a mere $500) the invention of the vacuum tube. His impulse has since been followed by countless Stanford alumni. In the 1980s, Sun Microsystems and Silicon Graphics led the charge. More recently, former Stanford students made headlines with Yahoo, Excite and VA Linux. And just one week after my visit to Stanford, the two former Stanford computer science graduate students behind the search engine Google hit it big, scoring $25 million from venture capital superstars Kleiner-Perkins Caufield & Byers and Sequoia Capital.

From $500 in 1909 to $25 million in 1999 -- it's no wonder that, as Henry Lowood, Stanford's curator for the History of Science and Technology Collections notes, "it's a lot easier to get students to take [a course in the history of Silicon Valley] than it is get them interested in early enlightenment Germany."

"For a not-insignificant proportion of the students," adds Alex Pang, project manager for SiliconBase, Stanford's online archive of Silicon Valley history, "the course is a how-to."

But what exactly do you study? Just what constitutes the history of Silicon Valley? It's a question that vexes computer age historians. One might imagine that chronicling an era as recent as the flowering of Silicon Valley might be relatively easy, but the sad reality is that historians are already losing access to core materials.

Vast amounts of digitally stored information are contained in media formats already unreadable by modern technology. Web sites vanish almost as quickly as they are created. Even worse, the valley is now purposefully destroying itself. Local corporations, spurred by anxious lawyers, are now demanding that corporate e-mail be "scrubbed clean" every few months.

"We run the risk of, 200 years from now, knowing less about this place and this point in history than we know about 15th century Germany, where the printing press was invented," says Pang. "And arguably we are in a time that is seeing the development of technologies that ultimately are going to be as important as movable type. If the basic materials get lost it will be a genuine tragedy."

There's also the problem of what Keller calls the valley's "lack of historical self-consciousness." In an environment in which product cycles are measured in terms of months, no one has time to consider what might be worth saving, or what future generations may find useful. As Saffo jokes, "The difference between the valley and everywhere else is that other people have nostalgia. We have prostalgia -- a sentimental attachment to things that don't yet exist."

But as the pioneer generation of Silicon Valley begins to retire and reflect upon its past, prostalgia may finally be on the wane. The rise of the Net, combined with the computer's victorious penetration into the fabric of nearly daily life, is encouraging a new interest in the history of computing.

Next page | When history disappears right before your eyes