The sounds of libraries in the past were not always terribly easy to pick out: the rustling of paper, the occasional scratching of pens, the odd scrape of a chair and whispered conversations. Nowadays, though, there are more sounds in the background. In a remote section of the venerable library at the University of Göttingen, founded in 1734, you hear hissing, clicking and whirring -- the sounds of Old Europe meeting new media.
An old volume lies open under glaring lights, as if it were spread out on an operating table. A librarian focuses the lens, checks the image on a computer monitor, and activates the scanner. Then she taps the left pedal with her foot, a glass plate rises with a hissing sound, she turns the page, and the plate lowers itself, whirring, onto the next page. Click. With this process, up to 300 pages an hour are being uploaded from the book world into the cyberworld as characters made of printer's ink are converted into 1's and 0's.
The work in the University of Göttingen library's digitization center is moving along at a steady clip. The project's nine full-time employees have already digitized more than 4 million pages from periodicals and thousands of volumes of books and transferred the data to the center's server, from which it can be downloaded in a matter of seconds through any computer connected to the system.
But while it may at first sound like a lot, library director Elmar Mittler is far from satisfied. "It's enough to drive you to despair," he says, speaking so quietly that one has to lean forward to understand him. But then he slaps his hand against the table and adds: "It's hard to believe how Germany has wasted its opportunities with digitization. Others are moving ahead, while we write reports."
Mittler, 65, is arguably one of the leading experts in this field. In the mid-1990s, he developed one of the first German catalogs available on a digital network. In 2000, his team unveiled the electronic version of one of the 15th-century Gutenberg Bibles on the Internet, all 1,282 pages reproduced in high quality and full color. Although USA Today celebrated the site as perhaps the best Web site to mark the 550th anniversary of the printing of the Gutenberg Bible, the German public has hardly even noticed.
But that could soon change, especially in light of the newfound popularity of an old medium. After online games, blogs and podcasting, the big Internet companies have now discovered something of a totally different nature, at least for them: the good old book.
Especially startling was search engine Google's recent announcement of its intention to make the world's printed works available online. The company plans to digitize about 15 million works by 2015 and store them in a database, at a projected cost of up to $200 million. The tally of works digitized so far, according to Google, is already up to "several hundred thousand." The implications of the project are enormous. For example, someone who enters the term "Johannes Gutenberg" in the future will be referred not just to Web sites, but also to text passages in paper books, available in fully scanned, full text form, at least in the case of "generally free" books no longer under copyright protection. With new books, however, readers will have to make do with "text snippets" -- anything more would constitute copyright infringement.
At least a portion of the project's cost will be offset by targeted advertising that will provide links to libraries, publishers and bookstores for users interested in specific books.
Internet catalog merchant Amazon already offers a similar service. More than a hundred thousand English-language books can now be searched online for keywords, in what appears to be only the beginning of a more ambitious venture. As of July 2005, Amazon's database also includes German-language books from 120 publishers. The German trade publication "Buchreport" (Book Report) succinctly sums up the pull of the Internet, at least when it comes to books: "Those who find what they want read it." Amazon's book search engine has apparently increased sales of detective novels by 5 percent and comics by 48 percent.
The American Internet giant's entry into the business of book search engines signals the beginning of a deep-seated media revolution: While the printing of books makes dissemination of knowledge on an industrial scale possible, it is only the full-text search that automates access. In the past, for example, finding that specific passage from a book read long ago meant hours of flipping through hundreds of pages. Now, it will become as simple as plugging keywords into a search function.
Book search engines are still far from efficient, and the results are often laughably incomplete. But the potential is vast, with today's efforts representing only a first step toward a sort of universal library whose portals are open to anyone, day or night, whether in university towns like Göttingen or in an Internet cafe in Johannesburg.
But while the public appeal of this digital-library dream is strong, negative reactions in the professional world -- where librarians, publishing house executives and booksellers feel steamrolled by the marketing muscle of the world's Amazons and Googles -- are equally vehement. Critics are especially offended by Google's brusque approach. Instead of asking publishing houses for permission to scan their works, the company simply went ahead and did it. This approach is a "systematic violation of copyright law," fumed the Association of American University Presses.
Indeed, a wave of litigation is on the horizon, and authors are likely to be among the litigants. Anticipating trouble, Google recently announced that it was temporarily putting a stop to the scanning of books that could be problematic in terms of copyright law -- at least until November. According to an irritated Jens Redmer of Google, "the press has frequently portrayed this as an interruption, but the truth is that the disputed works make up an ever-shrinking slice of the pie."
But critics haven't relented. Their best-known spokesman is in Paris, where he occupies an office in one of the glass towers of the Bibliothèque Nationale de France (the French National Library, or BnF). Jean-Noël Jeanneney, director of the BnF, warns that the creation of international book-scanning policy shouldn't be left to American private companies. In a French-language polemic Jeanneney published in April, titled "If Google Challenges Europe," he argued that the dominance of English-language literature in the U.S. databases might not only displace the French language, but could also edge out European academic and legal traditions.
Other skeptics fear that book searches will no longer be free if Google runs into financial troubles. Or, even worse, search results could be manipulated to favor the highest-bidding publishers. Another concern is whether Google's ranking method, which automatically makes popular hits even more popular, will lead to a monoculture.
Jeanneney is clear about what he wants to see happen: European libraries should head off Google with their own digitization campaigns. In April, half a dozen heads of state asked EU commission president Jose Manuel Barroso to assemble a plan of action. But wheels tend to turn slowly when it comes to European bureaucracy.
In fact, it's questionable whether Europeans are even capable of staging a counteroffensive. Although a European library portal (called the European Library) already exists, all it offers is a confusing patchwork of individual national catalogs and individual digitized books in a wide variety of formats, but no comprehensive full-text search feature. Individuals occasionally press ahead with their own projects. Jeanneney, for example, had a collection of about 70,000 French works digitized at the BnF. But access to the database, which is called Gallica, is slow, complicated, unrewarding and daunting.
There has also been some movement in the German private sector. The Börsenverein des Deutschen Buchhandels (German Booksellers and Publishers Association), caught off guard by the American book search engines, is considering developing a decentralized cluster of publishing house servers that would make it possible to download current books in digital form. "Google could access our data, for a fee," proposes Matthias Ulmer, the director of the Full Text Search Online committee assigned to the project.
But when Ulmer presented his idea at a conference of book dealers in Berlin in June, the audience's mood remained skeptical. Besides, industry observers doubt whether struggling publishers are even capable of standing up to Google's technology experts when it comes to search speed, availability and logistics.
This technological deficit is all the more astonishing when one considers that Germany originally led the international pack. As early as 1997, the German Research Association (DFG) was spending about 28 million euros to fund more than 90 digitization initiatives throughout the country.
But when experts evaluated the effort in January, their results were devastating: Hardly anyone knows about or uses the Web sites that have been receiving the funding, many projects are "virtually invisible in the Internet," the range of available products is fragmented, and there are "only minimal synergistic effects." If these problems cannot be corrected, the report concludes, "continued funding should be at least significantly limited."
One major point of the criticism is that what's missing is a shared portal and, even more importantly, an understanding of user needs. Until now, says Ralf Goebel of the DFG, digitization programs in Germany have emphasized class over mass. In other words, digitization has centered on scanning valuable works for conservation purposes, both to protect them and reduce the need to lend them out. By comparison, the Google offensive seems to be the epitome of user friendliness. According to Göttingen expert Mittler, "the people at the DFG had neither the foresight nor the patience to keep developing all of their wonderful digitization projects to the point of marketability." Even Mittler's digitization center hasn't managed to make its scanned texts searchable as full texts. The reason is twofold, says Mittler: a lack of funding and limited political will.
And yet, such a high-tech and user-friendly project could certainly be a worthwhile investment. A modern research library is a marketing tool that can be used to attract researchers and students, in much the same way that the University of Göttingen's outstanding library attracted some of Germany's preeminent thinkers -- Lessing, Kant, Herder, Goethe -- more than 250 years ago. German author Heinrich Heine even immortalized one of the reading rooms in the famous library dream sequence in his novel "Harzreise" ("The Harz Journey") in which the dreamer is chased through the library by a horde of self-important scholars.
Sometimes, when Mittler walks through the historic reading room at night, he thinks of Heine's dream scene and is reminded of the German digitization landscape, with all of its little systems, projects and ivory towers.
He believes that it's still possible to wake up from this library nightmare: "If we want to ensure that German literature doesn't perish completely on the Internet, we must launch a major digitization project in Germany by early next year at the latest."
The more likely scenario is that the competition between America and Europe, between the private and public sectors, will ultimately turn into a cooperative venture. Libraries and publishing houses could supply the content, which they wouldn't be able to digitize on their own, while Google would provide funding and technical expertise.
Google is already evaluating the inventories of five libraries in the United States and Britain and, in return, is turning over digital copies to the libraries. "If we received a similar offer, it would be hard to say no," Mittler admits.
It's a decision that the officials in charge of Oxford University's venerable Bodleian Library have already made. Since early August, Google has been setting up its first European digitization production program on the library's premises, with scanning set to begin in October.
But Google book expert Redmer says that the company currently lacks the available capacity for additional projects -- in Germany, for example. "After all," he says, "we're not scanning for the fun of it, but out of necessity. We must set priorities."
There is one bit of consolation for Redmer: The Oxford library has books in many languages; some are even in German.
Translated from the German by Christopher Sultan
- - - - - - - - - - - -
This article has been provided by Der Spiegel through a special arrangement with Salon. For more from Europe's most-read newsmagazine, visit Spiegel Online at http://www.spiegel.de/international or subscribe to the daily newsletter.