The Internet is, by its very nature, a transitory medium-pages come and go. But if you had a publicly available Web page in the past three years, chances are that a copy of it is in the collection of the Internet Archive, a nonprofit group that saves “snapshots” of the Internet.
The Archive was founded by Brewster Kahle, whose San Francisco-based Web browser company, Alexa Internet, collects the snapshots every two months and donates the digital tapes to the Archive. As of May, the Archive was in excess of 13 terabytes (a terabyte is 1 million megabytes); in comparison, the Library of Congress holds the equivalent of about 20 terabytes. The Archive is stored in two separate machines in different locations. “It’s too important to have in one place. An earthquake could cause destruction of a collection that’s as large as the largest library ever built by humans,” says Kahle.
But it is proving easier to save the information than to sort through it for any useful purpose. While recent data are stored on disk for quick retrieval, the bulk of the archive is in a library of digital tapes that are too slow to search effectively. Currently, the only way the public can get at it is through the Alexa toolbar (downloadable at www.alexa.com), but, at the time TR went to press, only about the last six months of snapshots were available. When the reading room for these massive stacks is finally built, however, the Archive will be quite a collection.
Become an MIT Technology Review Insider for in-depth analysis and unparalleled perspective.Subscribe today