Linking the World’s Data

Abstract

key to this shift was the huge quantities of documents authored by countless authors in nearly every country. The need to find information in those millions, then billions, of scattered documents naturally led to the first search engines. But of course, that is not how we found information on the early Web. We “surfed” the early Web by starting with a page we knew and following one interesting-looking hyperlink to another to another to another. We spent hours playfully exploring the new medium. We might start by looking at Vatican artwork and find ourselves hours later having read something of the history of an artist, the major works of that artist’s hometown, the contents of a building in that town, a biography of the mayor, a quote from his favorite philosopher, the broad brushstrokes of utilitarianism, its relation to Epicurus, epicurean delights, and a wonderful new recipe for kangaroo rogan josh. It was exciting and stimulating. The Web’s ability to create a path from the Vatican to Kashmiri cuisine was unprecedented. People could not follow all the links on the growing Web. Computers, though, are not people. Computers are dumb but fast, uncomprehending but with vast memories. Could computers trace the filaments of the Web and create maps for us to use? They could. Search engines reduced the intellectual smörgåsbord of Web surfing to a niche activity. We no longer “followed our noses” from link to link. Link following was too slow and too undirected. Search engines were simply a more efficient way to find what we were looking for. Linking the World’s Data