Filtering Information On The Internet; March 1997; Scientific American Magazine; by Resnick; 3 Page(s)

All over the world, libraries have begun the Herculean task of making faithful digital copies of the books, images and recordings that preserve the intellectual effort of humankind. For armchair scholars, the work promises to bring such a wealth of information to the desktop that the present Internet may seem amateurish in retrospect. But many technical, economic and legal obstacles still make that promise an uncertain one.

To steer clear of the most obviously offensive, dangerous or just boring neighborhoods, users can employ some mechanical filtering techniques that identify easily definable risks. One technique is to analyze the contents of on-line material. Thus, virus-detection software searches for code fragments that it knows are common in virus programs. Services such as AltaVista and Lycos can either highlight or exclude World Wide Web documents containing particular words. My colleagues and I have been at work on another filtering technique based on electronic labels that can be added to Web sites to describe digital works. These labels can convey characteristics that require human judgment-- whether the Web page is funny or offensive-- as well as information not readily apparent from the words and graphics, such as the Web site¿s policies about the use or resale of personal data.