By Burt Lum

With little fanfare, www2002, the 11th International World Wide Web Conference (www2002.org), blew through town. Much of the material was so bleeding edge it flew above my head. Yet the topics discussed will have broad impact and will touch each of us in profound ways.

One prime topic of conference discussion was the concept of a semantic Web. This, according to Tim Berners-Lee, the inventor of the Web, is what the Web is evolving into.

So what exactly is a semantic Web? If we start with the Web as we know it, it basically consists of documents containing text and images. We reach these Web sites by typing a URL (universal resource locator). This same practice is performed for the millions upon millions of Web sites existing on the Net. As you might imagine, this available resource of information is tremendous but highly unstructured, and for the most part, nonrelational. Current search methods tend to be contextual, limited to one- or two-word searches.

The semantic Web aims to change that. The main concept behind the Semantic Web is to create a set of information about a Web site that a computer can understand and interpret. This goes beyond word searches.

Imagine a standard set of descriptors for a Web site that a computer program, called an agent, can interpret and return results from. This information set would become a standard and would be common among similar Web sites.

One example of the semantic Web in action would be scheduling and booking your complete vacation plans. Through a computer agent, you could set your flight plans, book your hotel, rent a car, buy tickets to a concert, find all of the souvenir stores in the vicinity, etc. This would be accomplished with information from the semantic Web, accessible by application agents.

For more information on this topic, check out www.w3.org/2001/sw/. There also is a link to an excellent primer on this and other leading-edge topics from Scientific American magazine called "The Future of the Web."

Reach Burt Lum at burt@brouhaha.net.