The Semantic Web, which some call Web 3.0, refers to an Internet where content has structure and allows for meaningful (semantic) queries. So the Semantic Web gives structure for machines and meaning to human readers.
At present, web pages mostly have plain old text with very little structure. Queries in a regular search engine give us loads of results, but they are sometimes not what we want. With the Semantic Web, we could ask for, say, a list of holiday packages to a specific destination, within a certain time frame, leaving from airport X, for a price in the range of Y-Z, etc. Or, you might search for homes near a particular location, or within T miles of a public school. The choices are virtually endless.
So what do we need to support the Semantic Web? Structure. Web content would have to have some sort of structure, and would have to be XML compliant. By having structure, the Web’s content becomes like a database. And structure is what web 2.0 feed mashup applications like Yahoo Pipes and OpenKapow can offer. The Semantic Web isn’t quite here yet, but these apps are pushing us forwards.
What we still need is more powerful query widgets in web application builders like Teqlo. Then, anyone wanting to do a query that they’ll use at least a few times can just plug and play a few widgets, run the custom application, and view the results.
Of course, this opens up the possibility of mobile Semantic Web applications – maybe even voice-driven – via smart phones connected to the Internet. Such apps would be ideal for the tourist or occasional traveller trying to find markets or other places to visit. These sorts of applications will add tremendous value to online content. I don’t know about you, but I can’t wait for true Semantic Web apps.
Originally posted on March 26, 2007 @ 8:50 pm
Aviva Gabriel says
Can you be more specific about the meaning of “structure” in this context? I understand (a little) about databases, but what are some of the mechanisms for creating a “database-like structure” for web-based information? Since the context of word usage is vital to establishing meaning, how do you “assign context” or multiple contexts to words? This kind of task is enormous, and seems to require making predictions about every single possible context in which any given word or term might be used. How does this get done?