Saturday, April 07, 2007

Semantic Web is closer to be real, isn't it or is it?

A recent post by technical evangelist Robert Scoble brought a small hype of semantic web again. His article was about the new achievement done by Radar Networks, which was founded by Nova Spivack. The following is quoted by Robert's post.

Basically Web pages will no longer be just pages, or posts. They’ll all be split up into little objects, stored in a database (a massive, scalable one at that) and then your words can be displayed in different ways. Imagine a really awesome search engine that could bring back much much more granular stuff than Google can today. Or, heck, imagine you could view my blog by posts with most inbound links.

Robert have expressed his excitement about watching the semantic web demo with Nova. Although I have not gotten the chance to experience it by myself, I can briefly feel what is going on and which technologies Radar Networks is now employing. This expressed scenario is a certain picture that current semantic web technology can support. Here at our research lab at BYU, we cooperate with some DERI researchers together and work on a project about enabling the semantic web to be real. In fact, the paradigm of our plan is close to the scenario that is expressed by Robert. So I feel familiar when I watch Robert's expressions.

The idea of semantic web has been discussed for years. The web is certainly moving towards holding more and more machine-processable semantics. But until now, the research of semantic web is still mostly limited in labs. What are the reasons?

According to our beliefs, the reason is not due to technical difficulties. Technical difficulties are severe problems, but not the deadly ones. The real difficulty is about who are going to take the control of semantic definitions. Are these definitions controlled by few elites or by the public? This is a grand question.

Many people have predicted the rise of "semantic Google." But I would say that there would never be a "semantic Google." Again, my argument is not due to the technical reasons. The problem is about who is going to be the owner of such a "semantic Google." If it is a US company, I must say sorry to it that countries such as China, Russia, or even European Union would ban it and build their own semantic giants because any major country in this world would not endure its "semantics" being controlled in the hand of another country. This type of threats is intolerant to any independent country.

In fact, we may even not need to raise this problem to the level of nations. Even within individual persons, no one would like to be forced agreeing on the semantics defined by another person. This semantic-definition problem is indeed the most crucial problem to the prevalence of the semantic web.

What is the solution? In fact, the success of Web 2.0 has shown a pragmatic resolution. We need to allow the public to define semantics by themselves. This is the only way that can promise the prevalence of semantic web. Any web user can apply his machine agent to understand the web based on his own understanding. This is what a pragmatic semantic web should deliver. On such a pragmatic semantic web, no "semantic Google" can exist due to the massive diversity of human understanding. A collaborative web search model will replace the current centralized search model. In fact, we have invented a new theory of collaborative web search on the semantic web, and hopefully it could be released soon.

In summary, Nova Spivack and his company Radar Networks have done great achievement on realizing the semantic web. We must congratulate them! Great work, Nova! On the other hand, however, unless they can show that their solution has properly solved the semantic definition problem, the age of semantic web is still not there yet.

1 comment:

Wayne Smallman said...

No one company can really be the primary driver behind a Semantic Web, not even Google, for all their might.

This is something that's got to come out of the World Wide Web Consortium and be supported by all of the big players in the web.

Yes, "Web2.0" is sort of the poor mans Semantic Web, but it's only a stop gap.

Lots of small applications are needed as intermediaries to support the storage and the transaction of data needed to make Web2.0 work, which isn't ideal...