Web search gets semantic

This article was originally on a blog post platform and may be missing photos, graphics or links. See About archive blog posts.

As far as it has come in its young life, the Internet still has a ways to go before search engines realize their potential to intuitively supply the answers we seek.

That is the still elusive promise of the ‘Semantic Web,’ a term coined by World Wide Web creator Tim Berners-Lee. San Francisco start-up Powerset recently unveiled its search engine, which tries to figure out the meaning of words and sentences by breaking down their relationship to each other instead of relying on keywords.

For now, Powerset is focused on Wikipedia’s online encyclopedia, offering users detailed answers to conversational questions (Which dog sheds the least?). Powerset plans to eventually expand to other websites using natural language phrases or questions. It plans to make money the old-fashioned way that Google and Yahoo do: by selling advertising next to search results.

Powerset is just one of many focused on the semantic in San Francisco. To breed camaraderie, Powerset hosted a semantic technology happy-hour several weeks ago, inviting South of Market neighbors Radar Networks and Metaweb Technologies.

‘2008 is going to be the year that semantic technology really takes off, and these are the companies that will be leading the way. Our plan is to continue these happy hours, rotating to Metaweb next time, every few months,’ Powerset product manager Mark Johnson said.

He and his semantic pals will also hit the Semantic Web Conference, which begins Sunday in San Jose. Expect more semantic banter and less wanton swilling than during the happy-hour, since Johnson and Powerset co-founder Barney Pell are both speaking on panels.

If Powerset and others one day succeed in their semantic quest, Web surfers may someday find what they need and what they want.

Now that’s an idea I can relate to.

-- Jessica Guynn