What if your Google search for "Paris Hilton" listed your top result as the Hilton Hotel in Paris, because it knew your interests were not in the other direction? This is the current dream of Tim Berners-Lee, the man who invented the (first) World Wide Web.
He calls his dream "Web 3.0" or the "Semantic Web," meaning it understands user context. He and many other experts believe that the Web 3.0 browser will act more like a personal assistant than a search engine. As you search the Web, the browser records your interests in your local storage. The more you use the Web, the more your browser learns about you, and the more relevant will be your results.
Current advertising and public relations startups are already thinking along these lines in fields all the way from clothes shopping, art galleries, online advertising, to managing press releases. In some ways, these aren't that different from the old Amazon.com "recommendation engine," which suggests new products based on your surfing and buying habits, but they go much further.
Someday you will be able to ask your browser open questions like "Where should I take my wife for a good movie and dinner?" Your browser would consult its intelligence of what you and she like and dislike, take into account your current location, and then suggest the right movies and restaurants. If you are the first to deliver this, your startup can be the next Google success!
But some are skeptical about whether the Semantic Web -- or at least, Berners-Lee's view of it -- will actually take hold. They reference other technologies also trying to reinvent the online world as we know it, from 3D virtual worlds to intelligent avatars. Web 3.0 could mean many things, and most of the possibilities have not yet been invented.
The Semantic Web isn't really even a new idea. This notion of a Web where machines can better read, understand, and process all the data floating through cyberspace first surfaced in 2001, when a story appeared in Scientific American. This article describes a brave new world where software "agents" lead the way in performing Web-based tasks that elude most humans.
A current example is the GetGlue from AdaptiveBlue. If you visit a movie blog, and read about a particular film, it immediately links to sites where you can buy or rent that film. Another example is WolframAlpha, an amazing computational engine that went live recently, which creates intelligent results, graphs, and reports from any natural language question.
But we are a long way from agents that can do full natural language processing and think on their own (artificial intelligence). A recent startup, Alitora Systems, provides software to enterprises based on a natural language processing (NLP) engine.
It builds knowledge statements from unstructured media files -- that's a particular challenge for the life sciences where high-value knowledge about many things, such as the relationship between genetics and disease, lies hidden within journal articles, research papers, clinical trial data, FDA websites, and even graphical data.
But extracting information from even less structured data such as Twitter feeds is a very different and sometimes more difficult knowledge extraction problem. The objective is the same; assimilating unstructured data, giving it some robust analysis, and offering the extracted knowledge across a collaborative network.
Just think of the fertile ground all this opens for startups! If you're looking for that "million dollar idea" to build a plan around, here is your chance. But don't wait too long, because the din for Web 3.0 is getting louder and louder. Catch the wave soon or it will pass you by!
Follow Marty Zwilling on Twitter: www.twitter.com/StartupPro