Anyone can search the Web. Now, Google aims to create its own invisible Web
Google launched a beta version of Google Base on Wednesday. The tool, in testing since October, lets users upload all kinds of content to Google’s servers.
It is an extension of Google’s existing content-collection efforts, with a goal of broadening Google search results, according to Salar Kamangar, vice president of product management.
“We think of it as an extension of the ways we have for collecting information,” he said. “The prior method was one where we pulled information from the Web crawl, and that will continue to be our primary method. But we’re experimenting with methods of pushing information … where people can create content and push it to us.”
Google Base lets users describe the content to be uploaded and assign attribute tags.
When someone searches Google Base, each result will include a list of attributes to help the searcher hone in on the more relevant result.
The attributes are a way for Google to study whether tagging can increase search relevance, Kamangar said. “We are thinking about this as a way to have more search results in our regular properties, and experimenting with attributes to see if they might improve search results,” he said. If it does, the search provider may extend tagging to organic search.
Gautam Godhwani, CEO of SimplyHired, a career resource site that aggregates job listings from other sites, said Google Base was a very significant shift from Google’s old mission of organizing the world’s information.
“We’re seeing the movement toward integrating data that either doesn’t exist online at all or isn’t available in a meaningful structured way at all,” he said. “In the process of creating structured data, Google is saying, ‘We don’t just take data from the Web, we’ll take data from anywhere.'”
To seed the database, Google worked with several information providers, including ArtNet, an online directory and marketplace for art; Career Builder, a job listings site; CollegeBoard.org, a membership organization that hosts college-planning resources; StepUp, a company that lets people search product lists of local retailers; and the World Resources Institute, an environmental think tank that uploaded links to data on sustainability issues.
The company said Google Base demonstrates its efforts to make more types of information discoverable online. The move highlights a growing trend in search: proprietary content in the index.
R&D aimed at improving search relevance shows diminishing returns. Yahoo , MSN and AOL Search has been able to wrest share from Google despite launching their own search technologies. Instead, the top four providers have been working to amass unique content.
AOL is bulking up its multimedia search index with 15,000 exclusive licensed video assets from Time Warner.
MSN is in the process of scanning 25 million pages of books from the collection of the British Library.
And the beta version of Yahoo’s My Web 2.0 search technology takes into account tags and ratings of Web content produced by those in the searcher’s network.
Kamangar said Google plans to allow third parties access to the database content. When asked whether MSN’s and Yahoo’s crawlers would have access, he said, “We’re putting this out and making decisions later. It’s certainly possible. That’s the principle: We do not want to lock this information.”
Peter Zollman, principal of Classified Intelligence, a market-research firm focused on the classified listings industry, considers Google Base as the basis for a full-blown classified listings service, another service that Google is rumored to be brewing in its basement.
“Google’s product is not a sure thing,” wrote Zollman in a research note Wednesday. “But it still poses an enormous challenge to traditional classified advertising publishers who use the pay-to-advertise model.
“Since all Google Base ads are free, the value of the contextual ads around it will soar, and the price pressure on traditional classified advertising publishers will intensify.”
