The classic Internet search model, where users type in a query and are confronted with page upon page of links ranked by someone else's idea of relevance, is giving way to a more efficient paradigm in which results are informed by who the searcher is, who his or her friends are, where they live, and other user-centric data.
The classic Internet search model, where users type in a query and are confronted with page upon page of links ranked by someone else's idea of relevance, is giving way to a more efficient paradigm in which results are informed by who the searcher is, who his or her friends are, where they live, and other user-centric data.That, at least, is the belief of panelists who spoke Tuesday at the Web 2.0 and Interop conferences at New York City's Javits Center.
Representatives from Facebook, Yahoo, and real-time search aggregators OneRiot and Collecta said the change is occurring because the classic page-rank model fails to fully leverage the power of social networking and other Web 2.0 tools and thus doesn't always deliver what the user is really looking for.
For instance, Michael Jackson's death earlier this year stirred up millions of queries on the pop legend. But most searchers didn't want more news about his demise-they wanted videos and other media that could help them remember Jackson.
"The most socially relevant result was the Thriller video," said OneRiot general manager Tobias Peggs. OneRiot pulls information from the Web, blogs, Twitter, and other sources to deliver contextual search results. Users have the option of hitting a button for "real time" updates, mostly from social networking sites.
Facebook engineering manager Akhil Wable agreed that sites like the one maintained by his employer have a strong role to play in improving Internet search. "It's more than just metadata when your friends are talking about it," said Wable.
The problem that older, traditional search sites have when it comes to integrating Twitter feeds and the like is that they still need to serve a more general audience, for whom the bulk of information from social networking sites may not be relevant.
"Topics on Twitter tend to be fairly geeky," said Yahoo search architect Vik Singh.
Yahoo studies have revealed that less than 2% of Tweets are related to hot topics that are trending on major search sites like Google and Yahoo, according to Singh. Still, he agreed that even the big search engines can't ignore the social Web when it comes to improving results.
Collecta CEO Gerry Campbell said the best search results rely on numerous sources. "I pull information from places that I would never see on my socialgraph," said Campbell. The Collecta search engine pulls data from blogs, microblogs, social networking sites, and even photo sharing services.
5 Top Federal Initiatives For 2015As InformationWeek Government readers were busy firming up their fiscal year 2015 budgets, we asked them to rate more than 30 IT initiatives in terms of importance and current leadership focus. No surprise, among more than 30 options, security is No. 1. After that, things get less predictable.
Join us for a roundup of the top stories on InformationWeek.com for the week of December 14, 2014. Be here for the show and for the incredible Friday Afternoon Conversation that runs beside the program.