Mon
16 Feb 2004
2:53 pm
Washington Post: Search For Tomorrow
Posted by shimon under business/companies/Google , computers/AI , computers/search , computers/semantic webNo Comments
A well written article with a pretty good analysis of the strengths and weaknesses of internet search, starting with the typical lines from librarians and proceeding to describe some of the potential of personalized search.
The way one part links to another reflects the preferences of Web users — and Google tapped into that. Google, in detecting patterns on the Web, harvested meaning from all that madness.This points the way to one of the next big leaps for search engines: finding meaning in the way a single person searches the Web. In other words, the search engines will study the user's queries and Web habits and, over time, personalize all future searches. Right now, Google and the other search engines don't really know their users.
Unfortunately, it's hard to gain much inspiration from this article since it mostly repeats the adages of semantic web and "intelligent agents" research. In case you hadn't noticed, neither of these fields has made visible progress in search or any other realm. Where AI and explicit metadata techniques were once thought to be critical, approaches that seemed overly simplistic are closing in on the AI vision very quickly. The best example is Google, but AI/semantic web is losing out in other counts. Like right now it seems that a more feasible way of automatically figuring out schedules between people around the world is for Microsoft to achieve 100% market share with Outlook. I certainly don't look forward to that but the semantic web community doesn't seem to understand that it would take years of work by a marketing force as leviathan as Microsoft to achieve the kind of widespread consensus they require.
Not that intelligent agent research couldn't be useful at some point, but if Google wants to bank its future on knowing its users, it's own Blogger and Orkut services are much more valuable.
Full Entries RSS