US librarians vs Google
p2pnet news view Advertising | P2P:- US Librarians are taking Google on.
But there’s a difference. They’re aiming at providing a vehicle through which users will find information rather than a user exploitation platform created solely to suck in advertising dollars.
Called Reference Extract, it’s, “envisioned as a web search engine, like Google, Yahoo and MSN,” says site.
However, “unlike other search engines, Reference Extracts will be built for maximum credibility by relying on the expertise and credibility judgments of librarians from around the globe,” it says, going on:
“Users will enter a search term and get results weighted towards sites most often referred to by librarians at institutions such as the Library of Congress, the University of Washington, the State of Maryland, and over 1,400 libraries worldwide. This grant will support planning for Reference Extract and building the foundation necessary to implement it as a large-scale, general user service.”
Behind it is the Online Computer Library Center and one of the founders, Michael B. Eisenberg, dean emeritus and professor at Washington’s information school, is asking for people to submit ideas on the project Web site.
“Google is everywhere, easy to use, and somewhat effective in offering useful results,” Wired Campus has him saying.
“But, I can’t always trust the results,” he wrote. “Is there a way to improve on that?”
“How does this idea differ from DMOZ.ORG (The Open Directory Project)?” – asks a comment post.
“Isn’t it just DMOZ but limmiting the editor function to approved Librarians?”
“This is not a directory program where a group of folks (librarians in the case) go out hunting for good web sites,” answers says project partner R. David Lankes, director of the Information Institute of Syracuse, and an associate professor in Syracuse University’s School of Information Studies »»»
Such projects have two primary problems: selection and scale. in essence, which sites should be included and how can we get a bunch of them so the tool is worth using. Reference Extract as currently envisioned will build on librarians answering real questions around the world. They answer a lot of questions, handling, in part, the scale question. The selection is not based on some common arbitrary criteria, but selection is done by answering real users and their questions. This generates a lot of “semantic richness” with which a search engine and other services can be built. Reference Extract can not only get a lot of good sites in (that is getting constantly updated), but better select and target sites based on real questions and human conversations.
“I’ve got to say this gives me a slightly sinking feeling, to do with seeing colleagues girding their loins to fight unwinnable and unnecessary battles (I’m a professional librarian BTW),”" says Ben Toth in another comment, adding:
“For example, there is an initiative in the UK called Intute which makes similar claims to the ones you’re making. It’s been running for years but has never solved the problems of timeliness and scaling. As a result, Intute is incomplete. if you search it for cost effectiveness of statins you find 1 reference – a fairly good one. But it misses this reference – http://www.hta.nhs.uk/pdfexecs/summ1114.pdf – which is more up to date and more comprehensive, and found on Google.”
“I hear your sinking feeling,” Lankes responds, adding »»»
However, the point is not to go toe to toe and “bring down Google,” but rather to build a new set of functionality that will have its own search interface as well as be able to partner with folks like Google, Microsoft, Yahoo, etc. So, I could imagine a situation where someone goes directly to Reference Extract and searching. I could also see a user that searches another search engine and gets back the regular results with some highlighting credibility scores computed from a librarian’s citation pattern.
“No word on the technical architecture that would power the search engine,” writes Lisa Guernsey in Wired Campus.
“(The vision reminds me of The Librarian, the uncannily human-like software in Neal Stephenson’s famous science-fiction novel Snow Crash.)
“Entrepreneurs have been trying for years to beat Google at its game. Could the combined expertise of tens of thousands of librarians conquer the juggernaut?”
Wired Campus – Librarians Want to Out-Google Google With a Better Search Engine, November 10, 2008
Use free p2pnet newsfeeds for your site. It’s really easy!Subscribe to p2pnet.net | | rss feed: http://p2pnet.net/p2p.rss | | Mobile – http://p2pnet.net/index-wml.php
Net access blocked by government restrictions? Use Psiphon from the Citizen Lab at the University of Toronto. Go here for details.
