Tag Archives: reference extract

resisting google: not so futile

Not too long ago I mused upon the idea of how some search engine companies are trying to provide more  human interaction when one has an online reference question, by either doing the searching or providing suggestions on how to perform the search.  This quasi virtual reference seems to be catching on, and librarians are suddenly becoming more recognized for the credibility they provide in their reference work.

This sentiment is the impetus for a new project that aims to compete with likes of the great goog, Reference Extract.  The project, an ever-increasing collaboration of libraries, aims to differ from Google in the credibility taken from the shrewd linkages that librarians provide in applying sound information literacy principles. Said better than myself:

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.

The issue of credibility is interesting when compared to the measure of relevancy and popularity Google bases its index on.  The issue of credibility is more fully explained:

In essence linkages between web pages by anyone is replaced by citations to web pages by highly trained librarians in their daily work of answering the questions of scholars, policy makers and the general population. Instead of page rank, the team refers to this as “reference weighting.”

That is to say, it is no great leap to believe that working one-on-one with a librarian would yield highly credible results, but it also appears that gathering the sites librarians point to across these one-on-one interactions and making them searchable continues to yield highly credible results. Further since the librarians answer question on very wide range of topics, their answers can be applied to a general purpose search engine.

I find it clever that the organizers of RefEx measured their index by using the custom search engine provided by Google…beating it at its own game perhaps.

It is important to note that by using the Google Custom Search Engine service the exact same technology was used to search and rank the results, the only thing that varied was that one was an open web search, and one was limited to only those pointed to by reference librarians. So, even outside of the library website context the credibility of librarians is retained.

We may index less pages, but the ones we point to are more informationally literate. One question to walk away from with this: does less material indexed = more reliable?  Philosophically speaking, words like popular, relevant, and usefulness will cause debate; academically speaking, this justifies the librarian’s attempt to wean those frothing, zombie-like patrons away from The Google and more toward our subscribed databases, online resources and guides.  And with RefEx, Google’s helping us do it.