marc record tag cloud

From: Eric Lease Morgan <emorgan_at_nyob> Date: Tue, 13 Mar 2007 08:31:10 -0400 To: NGC4LIB_at_listserv.nd.edu

For a good time I created a MARC record tag cloud:

   http://tinyurl.com/ypqcyn

Tag clouds are graphic illustrations depicting the common elements in
a larger set of words. Using tag clouds you can see what words occur
frequently or not. Delicious uses tag clouds extensively. So does
LibraryThing. A person an also use tag clouds to navigate sets of
documents. Click on word. Get search results.

A service called Many Eyes by IBM allows anybody to create tag clouds
-- as well as other data visualizations. I did this with a set of
14,000 MARC records representing new books, and the result is
illustrated at the URL above. Based on what I saw, most of the
content the MARC records is administrative and does not lend itself
discovery. (If we spend more of our effort creating administrative
data than discover data, then what does that say about our values?)

I wonder what would happen if I removed the administrative data?
First off all, the data set size would probably diminish by 50%, and
second, the cloud would probably make more sense.

--
Eric Lease Morgan
University Libraries of Notre Dame

(574) 631-8604