In hunting for data to help model subject faceting for MARC records,
I've just been looking at Bill Moen's Zinterop report
(http://www.unt.edu/zinterop/ZInterop2/Documents/ZInterop2FinalReport_we
m4Dec2005.pdf). It occurs to me that with all our various projects
working on indexing MARC records in Solr, we should set up and
distribute a set of "radioactive records" to use in each project to
diagnose and compare indexing and querying behaviour. Probably we could
just use the Zinterop records (which are described in detail in that pdf
but aren't available for download anywhere I could find); but we might
want to enhance them with data suitable for testing our faceting
systems. Not sure what that would mean but I thought I'd throw it out.

If you were at Access '05, you heard Bill describe the Z39.50 testing he
was doing with radioactive records: records with known unique values in
all indexed fields, that could be used for automated testing of Z39.50
search functionality. The same approach might be very useful as we feel
our way towards a Solr MARC indexing system.

Has anyone already done something like this?

Peter

Peter Binkley
Digital Initiatives Technology Librarian
Information Technology Services
4-30 Cameron Library
University of Alberta Libraries
Edmonton, Alberta
Canada T6G 2J8
Phone: (780) 492-3743
Fax: (780) 492-9243
e-mail: [log in to unmask]