Metadata Committee Monday July 14 at 9am - Goldberg Conference Room.

Present: Marshall Breeding, Nancy Boggess-Korekach, Zora Breeding, Ronee Francis, Mary Charles Lasater, Steven Nordstrom, Bill Walker, Pete Wilson, Roberta Winjum.

• Review annual report drafted by Pete and Zora.
  • Some minor changes were suggested. Everyone seemed happy with the report.
• Continued discussion of the Dublin Core records for the Readex collection of government documents
  • ACTION: Although there are issues in mapping DC, it was agreed the Readex records should be mapped to Discover Library.
  • NOT DETERMINED: The Metadata Committee is ready to map collections to Discover Library. Approved collections include VU e-Archive, the Global Music Archive, and the Readex records. A group volunteered to help map records including Mary Charles, Steve, Ronee, Nancy, and Bill (or Linda Tesar).
  • Notes:The discussion continued on whether the Readex records should be put into Discover Library. Advantages would be they could be used for citations and searchablity by subject words, however, the records do not contain library of congress subject headings. Whether or not Discover Library can provide an effective search environment for different types of records was discussed. One possibility is to make use of the DC:Type field as genre or create a facet.
  • What is added to Discover Library will affect other collections added. An example of this is the DC:Coverage field in the Redex collection. It is more similar to the DC:Relation.isapartof – series field used in VU e-Archive and previously defined in the Application Profile.
• Issues related to non-roman scripts that are now beginning to appear in authority records.
  • ACTION: Mary Charles offered to contact Marcive to see if there is a way to identify the records
  • ACTION: Nancy Boggess-Korekac will consult Dale P. about storage of data.
  • Notes:Because Acorn cannot read Unicode, authority records containing non-roman scripts translate into functional links with nonsensical and unreadable words. With the idea we will be able to one day read Unicode, the question requiring an answer, is, can we store the non-roman fields in a way that can be recreated so data is not lost? Can we locate the affected authority records in order to strip the non-roman script from them?
  • Users have employed the nonsensical strings (aka garbldy-gook) to find content.

Next meeting scheduled for August 18th.