ALCTS - Association of Library Collections & Technical Services

Task Force on Metadata

Meeting Minutes

Sunday, January 16, 2000 — 8:00 - 11:00 a.m.
Four Points Hotel, Rio Grande Ballroom


Members present: Mary Larsgaard (Chair), Rebecca Guenther, Matthew Beacom, Brad Eden, Bill Fietzer, Judith Hopkins, Steven Miller, Mary Woodley
Approximately 20 guests.


Charge 4 prototypes:

Steven Miller and Mary Woodley gave a presentation on prototype systems that might fulfill the Task Force’s 4th charge to “recommend ... ways in which libraries may best incorporate the use of metadata schemes into the current library methods of resource description and resource discovery.” They defined “Prototype” as “a virtually seamless access to information and relevant retrieval of information from the user’s point of view.” They described different types of prototypes:

  • Those that search multiple databases vs. those that collect multiple metadata records into a single database or repository.
  • Those that search different metadata semantics vs. those that convert or map diverse elements into a single semantic standard searched by the interface.

Steven described NESSTAR (Networked Social Science Tools and Resources) which provides a common gateway into multiple databases using a single metadata semantic standard; AHDS (Arts and Humanities Data Service) which provides access to five databases using different metadata standards and cataloging practices, and CORC (OCLC’s Cooperative Online Resource Catalog) which currently uses two semantic standards: USMARC and an enhanced version of the Dublin Core to provide access to records in a web-based system.

Mary Woodley described three projects of the Getty Trust which have been retired plus its still-extant Getty Research Institute auction catalog records, as well as two Z39.50 projects: PHAROS (the California State University gateway for searching the online catalogs of the 23 CSU campuses plus electronic journals and reference databases) and SEARCHLIGHT which provides access to the California Digital Library.

For a fuller version of their presentation click on: http://library.csun.edu/mwoodley/Prototypes/

Bill Fietzer referred to a German prototype whose address I did not get. Mary Woodley noted that most of the prototypes mentioned on the Metamarda e-mail list prior to the January 16, 2000 meeting used only one metadata standard.

Matthew Beacom commented that in using the Z39.50 protocol the results obtained depend on how each user institution sets up its profile, e.g., what USMARC fields are defined as titles.


Dublin Core qualifiers:

According to Rebecca Guenther and guest Stuart Weibel discussion of qualifiers for the Dublin Core began at the 4th DC workshop in Canberra and continued through the October 1999 7th DC workshop in Frankfurt, Germany. There are several objectives for qualifiers:

  1. To refine the semantics of DC, e.g., to specify if the “Date” element represents a date of creation, date of coverage, date of latest modification, date when item was put on the web, etc.
  2. To provide for inclusion of information on encoding schemes used.
  3. To specify controlled vocabularies used.

Recent Dublin Core work has focussed on discussing proposals for qualifiers. Voting by the DC-Usage Committee is currently taking place on a set of official DC qualifiers (Voting will conclude by Feb. 11, 2000). Mary Woodley asked what happens if the DC-Usage Committee doesn’t approve a particular qualifier; is there an appeals process? Stuart Weibel said there isn’t a formal one but that one is informally built into the voting process. Voters have a choice of Yes, No, Abstain with provision for comments. Someone voting No or Abstaining on a particular qualifier can say why they are doing so, thus providing feedback on what the voters consider problem areas that probably should be re-considered.

Dublin Core qualifiers fall into two categories:

  • DC Interoperability Qualifiers (which would apply to all domains using the Dublin Core), and
  • DC Domain-Specific Qualifiers (which would apply only to specified domains using the Dublin Core).
If many domains adopt a particular qualifier that would be an argument for making it an interoperability Qualifier or even a new DC element. One possible new element might be Audience. Another might be Agent with qualifiers such as Creator, Publisher, Contributor. As an aside Stuart Weibel suggested that people not use the Source element; instead use Relation with an appropriate qualifier.

Development of qualifiers involves the development of registry software that will permit the registration of domain-specific software. According to Stuart Weibel no description of the registry is yet available; a Registry Working Group will be set up to deal with this topic.

Rebecca Guenther said that several different working groups were set up at the Frankfurt workshop, including a DC-Libraries Working Group. [To subscribe to dc-libraries: Go to: http://www.mailbase.ac.uk/lists/dc-libraries/join.html and fill out the form.]

This group will provide an avenue for input from librarians on such matters as what types of qualifiers are used in our legacy data, whether (and how) we can map the Dublin Core qualifiers to MARC and other metadata schemes such as TEI used in libraries, what qualifiers we won’t use, etc. Work remains to be done on identifying a possible controlled list of types that are at a more specific level than the one developed by the Dublin Core Type Working Group (which has top-level types corresponding roughly to material types).

Judy Ahronheim (U. of Michigan) asked about the 1:1 controversy. Stuart said that it asks “What are we describing?” “What is the granularity of the description?” The question should be decided by the organization creating the description.


Recommended changes to AACR and MARC21:

The Joint Steering Committee on AACR accepted some, but not all, of the recommendations of the CC:DA TF on ISBD(ER).

Rebecca Guenther said that MARBI will prepare a discussion paper for the 2000 ALA annual conference in Chicago. There are presently two Dublin Core elements/qualifiers that don’t have a home in MARC21: (others may be identified in the process of approving qualifiers).

  1. Date last modified (referring to the content being described, not the metadata; the 005 field in MARC21 gives the date and time information the metadata was last modified.)
  2. Relation.
According to Rebecca the Library of Congress modified the MARC-Dublin Core crosswalk before the Frankfurt Dublin Core meeting; it will have to be updated again after the Feb. 11, 2000 vote on qualifiers.


Final Report of CC:DA Task Force on Metadata:

Mary Larsgaard said that the first draft should be ready by March 1, 2000; the final draft by May 1, 2000. The content of the report will be based on the points of the charge, the summary report that Rebecca Guenther prepared in 1999 (http://www.ala.org/alcts/organization/ccs/ccda/tf-meta3.html), and the work of Mary Woodley and Steven Miller on prototypes. (see above)

The final report is scheduled to be submitted to CC:DA at the Chicago annual meeting in 2000 after which the Task Force will go out of existence. Its work may be carried on by the ALCTS Networked Resources and Metadata Committee.


The meeting was adjourned at 10:31 a.m.


Minutes prepared by Judith Hopkins