ARLIS/NA
30th / VRA 20th
Session
2
Common
Ground: Standards for Cataloging Images and Objects
Moderators:
Linda McRae,
University of South Florida
Lynda White, University of Virginia
Speakers:
Murtha
Baca, Head, Getty Research Institute, “Enhancing End-User Access On-Line Art
Historical Resources”
Elizabeth O’Keefe, Pierpont Morgan Library, “Sharing the Wealth: Controlled
Vocabularies for Libraries, Visual Resource Collections, and Museum”
Sherman Clarke, New York University Libraries, “Are You Content with Your Data
Content?”
Thornton Staples, University of Virginia, “A Hierarchical Metadata System for
Image Collections”
Recorders:
Maria Oldal, Pierpont
Morgan Library
V. Heidi Hass, Pierpont Morgan Library
Murtha
Baca stressed that the same kinds of tools that have been used for decades to
create “traditional” bibliographic records can be used to catalog visual
materials. She talked about the importance of the “five Cs” —
content, curation, cataloging, controlled vocabularies, and copyright — in the
creation of high-quality art and architecture imaging projects; her presentation
focused on issues relating to cataloging and controlled vocabularies. She
pointed out the importance of selecting or devising an appropriate metadata
schema or “container” for the material in hand.
Next, that schema must be populated with values from a “menu” of
appropriate controlled vocabularies and classification systems including, but
not limited to, the Art & Architecture
Thesaurus (AAT), Union List of Artist
Names (ULAN), Thesaurus of Geographic
Names (TGN), Thesaurus for Graphic
Materials (TGM), and ICONCLASS. There
are different solutions for taking advantage of the power of variant names and
broader terms to enhance end-user access to visual materials. Baca gave a brief
survey of some online tools and resources for cataloging art objects and their
visual surrogates: Categories for the
Description of Works of Art (CDWA), the Getty vocabularies on the Web, the
Library of Congress’ Thesaurus for
Graphic Materials, and ICONCLASS. She
stressed the importance of receiving training if these tools are to be used
effectively. She described two
forthcoming publications, Introduction to
Art Image Access (Getty Publications, Fall 2002), and the visual resources
cataloging guidelines (Cataloging Cultural
Objects) that are currently being developed by a VRA editorial team with an
advisory committee composed of members from the library, museum, and archival
communities. The reason why art
information professionals take the time and trouble to apply descriptive
metadata to images is to enable users, both expert and non-expert, to find what
might otherwise elude them.
Elizabeth
O’Keefe began by saying there was a substantial overlap among the indexing
terms required by catalogers of art books, art objects, and images of art
objects. Yet there is a great divide between the vocabularies used by book
catalogers and those used by non-book catalogers. Book catalogers are committed
to Library of Congress Subject Headings (LCSH). Image catalogers are unlikely to
embrace LCSH for many reasons; for them, AAT is the logical choice for object
terms. Names are an entirely different proposition. In the library world the
standard currently used to establish names is the cataloging code AACR2. Art
Name Authority Component (NACO), coordinated by Sherman Clarke, was established
in 1993 to enable art catalogers to contribute names to the Library of Congress
Name Authority File (LCNAF). LCNAF would be equally valuable to catalogers of
art objects and visual materials, if they adopted AACR2 as their standard for
formulating names. This decision need not carry with it an obligation to use
AACR2 as a guide to choosing main entry. The use of a single set of rules for
name formation ensures consistency and predictability within the whole universe
of names occurring within a database. The other argument in favor of using AACR2
is that there is nothing analogous to it within the object and image-cataloging
world. The Union List of Artist Names, the Grove Dictionary of Art, and the
Thesaurus of Geographic Names are extremely valuable as reference sources, but
they do not provide for the wide range of names covered by AACR2, and none of
them attempts to be an authority file. Using AACR2 for names would also enable
object and image catalogers to benefit from the research of others. Using AACR2
for names would bring the cataloging guide currently under development by the
VRA into line with library cataloging guidelines for the formulation of names.
This would save the drafters of the code a great deal of time and enable our
end-users, as well as our cataloging staff, to move more easily from one
database to another. This might also some day lead to a VR NACO, which would
funnel the names used in cataloging visual resource collections into LCNAF.
Sherman
Clarke began by describing the International Federation of Library Associations
and Institution’s (IFLA) 1998 Functional Requirements for Bibliographic
Records (FRBR) and their original focus on bibliographic records. He defined
metadata (description of the item) and data content (terminology that is used
rather than the fields and records where data is entered). He continued with a
detailed description of the areas FRBR tackles, including why bibliographic
records are constructed and the three groups of entities represented by or
included in records. Clarke then discussed the divergence between records for
books and for visual resources. He analyzed the approach of library catalogers
and the tools they use to construct records, and their unsuitability for the
description of visual materials. The VIA database at Harvard, which has records
for group, work, and image was described as a successful approach to the problem
of identifying works and their different manifestations. He concluded by
reminding us that our focus should be our users, and their need to find,
identify, select, and use what they are looking for.
The
final speaker of the panel was Thornton Staples. The General Descriptive
Modeling Scheme (GDMS) is a project at the University of Virginia to create a
formal information structure that can be used to construct descriptive models of
real-world or imaginary phenomena to create contexts for collections of digital
resources. The underlying data structure is provided by an XML DTD, which allows
the model to be as hierarchical or as flat, as is appropriate, and provides ways
to cross reference data within or among models. Staples started out with giving
a brief overview of the history of the project and then went on to present the
basic element set of the scheme. The presentation was supported with elaborate
graphics showing the structure and examples of the scheme. More can be read
about GDMS on the GDMS Research and Development Site at http://www.lib.virginia.edu/dlbackstage/resndev/gdms.htm.