Interoperability of Metadata for Thematic Research Collections: A Model Based on the Walt Whitman Archive

Topic:
Created by scholars in collaboration with librarians/archivists,
thematic research collections are directed primarily at other
scholars, though they are also used by students and by the
general public. Their development and preservation pose several
challenges: 1) they require very detailed encoding and metadata
to serve their demanding audience; 2) they often involve
materials which are dispersed; and, 3) critically, they
incorporate a variety of different types of materials with
different digitization and metadata requirements. Digital
thematic research collections often include texts, images, finding
aids to archival material, and administrative records. While
standards have been developed for each of these data and
metadata types (e.g., TEI, TIFF, EAD, and MODS), there has
not yet been a disciplined effort to integrate the standards and
explore where overlap and commonalities are, and to document
best practices for recording appropriate metadata with a
minimum of duplication.
These issues are crucial to creators of digital thematic research
collections because 1) the primary resources are held by the
archives and libraries, and many of the repositories are
digitizing materials and making them available to scholars,
even though the digital representations frequently are not as
richly represented as those which the scholars will ultimately
create; 2) archives and libraries will ultimately be responsible
for the collections that are created; and 3) if publishers are to
have a role in the publication of digital thematic research
collections, standards will be essential for portability, and these
standards will by and large come out of the library world.
That said, thematic collections are generally far more complex
and dependent on intricate metadata interrelations than digital
objects created for most digital libraries. This complexity is
due in part to the interpretive aspects of projects in which
scholars are deeply engaged with the materials. For example,
the poetry manuscripts section of the Whitman Archive
incorporates images, transcriptions with scholarly annotations,
copyright and administrative information, and finding aids. All
of these materials are encoded, presented through XSLT
stylesheets, and made available via various search options. The
Whitman Archive uses an array of data and metadata standards:
TIFF, TEI, EAD, and MODS. Because of its complexity, the
Whitman Archive is a good case study for investigating the use
and adequacy of Metadata Encoding and Transcription Standard
(METS) as a bundling tool for coordination of the standards
found in digital thematic research collections.
In 2005, the Institute of Museum and Library Services awarded
a two-year grant to the University of Nebraska-Lincoln to
explore interoperability of metadata for thematic research
collections and to demonstrate the power of METS as an
ingestion package. The current project builds on an earlier grant
project to create an integrated guide to dispersed manuscripts.
One hypothesis of the current project is that by standardizing
the way metadata is encoded, creators of digital thematic
research collections can make their work more sustainable and
ingestible; in other words, digital scholarship will be collected
in a form that will allow libraries and other entities to
incorporate into their holdings digital thematic research
collections created at diverse locations by a variety of scholars.
As the research team’s work has progressed, the group has
naturally experienced evolution in its thinking and approaches
to metadata. Members of the project team will report on project
processes and findings.
Organization:
Katherine L. Walter, Co-Director of the Center for Digital
Research in the Humanities (CDRH) and co-project
director for the interoperability of metadata project, will provide
a brief introduction to the topic, including a description of thematic research collections such as the Whitman Archive, the
varying types of metadata that may be involved, the impetus
to study integration of overlapping standards for such
collections, and the project goals originally set by the research
team. Walter will introduce speakers, moderate the panel, and
invite questions from the audience at the end.
Brett Barney, Senior Associate Editor and Project Manager of
the Walt Whitman Archive, will describe the team’s approach
to exploring what constitutes redundancy in metadata and its
exploration of questions such as what is really redundancy and
what is not? When is it desireable and when is it not? Does
redundancy stymie ingestion or aid it? His report will pose
conclusions regarding the value of redundancy and its
drawbacks in the ingestion of thematic research collections into
open-source library catalogs. Barney will also describe the
thinking of the team in selecting different kinds of representative
digital objects in the Whitman Archive for the ingestion
demonstration project.
Julia Flanders, Women Writers’ Project, Brown University,
will describe issues raised for scholars in contributing all or
part of a thematic research collection into library catalogs.
Questions explored will include what responsibility does an
acquiring library have to preserve or honor a scholar’s intentions
in creating and shaping such a collection? Is a library in fact
ingesting a collection or simply objects in a collection? This
will include thoughts on preserving contextual aspects of
thematic research collections, navigational features, and
interface. As a member of a standards board, she will speak
about the value of this kind of research for standards bodies.
Terence Catapano, Special Collections Analyst/Librarian,
Libraries Digital Program, Columbia University, will describe
the importance and adequacy of METS as a publication and
ingestion package, or in OAIS terms, how a Submission
Information Package (SIP) is derived for ingestion into a Digital
Library, and the development of a METS Profile for Thematic
Research Collections. Catapano will briefly describe work of
digital library projects that helped inform this study, such as
the DLF MODS Implementation Guidelines for Cultural
Heritage Materials, Descriptive Metadata Guidelines for RLG
Cultural Materials, DLF Aquifer, and others.
Daniel Pitti, Associate Director of the Institute for Advanced
Technology in the Humanities will discuss whether metadata
should exist in different forms at different stages; and describe
ingestion of the Whitman Archive into FEDORA, what this
means for other thematic research collections, and reflect on
the thoughts the FEDORA group at the University of Virginia
has with respect to preservation of such collections.