Session Proposal


Worlds Collide: A Repository Based on Technical and
Archival Collaboration

Erin O'Meara
Electronic Records Archivist, University of North Carolina at Chapel Hill

Gregory Jansen
Repository Developer, University of North Carolina at Chapel Hill

The failure of many institutional repositories (IR) to acquire large sets of faculty
publications has shown that the traditional IR model is not sustainable without a shift in
academic publishing. The Carolina Digital Repository (CDR) aims to be more than a
traditional IR and instead of focusing primarily on open access publishing, it will acquire,
preserve and make accessible a range of at-risk scholarly output, such as datasets,
faculty papers, university records and other faculty research projects.

Research libraries have begun to focus on their special collections as core to their
mission and a key contribution to the wider world of scholarship. We extend this
emphasis on collecting unique materials to include the new IR as a strategic direction.

Repository implementations suffer from a disconnect between technologists and digital
curators in terminology and approach. This barrier has inhibited the development of
comprehensive solutions that meet technical needs and deeply incorporate curatorial
requirements. Overcoming this barrier demands work to establish common language
and mutual understanding of concepts from each discipline.

The key output of academic institutions, besides the conferring of degrees, is the
development of new knowledge. Through preservation a repository supplies a key
missing ingredient in the life cycle of knowledge production, a persistent point of access
that fills the growing gulf between modern modes of digital scholarship and traditional
publishing. The new IR model addresses the institution's role in providing stewardship of
these assets.

The Carolina Digital Repository (CDR) is an institution-wide initiative at University of
North Carolina at Chapel Hill (UNC-CH), spearheaded by University Libraries. It is a
preservation framework for material in electronic formats produced by members of the
UNC-CH community. Following standards developed in the OAIS reference model, the
CDR employs Fedora for data content models and uses iRODS as a data store.

In this paper we will explain how the repository works as a whole to meet these
challenges. We will show how our repository was made possible through a sustained
commitment of institutional resources and the deep involvement of motivated experts
with archival, librarianship and software development backgrounds.

We will trace how the functional needs for preservation activities were brought into the
project and satisfied without sacrificing flexible access and ingest work flow. We will
explain how we blended and extended the models set forth in Open Archival Information
System (OAIS), Fedora Commons and the Integrated Rule-Oriented Data System
(iRODS) to meet these requirements without creating a muddle of concerns in code and
subsystems. Lastly, we will explore the stable measures taken and the ongoing project
to create data management and object integrity safeguards used to mitigate risk of data
corruption and loss.

A key recommendation from the CLIR report for Reconceiving Research libraries for the
21st century is:



Collaboration should under-gird all strategic developments fo the university,
especially at the service function level. Greater collaboration among
librarians, information technology specialists, and faculty on research project
design and execution should be strongly supported. (CLIR 2008, 10)

The CDR has been a collaboration of four partners, University Libraries, Information
Technology Services, the School of Information and Library Science and the Data
Intensive Cyber Environments (DICE) research group. From conception to deployment,
on committees, the project team and working groups, the initiative incorporated
expertise from all of these disciplines.

The CDR's design is guided by archival theory that includes principles of authenticity and
trust. OAIS provided a conceptual model that informed the design of system layers and
components. Certification tools, such as Trustworthy Repository Audit Checklist (TRAC),
supplied functional requirements. The resulting implementation is a distinct best-of-
breed software architecture aimed at teasing apart the concerns of ingest work flow,
durable access and data preservation. These worlds met as we adopted and adapted
numerous preservation standards, including OAIS, METS, MODS and PREMIS.

The CDR is set to go live in April 2010. We will have several pilot collections populating
the repository that span disciplines and collecting efforts. Our goals for the first year of
deployment include an enhanced user interface with greater search and access control,
enhanced metadata and curation of objects, and audit of current preservation activities
within the repository.

References

Council on Library and Information Resources. No Brief Candle: Reconceiving Research
Libraries for the 21st Century. August 2008.


	Worlds Collide: A Repository Based on Technical and Archival Collaboration
	References