Evolving an in‐house system to integrate the management of digital collections
| Published date | 01 April 2006 |
| Pages | 241-260 |
| DOI | https://doi.org/10.1108/07378830610669619 |
| Date | 01 April 2006 |
| Author | Marilyn Lutz,Curtis Meadow |
| Subject Matter | Information & knowledge management,Library & information science |
CONTENT MANAGEMENT SYSTEMS,
PART 2
Evolving an in-house system to
integrate the management of
digital collections
Marilyn Lutz
University of Maine Library, Orono, Maine, USA, and
Curtis Meadow
Trefoil Corporation, Orono, Maine, USA and
University of Maine, Orono, Maine, USA
Abstract
Purpose – To describe the evolution of a content management system at the University of Maine
Library that would function as a universal, extensible metadata repository, thereby eliminating the
need to build separate databases for new digital collections, and facilitating both end-user access and
the management of electronic resources in an integrated technology environment.
Design/methodology/approach – Beginning with the development of a prototype system that
mapped EAD encoded finding aids to a relational database, this paper discusses the evolution of this
prototype into the design and implementation of a RDBMS (and continuing development of an
object-oriented database management systems (OODBMS) system) to actively manage digital objects
and associated metadata. The key to the system design is metadata: extracting and mapping,
transforming, and managing the processing of MARC-based metadata into non-MARC schemes to
build digital collections. Other relevant CMS architecture issues discussed are the design of a
functional bibliographic structure and utilities for metadata harvesting and indexing.
Findings – Provides information on the use of the Dublin Core Abstract Model and a flexible and
adaptable collection-centric approach in the overall CMS architecture as implemented on a non-MARC
RDBMS, and provides an explanation of the advantages of an object oriented database system over the
complexity of evolving relational database tables.
Practical implications – A useful source for the development of an in-house CMS, and a
contribution to the growing body of literature about the transformation of MARC-based metadata for
database design.
Originality/value – This paper is a case study of actual work conducted at the University of Maine
Library. The RDBMS manages digital collections; the OODBMS manages digital video and other
multimedia resources.
Keywords Contentmanagement, Informationsystems, Databasemanagement systems,Digital storage,
Digital libraries
Paper type Research paper
Library environment
As early as 1991,library staff began creatingdigital resources, althoughat the beginning,
they did not adopt a strategic plan for the digitization of collections. While choosing a
collection for digitization was not entirely serendipitous, it depended largely on the
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0737-8831.htm
Evolving an
in-house system
241
Received September 2005
Revised November 2005
Accepted November 2005
Library Hi Tech
Vol. 24 No. 2, 2006
pp. 241-260
qEmerald Group Publishing Limited
0737-8831
DOI 10.1108/07378830610669619
uniquenessof the resources and theavailability of fundingfrom grants. Staff did adhereto
and adoptstandards or “de facto” standards(where none existed) for eachproject to insure
long-term accessibility and the usefulness of resources. Over thecourse of a decade, this
“project approach” to building digital collections resulted in the proliferation of many
separate databases, based primarily on varying formats and metadata requirements.
Simultaneously, an information technology infrastructure evolved to support access to
and distribution of digital content to the universitycommunity of users.
With initial success in creating digital resources, the library staff’s traditional
collection development and management responsibilities shifted to a planned approach
in creating digital content, in providing access to digital content the library neither
owned nor managed, in preserving and archiving content not accessible to users, and
in serving as an access portal. This shift in digital collection development occurred as
standards developed and new technologies evolved to support the integration of vast
amounts of disparate information. User demands for “one-stop shopping” to meet
information needs pushed the envelope. To accommodate this change, the library’s
information technology environment required a new architecture in which digital
objects and metadata make sense within the context of individual projects, but also
serve as multi-purpose “building blocks” for reuse in the context of other services.
Overthe course of a decade, the library’sdigital collections grew,and so did the need for
a solutionto the “many databases” problem(Figure 1). At the Universityof Maine Library
Figure 1.
Fogler Library digital
collections 1991-2004: “the
many databases” problem
LHT
24,2
242
To continue reading
Request your trial