Files in this item



application/pdfurban_richard.pdf (612kB)
(no description provided)PDF


Title:Principle paradigms: revisiting the Dublin Core 1:1 Principle
Author(s):Urban, Richard J.
Director of Research:Renear, Allen H.
Doctoral Committee Chair(s):Twidale, Michael B.
Doctoral Committee Member(s):Renear, Allen H.; Palmer, Carole L.; Furner, Jonathan
Department / Program:Library & Information Science
Discipline:Library & Information Science
Degree Granting Institution:University of Illinois at Urbana-Champaign
Knowledge organization
Knowledge representation
Digital libraries
Dublin Core
Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH)
Aggregations of digital collections
IMLS Digital Collections and Content
Cultural heritage
Semantic Web
Abstract:The Dublin Core 1:1 Principle asserts that "related but conceptually different entities, for example a painting and a digital image of the painting, are described by separate metadata records" (Woodley et al., 2005). While this seems to be a simple requirement, studies of metadata quality have found that cultural heritage metadata frequently does not conform to the Principle. Instead, representations commonly appear to make statements about multiple related resources, such as a painting and a digital surrogate that depicts the painting. Although these "violations" of the Principle are assumed to reduce metadata quality, they are widespread in cultural heritage metadata, and metadata creators indicate "a great deal of confusion" about what the Principle means and what constitutes a violation (Park & Childress, 2009). A conceptual analysis of the 1:1 Principle reveals that it is the product of an encounter between two di fferent paradigms, with distinct approaches to how descriptions function, that have dominated the development of Dublin Core. The knowledge organization (KO) paradigm draws from a century of practice developing bibliographic representations and rules for description in libraries, archives, and museums (LAMs). This paradigm is primarily concerned with the organization and classi cation of document surrogates. It does not provide a formal account of how bibliographic languages describe and reference the resources they represent. In contrast, the knowledge representation (KR) paradigm is influenced by recent computer science, linguistics, and philosophy. This paradigm is primarily concerned with supporting automatic inference and data integration, mobilizing formal semantic theories to provide descriptions with grammatical structures that can be computationally modeled. It provides an explicit and formal, although not untroubled, account of reference and description. Further analysis of how discussion of the 1:1 Principle has been shaped by these two different approaches to description shows that 1:1 Principle problems are as much about the design and function of representation languages as they are about errors made by metadata creators. The Principle is most directly derived from the formal theories of description and reference that use proper names (or identifiers like URIs) to refer directly to resources. Bibliographic records with fixed syntaxes, but informal, colloquial semantics, can successfully communicate descriptions that are meaningful to human interpreters and enable syntactic interoperability between systems. However, when viewed through the lens of formal semantics, bibliographic representations may appear incoherent|as failing to unambiguously reference any resource, or provide a single, shared semantic interpretation of descriptions. These problems are exacerbated by the need to make subtle ontological choices about descriptions of resources involved in equivalent, derivative, or descriptive relationships. Because knowledge organization representations rely on relatively informal semantics, often not more than a sentence or two of natural-language prose, the heuristics for identifying 1:1 Principle violations depend heavily on the implicit and informal, shared conceptual framework of cultural heritage professionals. A formal interpretation of these heuristics, therefore, requires more than operational definitions of the Principle; it requires highly expressive ontologies that make that understanding logically explicit. Moreover, even formalizing traditional ontological distinctions will fail to identify violations when contingent historical facts play a role in interpretations of metadata records. While these difficulties make fully reliable methods for automatic detection of 1:1 Principle violations impossible in principle, they usefully reveal obstacles that will require attention as the cultural heritage community moves towards adopting more formal representation practices that are the basis of the Semantic Web and Linked Data. Recognition of the fundamental differences between the knowledge organization paradigm and the knowledge representation paradigm can lay the groundwork for reconceptualizing how we represent related resources.
Issue Date:2012-05-22
Rights Information:This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License. To view a copy of this license, visit or send a letter to Creative Commons, 444 Castro Street, Suite 900, Mountain View, California, 94041, USA.
Date Available in IDEALS:2012-05-22
Date Deposited:2012-05

This item appears in the following Collection(s)

Item Statistics