Archival description has relied on standards that brought consistency and structure to the management of historical records. Frameworks such as Encoded Archival Description (EAD) and Machine Readable Cataloging (MARC) enable archives to create finding aids and catalog records that support discovery. These standards served the profession well during the transition from paper-based catalogs to digital collections.
However, the evolution of the web and information systems has revealed limitations in traditional descriptive models. As researchers expect interconnected, searchable, and machine-readable information, archives face a significant transformation. The profession is moving from static text-based description to Linked Open Data (LOD), an approach that enables information to become part of a broader network of knowledge.
Limits of Traditional Description
Archival description was developed with physical collections and early digital environments in mind. EAD, for example, mirrors the hierarchical arrangement of archival materials by describing records from the fonds level down through series, files, and items. This structure reflects the archival principle of respect des fonds by preserving the relationships between records and their creators.
While this hierarchy remains valuable for understanding provenance and original order, it also presents challenges in a digital environment. Much of the descriptive information in finding aids is plain text. Names of people, organizations, places, and events appear as character strings that can be reliably interpreted only by human readers.
This limitation affects discovery across institutions. Consider a historical figure such as Abraham Lincoln. In a finding aid, his name appears simply as text. If multiple archives describe collections related to him, each institution may use slightly different forms of his name or apply different descriptive practices. Researchers searching across repositories must rely on text matching rather than identifying individuals. As a result, valuable connections between collections remain hidden.
The problem extends beyond personal names. Organizations, geographic locations, historical events, and subjects face similar challenges when represented only as text. Traditional metadata remains confined within individual systems, making it difficult for archival information to participate in the interconnected environment of the web.
From Text Strings to Identifiable Entities
Linked Open Data offers a different approach to describing archival resources. Instead of relying solely on text strings, LOD assigns Uniform Resource Identifiers (URIs) to entities such as people, places, organizations, and concepts. This represents a shift from describing names as text to identifying entities through persistent references.
The underlying technology that enables this approach is the Resource Description Framework (RDF). RDF expresses information as statements known as triples, each consisting of a subject, predicate, and object. For example, an archival description may identify a letter as the subject, connect it through the relationship “was created by,” and link it to the URI representing Abraham Lincoln as the object.
Because the URI identifies the individual rather than recording a name as text, any archives that uses the same identifier creates a connection between related collections. People and machines understand these links, allowing software to discover relationships across repositories.
New Descriptive Model
The adoption of Linked Open Data calls for a new way of thinking about description. Recognizing this need, the International Council on Archives developed Records in Contexts (RiC), a conceptual model designed to expand upon descriptive standards.
RiC views records, creators, functions, activities, and places as interconnected entities. This approach acknowledges that records exist within multiple contexts and that their significance cannot always be captured through a hierarchical structure.
The accompanying ontology, RiC O, provides a framework for expressing these relationships using Linked Open Data technologies. By representing archival information as a graph of interconnected entities, RiC offers greater flexibility while preserving archival principles such as provenance and context.
This model reflects the complex ways in which records are created, managed, and accessed in digital environments. It also enables archives to integrate with libraries, museums, and other cultural heritage organizations that adopt semantic web technologies.
Future of Discovery
The transition to Linked Open Data represents a significant undertaking. Legacy metadata requires extensive remediation, staff must develop new skills, and institutions need tools that support semantic data creation and management.
Despite these obstacles, the benefits are substantial. Linked Open Data enhances discoverability by connecting resources across institutions and disciplines. It improves interoperability between information systems, reduces duplication of descriptive effort, and creates opportunities for richer scholarship. Most importantly, it enables archival collections to participate in the global knowledge environment.
As archives preserve and describe documentary heritage, Linked Open Data offers a path toward more connected archival description. The transition from strings to things represents an evolution in how researchers understand archival knowledge in a contemporary environment.
0 Comments