OCLC Research conducted an international linked data survey for implementers between 7 July and 15 August 2014. This is the second post in the series reporting the results.
The survey received responses describing 40 linked data projects/services that are in production and accessible to others.
American Antiquarian Society’s General Catalog, which includes the North American Imprints Program.
Archaeology Data Service’s STELLAR (Semantic Technologies Enhancing Links and Linked data for Archaeological Resources) project to enhance the discoverability, accessibility, impact and sustainability of ADS datasets.
British Library’s British National Bibliography, released to expose linked data in bulk, to explore feasibility of service delivery with semantic web technology, to develop our knowledge of semantic web technology and make our data available to new audiences.
The British Museum’s Semantic Web Collection, providing Sparql EndPoint point using the CIDOC CRM ontology for the Museum’s entire collection.
Colorado College Tutt Library’s TIGER catalog that connects to a Linked-Data back-end semantic server.
Colorado State University’s archived datasets from the NSF-funded Shortgrass Steppe-Long-Term Ecological Research station in northern Colorado, linking both the data and everything associated with the data- images, technical reports, theses and dissertations, articles, etc.-in its digital repository.
DANS’ (Data Archiving and Networked Services, Royal Netherlands Academy of Arts and Sciences)’ Cedar project which takes Dutch census data as its starting point to build a semantic data-web of historical information.
Europeana, which aggregates metadata for digital objects from museums, libraries, archives and audiovisual archives across Europe. It has developed the Europeana Data Model (EDM) which is based on the principles of the Semantic Web.
Fundacción Ignacio Larramendi’s Polymath Virtual Library, which brings together information, data, digital texts and websites about Spanish, Hispano-American, Brazilian and Portuguese polymaths from all times. It aggregates information about the thinking, philosophy, politics, science, etc. from Spain, Hispano-American, Portugal and Brazil written in any language (Latin, Arabic, Hebrew, Spanish, Portuguese …) and at any time (since Seneca in the first century BC to the present).
Library of Congress’ id.loc.gov, a Linked Data Service providing access to LC authority and vocabulary data for consumption by anyone using the data, and BIBFRAME, the technical website for the Bibliographic Framework Initiative hosting the vocabulary and tools to assist with evaluating the nascent vocabulary and model by viewing actual data encoded using the vocabulary.
Missoula Public Library’s Newspaper Index.
North Carolina State University’s Organization Name Linked Data, a tool to manage the variant forms of names for journal and e-resource publishers, providers, and vendors in its locally-developed electronic resource management system.
Norwegian University of Science and Technology’s special collections data, to provide a functional workflow and website for digitized manuscript production and delivery based on RDF and linked data, an index to various linked data projects, and personal name authorities from the BIBSYS consortium.
OCLC’s Dewey Decimal Classification in multiple languages and multiple editions as linked data; FAST (Faceted Application of Subject Terminology) Linked Data, for sharing the FAST authorities so they can be used elsewhere; ISNI (International Standard Name Identifier) providing identifiers, name, and links to sources; VIAF (Virtual International Authority File), which merges name authority files from around the world, primarily from national libraries; WorldCat.org, representing the world’s largest network of library content and services-making it available as linked data was a proof of concept to learn about data mining, semantic content publishing, data modeling and gap in technologies; WorldCat.org Works, high-level descriptions of the resources in WorldCat, containing information such as author, title, descriptions, subjects etc., common to all editions of the work, plus links to the record-level descriptions already shared in the experimental WorldCat Linked Data.
Oslo Public Library’s library catalog data converted from MARC to RDF linked data, enriched with information harvested from external sources and constructed with SPARQL update queries, to fuel their own services depending on more complex queries and data not included in MARC records, such as FRBR relations, book reviews, cover images etc.; collection of book reviews written by Norwegian libraries and linked to bibliographic data; and a web directory for pupils with links to selected webpages, described with metadata and connected to a topic structure linked to DBpedia resources.
Public Record Office, Victoria’s PROV wiki, a very early attempt to structure archival metadata using semantic mediawiki technology and to provide an alternate gateway into a small selection of the holdings that was also machine readable.
Research Libraries UK Linked Open Data, a transformation of 20 million bibliographic records from 34 UK libraries into linked open data format to encourage external, non-library sector use of bibliographic data; to profile their collections internationally; and to give impetus to libraries’ consuming linked data to create their own services and develop their staff skills.
Smithsonian Libraries’ Books Online, scanned books from their collection that are presented with RDFa bibliographic data to make their content more reusable and available to others.
The European Library, a large set of bibliographic data contributed by 48 European national libraries and more than 50 research libraries. It has enriched its data using VIAF, ISNI and GeoNames and implemented the results of aligning LC subject headings, RAMEAU and SWD from the Multilingual Access to Subjects (MACS) project, adding the URIs from these three vocabularies to the bibliographic data.
University College Dublin’s Digital Library, where all resources captured and managed within the UCD digital repository environment are exposed as linked data, their approach to managing metadata for digital collections, to provide maximum flexibility in resource discovery and dissemination.
University of Alberta, partner in the Pan-Canadian Documentary Heritage Network’s Out of the Trenches, a “proof-of-concept” to showcase a sampling of the network’s wealth of digital resources using linked open data and principles of the semantic web to maximize discovery by a broad user community.
University of North Texas’ UNT Name App, providing authority services for the UNT Digital Libraries’ ETD Collection and UNT Scholarly Works; supports linking between local authority and other systems including VIAF, id.loc.gov, wikipedia, ORCID and local faculty profiles which can be used in metadata creation.
Yale Center for British Art’s art collection data set, a Linked Open Data resource to build an environment for the development of British Art scholarship.
[Originally posted 2014-08-29, updated 2014-09-04]
Karen Smith-Yoshimura, senior program officer, topics related to creating and managing metadata with a focus on large research libraries and multilingual requirements. Karen retired from OCLC November 2020.