Summary and background

Shifting the library community from the use of complex strings as authority values to URIs with traditional components asserted as properties will align library practice with the semantic web community. This work package covers three aspects of external identifier linkage: (1) incorporation of external identifier lookup and local identifier creation into the cloud-based editing environment of WP1; (2) adaptations in cataloging policy and practice; and (3) an in-depth exploration of how Wikidata could serve as a platform for publishing, linking and enriching library linked data.

Incorporation of external identifier lookup and local identifier creation into the cloud-based editing environment of WP1

Building on the authorities work done in LD4L-Labs, the cloud-based editing environment (WP1) will include lookups to external authorities and creation of local authorities. These services will provide reliable and performant lookups for a number of authority sources. In addition to being integrated into the cloud-based editing environment, these services will be made available to the community for other integrations.

The linkage to external authorities will be supported by the developers and UX designers at Cornell, the Iowa team, and the Wikipedian in residence.

Adaptations in cataloging policy and practice

Harvard technical services staff will provide leadership and engagement in working with PCC to:

  1. Develop identifier creation and management workflows that provide clarity for the library metadata community regarding data input standards that result in effective look-up and matching.
  2. Contribute to tool development roadmaps by identifying and prioritizing technical features that would enable efficient and effective metadata production workflows. Workflows would include but not be limited to the creation of Wikidata and ISNI entities.
  3. Promote and guide adoption of expanded and diverse vocabularies and identifier registries that have been (or can be) released as effective linked data, such as FAST, the Getty vocabularies, the Library of Congress's faceted vocabularies, ISNI, VIAF, and GeoNames. Work will include developing metadata interoperability guidelines for tool development and data input guidelines for metadata practitioners.

Harvard will also develop metadata partnerships with non-library organizations who have a role in the information production ecosystem to identify and pursue creation of high-value metadata pipelines for automating linked data production through creator-contributed metadata. Examples could include a North American university thesis/dissertation metadata pipeline and a university press author-metadata pipeline.

Exploration of how Wikidata could serve as a platform for publishing, linking and enriching library linked data.

Cornell and Iowa developers and technical services staff at Cornell and Harvard will collaborate with the project’s Wikipedian-in-Residence in to understand how best library data can be added to Wikidata and how Wikidata authority information can be leveraged in the library environment. Cornell and Iowa will develop reconciliation services that can be integrated with cataloging workflows or run as separate semi-automated batch processes to improve existing descriptions through the addition of links to related entities. (Stanford may leverage this reconciliation work in WP 2: Reuse of Existing Metadata as part of its conversion pipeline.) It is this reconciliation and linking that will provide the ability to leverage external context for improved discovery. UX design and refinement of the authority lookup interface will be a significant component of this work.

  • No labels