SPIKE: BIBFRAME Data Import RDF

EBSCO’s contract with the Library of Congress (LC) requires supporting the ability to import BIBFRAME data into the library’s data graph (vs. transforming data that originates from a MARC record).

The primary use case for this functionality is when LC receives metadata from a vendor / supplier. Currently, LC receives metadata as MARC records. With this feature, LC would be able to receive metadata natively in BIBFRAME for building out its data graph.

The purpose of this card is to investigate technical aspects and considerations for receiving and processing BIBFRAME data as a new input in order to provide a t-shirt size estimate for completing the feature outright, and to consider what can be delivered in the 2025 calendar year.

The Data import feature is related to the Hub entity feature, insofar as importing BIBFRAME metadata for Hubs (via the Network Development department at the Library of Congress) represents a first, concrete step toward supporting data import of BIBFRAME RDF metadata more broadly.

Assumptions

• Assume incoming BIBFRAME resources describe new works and not updating existing ones.

• For the Thin Thread, only a subset of such components will be required (e.g. focus on the most common BIBFRAME components for the Thin Thread; then build out support for the standard)

• Incoming BIBFRAME resources will be self-contained, e.g. inclusive of components for both Work and for Instance as well as accounting for additional fields like Administrative Metadata.

• Incoming BIBFRAME resource descriptions will have a range of specificity → so that we will have to account for use cases where values for some components in the Work or Instance may be empty

Additionally

• We will need to understand requirements around the assignment of identifiers. Currently, identifiers for Work (id.loc.gov/resources/works/######) and for Instance (id.loc.gov/resources/instances/######) leverage the identifier assigned through LC’s Voyager system.

  • What identifier will LC want to apply for newly created Works and Instances?

  • How will that identifier be applied for incoming BIBFRAME resource descriptions?

• There will be overlap between this card, and dev required for processing Hub entities created by LC’s Network Development (NetDev) team.

  • Where possible, it would be advantageous to leverage the work completed for Hub entities for the data import Thin Thread

• Although the scope of the Thin Thread is limited to new works, we may still need to account for how resources will be updated