BML Iteration One
The first proof of concept for the project started in September of 2016 by extending BIBCAT. Sample records from participating institutions included Denver Public Library, University of Wyoming, the Marmot Library consortium, Colorado College, the Colorado State Publication Library, History Colorado Museum, and the State Library of Wyoming.
Based on these sample records, we took the approach of creating default RDF rules that then can be added to with custom RDF rules for specific jobs or workflows.
An lightweight web front-end that provides a ResourceSync feed that links to individual records that are in the DPLA Profile version 4 JSON format
All of the samples were ingested into a single triplestore with varying ingestion times.
Using Docker and Docker-compose speed up the development and deployment time for the application
Problems and inconsistencies in the source metadata can be ameliorate with specific RDF Rules but more work is needed by the source institutions to improve the quality of their provided metadata.
Some of the institutions do not have a publicly exposed URLs for their digital objects, which is a requirement for DP.LA inclusion. This is still an open issue.