Unifying Fragmented Catalogues: A Prototype Framework for Library Discovery Using SLiMS, VuFind, and Python via OAI-PMH

The paper describes a model designed to overcome the fragmentation of bibliographic metadata among the different information systems kept by libraries. The bibliographic records were first obtained from the open-source integrated library system SLIMS (Senayan Library Management System) via the Z39.50 and Search/Retrieve via URL (SRU) interfaces. SLIMS was registered as a data provider with the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH), and its metadata was transferred to VuFind in order to create a single discovery interface consisting of 18,395 records. The collection was also expanded by including 38,113 bibliographic records in MARC 21 (Machine-Readable Cataloging) format that were obtained from a number of web-based repositories, thus increasing the amount of material available for discovery of the collection. The metadata was harvested from the OAI-PMH server using the Sickle OAI-PMH client within the Python environment of the Jupyter Notebook. The model was built using only open source components on a Linux platform. The findings indicate that it is possible to develop a single search interface based on a traditional library catalog which will allow for federated search and faceted browsing. The innovation of the system lies in the combination of protocol-driven harvesting and programmatic metadata processing, this enabling bibliographic records and item-level metadata to be stored in a single, structured, machine-readable format. Therefore, the model supports the objectives of open catalogs and open science, while also providing a metadata foundation for possible intelligent indexing and retrieval.