Session 2B | Achieving Unified Search across Digital Repository Platforms
"Texas A&M University Libraries have begun going live with production exhibits in the new open-source Solr AGgregation Engine (SAGE). SAGE has been in development since 2019 and consists in two complimentary feature sets: (1) The aggregation of multiple Solr indices into a target index with arbitrary fields, and (2) Curator-configurable views of any Solr index with custom filters, facets, and display fields.
The ubiquity of Solr indices in library applications like VuFind, Blacklight/Spotlight, DSpace, Fedora, and others, juxtaposed with the tantalizing prospect of one search interface across the myriad of library holdings, led readily to the concepts behind SAGE. Once the development team implemented the aggregation functionality, it proved straightforward enough to make a configurable display of the fields of the bespoke Solr documents SAGE was writing. Soon, the development team was demonstrating rough views of synthesized collections of DSpace, Fedora, and Spotlight documents.
However, when it came time to prepare these views for curatorial management and public display, numerous issues arose for the product owners. Among other things, the views posed problems with formatting and normalization of metadata, uniquely identifying objects, and providing viewers for content like images, PDFs, and A/V. Resolving these issues to the satisfaction of product owners has yielded the first production hybrid collection in SAGE, the Apfelbaum collection of World War I Postcards. In this presentation, we will describe the means whereby this content from DSpace and Fedora repositories was brought together in a harmonious view."