Project Blacklight
The Prototype Project
Blacklight is UVa’s prototype Lucene/Solr based faceted discovery tool, first shown at the 2007 code4lib meeting . The name comes from Solr + UVA = Blacklight. This tool is designed to allow users to more easily filter content and build queries that are highly focused and reusable, resulting in more accurate results with less effort and frustration. It takes advantage of the rich metadata in the bibliographic records and structured XML objects. Constraints based on browse facets can be used to narrow results, and results are ranked by relevance as they are displayed. You can explore the experiment at: http://blacklight.betech.virginia.edu/. Blacklight has so far indexed ~3.7M MARC records, a ~500 text object subset from our Digital Collections Repository, and 320 Tang Dynasty Chinese poems from the Chinese Text Initiative. The MARC records are indexing through a ruby script directly from MARC, without transforming into marcxml first. The Digital Collections texts and Tang poetry are indexed from TEI xml. Blacklight was developed by Erik Hatcher, with help from Library staff Bess Sadler, Bethany Nowviskie, Erin Stalberg, and Chris Hoebeke. It is based on a tool called Collex that was developed for the NINES project based at UVa. BlacklightDL
We additionally have a BlacklightDL experiment for our Digital Collections Repository, which indexes over 10,000 TEI texts, ~23,000 GDMS images, and ~4,000 EAD finding aids. The development of this prototype experiment was undertaken by OpenSource Connections and Library staff Bess Sadler and Matt Mitchell. As we consider moving these experiments into production, the two projects will come together we plan for a single discovery resource that supports searching and browsing across many types of content while also providing deep functionality for specific content types, such as images, full text, and music. SpecificationsSolr is running through Jetty, with apache doing load balancing out front. Sessions get written to a mysql database, with capistrano for deployment and versioning. |
