The EBSCO Discovery Service is EBSCO’s own next-generation resource discovery system, built on the already-very-familiar EBSCOhost database platform.
EBSCO’s particular ‘angle‘ for EDS is that its content is built up out of a lot of high-quality, ‘scholarly’, subject-indexed content (similar to the individual bibliographic databases on EBSCOhost), which they are keen to push as superior to basic ‘Google-type’ keyword-indexed searching, where the quality-assured, ‘information literacy’ aspect to resource discovery may not be as strong.
(Enough scare quotes for ya?)
- Information about EDS can be found on EBSCO’s website at: http://www.ebscohost.com/discovery
- Online demo/presentation about EDS
- RSS feed of press releases/information about EDS
- EBSCO support site for EDS
- See also the ‘Discovery‘ pages on on the SCONUL Higher Education Library Technology (HELibTech) wiki
Features of EDS:
- Highly customisable/’brandable’ – logos, colours, background images, text/field labels;
- Uses the same administrative interface (for back-end configuration) as EBSCOhost;
- Integrates with EBSCO Electronic Journals A-to-Z and LinkSource (i.e. Find it @ Lincoln) for access to full text via OpenURL;
- Harvests MARC records from local catalogue, and repository etc. records (via OAI-PMH, presumably, although I forgot to ask);
- Content: as well as the library’s own local collections (above), EDS searches a central EBSCO ‘base index’ of content/metadata from ~20,000 providers, plus content from those EBSCOhost databases to which the library subscribes; it also contains a lot of enhanced book metadata (cover images, subject headings, reviews, etc.). See EBSCO’s website.
- It’s possible to set up a public, ‘guest’ version of EDS to search catalogue, repository, and the main EBSCO index – then allow your own users to log in and search the more complete content including subscription databases (though EBSCO suggest that few libraries actually provide guest search in practice, despite asking for it to be made possible!); it’s also possible to use EDS to create custom search interfaces for groups of packages/databases (or even for individual databases) – e.g. subject clusters;
- Users can extend their search out to remote databases (i.e. those not included in EBSCO’s central base index + local databases) via a traditional metasearch facility (related: EBSCOhost Integrated Search);
- It’s possible to limit the default search to full-text items only (making use of the coverage information held in the A-to-Z/LinkSource knowledgebase) – however EBSCO advise that most subscribing libraries don’t do this – instead starting their users off with searches of the complete EDS collection, then later on allowing users to narrow the search results down to full-text-only, if they want to;
- Various APIs, HTML widgets, and other extension tools available through an ‘EBSCOhost Integration Toolkit’ (http://support.ebscohost.com/eit/) – N.B. some of these can also be used with the existing EBSCOhost databases;
- Developer community of library people extending and customising EDS – example blog posts here and here;
- While the advanced search options and user interface are highly configurable, there’s no facility to adjust the search ranking algorithms – i.e. the relative placing of items/collections against each other in search results (as is possible in e.g. Ex Libris Primo);
- FRBRising of search results will be introduced in 2012;
- EBSCO will offer libraries free trial access to EDS, including MARC record harvest where possible.
UK HE libraries using EDS include:
- University of Liverpool
- Heriot Watt University, University of Edinburgh, University of St Andrews, University of Stirling (through the Scottish Digital Library Consortium)
