Between 2006 and 2009 the National Library developed four search service prototypes to support our investigation of improvements to discovery services. These prototypes have now been overtaken by the release of Trove in December 2009, but they are described below as an historical record.
- The NBD Prototype
- The Beta Catalogue
- The Single Business Discovery Service Prototype
- The Australian Newspapers Beta Search System
1. The NBD Prototype
The NBD Prototype is a simple demonstration of searching MARC bibliographic records using Lucene for storage and indexing. The database being searched is a copy of the Australian National Bibliographic Database (NBD). The same data (more up to date) is also accessible in a public form through Trove.
The prototype was established by Kent Fitch in January 2006 to test how a simple Google-like search could deliver effective results when searching a large bibliographic database. It explores concepts like relevance ranking, clustering, spell-checking based on corpus content, the augmentation of indexes with decoded data, the deployment of authority files, FRBR and ways of extending the corpus with information from Amazon.com, Google Books, etc.
A full description of the prototype and its underlying technology is given on the NBD Prototype page.
2. The Beta Catalogue
The "Beta Catalogue" was a prototype (released in February 2008), but it has now become simply the National Library catalogue, a fully operational offering. It is a re-engineering of the National Library's former catalogue, which used Ex Libris Voyager. The new catalogue uses the VuFind open source toolkit developed by VillaNova University in Pennsylvania. It supports features such as relevance ranking, faceted refinement of search results, links to related information in web services such as Amazon and Wikipedia. In implementing the new catalogue, the Library developed a feature integrating it with the Ex Libris Voyager electronic call slip service.
3. The Single Business Discovery Service Prototype
This prototype has now developed into Trove. It was first released in May 2009, and was replaced by Trove in December 2009.
A description of the Single Business Discovery Project can be found in this paper, which was prepared for the October 2008 meeting of the Libraries Australia Advisory Committee.
In December 2008 the Project Team undertook a "card sorting exercise" with a small sample of end users, with the aim of developing a better understanding of the collection groups and terminology that would be meaningful to users. The results of that exercise are found in this report.
4. The Australian Newspapers Beta Search System
The Newspaper Digitisation Program is described here. As part of this Program, the Library built a prototype Newspaper Search and Delivery Service in November 2007, which was released to State and Territory Libraries for testing. This was then developed into a Newspapers Beta service which was released to the public on 25 July 2008. Since late 2009 that service has no longer been a beta service. From about April 2010 it will exist only as a Collection View within Trove.