Behind the scenes at OurOntario.ca
The interface allows for keyword search, advanced search, and browsing of media types (audio, image, text, video, object and collection) and by the name of the contributing organizations. Search results can be limited to or refined by media types, contributing organizations, locations, collections, and by item types. Another option for narrowing a search is to conduct another search limited to the current set of results.
We continue to explore the ways in which we can assist the user to make sense of their search results and to provide as much information as possible in the context of a particular search. This will be an interesting challenge as the portal continues to grow in content and the user is presented with an ever-increasing number of search results.
Technologies and Platform
The portal is being developed and funded by Knowledge Ontario as a province-wide initiative. It is based on open source platform that includes Apache Tomcat, Apache Cocoon, Apache Solr and Apache Lucene.
How It Works
A meta-search layer to the user interface allows distributed digital collections from a range of institutions to be searched from the Our Ontario index. For managed collections, we obtain the metadata records held on the participating agency’s website. The metadata records are ingested, normalized and indexed on our servers. Searches are conducted on our local copy of the metadata provided by the contributing partner, and produces faceted search results. The output passes though XSL and CSS forms which offers control over presentation.
We harvest metadata from standard and non-standard content – the searching options reflect what archives, museums and libraries have in common in their metadata to provide a Dublin Core basic record. In a specific context we can leverage a much richer discovery experience, such as providing browse, search and faceted results to scoped sets such as government documents and newspapers.