David Clover 27 February 2011 10:37:25Following a recent (free) session in Birmingham about the Talis platform, Henryk and Jef are looking at the possibilities of exploiting 'Linked Data' using RDF SPARQL queries. The University has started to publish publicly accessible data in this form on its website at http://data.open.ac.uk as a SPARQL endpoint. The data sets offered by the University currently include information about courses, research publications and podcasts. From our point of view, we should be able to use these University data sets to enhance and add value to Faculty and Departmental websites which describe our people, their research and our courses and materials.
Linked Data is all about using the Web to connect related data that wasn't previously linked, or using the Web to lower the barriers to linking data currently linked using other methods. It relies on interrelated metadata schemas which can appear very abstract and initially hard to grasp. But it also means that when you have found one thing on the web, the interactions of metadata can quickly find context and provide additional value for any given query, automatically drawing on many sets of data. It can work well even if the other data has never been conceived as being related to the set you are working with. This sense of the 'loose coupling' of data is very familiar to us as it underlies the database model that provides the IBM Domino platform that we have been using extensively for our office and web productivity applications since 1997.
This is an exciting new area - the recently released website http://data.gov.uk/ is opening up a number of important data sets held by Government departments for general use.
Sometimes known as the 'Semantic Web' it's been in the academic domain for some time, but the techniques and concepts are starting to find their way into real life applications. A fine example of how Linked Data can be used to populate informational websites is provided by the BBC at its 'Wildlife Finder' website which is built on the same principles in collecting the data from a wide variety of unlinked sources across the World Wide Web with the results being assembled in real time 'on the glass' to satisfy the users query. Link Sailor also uses linked data to collect information from multiple sources.
Using some data from the Open University's published site and working at a very basic level, Henryk has already undertaken a Rapid Development activity which shows how we can use use a Domino XPage to re-present some data about Open University course-based Podcasts. As yet this is not a fully dynamic application drawing on all the Linked Data's features, but it could readily be extended once we understand more about the underlying metadata schemas and how to interrogate them.
- Comments