Online Information Conference
Day 2 - Wednesday 3 December
TRACK 2: ORDER OUT OF CHAOS- CREATING STRUCTURE IN OUR INFORMATION UNIVERSE
Knowledge structuring in a semantic world
USING SEMANTIC WEB STANDARDS TO FACILITATE EFFECTIVE INFORMATION RE-USE
The government has much information to which people want access either to find and use or, more importantly, to re-use. Increasingly, the most useful services are those which combine data from different sources, mixing public, private and user created content. There are a number of options for enabling this kind of re-use using the web as platform to deliver data, including APIs for structured data using XML, JSON or YAML. With new Semantic Web standards such as RDFa, it is now possible to mark-up textual information inside documents, in effect turning a traditional website into an API. The government has been exploring the use of semantic markup inside XHTML documents in order to facilitate access, use and re-use of data. Apart from the data held in databases much of the interesting information that people want to re-use is semi-structured. Here there is commonality within certain types of information but also considerable variance. The creation and dissemination of such semi-structured information is widespread throughout the public sector. Examples include the description of job vacancies, government consultations and official notices such as those published in The London Gazette. To unlock the potential of these rich data sources we are applying semantic markup to the content where it is already published. Both the government and the web are inherently distributed so are natural partners. There is great appeal to obviating the need to create and manage a central database for each type of information, but provide for aggregation across the government's web estate. Using semantic markup delivers: a means of bringing order to allow re-use of information within documents; flexibility in structure, while ensuring consistency of content to be accessed; some element of future-proofing. The aim is to release the information so that third parties can be creative in developing or extending new services to their audiences. The government wants, respectively, the best potential candidates to see job vacancies, the best response to consultations and to maximise the value previously held captive in print in the London Gazette. The principle of serendipitous re-use means that the very best uses of government information will be all the things we cannot think of but that others will develop. Enabling such re-use through the Semantic Web will, we hope, deliver a measure of personalisation as online services become increasingly tailored. We are using simple and pragmatic approaches where possible, re-using what we can, such as established microformats with GRDDL transformations, inventing where we must, with RDFa and GRDDL. The paper will report on these three uses of Semantic Web standards and draw out lessons learned from their implementation and consequent benefits.
Speaker
