On Friday 25th October 2013 I attended a workshop on Making Data Discoverable put on by Data.ac.uk who work on linked open data development for academia. The session was held at the very shiny offices of the Open Data Institute (ODI) in the City in London.
There were few of us participants and several presenters from University of Southampton (see University of Southampton Data Service) so we had a very hands-on session working on live data with people who wrote much of the software and protocols we were using.
Rather than starting with how to make data available, we started looking at how we could simply access linked data: we learnt the basics of the Resource Description Framework (RDF) and how it can be represented in Turtle – Terse RDF Triple Language. Then we dove in to an example where we wrote just a few lines of PHP to follow links in data about a publication to get where it was presented and what other publications were presented at the same event. We used the Graphite PHP library which our presenter had written. After getting my head around what information was being pulled in when by the commands and drawing myself a diagram I found it quick to understand.
We had lunch and plenty of coffee and then spent the afternoon looking at how we might present what open data the university offers in a machine readable form through Uniform Resource Identifiers (URN) and the new Organisation Profile Document (OPD) proposed standard for which they have an Organisation Profile Document (OPD) Checker. We ended with some demonstrations of how linked data about equipment , facilities, buildings, people etc. can be visualised and presented.
If there’s one thing I want to share about open access and open data is that others can quickly put together useful resources from structured data – often answering questions the original researchers had never considered.