IWMW blogs

Looking at Linked Data

“Linked data” is the buzz word of the hour, and apparently the government are using it on data.gov.uk, but what the hell is it and why would a university web manager need to know about it? There’s plenty of technical documentation already, but very little explanation for people who don’t read XML for fun.

Linked data is a way to make your data easy for other people to work with, and easy to integrate with other people’s data. Publishing some of your data in this way doesn’t just make things easier for people outside your organisation, but also inside. Some good examples are your (public) staff contact database, your university structure, projects & publications.

There’s a whole bunch of scary issues, such as data protection, licenses and so forth, but there is also some very low hanging fruit. For example, publishing the room & building of every academics office would be hugely controversial, but just publishing the list of buildings at your organisation, with name, number, campus, lat, long & primary inhabitants would be a hugely useful resource which would require very little effort to keep up to date. Let a student mash that up with their timetable and you’ve got an iPhone app that shows freshers when their next lecture is and points them in the right direction!

For good Linked Data, good design of URIs is vital. A URI usually looks like a URL, but often identifies concepts beyond the scope of web pages, such as people (http://id.ecs.soton.ac.uk/person/1248) or the concept of income tax (http://dbpedia.org/resource/Income_tax). In our session on Looking at Linked Data we’ll be covering a bit of best practice on URL/URI design which will help you be ready for the next years of the web. Sir Tim calls them “cool URIs”: http://www.w3.org/TR/cooluris/

Hopefully you’ll also learn the difference between URLs and URIs in a way which won’t give you a headache