Southampton Open Data Blog

More RDF

March 11, 2011
by Christopher Gutteridge

I’ve improved the back-end tools which provide RDF when you request a .rdf or .ttl file. By default the system just gives an facts which have the current resourece as the start of the fact. This sort of sucks as when looking at a building it’ll tell you BUILDING-X  is within SITE-Y and BUILDING-X is called “The building of advanced science stuff”. What it won’t do is give any that go backwards, eg. if I know ROOM-Z is within BUILDING-X, it won’t mention that by default.

So I’ve mae a way to make it relatively easy to add this information. I can also tell it to follow several hops to find all the useful information. The art is going to be, for each class of item in our system, working out the balance between utilitity and brevity. The very simple rule of thumb is to get all the information you need to display an HTML page about that thing.

And that example leads to another point, I really need to give an example of every type of data item under the hood. This site is all iceberg-like right now. Only I know for sure what lurks in the SPARQL… I’ll get to it, I promise.

Friends, Romans, Countrymen…

…send me your data. But maybe don’t hurry, as I’ve got a back-log already! Yesterday I got an email about some data we have a legal obligation to publish… am I the right person? I guess I am! But it’s not my only responsibility and I had to put all my other work on the back-burner to get this site up, so things will now move slower but always forward. Maybe a little sideways.

What I won’t except are things which we have no hope in hell of keeping up to date, so we only really want data which is already someones job to maintain. I’ve made a couple of exceptions, most notable that the building position and footprint data is created by volunteers — however this data moves very slowly. We’ll learn what works as we go, this has never been done before!

Research data is a world of complicated and awesome all by itself. We’ll never add it to this site. It will want a very different form of collection and curation. If there’s research data that you want to publish right now, and it’s not crazy big, I recommend you put it in – this will give it metadata, a license and a permenent home on the university web.

Categories: Uncategorized.

Leave a Reply

Your email address will not be published. Required fields are marked *