Oooo, data
by Christopher Gutteridge
On Wednesday I gave a well-recieved talk to the university ‘Digital Economy’ research group (a virutal group containing people from all over the university).
Yesterday I had the fun problem of lots of people getting in touch with ideas! For the next couple of months I still can’t put my full focus on the Open Data, but here’s some of the interesting things going on behind the scenes:
- Facilities / Equipment dataset to describe our cool toys. I’ve got people interesting in contributing to this from all over the university. You can see a preview here. The idea is to help the left hand know what resources the right hand has, and who’s allowed to use them. I’ve had provisional interest in this from medical imaging, the high voltage lab, the nano cleanrooms, archaeology, civil engineering and chemistry.
- Disabled Go reports – someone pointed me at this site which has detailed reports on disabled access for 98 of our buildngs. Most of the data is too detailed to map into RDF, but what I was hoping to do is (1) just provide a link to the reports for each building from our data and /building/ pages. That alone gets far more value out of it and maybe (2) pull out the headline data, eg “has disabled loo”, “allows guidedogs”. We’ve been in touch with them and it sounds like they are pretty postitive about the idea. I still need their permission to provide that information under OGL or another open license.
- Catering have updated all the menus to include coffee & other hot drinks (it was missing before), after noticing the the opendatamap didn’t have any results for searching for ‘coffee’ (the horror). Problem is, the menu says “Filter (Large)” now so still no match for coffee! We’ll either rename it to “Filter Coffee (Large)” or consider adding a “Hidden Labels” field to help searches.
I got asked what the success criteria for the Open Data project was. This is very difficult to define but for me it will be when the open-data-service is so much part of business-as-usual that people on longer want an enthusiastic hacker running it! I’m looking forward to talking about the good ‘ole days when open data was a new frontier and nobody even had an ontology for coffee types or bus timetables yet.
The Open Data is starting to get put to use to:
- People are using the bus times pages (I need to make the interface better, I know!)
- Our upcoming campus mobile phone app will use some of the location data
- I’ve been asked how the service could aid with student induction– eg. help people find what’s available, and where it is.
The other thing ticking along is getting live hookups to databases. Right now it’s all done with one-off dumps, we want to be showing the living data. The dump-and-email approach is fine for getting started but now it’s time to do the far less glamorous job of making the back-end more automated. I’m still working on getting energy use data per building, and I’ve a lead on recycling data!
Good times.
One final thing, you may notice that the Open Data Map is now not quite as pretty, there’s a good reason for this. We noticed that we may not own data traced using the Google Maps, so Colin has re-created all the data from the ordnance survey instead. There is slightly less detail, but the functionality is all still there.
The slides from my talk are available on EdShare. I’ve never uploaded to EdShare before — they’ve done a really great job at making a streamlined submit process. It’s far better than anything I’ve used in EPrints before, and I say this as the person who designed the EPrints 3.0 submit workflow!