Exploring Open Data at the Gurteen Knowledge Cafe

Gurteen Knowledge Cafés do not come along that often and when they do it is often at short notice and I sometimes find that I have already booked something for that evening, even though I usually book my theatre at fairly short notice too. When I can get to one then I do because I always find them stimulating, which is what happens when you get a room full of bright enthusiastic people together and hit them with a conversation stick.

The theme of this event was Transparency and Open Data, a hot topic indeed. Making it hotter was having it hosted by the Department for Business, Innovation and Skills (BIS) who were at the forefront of opening up government (our?) data for wider use.

As usual the evening started off with some casual networking and then some more formal speed-networking before we sat down to hear the speakers kick-off the discussions.

The usual caveat; what follows is a mixture of what various people said and what I wrote at the time modified by my further thoughts when writing this up. None of the comments are attributable to anybody and all the errors are mine.

The website Data.gov.uk lists all of the government data sets. There were 2,4000 of them! Data was being opened to improve transparency, enable internal improvement within the government and to boost the economy. A value of £1.8bn had been put on this but I remain very sceptical on that.

A big issue with sharing data is understanding how it is structured and what the values actually mean. For example, is a postcode one item of data or two and does a gender value of "0" mean male, female or unknown? There are data standards in some areas to let organisations co-operate, e.g. banks on sending payments, but there are few of these examples and they cover quite narrow and very specific fields.

Another big issue is the accuracy of the data. I have not had proper data on the 371 bus times for months with my two nearest stops not included in the feed.

This issues of what the data means and how accurate it is brings us back again to that all important topic of Information Literacy. Open Data is only useful if it is understood properly and is accurate.

Open Data has many passionate advocates and great claims are made for it but the success stories are few and relatively minor. Things that have worked reasonably well include maps, transport routes and timetables, and data from the financial markets.

The government's approach is to throw its data out there in the expectation that somebody will want to use it but this is a shotgun approach and most of the data misses. A better approach might be to let potential users approach government bodies and ask for the data that they want to use.

For example, I would be very interested in a simple way of accessing planning data across local authorities (within just one authority would do for a start!). It is incredibly hard to find out simple information like how old a building is and how tall it is.

Personal data is another important issue and this has two sides to it, e.g. the doctor and the patient where one needs information about the other in order to help them.

The various table discussions and the large circle discussion that closed the evening where interesting and stimulating and I was an active participant in both (no surprises there then). It felt a little as if we were going around some old circles at time and while the conversations were fun I am not sure that I got much new out of them.

At some point in the evening I wrote "Knowledge is Beautiful" and that seems like a good note to end on.