A short while ago I sent some questions about the recent Eurostat Hackday to Jonathan Gray (@jwyg). He ist the Community Manager of the Open Knowledge Foundation. “Please provide some short answers”, I wrote, “maybe 2 or 3 sentences”. That didn’t work out so well – Jonathan wrote back:” As the saying goes, ‘if I had more time I would have written a shorter letter’.” Nevertheless he’s painting a thoughtful picture about OpenData in the EU and its economical impact in his answers. So it would have been a waste to publish just a translated shorter version in the German OpenData Blog over at zeit.de.
„Useful and interesting things” could be done with statistical data of the EU – that’s been said at the Eurostat Hackday website. What kind of things?
Jonathan Gray: The Eurostat datasets contain an enormous amount of information about EU member states – from tonnes of fish produced to total length of railway lines. A lot of work has gone into harmonising and publishing the data, which is gathered from statistics departments all over Europe. However for the ordinary citizen it may not be fantastically easy to get stuck into the data to start to understand what it means. Luckily the data is basically ‘open’ in that anyone can reuse it for any purpose as long as they attribute the source. This enables other people to do interesting things with it.
The Eurostat Hackday is all about making things which give people interesting and meaningful ‘hooks’ into the Eurostat data, whether this is a simple website with information presented in ordinary language, graphics for print, or all singing all dancing interactive web services which visually represent the data or allow people to ask questions of it.
Much of the data contained in Eurostat is absolutely fundamental to understanding current EU policy-making. For example, one of the things some of us put the 2020 energy reduction targets into the context of past demand and supply. We want to make it easier for citizens, decision makers, NGOs, researchers and others to understand what these targets mean, how difficult they will be to achieve, and how whether we are currently on track. There were also people focusing on migration data, and lots of other things.
How come most OpenData Apps seems to be developed in Apps4-Contests and Hackdays – why is there no business going into this sector yet?
JG: On the one hand the idea that open data can create economic opportunities is not such a new thing. There are already plenty of businesses which are using and deriving value from open data. Many companies have been reusing public sector information (PSI) in information products and services for decades before the phrase ‘open data’ became popular. Indeed a lot of the initial interest in formulating explicit PSI reuse policies from the European Commission was strongly influenced by, e.g. geospatial data market activity in the US, which was partially the result of a policy which made certain Federal government information exempt from copyright. If you look at some of the rhetoric from the US in the 1990s, while some of the language is extremely dated (we don’t talk about the ‘information superhighway’ so much any more) some of it is astoundingly similar to the kinds of things that people are saying now. The European Commission is currently supporting evidence based studies to evaluate the impact of liberal PSI reuse policies (i.e. open data policies). So there is a lot going on in this area.
On the other hand this is still all very new, and things are changing very quickly. Only in the last 18 months have we started to see such strong, explicit and widespread commitments to opening up government information for all to use. New digital technologies (on the web, on mobile devices, and so on), enable us to do a lot more powerful things with open government data a lot more easily. Whereas PSI used to be the preserve of many more traditional information services (business intelligence companies, property information services, etc), there are now lots of new businesses, new revenue models, new markets, and very different kinds of opportunities. In that sense things are just getting started, and it may be a little while before we start to see open data policies having a really big impact.
It is also quite difficult to evaluate the impact of reusable information on businesses, in the same way that it was probably difficult to evaluate the economic impact of the internet a few decades ago. How do we assess the impact that roads have on our economy? I’m also keen to see a bit more of a rapprochement between economic and social arguments about open data, especially at EU level. In countries such as the UK it seems to have been the winning mix of economic and social incentives (transparency, accountability, betterpublic service delivery) that has put such wind into the sails of open government data initiatives at local, regional and national levels.
What is is the state of OpenData in the EU? What must be done?
Open government data initiatives are popping up all over Europe at the moment: from Helsinki to Zaragoza, Manchester to Paris. There are also lots more things happening behind the scenes, and initiatives pending announcement. Taking a step back, there is currently a lot more interest at grassroots level (web developers, civic society, media) than within government in several European member states. There is clearly more interest in certain member states than in others, but this interest is very gradually starting to spread and I’m fairly confident that in the next few years we’ll see at least some government information being opened up in most member states. There are currently problems with exclusivity arrangements in several EU member states, whereby some companies are given an effective monopoly on the information by the state. But this is something that is being addressed.
The European Commission is pretty on the ball in this area, not least due to a few really amazing civil servants who really understand the potential of open data. You may have seen the recent announcement that there is interest in building a pan-European open data portal. The Open Knowledge Foundation is also doing work in this direction as part of the EU funded LOD2 project. There’s a lot of potential for member states to exchange expertise and best practices in this relation to PSI. There’s also a lot of potential for technical expertise (and even bits of code) to be shared across member states, for example producing open-source applications which are intended for reuse across a number of different European cities.
One thing that will be really important not just in the EU but internationally is the development of a common legal and technical
understanding of what we mean by ‘open government data’. These should at a minimum mean that information is released under an open license (opendefinition.org), ideally in a machine readable format. The technical stuff is of course very important, and lots of people have different views about the best way to do this, but at the OKF we take the view that if datasets are not legally open for anyone to reuse, then one can’t really get started. We’ve had few conversations about this kind of thing with folks in the Whitehouse and in the Cabinet Office. It would be great to see stronger agreement about this internationally.
There are a lot of Hackdays recently – isn’t this frequency a little bit too much?
Indeed! For months we’d decided to schedule a small informal hackday for several people to get stuck into the data. It was basically an excuse to do something we’ve been wanting to do for ages anyway. The Eurostat data is so rich and there’s so much potential to build useful and interesting things with it. The event was not meant to be on the scale of, for example, the amazing international Open Data Hackathon that took place a few weeks ago. We made a few calls and sent a few emails out and the event has grown since then. We’ve had unexpected participation and support from numerous NGOs, projects, and from official bodies like the Greek Prime Minister’s Office, and from Eurostat itself, which has kindly agreed to set up a dedicated helpline for the day. But generally this is meant to be pretty small, and pretty focused. Also hopefully hackdays don’t always have to be big high profile things, they can also be a few people sitting around, having a go at making something.
Picture: Jonathan Gray (CC by:sa)