Last modified: 27 February 2017
Cardiff is getting good at data. The ONS are in Newport, Cardiff University and Nesta collaborate on Y Lab, the city hosts the Welsh Parliament (whose needs for data insight is growing along with its powers), ODI Cardiff run by Sartori Labs has begun delivering projects, and, as part of a city region containing Bristol, it is home to many talented people and successful companies.
So it was great to visit for my first Open Data Camp. I think it was the first Open Data Camp that anyone from ODILeeds has attended, although Edafe always represents our city well.
ODCamp is a typical unconference. People pitch sessions, the organisers pick and order them by interest, and then conversations continue in separate rooms. Because of AirHack on the Friday and Saturday in Leeds, I arrived late on Saturday. I was just in time for a beer-fuelled summary of the first day from Kathryn and Esko.
On Sunday I stayed for the whole day and over both days I met in real life for the first time a number of people I only know from the internet.
My blog is my notepad. I share it because it’s easy to share and sharing often brings rewards. So here are my thoughts, as I think of them, and as written on the train back to Birmingham. You may need to add your own tact because I haven’t been at home since Wednesday and I’m too tired to find my own. Oh and just so you know, I run my own business (imactivate) that does a lot of work with open data. And I’m an associate at ODILeeds, where my business and other businesses in Leeds work together as ODILeeds to win and deliver bigger projects. So you’ll see a lot of strategically-placed “we” and “I”s.
We’re achieving a lot on data in Leeds. At events like these I’m always reminded of that. I’m also reminded that we’re doing it in quite unusual ways.
Most of what we do adds value. But some of what we do contributes to the mess of data portals and data standards that make other people’s lives harder. A lot of what we’ve achieved has been possible because we’ve tried a lot and failed quickly. We’ve relied on good people and trust above formal processes. We’ve not always followed the rules. A lot of people are wary of this approach. I understand.
So this year I will find more opportunities to contribute to the shared data infrastructure that others have built and that now makes much of what I do possible. We already do this. We contribute to Open Street Map and to Wikipedia. We make free tools that help people to access data held in standard formats. I build free and open tools that generate data in standard formats, and that convert between standard formats. I publish open data.
But I could do more. I think that The Netherlands’ bottom-up approach to setting standards works better than the UK’s top-down way of doing so. But to emulate that we must play a bigger role at ODILeeds in convening the meetings that set these standards. I hope to set a de facto standard for bin collections data in this way later this year. Cleverer people than me have designed it, I have proved that it works, but now I need to learn how to do the community part of documenting and promoting it.
I hope that our success at ODILeeds inspires others, but I note that we could not have achieved what we have without support from efficient, open, and entrepreneurial local governments, support from the high concentration of data companies that exists in Leeds, and a dollop of good luck.
It takes a lot of hard work to take advantage of good luck, but you need the luck all the same.
I think they’re increasing efficiency and improving services in cities across the UK. In 2017, I’m going to try and sell them to you. They will be much cheaper to buy and maintain because of open data, but they will not be free.
The UK led on open data early on. Our centralised structures made it easy to release more datasets back when national governments were the main publishers.
But national datasets are often of little interest to people. Usage has often been disappointing.
People receive most services on the local and regional government scale and there is rarely the capacity here to achieve the same success that the UK has achieved nationally. The UK is now falling behind less centralised places like France and The Netherlands when it comes to getting value from open data.
The temptation of the UK state is understandably to look at what worked nationally and emulate that locally. I think this would be a mistake. Centralising voter registration took power away from local government and delivered a poorer result. The funding that exists to make GDS a success nationally does not exist at local level, and will not be found. Local government needs a much leaner collection of processes than GDS has developed.
The desire to avoid postcode lotteries is frustratingly strong in the UK. Yet I remain convinced that allowing some places to succeed ahead of others, and then using national money (not free experts on a train up from London) to spread their success more widely will deliver the data-powered improvement in public services and public debate we want.
Much of the best work in open data is done by small companies like Epimorphics. They bring new skills and ways of working, they help government agencies to get the most from their data, and then they share the work that they’ve done and the experiences that they’ve gained. I’ve been very impressed by their work with Land Registry.
Epimorphics are one of many fantastic sponsors of ODCamp. They came to the event both to learn from others and to share what they’ve been doing. They were also trying to sell stuff, but they were much more interested in what was happening with data than in selling their services to more people. This is fantastic.
I am not sure why larger consultancies were not in Cardiff this weekend. They still sell a lot to governments of all levels and yet they were notably absent.
ODILeeds could not exist without its supporters and sponsors. Among these are KPMG, Arup, BJSS, and Bloom – all big companies who sell data products to governments. They are playing their part in supporting the UK’s open data ecosystem, and they tell us that they get good value from doing so.
I’d like to see more examples of this.
So maybe ODCamp 5 will be sponsored by Fujitsu, IBM, and BT. And maybe they’ll use the event to find partners that will help them win contracts and work with clients in the open. You never know, stranger things have happened.