Open data – forward strategy
A few people from central government have asked to talk to me recently about open data as they look at their strategy beyond legislation. I thought I would write down repeated themes to save on more meetings. The conversation is normally about how data can be used for the public good as opposed to, say commercial re-use and generally geeing people up to reuse open data that one bit of government sweated blood to get another bit of government to publish.
For open data to work it needs to be relevant to people’s needs. The focus in the UK hitherto, in line with the Berner’s Lee thesis has been on getting data out – publishing gigabytes of the stuff. This is rational and as it were pump-priming, egg-chickening or market-making. But it also plays to politician and civil servants’ comfort zones. They mainly work with Whitehall types and data experts and fairly learned salaried people with 9-5 time to spare to come to long tedious meetings usually in central London. If you look at reuse of public data as a market this is the supply side. There’s still plenty of work to do but the supply side of data now has plenty of people focussing on it in Whitehall, if not in local or national governments.
People’s needs, the demand side needs much more work to get a public social economy of data working, by which I mean, helping regular folk find and use public data to achieve social outcomes – say a campaign on local development, for a new school or stopping arson. This is often well outside the Whitehall comfort zone. If you are a civil servant or a government minister you often don’t fight scrappy grass roots campaigns that are the life blood of civic action in the UK. Attempts to stimulate the demand side hitherto have been the ubiquitous ‘hack day’. These are great in a new sector that hasn’t played with open data before but in the main pander too much to the geeky. And are overwhelmingly in London.
What we see very little of by central govenrnment is outreach to campaigning groups or campaigns around the country with almost no geek credentials to help understand that data exists that can help them and how to use it. Obviously most of these campaigns are oppositional in nature, reflecting the design of the British system that loves to set up a fight rather than a consensus. This makes it hard for neutral civil servants and Ministers to engage and the phrase ‘I’m from the Cabinet Office, I’m here to help’ is never entirely plausible. But there are ways to skin this cat.
Here’s some random observations:
- in general proponents of open data who are serious about a proper social economy in the stuff need to commission some market research into the user audience and meet their needs.
- pick a major policy area where there will be huge amounts of citizen activism in the next few years saturate it with public data and tell people you have done so. The obvious examples are planning and alcohol/entertainment licensing. I’ve fought dozens of planning campaigns – the citizen case is invariably weakened because the developer (buys data) and local authority (collects data) have hegemony over the data you need to make your case. Correcting this power imbalance by putting data into the hands of the public as they make their neighbourhood plans and fight the all important test cases could be a massive boost to the big society. It would be great to see the police publish open data on crime stats for individual pubs – at the moment the information commissioner’s office, which exists in a sort of alternate reality at times is apparently blocking this
- communicate – I have met very few grass roots activists who know enough about open data, the ease with which it can now be obtained etc. There’s a huge communications job to do, but no reason why, in areas where the government wants to see more activism such as planning that this can’t be done.
- clarity is all – the open data sector is bedevilled with geeky jargon and tech mumbo jumbo. The semantic web in particular confuses people and it is hard to prove that it actually helps on the ground, so move semantic stuff that from front and centre to the rear
- focus hard on intermediaries who can build consumer facing applications that last, not flash in the pan hack day lash ups that go nowhere. MySociety as one of the former for instance.
- most/many people who are literate enough to run a campaign can use a spreadsheet. The numbers of them who can use an API ‘tends to zero’ as the statisticians would say. So present data in spreadsheet-szed chunks as well as behind an API.
- have a look at the dull but worthy Knight Foundation/Monitor community information toolkit – it is turgid as presented, but you can see something in the framework. [update bullet added 24/02]
Someone called Randi Ryterman has just popped up in my twitter stream retweeted by one of the god fathers of UK open data Andrew Stott CBE
‘Technology is the easy part. Citizens need to be mobilised. Governments need to listen.’
Which sums it up.
Disclaimer and context stuff – as a Cabinet Office civil servant I commissioned and helped drive through a thing called the Power of Information Review in 2006-7 which sort of kicked off the modern central government drive for open data. I am no longer a civil servant but am on two advisory panels on open data for the current government – on Local government Data and Crime/Justice where I rail on about the need to focus on users, not data suppliers. I worked with NESTA on their ‘Make It Local‘ programme that did some stuff with local data.