The thing that really differentiates the Public Service Launchpad from other accelerator programmes is that teams are required to be building things that have a social purpose. As local government officers, we already have lots of that – it goes with the territory – but as with the product development there was a process of focusing that we went through to allow the specific social purpose of our project to emerge.

Partly this mirrors some of the steps we took in product development and is about understanding what is not happening that should be. And partly it is about taking a deliberately different view of something you already know about to challenge the way it is being done now. Both of these things happened to us – largely unconsciously – as we worked away on our product development. When we looked at existing data products, for example, it was immediately clear that applications for people without high levels of IT literacy were quite thin on the ground, and that the open data movement had a focus on creating and releasing data sets without necessarily worrying too much what they were being used for.

Just as traditional business development had a “build it and they will come” mentality and traditional software development has a top-down “waterfall” ethos, so our initial approaches to open data seemed very traditional. We seemed to want Government to just release its data and surely some clever people out there would do great things with it.

About half way through the programme my co-founder Lucy was involved with a hackathon that the council provided some data sets to. A lot of effort had gone into persuading the data custodian to open the data, it was good quality and – we thought – valuable. But the developers weren’t interested in it. Similarly, I don’t know the statistics but I feel sure that too few of the data sets on data.gov.uk are used for public benefit (and how are we tracking this anyway?).

There are two problems with this: firstly that there is a lot of waste (as people open data sets that never get used), and secondly that it skews the use cases of open data in favour of business (as only data sets that can be monetised get used, and only then by people who already have money). We aren’t against open data at all on any level: we just think that we aren’t making the best of it.

In a nutshell, we think that something more like a design approach (understanding what communities need, opening the data in a way that they can use it, testing this with them and iterating for continuous improvement) would result in more relevant data sets being opened for higher levels of benefit all round.

Around week 9 of the accelerator programme we started the Plain Data Campaign and, with the help of some of the programme mentors, began to think through what this meant. Crucially we needed to test it against what was already happening and ensure it was complementary to the great work that is already being done by organisations like the Open Data Institute – and lots more still needs to be done.

We’re now convinced that more effort needs to be made to work with the people we are trying to benefit to understand whether opening particular datasets is worth the trouble. We think that open datasets need to have purpose and that purpose ought to be regularly tested – otherwise the stories told by open data will be determined by those with the skills or funding – and we think that is dangerous. We think this is crucial if everyone is to benefit from the potential of data science and not just those with the brains or the money.

On our campaign blog you will find some further explanations and a sign-up form. Please sign up and help us shape the campaign for better, not just more, public data.

Next: how does a local gov team get accelerated

Advertisements