This article first appeared on Computer Weekly on 12 May 2015.
Dan Hubert was driving into London’s West End to see a show.
He circled the roads around the theatre trying to find a parking space. The only one available was on a single yellow line. He looked for a sign showing whether the bay was residents’ parking only, and if not, what the time restrictions were. Finding nothing, he assumed he’d have to move on. Fortunately, he spotted a traffic warden who said he was, in fact, able to park there.
It was then that Hubert realised London must be filled with hundreds of parking spaces that go underused because people are confused about the rules for each street and are cautious about getting a ticket. Their confusion is understandable: there are 1,800 individual controlled parking zones across the capital, each governed under different rules by the 32 boroughs.
Hubert’s revelation was to create AppyParking, an app that can give a user all the information they need about available parking spaces anywhere in the capital. To build it, he contacted each borough to ask for data on their street parking restrictions. After considerable negotiation, he sourced the information he needed. But it was here the problems started.
First, he discovered that every borough records their information in different styles, (including hard-to-use formats such as images of maps in PDF documents). Undeterred, Hubert invested his own time and money building ‘The Parking Platform’ – a tool that could handle all the different data styles to feed into the app. To date, he and investors have put up £250,000 for the project.
Second, while some councils have improved the quality of their data in the two years since Hubert started his venture, others such as Kensington and Chelsea, Barnet, Havering, Lambeth and Barking and Dagenham, have said they will no longer share their data – in some cases because they may wish to develop their own parking apps. As he seeks to raise a further £1 million to extend AppyParking across the rest of the UK, Hubert wonders whether councils’ reluctance to share their records could derail his plan.
Alongside increasing transparency and improving public services, one of the government’s core stated objectives of open data has been to boost innovation by data-driven businesses. The experience of Dan Hubert points to three flaws in the approach of some local authorities that put that objective at risk.
No guaranteed supply of data. Few entrepreneurs or investors will be willing to risk their own time and capital to build products if the data they rely upon cannot be guaranteed. Yet it’s entirely understandable why no such guarantees are made. It’s because the UK’s ‘get it out and see what people do with it’ approach to open data is financially unsustainable. Releasing open data is not cost-free. It can require investment in new or updated software, potentially the creation and maintenance of a web portal or the development of APIs. Most importantly, it requires the resource time to annotate the data and to provide sufficient support for those who wish to use it. The fact HM Treasury may eventually receive extra tax revenues from data-fuelled businesses does not help the budgets of the organisations who provide the data in the first place. Given the current pressure on public sector finances, it is not implausible that some councils may cease providing open data altogether if they see it as nothing more than a cost centre.
Too small scale to be viable. Though a small number of local authorities make use of combined open data portals, such as data.gov.uk or the London Data Store, many more release open data separately and in non-comparable formats. This makes it incredibly hard for potential developers to create a viable business model from an open data product. Given that the typical price of an app is around 69p, and revenues from advertising are just a few pence per click, developers need a larger potential customer base than the residents of just one local authority. Equally, the case of AppyParking – where some local authorities wanted to create their own parking apps – shows how councils ignore the fact that citizens do not conveniently live out their lives in a single local authority area. Who wants to have 32 parking apps just to be able to drive around London?
Poor data quality. While some – like Hubert – can work with poor quality data, it raises the barriers to developers making use of data. Without further standardisation, there will be an artificial limit on the number of businesses that have the time and resources to build useful products.
The solution is surely for councils to start from the principle that they are the primary consumers of their own data. They should learn from New York City’s Mayor’s Office of Data Analytics (MODA) – a team that brings together and analyses public sector data from across the city to enable authorities to target their resources more effectively and efficiently. By using their own data to save money, councils would have a direct financial interest in ensuring the accuracy, timeliness and usability of their own information. As in New York, a subset of that data could then be released on an open data portal covering the whole city for use by businesses and citizens. The MODA model removes the disconnect between the payers and beneficiaries of open data, making it financially sustainable for the long term as well as improving the quality of the open data available to businesses. (The MODA model will be explored in detail in an upcoming Policy Exchange report.)
The UK has done much to deserve its reputation as a world leader on open data. Local authorities from Camden to Leeds are doing exemplary work to deliver results. Yet if they are serious about spurring lasting innovation, all of them need to ensure they meet the basic needs of the entrepreneurs who could help them. Putting in place their own sustainable business model for open data would be a good place to start.
Find me on Twitter