Post: 9 December 2015
Open Knowledge’s Global Open Data Index for 2015 was released today. After two years at number one the United Kingdom was narrowly out-scored this year by Taiwan.
According to this year’s Index the top ten countries for release of key open data are: Taiwan (1), United Kingdom (2), Denmark (3), Colombia (4), Finland (5), Australia (6), Uruguay (7), United States (8), Netherlands (8), Norway (10), and France (10).
As part of a series of insight posts Open Knowledge asked me to contribute my point of view on the current state of UK open data. You can read my post on the Open Knowledge site, and there is a version below annotated with some links.
For a third year running the United Kingdom has come out at or near the top of the Global Open Data Index. Unlike many of the countries that did well in previous years, the UK’s overall standing has not been greatly affected by the addition of five new categories. This demonstrates the broad scope of the UK’s open data programme. Practitioners within UK government who work to develop and release open datasets have much of which to be proud.
However the UK’s role as an open data leader also carries the risk of overconfidence. Policymakers can easily be tempted to rest on their laurels. If we look in more detail at this year’s submissions we can find plenty of learning points and areas for further development. There are also some signs the UK open data agenda may be losing momentum.
The biggest gap this year is in election results data, with the Electoral Commission dataset disqualified because it only reports down to constituency level. The criteria have changed from previous years, so this decision may seem a little harsh. But globally most electoral fraud takes place at the polling station. The UK is a mature democracy and should set an example by publishing more granular data.
There is a similar weakness in UK public data on water quality, which is available only at a high level in annual reports from regulators. Environmental data in general has been a mixed bag in 2015. Ordnance Survey, which maps most of the UK, published the first detailed open map of the river network; and the environment department Defra announced an ambition to release 8,000 open datasets. However there is a noticeable absence of open bulk data for historical weather observations and air pollution measurements.
UK progress on open data is also held back by the status of land ownership data. Ownership records and land boundaries are maintained by Land Registry and other government agencies. But despite (or perhaps because of) the considerable public interest in understanding how property wealth is distributed in the UK, this invaluable data is accessible only on commercial terms.
In other categories we can see deteriorations in the quality of UK open data. National Archives is struggling to maintain its much-admired Legislation.gov.uk dataset. The latest version of Contracts Finder, an open search facility for public sector procurement contracts, no longer offers bulk downloads. Government digital strategy is turning steadily towards APIs and away from support for analytic re-use of public data.
Can the UK sustain its record of achievement in open data policy? Most of the central funding streams that supported open data release in recent years came to an end in 2015. A number of user engagement groups and key initiatives have either been wound up or left to drift. Urban and local open data hubs are thriving, but political devolution and lack of centralised collection are creating regional disparities in the availability of open data. Truly national datasets, those that help us understand the UK as a nation, are becoming harder to find.
UK open data policy may play well on the international stage, but at home there is still plenty of work for campaigners to do.
Image: Open Data stickers by Jonathan Gray (CC0 1.0)