Data.gov.uk explains ‘Public Data‘ as ‘the objective, factual, non-personal data on which public services run and are assessed, and on which policy decisions are based, or which is collected or generated in the course of public service delivery.’
The release of ‘public data’ will be regarded as one of the most significant government policy programmes of this century because of its economic, political and social disruption. Although it is still in its infancy, the UK’s ‘Public Data’ initiative, which was started by the Labour government, has survived the transition to the Coalition. The Coalition have consciously nurtured ‘Public Data’; if anything the Government has turned things up a notch.
Try to choke it in its sleep or hang it in full-view, in truth there would be no way that any government could get away with killing off ‘Public Data’ once it has taken root. The release of data into the public domain by governments is, after all, the latest in a long term trend toward more openness and transparency in public administration. It is one perfectly in tune with and enabled by the digital age.
But while it might intrinsically feel like one of the most significant things to happen to UK democracy, hang on. How do we know? Where’s the proof? Can we say with confidence that the release of data held by government is good? With questions like that flying around, the Oii’s ‘Internet, Politics, Policy 2010: An Impact Assessment‘ conference (held in September) would seem a good place to look for answers.
Unfortunately, despite data release initiatives happening here in the UK as well as the USA and a host of other developed nations, there was precious little discussion in the conference papers of why they were happening or what the impact was as a result.
One paper did catch my eye. ‘The Trouble with Transparency: A Critical View of Openness in e-Government‘ by Frank Bannister and Regina Connolly sounded exactly the sort of academic analysis that’s called for.
Bannister and Connolly state in their abstract:
Transparency in public administration is generally held to be desirable, something to be fostered and enabled. This long-standing idea has gained considerable further momentum with the emergence of e-government and the affordances of computing in general and the Internet in particular. This paper examines the argument that transparency may, in certain and not uncommon circumstances, be inimical to good government and good governance… It is (sic) suggests that in an electronic age, the scope and nature of transparency needs to be carefully managed and that expectations of the benefits of ICT enabled transparency may be overblown.
Yet, while Bannister and Connolly certainly have an excellent grasp of the literature on transparency, they are not very comfortable with the technology. There is some discussion of whiteboards, podcasts and something called ‘picanets’ but they have overlooked any of the public data initiatives, all of which share their genesis in digital media.
That said, this is still a paper worth reading.
Firstly, it provides two handy models for practitioners and observers alike.
One summarises the factors influencing transparency policy:
Another explains the categories of transparency and their relationships:
- Data transparency… This type of transparency is concerned with the facts and figures of government. Although expressed as ‘what?’, it may include ‘who?’, ‘when?’ and ‘where?’ as appropriate.
- Process transparency… This means making available information on the steps in various processes of government from policy formulation to issuing of a dog licence. The primary question is here is ‘how?’ though ‘where?’, ‘who?’ and ‘when?’ may also be relevant.
- Decision/policy transparency… This encompasses the requirement to explain the rationale for decisions and/or the actions and policies of government. The primary question here is ‘why?’, though other questions may also be of importance depending on the circumstances.
Bannister and Connolly explain that ‘in many practical situations, two or three of these will be applicable; process transparency will usually, though not necessarily, require data transparency and decision transparency will require, though again not necessarily, both process and data transparency.’ They provide a summary table of these interactions:
More importantly than these guides, is the contrarian position the paper takes. It reminds us that there are a number of competing futures resulting from the release of data by governments to promote transparency and openness; in this case, one where government and public servants clam up rather than open out (because of costs, systemic risks, the problem of conformity, problems of misinterpretation and misunderstanding, the blurring in the public mind of the roles of elected politicians and unelected public servants, and the rights, privacy and dignity of public servants).
A debate about the ‘good’ represented by the release of data held by governments needs to be had. It is a debate renewed by the latest official ‘public data’ release and hurried on by unofficial release, such as the recent Wikileaks activities.
That debate should be a public debate, as called for by Simon Jenkins in the Guardian. But it must be underpinned by more academic research and critiques. We need more papers on the theory – like this one from Bannister and Connolly – and also research papers presenting data on the data.
Ironically, for this to happen the academic world itself needs more openness. It is still too hard for the public to access high-quality academic work, which is why the efforts of the Oii to get itself and its peers ‘out there’ into the public domain are to be applauded.