Making Public Data Public – are we asking the wrong questions?

Published : 18 March 2010

Within the space of the past 36 hours I happened to be present at three different events which illustrate the contradictions at the heart of the data and information debate.

A medical innovations conference where there was an enormous focus on licencing IP and securing the rights and protections which, many have always argued, provide security for the long term future of innovation funding. 

A meeting to look at the beta test site of the UK IPO’s new website which aims to guide public sector officials on how properly to identify, then licence and exploit their intangible assets (as urged by the Operational  Efficiency Programme).

Then finally by another engaging and lively debate at the LSE featuring the Sancho Panza of the Making Public Data Public debate, Professor Shadbolt himself.

It was this latter debate which has caused these reflections. A concern that we are asking the wrong questions, perhaps going in the wrong directions, and destined for continued muddle.

The medical innovations conference focussed my thoughts on Open Access and the Pharmas – and the age old question of how innovation is best fostered and paid for. The Novartis representative making a very good case – as you would expect – for the private sector patent protection model; the public sector making the (unfortunately weak) response that “something” better had to take its place, usually at tax payers’ expense. Shades of that twenty year old debate about Open Access, which has the publishers cornered but biting back with the challenge: show us a better self financing model for peer review?

Fast forward to the Making Public Data Public debate. Professor Shadbolt made the point, fresh from his visit to the US, that many essential public services – he quoted one small town where law enforcement had effectively been sub-contracted to private sector providers – may not be in the public sector in the future. Schools, hospitals may already be going the way of utilities.

And I thought – is this really about “public” data (i.e. owned directly by the public sector) or “essential” data, without which innovation cannot flourish.  After all, at the heart of Making Public Data Public initiative there are two essential core datasets – the Postcode Address File and the basic geo-spatial references within Ordnance Survey - which if they are not made available then, as St Paul said about Easter, “then is our preaching vain”. Yet one is in control of a commercially driven organisation ; the other in the protected limbo of a Trading Fund. Government may change these circumstances partially in coming weeks, but there is a wider context.

Should all publicly sponsored data should be free? And come to think of it, why stop there? Doesn’t democratic logic suggest publicly paid for software and research as well?  As it happens, the most perspicacious commentators on this issue (see Barr and Roper) and many others don’t see it as an issue of data being free because it is created at taxpayers’ expense. They recognise that data (public and private) should be “free at the point of use” if it is both widely used across the economy and is a “natural monopoly. Other public datasets might be charged for.

I finished my 36 hours pro bono public sector marathon  appropriately on St Patrick’s night and was left to ponder whether in fact we are asking the wrong questions:

Is it all public sector data, regardless of its core utility or “contestability”? or is it essential core data, whether in the public sector or not?

Is it a matter of its being free – or a pragmatic choice of who pays for it and where in the cycle?  Taxpayer, producer/registrant or user?

The Making Public Data Public advocates are on firmer intellectual ground when the cause is dressed in the language of democratic transparency. Seen from the viewpoint of stimulating innovation and paying for it, the viewpoint is more opaque.