Excuse the corny link to Nelson Mandela’s autobiography but this morning it feels as if we are on a pretty long walk to the nirvana of open data and that we are going to need the patience and determination of Nelson to get there.
Last week my friend Giusseppe Sollazzo published a report “Open data in the health sector – Users, stories, products and recommendations“. Giusseppe suggested that I might be somewhat critical of his report
@StevenFeldman looking forward to your savaging of my health open data report… 🙂 (Or giving some feedback) https://t.co/K0wh7uxuU8
— Giuseppe Sollazzo (@puntofisso) February 23, 2017
Not at all, it’s an interesting read on the usability of open data in the health sector based on a series of fairly in depth interviews with users of the data – worth a read if you are an Open Data Geek. But …. Of course there is a but. Something was niggling away in the back of my mind, why was my skeptic dial turned up to super-cynic level? Then I realised that we are in 2017 not 2009 (when Gordon Brown made Open Data the default option for public sector data), 8 years on and the NHS commissions this report
NHS England have commissioned Open Health Care to gather evidence from the community of health open data users to help understand the needs of people using data outside NHS England. This report is intended to provide clear recommendations and evidence of user need to inform NHS England’s open data policy and programme development.
Hallo! Aren’t they a bit late? OK, be fair and accept that understanding user need is a continuous process and the NHS needs to update its Open Data strategy to take account of change.
The report makes two recommendations based upon the feedback from users:
Recommendation 1
Work towards a single authoritative online point of access for datasets in the health sector, and proceed with a programme of reducing duplication in data publishing. This online service should include both direct access to open data, and a clear process to access shared data.
Recommendation 2
Establish an expert and reachable data support team with a remit covering the entire health system, not limited to individual institutions. This team should be encouraged to build strong links with the wider data ecosystem and lead a health data user group.
I thought that data.gov.uk was meant to be the access point, it does list 4,594 results when you search for ‘health’ of which 4185 are published. I am sure there are explanations for multiple points of access but after nearly 8 years shouldn’t this have been resolved. Furthermore, duplication of data publishing suggests, inefficiency and potential for confusion re versioning and authority.
More worrying from my skeptical standpoint is the second recommendation to “establish an expert and reachable data support team”. We seem to have drifted a long way from the initial arguments made in favour of Open Data. Back in the day (St Gordon’s day 2009) there was a belief that all that was needed was to publish data that was created as a by product of the ‘public task’ of a government department or agency and that armchair auditors would be able to analyse that data and shine a light on inefficiency within government, citizens would be able to hold departments to account for their decisions and priorities and that billions of pounds of economic value would be generated by innovative new uses of Open Data.
“Open Data enables accountability; it improves outcomes and productivity in key services through informed comparison; it transforms social relationships – empowering individuals and communities; and it drives dynamic economic growth” Making Open Data Real, Cabinet Office, 2011
All of this was meant to happen without any significant cost to the public purse. After a few years it became apparent that some public funding was needed to prime the pump, e.g the Data Strategy Breakthrough Fund of £7.5m, no doubt other public funds have also been used to both meet technical costs of publishing data (it doesn’t publish itself) and running API services. There are also people within numerous departments whose roles are largely or wholly associated with the ongoing release and advocacy of Open Data, now this report is recommending a team within the NHS. The creation of a cross NHS team could provide an improved service at a lower cost if these roles already exist in different parts of the NHS but it does not negate the thought that publishing and supporting Open Data within the NHS (and by extrapolation across the whole of the public sector) is an ongoing cost that probably wasn’t anticipated when the Open Data initiative was launched.
In 2012 the NAO Cross-government Review: Implementing Transparency commented that
“Few departments are tracking benefits of transparency – a key requirement in monitoring success and learning what works.”
“The Government cannot maximise the net benefits of transparency without an evaluative framework for measuring the success and value for money of its transparency initiatives.”
“Different types of data release have different cost implications, but neither the Cabinet Office nor other departments routinely collect data to monitor the additional costs of transparency initiatives. The Government has assumed that the standard releases required of all departments incur zero or very low costs, because they are releasing information already held.” (my emphasis)
So, 8 years down the line it is recommended that the NHS should consolidate Open Data publishing via a singe point of access and provide a support team. Presumably this will cost money as it has done at Defra and other large departments. This may well be money well spent both in terms of the internal efficiencies generated and the innovative economic value created by Open Data entrepreneurs but, and in my opinion this is a massive but, surely we need evidence to support this expenditure?
Is it time for an update by the NAO? The economic environment is challenging to say the least, barely a day goes by without further examples of the funding challenge that the NHS is facing. The National Institute for Health and Care Excellence is tasked with evaluating the value for money of clinical treatments and medicines, perhaps we need to take a deep breath and evaluate the costs and benefits of Open Data?
8 years in it would be good to see the evidence of efficiency and economic benefit to offset the costs of Open Data release and support. I fear we may still have a long walk to Open Data nirvana.
One thought on “Long walk to open data”
Your review of eight years of the post ” Berners Lee met Gordon Brown and helped distract attention from MPs expenses scandal” Open Data era is well observed.
The hijacking of a serious competition law and innovation argument about Government Trading Funds by Open Data activists has diverted attention from the need to create transparent public-private mechanisms to support quality checked and supported public datasets which can be the real basis of wealth creation and new services.
As you rightly infer, eight years on we have little measurable progress, funded monopolies continuing to harvest profits which would be the envy of the private sector, and more noise than achievement.
Whatever happened to the idea of the National Information Infrastructure and a coherent approach to building our national data infrastructure?