10 Do’s and Don’ts of Opening Up Public Data

Free our data!

The UK and many other Governments are taking the first steps towards freeing up the mass of data held on our behalf.   This is a crucial period in history for democracy because if handled right, the movement could usher in a new era of of mass participation and re-engagment in politics.  It is also a watershed moment for public services and social entrepreneurs.  What economic and social value can be generated with all this data?  How can we make more innovative products and services for less?

Science & Innovation 2010 Ordnance Survey Seminar

It is fair to say that from the the inside of a large public organisation freeing up data voluntarily is a lot more progressive that waiting for the the Freedom of Information Act requests.  Probably less time-intensive too.  However data won’t leap up and create any value by itself any more than a pile of discarded parts outside a factory will assemble themselves into a car.   The secret of successful open innovation using public data then is more likely to be with working people to solve some specific problem.  Simply releasing the data is not enough.   The overarching question that emerged at the seminar ‘Underpinning Innovation with Geography’ which I chaired last week was, in effect, what do organisations expect people do do with all this data?  It is a salutary fact that  only a tiny minority of the 272,677 datasets available on Data.gov in the US have been used in creating apps or web sites.  Why is this and how can data publishers help developers create value?

Do’s and Don’ts

Ordnance Survey are extremely interested in these questions – it’s the second year of their Geovation programme which supports open innovation using geographic data.  Other public organisations are opening up too  – Warwickshire County Council and the Governments’ own National Archive presented their own leading edge initiatives at the seminar.With an audience largely composed of policy makers, the Q&A session visited the practical, legal and moral issues that arise.  You can find copies of the speaker presentations here.  Here are the resulting themes presented as do’s and don’ts:


  1. Set out to create commercial or social value with data
  2. Make sure data quality is high
  3. Promote innovation using government data sets. Transparency is only a means to an end
  4. Enhance communication with the public
  5. Make sure your co-creators are incentivised
  6. Get organised, create a community around an issue
  7. Pass on learnings to other similar organisations (local authorities)
  8. Invent new business models
  9. Promote innovation using government data sets. Transparency is only a means to an end
  10. Be brave – people may do things with the data that you don’t like


  1. Just release data and expect people to understand or create with it.  Publication is not the same as communication
  2. Wait for FOI requests, put the data out first informally
  3. Avoid challenges to current income streams
  4. Go straight for the finished article, use rapid prototyping
  5. Be put off by the tensions between confidentiality, data protection and publishing
  6. Wait for the big budget or formal process but start big things with small amounts now
  7. Be technology led, be business led instead
  8. Expect the community to entirely self-manage
  9. Restrict open data to the IT literate – create interdisciplinary partnerships
  10. Get caught in the false dichotomy that is commercial vs. social

As ever, please let us know what you think.  There may be more than 10 of each!


Post a comment

You must be logged in to post a comment.