Killing Open Data Softly?

As usual there has been a little pause between blog posts, and it has taken a bout of the flu and utterly dull daytime TV to prompt this one.

Over the past few months I have starting thinking about what might be making doing Open Data a more difficult job. My first concern is linked data (stands back and prepares to be shot down in flames). Linked data is a great idea, don’t get me wrong, but it is really quite hard to do (even after a 30 minute whirlwind tour in Costa with Rob Styles from Talis). So many authorities are struggling to get to grips with the concept of open data (or get the concept but just won’t do it) this adds a layer of complexity and potential cost that may put some off. Let authorities learn to crawl, toddle, then walk, don’t go asking them to run a marathon yet!

I am also sensing the possibility that the wikileaks saga may put the concept of open data in a bad light. Those of us who think open data is vitally important will accept the Anonymous response for what it is,  those who haven’t yet seen the light  may see the whole thing as just an inconvenience and would rather buy their presents through PayPal than hold our governments to account.

Finally, (and I really will get some critisism for this!) what about the people who are clamouring for our data? We have been getting some great results from our Hack Warwickshire competition, but not that many. Someone wrote a really good post recently, which I can’t track down at the moment, asking people to start showing what is being done with open data. It’s really hard to demonstrate to data owners of an authority why they should open up data if we can’t show good examples of how it might be used. I’m particularly interested in the correlation of data, proving why all kinds of data should be released – you never know what you might uncover!

My final thought: reluctance to share stems from early childhood. We’ve all sat there in classes shielding our work so other kids can’t see what we’re doing. Although back then it was because we didn’t want them to see our answers, now it’s because we don’t want people to see our mistakes.


A few thoughts on open data

I’ve been meaning to compile my thoughts on open data in Local Government into a blog post for a few weeks now. So here they are in all their glory:

During the course of the project at Warwickshire County Council (WCC), and through dialogue with other local authorities, I have some thoughts on the key blockers to the Open Data Agenda in local government:

  •  Fear – at all levels there is fear of the repercussions of releasing data. There are two contributing factors to this fear:
    • Fear that the data we are releasing may be inaccurate in some way
    • Fear that inadequacies in our services will be highlighted through the released data
  • Assumption of cost – we didn’t experience this at Warwickshire as the project was run on a very limited budget. However, there is an assumption by other authorities that there are large costs associated with this. By going down the route of a hosted solution, we minimised both set-up and running costs. We wrote our own open source application. For other authorities starting on this journey there are now a number of open source solutions available. Going forward, I believe we should create an Open Data Toolkit to enable local authorities to create their own open data site, provide a basic business case and suggested datasets to get them started.
  • Misunderstanding – there is a gap in the understanding of why we are releasing data and who we are releasing it to. We are opening up data with the hope that it will  be picked up and interpreted by individuals with the skills to provide a new view of what we are doing as an authority. Although anyone can download our files and review the data in them, the real value comes from the wider societal interpretation of our data.

 Practical steps to delivering an Open Data Project

 It is very important to be aware that this is not a project that will directly save money and it will not generate revenue. It will however

  • Provide an opportunity to ensure data being produced and collated by the authority is of the highest quality
  • Give us more opportunities to identify where we could do things better, and put them right
  • Enable us to reuse data internally so we have our own high quality “single source of the truth”
  • Deliver on the transparency agenda and the “right to data”
  • Enable us to work more collaboratively with the public

 My top tips would be

  • Start off with the easier datasets to get the momentum going, and look at what other people have published
  • Use open source software
  • Run a competition to spark interest in the local community – get some apps developed that prove the case for why open data is a good idea
  • Embed the values of open data wherever possible – the utopian state is that people have to fight for restricting their data, not fight against releasing it
  • Don’t take no for an answer – take different approaches for encouraging people to release their data if they are reluctant
    • Show that other people have done it
    • Show the applications that can be built (and have been)
    • Make them aware of the pressure from central government – release data now in a controlled steady fashion, rather than waiting to be pushed (like is happening with the financial data). The Right to Data is inevitable, so accept it with open arms