How it all began
It seems the key to triggering a successful open data project is to show the people that matter something shiny, like an iPhone, with a real example of what open data can achieve.
Jim Morton began a small covert operation to start opening up Warwickshire’s data under the guise of developing an iPhone application with news, events, jobs and location information from Warwickshire County Council (WCC). This was launched in January 2010 and by the middle of May 2010 had already been downloaded nearly 2,000 times.
Using the success of the iPhone project and the increasing number of good open data examples (e.g. ) we were able to kick-off our own open data project to create opendata.warwickshire.gov.uk. The business case and main benefits driving the project are:
- Transparency for the public
- Enhancing public communications
- Improving service delivery and enabling citizens to self-serve
- Contributing towards new ways of running public services
- Improving external contribution to WCC
- Enabling mash-ups of disparate sources of information to create new ways of looking at information
- Enabling 3rd sector organisations or individuals to develop applications aggregating data across organisational boundaries
- Reducing workload in areas like Freedom of Information (FOI), the Observatory and Public Relations
- Reinforcing our efforts to resolve data and information issues
Using the Identify, Represent and Expose principles outlined in Jeni Tennison’s blog, a small team of 4: Jim Morton, Steve Woodward and Terry Rich-Whitehead and myself began working on getting information out of the organisation and building a technical solution and set of standards that fitted with our ongoing work to introduce open, non-proprietary standards across our ICT architecture. We were keen to ensure the project made use of cloud technologies to deal with deal with any scaling/demand issues.
The open data site has been written using Ruby on Rails and is hosted on external platform-as-a-service provider Heroku. The database managing the sets of data includes a standard XML schema for the metadata associated with each dataset. It was very important to ensure this schema was aligned to that used by data.gov.uk to enable us to easily extract our metadata for inclusion in their data catalogue. The application will soon be open-sourced to enable other authorities to easily build their own open data site.
We began at the end January 2010 working 2 days a week, and by mid-April we had unofficially launched the site with over a dozen datasets. By the time we officially launched a couple of weeks later there were nearly 30, and this number is growing by the week.
Tips and Tricks
- Start with quick and easy datasets – the sooner you can get datasets up and build sample applications to demonstrate the purpose and benefits of open data, the more likely you are to encourage other people to give you their data
- For those reluctant to open up their data, there is one key question you need to ask: “If someone requested this information under FOI, would we have to give it to them?” If their answer is “Yes”, then you have a very strong starting point for persuading them to give you their data
- Ensure you have a good process for feeding back any issues with the data
- Use a standard [preferably open! — ed.] Creative Commons license to cover usage of the data rather than trying to write your own
As part of our work to publish the initiative, we are running a “Hack Warwickshire” competition between Monday 17th May 2010 and Friday 25th June 2010. This competition is challenging everyone to come up with new and innovative uses for our data and web services. The winner of the competition will be the proud owner of a brand new Apple iPad.