The decision to put government data into the open is made against a complicated range of factors and motivations but it often breaks down into three modes.
In our work with the MediaMill project we noted that many public bodies s operate across three ‘modes’ of release when it comes to open data.
- Required: Most local-government and public sector organisations are required to publish data about their operation. Perhaps the most obvious example of this in the UK, is The Local Government Transparency codewhich requires Councils to publish a number of data sets. Most councils will make this available, licensed as open data
- Reactive: Central government demands a huge amount of data related to key performance indicators and orgnaisations will often have their own reporting data to add to the mix. A number of councils have pipelines in place to publish performance data as open data through Open data portals. Some data may be created as a result of specific requests (the local government transparency code makes provision for demand led publication) or external pressures. Some public bodies have begun publishing all data generated by freedom of information requests for example. This often blurs the line between required and reactive open data publishing where the data itself isn’t required but its publication is enabled by broader transparency and reporting requirements.
- Responsive: A number of bodies have begun to pro-actively select areas to publish data they hold to generate a response from a broader community. This could be to support or encourage interest, stimulate innovative activities or directly address issues of concern. Most commonly these might be datasets to support a hack event or community initiative.
These three modes are not mutually exclusive. The required and reactivemodes in particular interrelate. On one level there could be an argument for trying to reduce the reactive part of data publication i.e. the logic of publishing data created by freedom of information requests is that it limits more freedom of information requests. However, its perhaps more constructive (although the technical and practical resources required can’t be underestimated) for organization to look at the expansion of what is required data for publication — publish more of what you have by default and you cut the need to react.
The responsive mode is in some respects an ideal position to aim towards. It’s likely that organisations with more mature community hubs for open data and a community willing to help identify and speculatively engage with datasets will see the value in this mode. More broadly it points to a longer term strategy for sustaining an open data commitment that balances the focus on statutory OGD and the a broader reading of open data.
Thinking more broadly, the three modes may not just be limited to public bodies. Many private organisations have requirements to publish data e.g company accounts but a more responsive approach as Chris Taggart, Co-Founder & CEO at OpenCorporates noted, open data could be part of an offensive tactic to:
- Mindshare/increase usage/ propagate message & brand
- Allow/increase community contributions
- Get to where you want to be faster Expose your competitors’ internal contradictions
What do you think? Do you have examples of the modes? Are they way off the mark? Any thoughts are always welcome.