[Spotlight]: Managing barriers for an Open Data Culture
Stories from Argentina and New Zealand
by Agustina De Luca
Getting different agencies to take ownership of data policies, share information in and outside of government, and promoting structural reforms to make sure a good data governance model is put in place are just some of the challenges open data champions within government face. So, while the end of the year is just around the corner, our Implementation Working Group held its latest meeting to discuss techniques from Argentina and New Zealand on managing barriers to instil an open data culture inside government.
In Argentina it was about focusing on what makes people tick. Agustin Benassi, Public Data Director of the Government of Argentina, shared what has worked in his country during the past four years. He laid out a strategy that divided efforts in building “soft” and “hard” incentives for government agencies to comply with their policy, although more often relied on the former than the latter.
As part of their arsenal of soft incentives, Argentina (i) encouraged each area to have their own open data portal or data published on their official website, with guidelines and support from the open data team. This helped increase ownership in the process, communicating it as, “This is your open data policy, we just provide guidelines and technical support”. Then, (ii) the team sent feedback on the type of data being published and rewarded policy achievement through featuring quality data in the national open data portal. This increased the (iii) visibility of often unnoticed backend work, because having data in the official portal helped grow a broader community and recognition from other stakeholders. For instance, the National Geographical Institute has been publishing maps and geometries for 30 years, but no one knew about it. After they adapted formats and got that data in the national open data portal, they gained wide visibility and more people started using this data.
As other agencies began to notice, they began (iv) making technological improvements to the data, mainly through adapting formats fit for the national API. Agustin shared how this was a powerful perk, since in some cases these were small changes with huge impact as more people would be able to access the data. Many public officials did not know what an API was until then! This in turn promoted (v) digital integration, countering international norms that centralization of portals was key to simplifying access for citizens.
Finally, they convinced other agencies through the potential to reduce their workload by (vi) expediting Freedom of Information (FOI) management, because if more data is published, then they would reduce FOI requests or even avoid them altogether.
If these carrots were not enough, Argentina had one stick up their sleeve — they worked hard to establish a modern FOI law which mandates government agencies to publish some data on a regular basis in open formats. If they don’t meet this, sanctions are provided.
After four years in office, Argentina is now facing a government transition and looking to ensure the sustainability of positive changes. They began by building manuals with clear and simple steps and exchanging knowledge with new officials so that they don’t start from scratch (for more on managing transitions, see our Resilient reform in government: lessons from open data leaders report).
On the other side of the world, we learned why New Zealand is called the place of ‘Open Spaces, Open Hearts, and, Open Minds’. While they don’t have a law mandating the proactive release of data and information, New Zealand has a softer mandate in the form of a “Declaration of Open and Transparent Government”, which is effectively a directive from Cabinet to agencies. As a result the open data team works by promoting the benefits of open data and achieves change by winning the hearts and minds of public servants who then want to make open data happen in their agency. They are working to embed open data in a broader data governance framework, considering data as an asset and including open data principles into everyday data management procedures.
Paul Stone, their Open Government Data Programme Leader and our IWG chair, recognized that having “some” data published does not mean that any of it is useful or understandable. That is why they work to provide, (i) good context, (ii) allow feedback and (iii) License it for reuse (such as with Creative Commons).
For New Zealand, context is crucial. How data is collected and for what purpose will provide extra information to allow better use and accuracy in its management, and helps mitigate concerns of people not understanding the data and using it incorrectly. Providing a feedback loop and inviting data users to contribute can improve data quality and help build an engaged community around the data.
As with any other tool, data can be misused by anyone, but we should distinguish between an ignorant from a malicious misuse. Whereas the first one is because the user didn’t understand it (increasing the case of providing context, describing variables and the collection method), the second one has a specific intention to make harm. In this case, we should not let one potential bad use prevent the opportunity for many good uses — instead, through using a Creative Commons licence, government agencies are protected from liability, and the responsibility for harm to others is on the data users.
Having the resources and technology available for open data policies is fundamental. In many cases, agencies lack resources, so Paul recommends starting with small quick win projects — promote an “open by design approach”, aligning data publication with the organization’s strategy and/or demand from it’s community, and publish on existing websites. Many agencies even have API capability but don’t know it, or don’t use it, so thinking about how to use available resources and technologies can go a long way.
Finally, according to New Zealand’s experience, privacy is maybe the biggest barrier, since safeguarding personal information is critical for maintaining trust. However there are tools available such as the Privacy Impact Assessment Toolkit provided by the New Zealand Privacy Commission, and the various statistical techniques to anonymize and confidentialize data (check out our blog on this!).
For both Argentina and New Zealand, shifting the mindset of a whole organization into open by default is the goal, but it doesn’t happen overnight. Transmitting messages such as “we are custodians of a public data asset”, or “ data will be improved as people start to use it,’’ are good tips to start shifting cultures and practices in government. Culture change takes time, but the key is to work out how best to make the change sustainable in your environment.