Stats NZ and DIA sponsored a visit to Wellington recently by leading international open data expert and Open Data Institute trainer Ellen Broad.
Ellen inspired and informed a bunch of open data enthusiasts at a Meetup titled: ‘Sorting fact from fiction: open data myths, mysteries and misconceptions’. Here’s what she talked about.
Open data is “a driver for scaling up really quickly” – and it’s often the fastest way to develop new services and products.
A good example of this is Citymapper, which was built off the back of an initiative by the City of London to make their transport data open. The city made the decision to move away from commercially licensing its transport data, to publishing it for anyone to use. Citymapper is now used for journey mapping in many cities, in many countries, around the world. For the City of London, open transport data has created a new range of businesses providing journey planners, transport data analytics and APIs, supporting economic growth.
Open exists on a spectrum of use, explained Ellen. “So we have lots of data sources that are openly licensed and available for anyone to use. And we also have datasets that for very valid reasons (perhaps because they have personal information or are commercially sensitive) are accessed in different ways, such as by approved researchers who have conditions attached about the way they use those datasets.”
The Open Data Institute data spectrum diagram illustrates this range of uses.
Ellen pointed out that people often think of ‘open’ and ‘private’ as binaries – so data can be either open or private.
However, where a dataset isn’t subject to data protection laws and can be anonymised so individuals can’t be identified (and the dataset can’t be combined with other data sources to re-identify a person) then it is no longer personal data – it can be published.
“We need a mature risk management approach if we are going to share data with different parts of government, researchers – accepting that occasionally things will go wrong and having the mechanisms in place to respond to those problems – rather than shutting everything down and saying ‘We cannot share, we cannot collaborate, we cannot open up because these risks may occur’. We need to move away from ‘Let’s not do anything’ to ‘How can we manage this safely and sustainably?’”
Ellen believes open data can bring direct, tangible economic benefits for a publisher – as well as indirect benefits, such as feedback on how your data is being used that gives you a sense of who is using it, issues they are having with it and how it can be improved.
Ellen worked on a project in the UK where the environment agency was given a directive to move from managing its LIDAR data as a revenue-generating dataset, to making it open. (LIDAR uses lasers bouncing off a surface to very accurately identify land contours. The agency used it primarily for flood modelling.)
Making the data open brought some important benefits for the department:
“Misuse is something that we struggle with whether data is open or not,” said Ellen.
“People misuse data all the time. Sometimes it’s errors, inaccuracies. It might be bias in the way that you collect data. Or it might be data that has been angled to present a particular truth.”
On the other hand, having data open can provide an opportunity for data users to help improve the data, such as by pointing out errors.
Ellen believes it’s essential that organisations move beyond just publishing data and forgetting about it.
“Communication is vital to a successful open data relationship – which includes trust and the feedback mechanisms to help you improve how you publish and encourage use of your data.”
The ODI’s Open Data Pathway enables agencies to look at the processes and mechanisms they need to have internally to support their open data initiatives.
“Ultimately open data helps you accelerate a better data management culture. It’s not really about just publishing a dataset. It’s about encouraging a culture of collaboration, team based working, feedback.
“The benefits from open data are potentially organisation changing because of the culture that open data encourages. Open data will drive government being more efficient and working collaboratively to solve common problems.”
Twitter: @ellenbroad