Jeni Tennison, CEO of the Open Data Institute, explains how better access to data can lead to a fairer society.
The conversation around data in the public sphere is often focused on its misuse by large companies, or large breaches placing individual information at risk.
While these are undoubtedly crucial stories, it is also important to highlight how the use of data and creation of greater data freedom can have a positive impact. The world at present is full of data silos and corporate monopolies, as well as financial obstacles for researchers and data scientists.
Jeni Tennison is CEO of The Open Data Institute (ODI), founded by Nigel Shadbolt and Tim Berners-Lee in 2012 to show the value of open data and the innovative ways in which it can be used. Siliconrepublic.com spoke to Tennison about her work and the work of the ODI in building more trustworthy and open data ecosystems.
Building more trustworthy data flows
“The vision of ODI is for individuals, communities and organisations to be able to make better decisions using data, and do that while being protected from any harmful impacts,” said Tennison.
While the ODI emphasises open data, it also encourages the innovative use of data that must be shared in more restrictive ways to protect privacy, such as medical information. In this case, it is about “making data flows work better”.
Tennison explained that there are three key processes when it comes to data: using it to make decisions, creating products leveraging data and stewarding that data. The latter is a crucial element of a more trustworthy data economy.
What is data stewardship?
Tennison described data stewardship as an “act of collecting data, maintaining it, making sure it is kept up to date, and then deciding how and whether to share it”. This is important as the collection and maintenance of the data is vital to retain the data value chain it creates.
Of course, many people would have legitimate concerns about who has access to data and how it might be used. The ODI is also a steadfast proponent of building ethical considerations into how data is collected, managed and used, as well as ensuring equity of access and broad engagement.
Three steps towards an open data model
Tennison explained that through this ethical stewardship model of data management, social justice goals can be achieved in three general steps. “First of all, one is that when people make decisions informed by data, we think those are better decisions and likely to be less biased and have less dogma in them, and that in itself can lead to social justice.
“When you have more open data – data that is more accessible by more people – then that means that civil society organisations, communities and individuals can use that to make decisions as well as the more powerful entities.”
Finally, when data is made more accessible, it creates transparency, which can lead to greater accountability in the right context. Tennison added that this model needs buy-in from decision-makers to work. “Data is necessary but not sufficient. You need to have decision-makers actually want to use data to make decisions.”
Data scientists can benefit
One area that could see a huge benefit from a more open culture is data science. More open data means that you can give data science teams access to greater volumes of data. It also removes barriers around cost and licensing, which Tennison said can be “big barriers when you are a small data scientist or when you just need it [data] to add a little bit of context to a project”.
Data scientists can also save time under a more open data culture as, according to Tennison, the bulk of the effort goes into accessing and cleaning data before it can be analysed. She made comparisons to the open source software movement and its creation of a collaborative ecosystem of developers.
While this all sounds positive, it will take a lot of work to get there. Infrastructure for said data needs to be developed. We need to ensure the data is “useful for us as a society to make decisions on, is there and available, being collected and maintained, [of] reasonable quality and is being looked after well”. Good data governance around it is an important step, which begins with policy and legislation.
Data literacy or capability is another key element, but Tennison said the definition must be wide. “Often when people talk about data capability or data literacy, they narrow down on to how to process data into a spreadsheet or something like that but we tend to think of it as much broader than that.
“[We] think in terms of people being properly equipped to make decisions around giving consent for data to be used; we think in terms of business leaders knowing that data can be used to make decisions and policymakers understanding how they can use availability of their data to achieve their aims.”
With regard to specific projects the ODI is engaged in, Tennison spoke about its partnership with Arup, a building and environmental design consultancy, with which it works on an open innovation model. Meanwhile, a project with Thomson Reuters around open identifiers, PermID, aims to see a move towards a more open data culture.
Tennison noted the importance of ethical use of data above all else, saying that the traditional ‘move fast and break things’ ethos is often incompatible with solid data ethics. “It is possible for bad practices to affect the lives of huge numbers of people we can’t anticipate.”
Communication with stakeholders and users is the fulcrum of this approach. The ODI created a tool called the Data Ethics Canvas to help organisations so that potential ethical issues associated with a data project can be flagged.
“It is beholden on data scientists, developers and business model creators to stop and think about what might happen if things go wrong, or if things go right,” Tennison stressed.
The new era of data has the potential to change many lives, but it is up to those working with the information to make these changes positive ones.