Data ownership, ecosystems and other thoughts from MyData 2018
I was in Helsinki last week for MyData 2018. It’s an annual conference run by the MyData Global Network, a community of people and organisations who think that “we, you and I should have an easy way to see where data about us goes, specify who can use it, and alter these decisions over time”. At the Open Data Institute we’re interested in similar things and wanted to learn more about what other people in the community were working on and thinking about.
After a couple of days of absorbing the eleven tracks, 51 sessions and 131 presenters (or at least the portion of them I joined), here are some of my thoughts from the conference:
The MyData community is diverse. It’s diverse in terms of the types of professions, skills and organisations that contribute to its discussions and events. This year the conference made a point of addressing the different ‘flavours’ of interest in personal data — business, legal, tech and society — which I thought were well-represented across the different tracks. I also didn’t spot a manel, although I did hear rumours of one session that had one.
Lots of people are working on the same things but calling them different names, and lots of people are working on different things but calling them the same names. This is probably true of any field where a disparate network of people and organisations are working on related issues. It feels particularly true right now of the work being done on approaches to data sharing and access around the world. I clocked data exchanges, data commons, data cooperatives, data trusts (more on them below), data markets, data marketplaces, data hubs, data portals, data platforms and more. There’s not always a substantial difference between these approaches; conversely, the same term is sometimes used to describe vastly different things. Without a better understanding of what these approaches mean and consist of, how can we effectively focus efforts on the same, or at least similar things, or consciously try new ones?

Data ecosystem mapping is popular! Data ecosystem mapping can help to visualise, understand and communicate how data is published, accessed, shared, and used by different people and organisations. The ODI has developed guidance on how to do it and we’re using it in our projects. We ran a session at the conference where we encouraged a packed room to map out different data portability initiatives in groups. Some of the people we spoke to think that ecosystem mapping will be useful in attempts to understand how personal data is collected, used and shared across different organisations.
Collectively we need to be clearer that increasing individual control over data does not equal data ownership. Many of the people I spoke to at MyData 2018 share the view that continuing to strengthen the rights people have over personal data is the route to increasing our control of it. However — and despite the fact that individual data ownership is problematic for a number of important reasons — the idea that we need to own ‘our’ data remains prominent. As proof of this, I was momentarily distracted from the conference by a Tweet about a new academic paper that almost certainly doesn’t provide “a robust economic model to prove why consumer data ownership is good for society”.
People are thinking about the effect organisational and legal structures have on data governance. For example, I learnt more about various data cooperatives that have been formed to manage personal data on behalf of their members, who own and have democratic control of them, and a share in their profits. Data trusts were described as legal structures that could be created to bring about independent, fiduciary governance of data. In a strange turn of events, Steve Bannon used the term ‘public trust’ in parallel to the conference to describe how the US could wrestle control of data held by Facebook, Google and others.
The MyData team know how to put on an unconference. Dubbed the ‘Open Space’, the conference had six and half hours of time for peer-to-peer learning and collaboration. The dozens of parallel sessions came together rapidly (and only slightly chaotically) and it was great to be able to dip into so many different topics.

Not everyone shares a broad interpretation of data portability. The right to data portability is described by the EU’s General Data Protection Regulation. As a result, lots of discussions about data portability often end up in debate around what data is and isn’t subject to the right. At the ODI we like to think of data portability more widely, as ‘the ability for people to share data held by organisations with third parties, including other organisations, services or people’. This is not least because other, sector-specific regulation seeks to make it happen (like PSD2 or the Australian Consumer Data Right) but also because voluntary initiatives like the Data Transfer Project can go beyond that which is within legal scope. We use a broad interpretation of data portability in our work, such as when exploring how data portability can be designed for when data describes multiple people in collaboration with IF. This interpretation wasn’t shared by everyone I spoke to though — I received some particularly strange looks when I asked “when I allow a new app on my phone to access my photos, isn’t that data portability?”.
Google has a team called the Data Liberation Front. I didn’t know that those working on products like Google Takeout and the Data Transfer Project were called this. I also learnt that they’re thinking about the types of services made possible by data portability beyond just switching — not just “migration from Google once an individual or company stops using their services” — such as those that complement or are unrelated to Google’s services.
A lot of experimentation is happening at a city level. Among a number of city-related sessions, ‘Data and Digital Identity in the Cities’ argued that “if for a long time cities were spectators rather than actors in terms of data use and management, this is no longer the case… many cities are embarking on projects for more innovative usage and management of data generated by citizens.” Understanding what the right level of control for cities to have in the context of data collection, use and sharing will become an increasingly important topic. Interestingly, there was a meeting of European city Chief Data Officers alongside the conference in Helsinki last week.
We project different meanings onto the ‘common good’. There was a lot of discussion at the conference about the use of personal data for the ‘common good’, the ‘public good’ or ‘public benefit’. For some people this meant opening up access to data held by the private sector. For others this meant enabling data sharing between public sector organisations. Or it meant individuals donating data for causes they believe in or research they want to support. In the UK, Involve have found “the term ‘public benefit’ is rarely, if ever, clearly defined” and have suggested five key features that a data sharing initiative designed to deliver public benefits should be able to demonstrate.
Cultural differences in attitudes and approaches to data was a prominent theme. Conversations about differing citizen attitudes towards the use of personal data, as well as approaches to regulation and trade, stretched across the different tracks and sessions I attended at the conference. In one Open Space session we asked whether the MyData community could be more proactive in engaging with technology companies and communities from parts of the world outside of the US and Europe (and especially China, given the growth of its technology companies and its export plans). It’s something to look for as the MyData Global Network evolves over time.
I enjoyed MyData 2018. It was engaging, thought-provoking and full of interesting people. Maybe at next year’s conference I will also go to the sauna (sorry, Esko!).
