Data Sharing Fireworks & your partner’s partner’s partner

by | Oct 21, 2019 | ICT4D |

I can think in frameworks. And even though I am a visual thinker, I can think in linear relationships. Whenever I try to think in networks, my brain hurts. And yet, it is a skill I need to develop.

Data Sharing can visualised as a one-way or two-way road connecting two nodes. It can also be visualised as a roundabout (multiple nodes). That works; I can do that. However, in reality, each node is rarely just a single node, but the start of an explosion of new nodes.

Therefore, when we think about data sharing, we need to think beyond the organisation we are talking with. Many international NGOs (INGOs) have offices in many different countries. There are variations of the governance models used. So when a local NGO in Burundi shares data with a country office of INGO, it often means it is sharing the data with the INGO’s office in Canada too. The local NGO may be ok with that.

But it may mean the data is also being shared with all the partners of every office the INGO works with. And the partners of the partners. And so on. This is where the ‘nodes’ turn into fireworks. The network diagram begins to explode and the trail the data leaves is very long.

To make it more challenging. The data shared may be viewed as harmless on its own, however when combined with various other datasets it may lead to deeply personal insights. The extremely difficult part is that this was never the intent of the data sharing. And it wasn’t the INGO who had the data analytic capability to create the insight. It was a partner’s partner’s partner.

So when our ‘informed consent’ process ask for consent of the person to share their data with ‘our partners’, I worry. When we are asked to sign data sharing agreements that state the other party will share the data with their partners, I worry. I see fireworks exploding everywhere and wonder what we are setting off.

This is a critical issue to wrestle with when considering data governance. It is full of ethical dilemmas. Data Trusts might be able to help with it, but again the discussion, the debate is essential.

Photo by Erwan Hesry

0 Comments

Submit a Comment

Your email address will not be published.