Data ownership and the semantic web: Interview with Professor Ruben Verborgh (Part 2)
–– 20 Oct 2019
In our first interview with Ruben, we talked about his campaign to get his data back from Facebook.
In part two we chat about the semantic web and the implications of this technology on privacy and data ownership…
Andy: Tell me about your journey to date…
Ruben: I am a professor of computer science at Ghent University Belgium. I’m mainly a researcher but I also teach. My main area of research is how we can deal with data on the web. I’m not interested in big data such as bringing lots of data together in one place. I was more interested in what happens if you don’t bring that data together. What if the data is distributed around the web and how can we make sense of it? Initially, I started to work with public and open data.
Andy: Most of your work focuses on the semantic web. What is the semantic web?
Ruben: Forget about all the buzzwords. The semantic web is simply about putting structure and data in such a way that machines can interpret it. Why do we need this? If we don’t have semantics then we depend on some organisations saying ‘this is how you store your own data.’
The semantic web essentially gives everyone the ability to control data as they want whilst still being interoperable.
Andy: What particular technologies are you researching at the moment that can enable that change you’re talking about a semantic web?
Ruben: I’m part of the Solid team where the big idea is to give everybody a silo or data vault – your Data Pod. Every piece of your data that you produce about yourself or other people produce about you, you store in your vault. It’s yours to control.
Solid isn’t the only Personal Data Store, but what sets us apart is interoperability; we don’t want to tell people or tell them what they look like. So you can structure your data in whatever way you want.
The problem with semantic technologies we still need to solve is that it’s interoperable even if we do our own thing.
There are two pillars. There’s the privacy perspective which is taking control of your data. But also it’s about innovation and if you’re an app developer, the first step is to get data, like harvest data from people and ask them questions so they can do something. Which means they are always in competition with Facebook and others.
But if people come to a brand new app that is their own data vault they don’t have to harvest their data, they just use what they have and then they’re done. You just leave the rest there. It’s not just beneficial for people but also beneficial for companies to not be forced into this data race which nobody will win anyway.
Andy: Are there ways in which the semantic web can improve data ownership?
Ruben: It’s mainly about interoperability. If you have a tool like Tapmydata, a big part of what you do is alignment. Because all those companies have different ways of representing people and whatever data it is. You need to somehow get back to a single data model. At the moment our data is all manual work. Every company has to do the same exercise over and over again. But the promise of semantic technologies is that search conversions can happen in a much more automated way.