polypoly polyVerse
2022-08-25

We interviewed Mira Mezini who is a professor of computer science at Darmstadt Technical University. Mira has received various awards for her research, including IBM Eclipse Innovation Awards, a Google Research Award and the German IT Security Award.

Tell us a little about yourself. What is your core expertise?

I am a computer scientist, a professor of computer science at Darmstadt Technical University. My core expertise is in software technology, more specifically programming languages and program analysis. I started my career with my Ph.D. in the area of [programming language] Smalltalk; I love it and that was the language in which I did my Ph.D. My primary work has been developing abstractions and languages for automating certain aspects of software development. Or the other way around, software analysis to discover, for example, security vulnerabilities or privacy violations in code.

Smalltalk was in some ways, a bridge to [polypoly initiator] Thorsten Dittmar.

Yes. I was giving a conference talk as a young Ph.D. student. At the time this conference was dominated by Smalltalk presentations since a lot of companies were working with this language. Thorsten attended my talk and approached me after afterwards; that’s how we met in 1997.

Wow, so you two definitely have got a history. Carrying on then, what, in your opinion, is the main problem with the current data economy?

Well, the main problem is centralisation. Most of the digital services we use every day are centralised; basically user interfaces to some functionality that run on central servers where data is being stored, and this is a problem. It’s a central point of failure from a security perspective. There’s also the matter of privacy. We don’t know what happens with our data. For me this begs the question, why aren’t we using the capacities that we have on the edge of the internet. All the devices on the edge are used merely as a way of interacting with the data that is stored elsewhere. This gives a lot of power in this kind of paradigm to some actors. They’re making a business of our data. There’s imbalance.

What motivates you to help solve this problem area? Did anything in particular trigger your interest?

I don't know whether there was something in particular that piqued my interest. I was working with my team on programming paradigms that would facilitate me programming decentralised applications. Then coincidentally Thorsten approached me at that time with his idea, which I found fantastic. I was really very enthusiastic about the idea of building a new paradigm. After you know the idea it seems like the most obvious thing in the world. But you have to come up with the idea in the first place. This is characteristic. After you come in contact with a good idea, it seems like a very simple and straightforward thing to do. So this idea of turning the data economy upside down was very exciting. And also the way he was planning to create this change by blending technology through different organisational structures. It wasn’t just about protecting consumers. The data economy needs to include business, and that has been accounted for too. Obviously we need to share data, but we need to share it fairly. So this is very intriguing. Ensuring data privacy and sovereignty, at the same time enabling businesses or enabling society to make use of the data. So this was very exciting, and that's why I like to contribute to this story.

Super. What is the best way to motivate everyday people to join the movement for a fair data economy.

I don’t know what to say about this, because you’d think that they have experienced so many bad things that there would be no need to motivate them. But on the other hand, such a huge change is difficult to implement because people are comfortable; they stay with the platforms they’re used to. So I think the transition has to be very easy. The same functionality that users get in their current platforms should be available in the decentralised versions. Technically it should be possible that people don't see a change, that they don’t notice the difference. Also, when people can monetise their data through the polyPod, that should spark change but we have a little ways to go before that happens.

Well, this next question is gonna be very easy for you. How did you become aware of polypoly? And why do you think polypoly will be part of the solution?

For part one of the question, it was gradual. As I said earlier I met Thorsten in ’97. He approached me because he noticed a project we were doing on naturalistic programming; basically writing programs in a way that are very similar to natural language. We talked about a business he had, then lost touch. Some time later he came to me with the concept for polypoly and I was hooked. And sorry, the second part of your question was?

Why do you think polypoly is part of the solution?

I think I already answered this a little bit. We need something like polypoly technically to compliment the policy side. Many people talk about data privacy and decentralisation, but do not connect this to business, to enabling business to interact fairly with user data. And the other part is the way polypoly is organised to include all aspects of the data economy. We need a system based on fairness and trust. I think this is important for making it work. I haven't seen this elsewhere.

Cool. And finally, each of our advisors has a particular skill set. Given your expertise, how would you advise polypoly to move forward?

My advice would come from the technical side, which we’re already doing. For example, I’m in contact with polypoly’s research lead through my lab. We talk to each other to exchange ideas about how to implement the platform right now.

And again, a little bit earlier, you mentioned that offering features and services that would make sense for everyday people. What do you think something like that might be, something on your wish list?

One of my former PhD students approached me with some ideas about how to reduce my carbon footprint. So maybe a feature that advises me in the every day life how to do something that keeps my footprint low. Possibly I could connect with companies, for example, that offer services or products that compliment the way I live, with my style of life. Possibly some of my data enables features suited to my needs.

Yes, one of the interesting things that I've been thinking about over the last year is how edge computing has a much lower carbon footprint than centralised solutions. So you have not just the data which might be on the device, but you also have the native processing power. At first I thought it was just wishful thinking, then I started reading academic paper after academic paper that lays out the science and the data on what one might call sustainable computing.