Insights with Cartesi’s Advisor — Serguei Popov (Co-Founder of IOTA)

Published in

Cartesi

6 min readDec 30, 2019

Recently we held an exclusive interview with our advisor, Serguei Popov. Serguei is a research mathematician working in the field of Probability Theory and Stochastic Processes. He graduated from and attained his Ph.D. at Moscow State University and is currently a professor at Unicamp in Brazil. Serguei is also one of the co-founders of IOTA.

Read below for insights into how Serguei got involved with DLT technology, met our Chief Scientific Officer, Augusto Teixeira, and his perspective on Cartesi!

Question: How did you get interested in and involved with blockchain tech in the first place?

Answer: A bit more than 6 years ago, I stumbled upon an article (on Computerra if I remember correctly) about Bitcoin. I found the idea simply genial, and started reading about it; in particular, I read a lot of material in the (in)famous forum bitcointalk.org. Already in 2013, there were a lot of altcoins, but these were mostly clones of Bitcoin (mainly with some changes with regards to the mining algorithm). One new project called Nxt caught my attention though: it was a pure Proof-of-Stake coin, quite a novelty for that time. I started to participate actively in discussions on Nxt, both on bitcointalk and later on dedicated Nxt forums. At that point the founder BCNext already became silent, but his colleague CfB (a.k.a. Sergey Ivancheglo) was quite active and participating (hmm… sapienti sat 😁). I had talked a lot to CfB who was kind enough to explain the Nxt protocol in a detailed way to me. Afterwards, I made some back-of-the-envelope calculations on the probabilistic properties of the Nxt blockchain and posted them. (A margin note: blockchains have a lot to do with a mathematical object called Poisson Point Process; as a probabilist, I know a fair bit about it). Also, an extended and revised version of those notes was later published in my article A Probabilistic Analysis of the Nxt Forging Algorithm. People looked at my calculations, and said “Wow! Looks great! I couldn’t understand anything, but it looks so convincing!” This is how I became “a (Nxt) community mathematician.” And then, at the beginning of 2015 it was time for David and CfB to start a new project that later became known as IOTA; which they invited me to be the mathematician of. Thus I made contributions to the basic protocol and wrote the first whitepaper. After that, many things have happened, but let me stop here.

Question: How did you meet Augusto Teixeira (Cartesi’s CSO)? Can you tell us a bit about the experience of working together with him?

Answer: It was in 2011, when he gave a course on random interlacements at the Brazilian probability school, together with Jiri Cerny. The course was quite interesting (although they have managed to scare off all of the students quite efficiently, at least one professor was able to survive!), and they motivated me to work on this model. Next year, I invited Augusto to Campinas in Brazil to discuss random interlacements where we had the “soft local time” idea, which permitted a very efficient decoupling of local pictures left by the interlacements’ trajectories on distant subsets. This resulted in this article, Soft local times and decoupling of random interlacements, which I consider my best math work up to the present time. Working with Augusto was always a pleasure: he’s a very quick thinker with a quite broad knowledge and skills. Hopefully, we’ll have opportunities to work together again in the future.

Question: How did you introduce Augusto to blockchain?

Answer: After publishing the aforementioned article, we had many further working meetings; both in Rio de Janeiro and Campinas. We tried to work on very challenging problems but unfortunately did not have much success with them. On the other hand, when mathematicians are stuck they tend to talk about other things in order to alleviate. So, I was telling Augusto about my blockchain/Tangle adventures. Again, I was impressed by how quickly he grasped all those ideas and came up with his own ones. By the way, the Cartesi concept was not the first new blockchain-related idea that I heard from Augusto; so you can be sure he still has some quite interesting cards to play.

Question: From being an early backer and advisor, how do you think the idea of Cartesi evolved?

Answer: Augusto had the original idea of creating a trustless marketplace for data scientists. Basically, it would allow specialists and companies to engage in a service agreement without the need to know each other or a reputation system. In order to implement that, he invited Diego Nehab to design and implement a reproducible VM capable of running Linux. With this architecture and the protocols Augusto had envisioned, they would be able to make generic computation verifiable by the blockchain. More specifically, the blockchain would be able to initiate routines in the emulator to verify the results sent by the specialists after they completed their jobs. The system would protect the two parties involved, only releasing the funds if the results were proven to be correct.

Later on, the team engaged in long discussions and realized they were tapping into a broader scope. Before Cartesi even had that name, they understood they could actually create a versatile layer-2 to overcome the problems of scalability of computation and infrastructure for blockchain applications.

Question: What kind of exciting application or use cases will Cartesi enable?

Answer: The original idea that brought the Cartesi team together was to build a decentralized AI marketplace. Although they have now come to the conclusion that the system is much more general than that and could have many other applications, I still would love to see this AI marketplace working on top of Cartesi. Such a system is extremely hard to build without their work and it would serve as a very good example of what a more decentralized world could look like.

Question: What’s the biggest change that Cartesi will enable in the blockchain space in your opinion?

Answer: Cartesi will promote broader adoption of public blockchains, facilitating the development of DApps (even for non-blockchain developers) and addressing the problem of computation scalability.

Question: Where do you see Cartesi being a relevant Layer-2 solution in 1.. 5… 10 years from now, when this industry has become heavily saturated with new projects?

Answer: Cartesi can attract a large number of veteran developers to easily code DApps without the need to touch smart contract implementation on the blockchain. Cartesi should also allow DApps to be portable across the most relevant blockchains. Because the DApp logic is implemented for Linux and deployed on the second layer, that will protect these developers against the shortcomings of blockchain projects during the next decade. As Cartesi fulfills this vision, applications can be easily ported across different blockchains without the burden and costs of code reimplementation.

Question: What are some hardships and risks you feel that Cartesi will help solve in the blockchain and crypto industry?

Answer: A big risk that the crypto ecosystem faces is the isolation from the traditional industries. While crypto is developed and used mainly by people who are already within the crypto environment, the adoption will be limited. I think Cartesi can help greatly with that, by making the development and usability of DApps much more similar to what people are already used to.

Question: Given that you are the co-founder of IOTA and inventor of The Tangle, do you see any applications with Cartesi on IOTA?

Answer: One of the basic principles of IOTA is freedom (see IOTA: Feeless and Free). In particular, the IOTA protocol permits the network participants to construct any sort of “structures” in the Tangle (the IOTA ledger). This, together with the feeless nature of IOTA, can benefit Cartesi a lot, in my opinion.

Question: How will DLT change our world in the next decade?

Answer: I’m just a “theory guy,” not really connected to the real world. But, as they say, the future always comes. So we’ll see. :)