I remember the moment I was interviewing Emin Gün Sirer when I was still a reporter at Bloomberg News. It was 2017 and he was telling me the story of how he was sick in bed but nonetheless had stumbled upon the bug in the DAO code that would eventually lead to a $55 million theft that rocked the Ethereum community. That small detail sent me on a journey that eventually led to my book Out of the Ether. Gün has been a leading academic in blockchain and distributed systems from before Bitcoin came into existence. More recently he founded Avalanche, one of the quickest growing blockchain networks that’s worth $21 billion as of this writing. Read on to learn how Gün got his start inc omputer science, what it was like growing up in a huge family in Istanbul and his vision for what blockchain technology means to the world. You can check out more about Avalanche here.

Matt Leising: So to begin, I know you're from Turkey because when we were driving around Ithaca in your BMW it was like you were on the streets of Istanbul. Where were you in the country?

Emin Gün Sirer: I grew up in Istanbul, indeed. And I ended up going to a government school with 55 kids for the first five years. And it was a fun time. Back then we had, I don't know, I can get into all these gory details of having to grow up with a TV with one channel and it was black and white. It was a very odd time. We were 30 years behind the rest of the world for some reason that I don't quite understand.

ML: Yes, please give us all the gory details. That's what we want. What do you mean by a government school? Is that a public school as opposed to a private school?

EGS: I went to a public school for five years. At the end of those five years, you take an exam and if you are in the top 110 or so then you get to go to the American High School, and there a couple of other foreign schools that people try to get into. I worked really hard from age nine to 11. You work like a race horse and you take this big exam with a million and a half kids. And I was number 20 something and got to go to a gorgeous high school, the American High School, the very first American high school outside the United States ever built and staffed with amazing people.

The American High School in Istanbul

EGS: And that's where I learned the English I speak, but more importantly, that's when I saw education geared towards hardcore science and if we're talking about origin stories, there was a little detail that maybe you'll appreciate that ended up being quite formative. So when I was growing up, we grew up in this drafty house and things were done to an engineering level considered acceptable in Turkey at the time. And so the windows closed, the doors closed, but they didn't really close. Nothing was really done perfectly.

EGS: And Turks are a nation of super fast moving individuals and they do things really well. They adopt technology really well, but when it comes to fine finishing work, that's probably not exactly the place.

EGS: So I remember going to this high school, and it's this amazing, impressive building on the side of Bosporus overlooking the water and it has giant doors that are probably like 15 feet tall and they close perfectly with the satisfying click <laugh>. And I remember thinking, oh, this is a different kind of engineering. It's a different culture. And ever since then, I've been fascinated with things that were done perfectly, things that you can count on. Things that give you guarantees about how they're going to work. It shouldn't just be like, ah, this is good enough. We'll manage.

EGS: But I was fascinated with building things that really worked a hundred percent of the time. And so, not surprisingly, I found myself fascinated by computers where you create your own universe internally.

ML: What did you like to do as a kid besides school and study? Were you an athlete? Did you have computer games at all? You said you had TV with one station, so what were you up to?

EGS: So I wasn't an athlete at all and we did not have computers back then because the very first computer I saw was when I was in sixth grade in my middle school. So what did I do when I was a kid? I don't know. I did kid things, played on the street, played really bad soccer, really terribly. I have good reflexes, so they put me as a goalie, which is the most miserable location you ever want to be in. Nobody says anything good to you and if you let the goal in, then <laugh> they really put the blame on you.

ML: Goalie is definitely the most thankless position in a soccer team.

EGS: It is so thankless. At that age I was infatuated with math. I figured out a little bit about electronics and I ended up going toward electronics, building my own computer, building my own micro controller. That's where I started getting into computer stuff.

“I wanted to build the world’s biggest platform for digitizing all of the world’s assets. ”

ML: What was your family life like? What did your mom and dad do? Do you have a lot of brothers and sisters?

EGS: I grew up in a huge family, my parents, my sibling and grandmother, great grandmother, uncles, all in the same big house. And it was a wonderful experience. I miss it quite a bit. Family sizes have dropped everywhere and also as I got older and people passed away, our family has grown to be quite smaller.

ML: And were your parents an influence on you in your love of science and math? How did you come upon that?

EGS: I grew up in a crazy house with eight, nine people at any one time, with giant family dinners and other uncles and aunts coming by to visit on any given day. The maternal line of the family was living with us. So we were the hub of the family. And so there was some commotion. I found myself isolating in my own universe of wires, electrical stuff and the computer universe. That was my respite. If I wanted to step away from the family craziness, I would go towards the computer stuff.

ML: Did the certainty of math appeal to you? You either get it right or you get it wrong. Was there something about that, or what do you think appealed to you as a young kid about math?

EGS: Absolutely. The certainty and the fact that there are clear metrics, the fact that you know the winner. The right answer isn't up for a popularity vote, there is a definitive answer to just about every question. And if you are bright enough or persevere hard enough, you'll be able to find it. So all of that was hugely appealing. I found myself really excited by math and of course physics and electronics.

The one and only

ML: So now that you're in the American High School, do you start thinking about college or where you go from there? And it sounded like you had more opportunity to interact with computers at that point.

EGS: Yeah, I would spend a lot of my time, all the free time I had, initially using the computers at the middle school. And then later on, I ended up getting a Commodore 64. So I was a big Commodore 64 guy. I was really into doing all sorts of low-level trickery with it, building hardware based on the Commodore 64. Then as I got older and towards the end of high school you get a choice, right? You stay in the Turkish system and take another exam with another million and a half people to try to get into your chosen area, or you apply to schools abroad. The American High School was a well-known high school to U.S. administrators at universities. So and I was doing well in school, so I ended up getting a scholarship to go to Princeton. In fact it was much easier to do that <laugh> than to get into that other [Turkish] exam place. But I went to Princeton and it was the start of my specialized education in computer science.

ML: From a personal level, I imagine the culture shock of Istanbul to Princeton must have been something incredible.

EGS: No, not really. You don't feel it, you're only 18. So I didn't feel like there was much of a culture shock. It was an extension of my geeky existence into a school where there were many other people just as geeky or even more geeky than I was. So it was wonderful in many ways. It was a lot of fun. I ended up seeing my first Internet-connected computer shortly after going to Princeton, and that was an amazing sight. There are things is in everyone's existence where you just get a glimpse into something really big. Like sometimes you have two mirrors facing each other and you put your head in there and then suddenly you get a glimpse into infinity.

EGS: Or maybe these days, kids are looking at Zoom windows where you have the same window inside of the infinity tunnel. That machine at Princeton that was connected to the Internet was exactly like that for me. I suddenly knew this was possible, I can now connect to whatever else, some other machine. I can send an email! At the time I was writing letters to my parents. If my parents knew how to collect the email on the other side, <laugh>, which they didn't, I theoretically could have sent them an email in my freshman year.

ML: So it's good to know that Istanbul and New Jersey are basically the same. And then when were you at Princeton? What years are we talking?

EGS: ‘89 to ‘93.

ML: Did that revelation of the Internet-connected computer take – what did kind of journey did that start you on?

EGS: In particular I had wanted to do AI [artificial intelligence] at the time, my very first year. I thought, ‘hey, I'm going to build super intelligent, super complicated programs that help people directly with their lives.’ But on the path there to Princeton and in general, even for the first part of the nineties there was a very clear signal that you could not miss. And I think everybody who's of a certain age will remember this. You would go to a bank. It doesn't matter where on the globe, whichever country you are in, you'd go to the bank. And with some real probability, they would say, ‘I'm sorry, our systems are down.’

EGS: Or you would go to the counter at the airport. And many or all of the screens would be blue. Things didn't work. There were those drafty old windows that didn't close. They worked some of the time, maybe 90 percent of the time, but they didn't really work. You couldn't rely on them. I realized very quickly that trying to build AI at an age when we did not have basic stuff figured out, the basic rails were not functioning, was a luxury that I couldn't afford. And the main thing I needed to do that the world badly needed was to actually build systems that work. And so I pivoted away from AI, into computer systems, and ended up looking at building very low-level software, like operating systems that work and never crash, operating systems that can be extended or distributed systems that have a life of their own, that live in the network and communicate with each other. So that was the big journey that that I launched into at Princeton.

“I remember coming away from that very first workshop thinking it’s going to be kind of tough for the Ethereum community to solve this problem. It’s not easy.”

ML: Did you on a personal level find it easy to adapt to a life in America coming from Turkey?

EGS: Yeah, it was pretty straightforward. I think kids at that age are the same everywhere. They have the same worries and same concerns, anxieties, et cetera. The U.S. versus wherever else distinction wasn't that big. The things I found hard in Princeton itself is, it’s an exciting place, but it's also a tough place for many people. And then of course being completely independent, leaving a big family behind and being alone, those are very hard things that everybody has to deal with at some point in their lives. I think those played a much bigger role on my psyche than anything else.

ML: And then did you go on to grad school?

EGS: Yeah, so towards the end of Princeton it was clear that I wanted to study more. I wanted to learn more. I wanted to do masters-level studies. The master's programs were very expensive and the Turkish lira was taking a huge dive at around that point. In my junior year at Princeton we woke one day and the lira had devalued by 30 percent essentially overnight. So a master's education was not going to happen for me.

EGS: But then I realized that if you apply to PhD programs and you get in, they're combined. You first get a master's degree, then you go on to get your PhD degree and they pay you along the way. So I was like, ‘whoa, this is cool.’

EGS: And that's how I got sucked into academics and stayed in it for the next 30 years.

ML: Did you stay at Princeton for that or did you go somewhere else?

EGS: No. I went to the University of Washington for my grad studies and after I graduated I went to Cornell University as an assistant professor in, I think, 2001.

ML: We've spoken about this a little bit briefly before, when we’re sort of leading up to the crypto era you had already been messing around with certain use cases, right?

EGS: Yeah, I built a system called Karma for ensuring that people who participate in peer-to-peer file sharing networks don't just leech. They don't just take resources from the network, but they also donate resources. So everybody was downloading files, nobody was putting up files for upload. And so my solution to this was, what if there was some magic Internet money that nobody controlled that you needed to use to download files? And if you ran out of it, then that would put an end to your leeching ways and you would now put up some files to get your Karma back.

Karma

EGS: So I ended up building the system, and the way Karma was created was through proof of work, but it was different from Bitcoin.

EGS: It did not have the brilliant insight that Satoshi [Nakamoto, inventor of Bitcoin] brought. He used proof of work as part of the consensus protocol. I didn't have that. And the proof of work in Karma was one time. So you do the hard work one time, and then you collect some coins and then after that, you can use those. So that's how that system came about. It's a fun system, it's cited very widely among academics. But it obviously did not turn into Bitcoin.

ML: What was it like for you when you first saw the Bitcoin white paper? Do you remember who introduced it to you, or how you came across it?

EGS: I came across it online and I thought, ‘hey, this is fantastic.’ You read the abstract and you’re like, ‘whoa, this is a person after my own heart.’ What a brilliant, elegant protocol overall, this is just a fantastic insight. Now I I'll be super truthful with you and tell you that the second thing that I did is exactly the second thing that any academic does when they see work like this. That was, ‘I wonder if he cites me?’ <laugh> He should really be citing Karma but Satoshi was not classically educated. I think he was a self-taught person. So he was citing some forum posts and so forth, but was not really in command of the academic work in the space.

EGS: So if he doesn't cite me, that's fine. And then the next thing, of course, any academic does, which I did as well, is, well, he makes a lot of claims, are they true? You know, there are no proofs and I have some gut feeling for why they might be true, but you need proofs. And so I started thinking about, okay, what are the attacks against this system? What are the assumptions that are lying deep inside of this protocol? So that ended up putting me on the path where we ended up looking at the security of the Bitcoin protocol.

“And if you don’t have a proof, then all you’ve got is a bunch of badly written code.”

ML: When you first discovered the white paper, was that right at the beginning of late 2008, early 2009?

EGS: No, it wasn't until 2011 or so.

ML: And once you started looking at the Bitcoin security protocol, how did you feel about the system in general? Obviously it's groundbreaking. When you kicked the tires what did you find?

EGS: I ended up kicking the tires for real with a postdoc, Ittay Eyal, who was visiting Cornell at the time. Ittay really liked cryptocurrencies and Ittay and I had been talking and he said, ‘well, look, there are these claims in this paper and there's a particular way that this mining protocol works. I wonder what one could do to trick this protocol to essentially make it behave not in the way that Satoshi intended?’ That started us down a path.

EGS: And the two of us started looking into whether or not Satoshi's claims were true. And it turns out that he had a brilliant insight, a brilliant protocol, but some of the claims that he made with regard to the security of Bitcoin were over generalized, over broad and flat out false. So that was really interesting. There had been academic work proving those claims.

EGS: So we ended up finding ourselves in a position where we discovered a way to trick the Bitcoin protocol into giving you more coins.

ML: Was the protocol changed after the issues that you brought up?

Read the selfish mining paper here

EGS: No. The protocol has not been changed one bit after we came up with this notion called selfish mining. Bitcoin is vulnerable to selfish mining still. And instead what changed around the protocol was a better understanding of the limitations and therefore a better understanding of the feasibility envelope and the unwanted things that are outside the envelope. These days you find the Bitcoin community gets very upset when miners get too big. They understand that the limit is 33% and no miner has at least advertised that they have reached that limit since our results were made public and accepted by the community. So people are fixing this problem simply by resorting to social means.

ML: You're probably better known for your role in Ethereum and have played a big role there over the years. What was it about Ethereum that appealed to you when Vitalik’s white paper came out?

EGS: The fact that the scripting language could be modified to enable arbitrary computations seemed like a fascinating new ability. I had worked in grad school on extensible systems. These are operating systems where you can inject new code into the OS. And Ethereum was essentially allowing anybody to inject new code into the blockchain. And that was a very exciting ability. A lot of the tricks that they used were quite prosaic. And so I thought, ‘hey, this is something I understand quite a lot about because I worked on it in grad school.’ I thought, ‘they're not using any of the advanced science that they could be using, but they are doing some interesting things.’ The next thought is, ‘I wonder what the community will make of this? I wonder what people will use this new ability to build?’ And here we are six years later, we're seeing an explosion of smart contracts now.

ML: To jump ahead you created Avalanche. What was that moment when you realized you wanted to do that? Was it a limitation in Ethereum or did you want to see something more robust in the smart contract world?

EGS: Sure. I think it was in 2016. So first of all it was Vitalik and Vlad [Zamfir] who told me about Ethereum 2.0 back very early, back in 2015 or 2016. We had a big workshop, a multi-day workshop talking about ways to build proof of stake networks that can work and how to make them scale. I remember coming away from that very first workshop thinking it's going to be kind of tough for the Ethereum community to solve this problem. It's not easy. The problems are really big. And the things that you need to do or will want to do are rather challenging things that nobody knows how to solve. And so it was clear to me then, as soon as I found out about the difficulty bomb, that it would to be postponed. Remember the difficulty bomb that would facilitate the transition into proof of stake? [The difficulty bomb is the way Ethereum will stop allowing blocks produced by proof of work so that the network can shift to proof of stake and stop blocks from continuing to be mined on the proof of work network.]

EGS: So when I spoke to them it was clear that what the world needed was a much faster substrate, a much more green, much more environmentally friendly, sustainable substrate for making decisions. A consensus protocol that's much better than the ones we had. So I started working on it.

EGS: My student at the time, Kevin Sekniqi, and I started looking into simulating these protocols, and we saw in simulation that these protocols are amazing.

EGS: They converge incredibly fast and they're they can accommodate millions of participants. This is stuff that’s really significant in the general space of distributed systems, but what we lacked was a proof for why they should work. So in 2016 we already had insights about the protocol behind Avalanche. We knew that there was something amazing there, but we didn't have the proofs that it would actually work every time. Many other people would've gone to market with that, many other people would've launched the system. But we chose to stay put and just banged our heads against this proof problem. That is, ‘does this actually work? Can you prove it?’ And if you don't have a proof, then all you've got is a bunch of badly written code.

EGS: Then we got the message from the Team Rocket folks [who he partnered with] that said, ‘Hey, this is, is something that we've been working on. And here is how we are analyzing these kinds of systems.’ And so that was the coming together of the minds between the people who provided the proofs for the paper and the people who had been working empirically on these systems

Read more here

ML: And to fast forward to today Avax is a top 10 cryptocurrency, if I'm not mistaken. [It was actually 12th as of this writing, with a market cap of $21.3 billion.] You're seeing a lot of gaming applications that are using Avalanche right now?

EGS: No, not that many gaming applications. It's mostly defi. So there's a lot of value in the total value locked in Avalanche. And there's a lot of NFTs and there's some games that are really cool.

ML: So basically what you wanted to do was a blockchain proof-of-stake system that was out of the box, is that a good way to say it?

EGS: I wanted to build the world's biggest platform for digitizing all of the world's assets. My vision is that everything we buy and sell has to be backed by a token and asset identifier on a blockchain. This certainly goes for every virtual asset, every financial asset, every piece of commercial paper, every single loan, et cetera, that should be on a blockchain. But it also goes for all of the machines and so forth that we buy. My car should be something that I obviously authenticate on the blockchain. Our mission is to digitize all things. And none of the chains that were in existence at the time can handle the load. They don't have either the technical background, the consensus protocol or they don't have the architecture. They don't have the ability to expand, the ability to accommodate multiple different kinds of chains under the same rubric. So that's what Avalanche brought. It didn't just bring in the consensus protocol as fast, but it also brought in a new architecture based on subnets that allows multiple people to do different things in a fashion that's isolated from each other.

ML: Are you keeping an eye on Eth 2.0 and how that's progressing?

EGS: Frankly, I'm not following Eth 2.0 closely. I wish them much luck. It's very hard for me to follow because the specifications change a little too frequently. And every time I read the spec and I quote it, the Ethereum maxis get very upset at me saying that I am purposefully misquoting something. And then when I say, ‘well, look, I just read this two months ago’ and they say, ‘it changed like a week ago, you should have known that.’ And I decided it's better if I don't follow this. Power to them, I hope they manage to do something great. I think I understand the specs, so we'll see how it turns out, but I do hope that they come up with something really awesome.

ML: Congratulations on Avalanche. It's an amazing accomplishment. I like to ask people at the end of these interviews if there's somebody in the crypto space you admire that you'd like to know more about who I could reach out to for a similar kind of interview.

EGS: That's a good question.

EGS: How about Do Kwon, do you know Do Kwon [the co-founder of Terraform Labs]?

ML: No I don’t but I’ll reach out to him.

EGS: He’s with Luna in Korea. He got served by the SEC for creating a security. He's got a whole bunch of cool things, interesting things on the Terra blockchain.

ML: Okay. Awesome. Well, Gün, thank you so much. It's always a pleasure to talk to you.

EGS: It's great to see you and thank you for having me on.