Q&A with ETH 2.0 Core Developer/Ringmaster Tim Beiko
We hear from the man herding all the cats as Ethereum core developers near the historic switch to ETH 2.0. But more than that we learn of Tim’s punk-rock t-shirt business, how a house painter who hoarded Iraqi Dinars introduced him to Bitcoin and what Tim hopes for a year from now in Ethereum.
Tim Beiko has always been a builder, whether it’s starting a punk-band t-shirt business in high school or marshalling the core Ethereum developers in the biggest overhaul of the blockchain since it went live in 2015. Tim earned a business degree at university in Quebec province in Canada and was introduced to Bitcoin by a guy who worked for the house-painting business he started after the t-shirt thing fizzled (told you he was a builder). Then he watched the DAO blow up in 2016 as he was just getting into Ethereum. He noticed, however, that people really wanted to use Ethereum during the initial coin offering craze of 2017 — even if it was not very user-friendly and transaction costs could skyrocket — and decided to help make the underlying infrastructure as sound as possible.
Beiko (pronounced “bake-O”) gives an update on the plan to shift the current version of Ethereum to the new version — a process known as “the merge.” A few other terms to know — the current version of Ethereum is called ETH 1.0 and is secured by proof of work, the same consensus mechanism used by Bitcoin. The new Ethereum is called ETH 2.0 and will switch to proof of stake, a system that relies on users’ pooled Ether to secure and manage the blockchain. The base layer of Ethereum is known as layer one, or L1. It tends to be slow and costly but very secure. To alleviate that, so-called layer 2 protocols, or L2s, have been developed that are much faster and cheaper than L1 yet have the same security of the Ethereum blockchain.
Matt Leising: First of all I think a lot of people would love to get an update on what you're doing as you're coordinating developers in the Ethereum community who are working towards ETH 2.0. Which is a huge task in and of itself, but you are kind of the guy to ask about how things are going and where things stand at this moment. Could you give us an update?
Tim Beiko: Sure. In early October we had complete specifications for the transition so we got everyone together from both the ETH 1.0 and ETH 2.0 teams for a week and worked on getting prototypes of this done. We managed to get kind of a short-lived test net up and going. We started on proof of work and transitioned over to proof of state. So that gave us good confidence that the general specification works and the general approach is sound. But obviously when you're working on that we kind of found all these little edge cases and tweaks that we wanted to do to bring it to production.
TB: What we're doing during November is we're trying to have these very short-lived test nets. That's where we'll just launch them, check that things work and then tear them down. The hope is that before the holidays we can have something a bit longer standing that we can leave up so application tooling, infrastructure and whatnot can use this and see how it works.
TB: One thing in parallel to that is we've started reaching out to the community and trying to explain to them what the changes to proof of stake entail for them, trying to understand the impact and then the applications and whatnot, just to make sure that when the code is actually ready we don't catch anyone by surprise.
ML: Okay. So this transition from ETH 1.0, which is what is live now, to ETH 2.0 is called the merge. Do you have a good idea of when the merge might take place?
TB: Next year, for sure. Realistically if we have a test net done by the end of this year there will probably be some small tweaks we want to do and I'm hoping that around February-ish we could have the code complete.
TB: If we have the code done in February, you can expect the merger a couple of months after that. So that's probably April-ish, maybe May. It's hard to give it a specific date just yet because if we find a major bug or something that takes us three weeks to fix, obviously, you know, that delays things by three weeks.
ML: It seems to me that this is the biggest change to Ethereum since it went live in 2015. Do you agree with that?
TB: Yeah, a hundred percent.
Keep up with what the Ethereum core developers are doing here
ML: Even back then, there was no network that was live that you had to worry about. And here there is. ETH 1.0 is live and going every day, every minute, and now you're trying to switch it over to this other one.
TB: I think that's something that's really interesting. Not only is the network live and has hundreds of billions of value, or maybe trillions if you count all of the tokens built on top, there's also like a massive set of validators on it too. There's 250 nodes.
TB: But there's probably one or two orders of magnitude more individual validators compared to the next largest proof of stake system. So it's hard that the system is already live, but it's also hard because we want this to be accessible to as many people as possible, but that obviously slows down because you need to coordinate across more people. It's not five validator entities that we can just reach out to and tell them, ‘run the software and I'll be on a zoom call while it's happening.’ It's really this massive set of users distributed across the world.
ML: To take a really big step back, how did you get here, to coordinating this enormously important change to Ethereum? You grew up in Canada. Can you tell me a little bit about where you grew up and what life was like for you as a kid?
TB: Sure. I mean, pretty typical. Grew up in Canada. Didn't move a lot. I'm from Montreal, but I just recently moved to the west coast. I've always kind of started things. I started a t-shirt business when I was a teenager, then I ran like an actual painting company for a couple of years.
ML: Like painting houses?
TB: Yeah. I realized painting houses that physical services don't scale well. You could probably manage like 10 or 20 painters yourself, but that if you need to hire a manager you just don't get a lot of scale. I got interested in technology at some point partially because of that.
Punk rock t-shirt guy
ML: What did the t-shirts you sold say on them? What kind of t-shirts were they?
TB: I used to be a big punk rock music fan, so it was a t-shirt brand – we would sponsor bands and stuff like that.
ML: Can you name some names?
TB: They were all small local bands. I don't think any of them would still exist today.
ML: So you weren’t selling Iron Maiden t-shirts?
TB: No, unfortunately didn't make it that big.
ML: What did your parents do when you were growing up?
TB: My mom's a doctor, a general practitioner. My dad used to be, well, still is an engineer and used to work kind of in sales for fiber optics. He didn’t like it too much, he always liked building houses so at some point he quit and now he builds houses.
ML: So you realize that inventory-based businesses don't scale very well. It's pretty hard. Were you into computers at the same time when you were a kid? Into video games?
TB: I was into a lot of online video games. I was in the top hundred of this game called AdventureQuest, or BattleOn, which was like popular when RuneScape was. I was pretty high ranked in Neopets as well. When I was a kid I would wake up at 5:00 AM and go play games on my computer. And I’d also find ways to hack them. I remember when Animal Crossing came out on GameCube. You could change the date on your game cube to speed things up. So I would go and plant a bunch of crops and then shut down your game, go to the settings, move it forward by a week, and then you can harvest your crops.
Wonder if this player hacked the game like Tim…
ML: That's great. Did you gravitate to anything in school? Were you a good student?
TB: I was okay. I kind of did the strict minimum that would get me by. I didn't have a particular thing. I like business and just doing projects.
ML: How old were you when you started the t-shirt business?
TB: Probably 15 or 16..
ML: And then you got into house painting a few years later?
TB: Yeah. definitely I was 17 because I remember -- 18 is the legal age in Quebec, a lot of things were complicated by the fact they had not quite reached 18.
ML: And were you headed towards college at this time? Or what was your plan?
TB: Yes, but Quebec [province] is weird. Basically high school is a year less than in the rest of Canada and the U.S., and college is also a year less. And you have these two years where you go to a separate school, which is really good if you want to be a nurse or something like vocational training. But for people who do want to go to college, it's this weird transition where it is like a university setting, you're not in high school, but it's also much less rigorous than university.
TB: So I did that and there was a massive student strike in Quebec, I think our schools were closed for like six to nine months. And at that point I basically dropped out of the physical school and signed up online to finish. But it was kind of cool because not having super high demands and then being able to do the coursework online was much better.
ML: Did you do that secondary education in computer science?
TB: No, for business. I didn't choose computer science until much, much later. Basically fast forward, I did that for a couple of years and towards the end of that I started to get interested in tech. And I had no idea how to do tech. So I had finished those two years and was signed up to go to business school at university and a friend of mine and myself decided instead of doing that, we were just trying to start a tech company. I would teach myself to code and she would teach herself design. So we did that for about a year. The company failed, it didn't work out, but I did learn enough programming to know how little I knew. So after the company failed I decided to go back and actually do computer science rather than business.
ML: What year are we up to about here?
The Iraqi Dinar leads to Bitcoin
TB: I think I started my degree in 2016, something like that.
ML: Had you come across crypto before, or was it through this course that you got introduced to it?
TB: The person who introduced me to crypto was actually one of the painting guys who worked for me. He did this as a summer job and he was always into all these weird schemes. The U.S. had obviously invaded Iraq [the 2003 invasion] and there were currency sanctions on the Iraqi currency. So he would buy bulk Iraqi dinars on E-bay and literally store them under his mattress. He would buy them and keep them in the hopes that one day they would go up when the U.S. left Iraq. And he always had these weird schemes and then the other weird thing which at the time felt equally weird was Bitcoin.
TB: So he was the person who told me about Bitcoin. I think I bought one Bitcoin when he told me about it, this must have been like 2014, it was about a thousand bucks. I just kind of bought the Bitcoin and forgot about it.
TB: I basically stayed in touch with that guy for two years and he would bring it up, but I wouldn't think about crypto between the times where I have conversations with him. And he mentioned Ethereum, you know, quite early on, I don't know when exactly either before launch or right after the launch.
TB: Then again, I didn't pay much attention to Ethereum, it felt like another overly complicated thing. And then when I heard about Ethereum again was when the DAO happened – but when the DAO was in the fundraising project, not the bug.
TB: I bought Ether the day before the DAO got hacked because I wanted to, and I think I did, contribute a bit to the DAO.
TB: After the DAO happened it felt pretty dead for a very long time. And at some point, you know, my coins were worth so little that it was not even worth selling them.
ML: So not diamond hands but like it's just not worth the trouble.
TB: Exactly. So I still kind of followed Ethereum but very loosely until I think it was late 2017. You started seeing some projects building on Ethereum. The ones I remember were Golem and Melonport, there were a couple others. And that felt exciting and was around the same time Zcash too. It felt like kind of a breath of fresh air in a crypto space that you had Zcash, which was fundamentally trying something new. Now you have projects using Ethereum. And that's kind of when I started paying attention again.
ML: And the price was obviously also going to a record too, right?
TB: This was before that. I think I bought it at like $20, roughly, and it hit like $20 again in March of 2017 or something like that, and I remember thinking like finally, you know, I'm even.
TB: At the time I was I was studying CS, I was focused on artificial intelligence and still thought AI was a safer career path than crypto. At that point it still felt like Ethereum could just straight up go away. Right? Like it was not guaranteed that this thing would stick around. You know, like $12 is much closer to zero.
ML: I totally agree. We're kind of skipping over the run-up to the record price to $1,400 in early 2018, but then there was a huge crash of course. And it was down to like about a hundred for a really long time after that.
TB: Actually, that's when I got involved more. In the summer of 2017 there were all these ICO's and the thing that I kind of figured is, ‘you know, even if most of them turned out to not work and a large portion of them were scams, it showed that there was massive demand for Ethereum.’ Right?
TB: And at the time using Ethereum was a terrible experience. When there were these ICOs the transaction pool would get clogged up for days. Gas prices, people think they're bad now cause they're high, but they were just straight up unpredictable [then]. And that's what I realized, that I would really like to work on the protocol itself.
ML: You wanted to work on the pipes of the system, make sure that the infrastructure was sound.
TB: Yeah. And then the thing I realized also is I was not even close to technically good enough to work on the code itself. At best you could describe me as a junior programmer and that would be a generous description. So over the next year I was interested in spending more time contributing.
“I’m feeling pretty good about Ethereum right now”
TB: And eventually ConsenSys was looking for a product manager for their protocol team.
TB: So that felt like a perfect fit. I’m comfortable being a product manager for very technical things but not being the engineer or the researcher myself.
ML: So taking that overview that you've had over the years into account, from back then in 2018 and 2019 at ConsenSys to now, how do you compare what the tech was like just a few years ago to now in terms of evolution and innovation and things like that.
TB: I'll focus on the protocol layer before the applications, but the protocol layer, I think that the mania in 2017 kind of caught everyone by surprise. And we spent a lot of 2018 and 2019 just patching things up. Making things work smoothly, making clients work more reliably. It was really only in late 2019 and 2020 that we started thinking about big new features.
TB: Especially on the ETH 1.0 side there was this thinking that, ‘oh ETH 2.0 will solve all our problems and we don't need to do these radical things.’ I remember it was at Devcon Prague [in November, 2018] we had a conversation about this and we realized that even if ETH 2.0 comes and delivers everything it's still a couple of years away and there's probably some pretty major things that we need to do to keep Ethereum sustainable.
TB: A lot of the things that we laid out there are still things we're working on today. So 1559 was one of them that we got done. The other big one is basically stateless. So the idea that the state of Ethereum grows unbounded and that makes nodes harder to run. And then the third one is just dealing with historical data. The fact that nodes have to keep all the data since forever is obviously a big strain on the network. And given that the data doesn't change anymore, there's better ways to like distribute it.
ML: What's your biggest concern about Ethereum right now in general?
TB: I'm feeling pretty good about Ethereum right now. Again, my views are at the protocol levels. The big problems we have is transitioning to proof of stake. I'm pretty confident about that. Dealing with state growth. I think we're on the right path. Like it's not going to happen tomorrow, but smart people are working on it. If I could accelerate something it would be better tooling and migrations around layer 2. I think the fees on Ethereum are quite high right now. And you know, I think that's something that's easy, at least for people like me, to feel like it's not part of our job because it's not the protocol itself, it's supply and demand. But I really want to make sure we're doing everything we can to push as much stuff on to L2 and to have people have a smooth experience of people using L2s. If you go on Coinbase and you buy a hundred dollars of Ether, you should probably never transact on layer one. Right? If it costs you $5 or $6 every time you send something 10 times, half your money's gone.
ML: Did you hear about this thing Constitution DAO?
TB: I heard about it. Yeah.
ML: I was thinking of joining and I was going to give him like $20 in Ether and then the gas fee was going to be $85. And I'm like, no, I'm not doing that.
TB: Yeah. I've actually done a couple of transfers where the gas costs is worse than the actual transfer amount. So yeah. It sucks.
ML: Doesn't that concern you? You think that L2s and ETH 2.0 is going to greatly alleviate that?
TB: Yeah. I mean, L2s work today. I was using Optimism and Arbitrum this weekend, transfers were like a couple dollars on each. It's important that at the protocol level we keep working on sharding. I'm not as much concerned about the tech being possible, but I'm more concerned about the deployment and the adoption and making sure that people are nudged in that direction. Like obviously if people want to use L1, feel free to, but we should make it clear that if you're not planning on sending a hundred thousand dollars you should probably just use L2. We should make it easy for applications to deploy on all the L2s and just make sure that it becomes a first-class citizen.
ML: Yeah. That's great. So where would you like to see Ethereum in a year from now?
TB: I'll be happy if we're on proof of stake. If we had something where the transaction count on L2 is at least higher than L1 – but ideally 10 or a 100 X higher than L1 -- or you can look at the usage, not necessarily the value transferred, because I think you might still see L1 be the main spot for large value transfers. But if most of the transaction-denominated usage is on L2 and we're on proof of stake by the end of 2022, yeah, I'm happy.
ML: All right, Tim, this has been awesome. Thank you so much for your time and sharing your backstory with us. It's fascinating.
TB: Thank you for having me.