Arbitrum Co-Founder on ETH Scaling Plans w/ Steven Goldfeder

0:00Your favorite neighborhood spot grows with Square. Indeed, my favorite neighborhood spot has quickly become Todd Snyder in Williamsburg. Todd Snyder is one of my favorite menswear shops and has supplied me with all the clothes I have needed this quite hot summer. Every business has different goals, but Square is the business platform that supports them all. From opening a new location, selling something new, or just expanding their reach. Indeed, I've seen it with Todd Snyder. In Square, also, you can get real-time insights, so don't wait for end-of-day reports. Go to square.com forward slash go forward slash realvision to learn more about how your business can grow with Square.

0:43That's S-Q-U-A-R-E dot com slash G-O slash R-E-A-L-B-I-S-I-O-N.

0:57Hey everyone, if you like this podcast, go behind the paywall to get privileged access to the smartest minds in finance. Join the Real Vision community and learn how to become a better investor. Visit realvision.com slash RVpod and use the promo code podcast10, that's podcast10, to get 10 % off our essential membership for the first year. Now, to the top analysis of today's crypto markets.

1:31What's up, everybody? Welcome to Real Vision Crypto Daily Briefing. I'm Ash Bennington. Today, I'm joined by Stephen Goldfeder, co-founder and CEO of Offchain Labs, the company that's developing Arbitrum to talk, of course, about all things Arbitrum. Stephen, welcome to Real Vision. Thank you. Great to be here today. Well, it's a pleasure to have you with us. So much to talk about. We're going to dive into all of that in just a second. But let's take a look at some price action here. We are in the middle of a bull run, ladies and gentlemen. Bitcoin right now trading at$30 ,425, trailing 24 hours.

2:04Well, it's down about half a percent, but trailing seven days. Bitcoin up over 15%. Same story over with Ethereum trading right now at$1 ,889, trailing 24 hours. It's basically flat, trailing seven days. Ethereum is up about 10%. Lots to talk about, but I want to also touch on Arbitrum right now trading on my screen at$1.16. Stephen, welcome to Real Vision. So much to talk about here. Let's get folks started with a little bit of general background. We're going to dive into the weeds in just a second. But for those who are relatively new to this space, what is Arbitrum? And give us a thumbnail sketch of the Ethereum scalability issues, challenges, and opportunities for L2s.

2:51So Arbitrum is a layer two scaling platform for Ethereum. There's a lot of big words there, but basically what it means is it is the way to scale Ethereum. It is the way to bring retail throughput and the capacity that really mainstream adoption needs to Ethereum. And this is not something that is a fringe belief in the Ethereum community. It's not something that only layer two builders believe. It's really something that the entire community believes. A few years ago already, Vitalik Buterin put out a blog post talking about Ethereum's roll-up-centric roadmap. And Arbitrum, as others are scaling technologies, are roll-ups.

3:27And the idea is that everyone's on board, both Ethereum Layer 1 and Ethereum Layer 2 folks, that the way to scale Ethereum is to get actual users and retail and mass adoption onto Layer 2s. And the real awesome thing we do is we take the security and decentralization from Ethereum and we bring that to Layer 2s. So together with the scalability of Layer 2s, you have a decentralized, secure platform that can scale to the demand that Ethereum has. So for folks who are relatively new to this space, I just want to give them a sense of context of the scale of Arbitrum because I think it's important for people to understand.

4:02First, I want to take a look at a chart from DeFi Llama. This is a table that you're about to see showing that right now Arbitrum is the fourth largest protocol by total value lock. This isn't just L2, folks. This is everything in terms of TVL on DeFi Llama. Second, I want to take a look at a chart, or rather I should say a table by L2Beat. This tracks the size and scale of L2s. As you can see on this table, Arbitrum currently number one by TVL. Again, it's the same number, about 6 billion. These are significant scale that we're looking at here. Stephen, talk a little bit about the adoption of Arbitrum and why, in your view, this protocol has been so successful.

4:46So Arbitrum is actually bringing real scale to users today. And that's something which, you know, early on, there were a lot of different protocols that were building towards a lofty and grand vision, and Arbitrum was chief among them. But Arbitrum was actually able to deliver that and still able to deliver that today. Arbitrum, among optimistic layer two protocols is the most advanced that has the security mechanism, also known as fraud proofs, fully built out and functioning in the system today. And that's half the story. So the technology is just very advanced and mature and secure. And the other half of the story, though, is the is the ecosystem itself.

5:22And it's a bit of a circular question because I guess the question is, how did the ecosystem get here? And I'm answering it with, well, there's a strong ecosystem. But that's actually how crypto kind of works, where a lot of what crypto does is enables really good, seamless interactions between different protocols. And you can just spin up and build and build a great experience that taps into other protocols. And one thing about the Arbitrum community, and this goes back to the earliest days of the Arbitrum launch, what we called an open, a fair launch, an open launch. The idea was there was no priority access to people.

5:54Everyone got access to Arbitrum on day one. that was August 31st, 2021 when it went to Maynet, all users entered at the same point and developers had a few months to build before that. And from that very point till today, it's a very inclusive ecosystem, very open and collaborative. And I think that, you know, that's a very big part of it. So it's the technical merits and the Arbitrum is by far the most advanced technical, you know, achieving layer two that exists today, bringing scale to users, but also a very broad and welcoming and diverse ecosystem that allows and encourages people to come in and build really awesome experiences.

6:29We should say when we're talking about the token, the token came later. You mentioned March of 2021, excuse me, 2021 more generally. I believe it's March of this year that the Arbitrum token dropped. Talk a little bit about the airdrop, the challenges that were had there and the current state of the token. Yeah. So just to put it in context, so the chain went live August of 2021 to users. And as you mentioned, the decentralization, which is when arbitrage DAO was formed and the governance token was a way to disseminate the governance power of the chain that came in March of this year just a few months ago.

7:03And from the way from my perspective this was about technical readiness. People like to ask the question of why and when did you launch a token then it was really about this is necessary to fully decentralize the protocol and as soon as that was technically ready that was the right time to do it. So basically in August of 21 the chain went live. One year later in August 22, there was the chain's biggest upgrade to date called Arbitrum Nitro, which massively increased capacity on the chain. So it allowed more throughput, more capacity, more transactions, and also at the same time lowered costs.

7:33A few months after that, validation was open to more parties. And today there are many institutional validators in Arbitrum, including names that are familiar probably to many like ConsenSys, Google Cloud, Quicknode, and many others. The Ethereum Foundation are all validating the network and then the next logical step was to decentralize. So what decentralized means, it means give the upgradability of the chain, give the power to make decisions for the chain's future both in terms of the resource allocation as well as in terms of the actual what technology is running on the chain. There is no centralized actor today and Arbitrum has self-executing governance.

8:09So whereas some chains their governance model has a disconnect between the votes and then there are some people with keys that can actually turn the keys and are supposed to listen to the vote. Arbitrum is unique among layer twos of governance today that is self-executing governance so the vote actually happens if the token holders vote hey let's do this technical upgrade that is the on-chain vote is what actually affects that so the token is uh obviously very important to the decentralization is what gives the gives the power to the hands of the community and um you know the arbitram foundation arbitram dao are the uh main players when it comes to the token and the current running current bodies that actually run the chain, whereas Offchain Labs was a launch partner there and really built a lot of the early technology that got us there.

8:56And together with the Arbitrum Foundation, together with Nansen, who helped a lot with actually designing the airdrop, there were a bunch of launch partners that got us to that point in March. Yeah, and there were some challenges there. Talk about the current state. Yeah, so it's interesting you bring up. So there were definitely some challenges early on once the DAO launched. And I'll just, you know, disclaimer that I'm really talking from a community member standpoint because I, at OffChain Labs, we don't have a privileged position via the chain or the token at all. But there were definitely some challenges in terms of some communication, I think, errors that happened in terms of the certain allocations of where the token went.

9:40And in particular, the Arbitrum DAO is unique in the sense that I mentioned before that it actually has a self-executing governance. One thing that it has, it has about 3.5 billion tokens in the DAO treasury. That's number one. It also has a significant amount of ETH because all the fees that come into the chain, some of them are paid out on layer one. That's the cost of the security. But the net fees actually go to the DAO treasury. And there's, you know, just in the few months since the DAOs existed, there's millions and millions of dollars of ETH sitting in the DAO treasury as well, all controlled, you know, by the DAO directly.

10:17There is also a foundation. When this is a foundation, you could think of it as the legs and the arms that represent the DAO. Because sometimes, you know, if you imagine if you want to have a deal, like a business deal with a Fortune 500 companies, they're probably not going to want to talk to a DAO. They're going to sign all sorts of confidentiality agreements and NDAs, etc. And they'll need often a body to actually represent, they can represent on behalf, but also have, you know, a in-flesh counterparty. And there was definitely a miscommunication in some of the earlier docs around how the foundation was funded exactly.

10:54But that's actually been fully resolved today. And recently there was these two AIPs, which are Arbitra Improvement Proposals, which approve the funding of the foundation. actually also include transparency requirements, included vesting. And I personally think that where it ended up was a lot better than it initially was because now the foundation has a really strong mandate and really strong controls, but also the community has transparency and will get information. And the coolest thing though is, this is definitely, I think, a hard time for a lot of people or it was a lot going on. It wasn't the easiest time for the Dow in its early infancy.

11:31But one thing that it set was a precedent anyone looking at the Dow and the foundation and any other players involved can just look to the very first proposal of the chain that actually failed. That was the initial funding proposal, the ratification, they say, I should say, of the foundation's funding. And that proposal failed. Anyone looking to where does the power run in the Arbitrum Dow Foundation can see immediately it's the Dow, it's not, you know, the service providers, it's not the investors that built it. It's not me, it's not Off-Chain Labs. Because remember, again, the first proposal failed and the Dow had its way and got the transparency that it asked for.

12:05So I think looking back, it's actually everything emerged a lot, a lot stronger. And there's a lot of good precedent for where the power lies in this ecosystem. Hey, everyone, we're going to take a quick pause and hear a word from our partners. We'll be right back.

12:21Yeah. And these are some of the challenges and frankly, just the sort of things getting hashed in a truly decentralized world these are some of the challenges we have more generally talking of which obviously a lot of competition in the space you mentioned optimistic roll-ups earlier let's talk a little bit about some of the mechanisms for running l2s the main sort of two fundamental principles are zero knowledge proofs and optimistic roll-ups arbitrum of course uses optimistic roll-ups talk a little bit about the way that arbitrum handles the scaling so the thing that roll-ups all have in common is they put data on ethereum just the transaction data and then they take the execution of that data off chain so when you think of a transaction you can think of it in two ways and there on the one hand it's just a blob of data zeros and ones that don't really necessarily mean anything to you looking at them but then if you look deeper those actually are instructions that say hey add these numbers store this value do this computation and so what we're rollups do is they put the basic data on Ethereum, but the execution of that data they take off chain and have somebody else do.

13:25We'll get to who those are, often called validators. And then they go back and report the result back to Ethereum. And the cool thing is, is they increase the capacity because all that computation they do off chain, Ethereum doesn't itself verify. Instead, Ethereum just verifies when they come back and say, hey, this is what happened. Now, in order to scale, though, what you want is the amount of data we're reporting back, the amount of computation we're doing is much more than Ethereum itself can verify. But we also want Ethereum to give to security here. So Ethereum needs to put its stamp of approval and say, hey, that thing that you did, I actually approve.

14:00I know that you got to the correct result. So you have this problem, and that's what rollups try to do. They try to take the computation off chain, but still convince Ethereum that things are correct. So Ethereum can use its security mechanisms to secure the rollup. And there are two approaches to do this, but they unify in in one reward, which is proving. They prove to Ethereum, say, hey, we'll prove to you that this is the correct result. And by the way, I'm gonna prove it to you in a way that's faster for you to verify than doing it directly. So you're doing all this work that Ethereum miners couldn't possibly verify in the capacity that Ethereum has, yet via these magical proofs, Ethereum is actually able to verify and say, hey, I can't actually run that directly because I don't have enough capacity, but I'm confident that that result is correct.

14:41And there are two different ways that these proofs happen, as you mentioned. One is called zero knowledge proofs, also often called validity proofs. And what that means is when you come back to Ethereum and say, this is the result, I'll give you a validity proof right here and now. I'll prove to you that this is the correct result. And these use zero knowledge proofs, which are mathematical proofs, which are really, really cool and have this property that it's easier to verify the proof than to do the work itself. It's harder to actually create the proof that the prover has a lot of work to do.

15:08But the verifier, so Ethereum can easily verify this proof, even though it can't do all that work. The second approach is called optimistic rollups, and they use something called a fraud-proof technology. And they're optimistic, as its name suggests. How are they optimistic? They come to Ethereum and say, hey, here's the result we did off-chain, but by the way, I'm not giving you any proof at all. And optimistically, they hope that it will get accepted and no one will challenge it. But if someone comes along and challenges them, there's a challenge window of now it's seven days typically. There's a challenge window where anyone can come along and challenge that, and then they'll engage the proving mechanism in the challenge case.

15:41So they're optimistic. They don't do the validity proof. They don't do the work up front. Optimistically, they do much less work. But if someone, if a challenger comment says, hey, I disagree with that, there's a very efficient proving process via these fraud proofs that will resolve that and get to who is, you know, who is telling the truth and what is the correct state of the role. Yeah. So computer science, like economics, is all about trade-offs. It all comes back to, in some ways, is Vitalik's trilemma between scalability, security, and decentralization. How do you think about those trade-offs between zero-knowledge proofs more generally and optimistic roll-offs?

16:15Absolutely. So if instantiated correctly, both of them will inherit the security and decentralization of Ethereum. So it really comes down to the third, and that's how the layer two paradigm is designed, which is the security and decentralization come from layer one. Okay, but how do we scale? Because you still need to scale, even if your value, say, security decentralization come first, users still need to use this for it to be valuable. So the scalability comes from layer two. And they really differ, I would say, in the scalability. But you can go a little bit deeper there. So I would say the planes that I would use to compare ZK rollups and optimistic rollups are one is cost.

16:50So it turns out that the cost of proving ZK rollups is actually significantly higher than the cost of proving or validating optimistic rollups. Because the ZK rollups, we mentioned that the verification of these proofs is relatively cheap. But the proving cost is actually much, much higher. And that's a cost you don't actually see on chain. On chain, the proof just shows up. But often you'll have to have someone in the background with exotic hardware in some cases, or a lot of hardware and a lot of work. And it often takes many, many hours to actually generate one of these proofs. So there's a cost aspect where optimistic rollups outperform.

17:25And you can look at actually, the nice thing is, this has always been theoretical, but now there actually are some ZK rollups that are live. you can look and the cost of transactions are higher on ZK Rollins generally and often some of this cost of the proving is even subsidized and not even showing up there. But then there's a second one which is compatibility. It turns out that optimistic roll of design allows you to use Ethereum's Geth stack directly. So actually in the early days of Arbitrum, Arbitrum had a custom node and in that Nitro upgrade that I mentioned before, it started using Geth directly and doing its fraud proofs a little bit differently.

18:01And what that means is you have deep, deep, deep compatibility with Ethereum. You know, not only on the surface, but also under the hood. It's actually running the same software on a very deep level. It has this like whole proving mechanism attached to it that Ethereum doesn't have, but it's fully compatible from a software perspective, from an integration perspective. And that's something which, you know, we've seen in ZK rollups don't have today because the difference is the model of computation. Optimistic rollups run basically on the hardware of your machine. You run get directly, you run the software directly.

18:32ZK rollups need to be translated into the language that these proofs run in. And these are called circuits or arithmetic circuits. And often, to give you a very concrete example, in an optimistic rollup, if you want to add two numbers, A and B, you will add those two numbers to your machine just like it would generally. In a ZK rollup, you would have to do often thousands or tens of thousands of different curve operations or operations in these circuits, these cryptographic operations, to translate to a different computation model. So the point is that the ZK proofs are very efficient, but they're relatively efficient to verify, but to generate these proofs, they're expensive.

19:11And to actually translate to the language that these proofs have, you have incompatibilities. And you'll see this often the ZK relaps have, you know, a number of features or pre-compiles that they don't support today. And that's another one. The third one is maturity. The technical maturity, Arbitrum has been running these fraud proofs on mainnet since its launch, actually since the launch of its test net nearly three years ago now, whereas Ezekiel Willems are first coming to market now. And they're doing great things, but they're just, on these three planes, cost, compatibility, and maturity, they're pretty well behind.

19:46I should say you have a PhD from Princeton University in computer science. Notwithstanding that, you've done an amazing job of explaining this to a level that business folks who don't have backgrounds like me in computer science can really understand. And I think it's very helpful for people to understand what's actually happening beneath the surface. One of the questions we hear at Real Vision, I think someone internally just brought this up a couple of weeks ago, is this question of, hey, look, there are all these great things that are happening on the Ethereum network right now. Costs are dropping in terms of gas fees.

20:16scalability is increasing why do we need l2s if ethereum is beginning to build out some of this functionality themselves they're expanding the core functionality of the l1 talk a little bit about the case for l2s going forward so a few things there one ethereum is actually in a few cases expanding the capacity and the feature set of l1 to support l2s so one uh example of this is something called EIP 4844, which is a proposal. Remember I said that what rollups do, they put all the data on Ethereum. It turns out actually that's the most expensive part of certainly Arbitrum. So if a transaction costs 10 cents, the majority of that 10 cents is actually just going to store the data in Ethereum.

20:57So Ethereum is doing work, the Ethereum community, and we had option labs are part of this. So besides for Arbitrum, we also developed Prism, which is the leading consensus client for Ethereum. So we're involved in both layers of the stat. But one thing that we're working on as a community is this data, cheaper data on Ethereum. But that's very much targeted at Layer 2s because it goes back to Vitalik's vision. It's not just my, it's not the L2s versus Layer 1. It's not like the Layer 2s and Ethereum are battling on this. Everyone in the Ethereum community is actually on the same page that, hey, we're going to move everything to Layer 2s.

21:31You have the Layer 1 development also gearing itself towards making things even cheaper for Layer 2s. And I think the other thing in a more like ecosystem, you know, broader view of things is, you know, things are probably, you know, suppressed today or more suppressed than they were at some points and maybe a year or two ago, if you look at different markets like NFTs and the like. And, you know, if you believe that, or if one believes that the market's never going to, the demand will not increase from here, or sort of like, you know, things are just going to stay stagnant, then you might say, well, we're doing okay today.

22:03But, you know, we believe very strongly and we're seeing increased signs of adoption from more Ethereum and crypto native projects, but also from big enterprises that were very, very early on in a multi-year or multi-decade adoption cycle. And we believe the demand is that it can keep and keep and keep going. Actually, I'll take back to one of the early experiences of us at Offchain Labs when we were raising capital in 2018 in our initial seed round. You have these people asking this question and saying, do we really need scaling today? And And I remember one investor saying, I have also 10 problems.

22:37And scaling is number 10 on that list, if at best. And the problem is that things come very fast. The prices get high and people are like, where are you guys? We need you right now. We need the scaling right now. If you don't put the time, the effort to build when things are more suppressed, you won't be ready when things begin to pick up even more. Again, we're already seeing very, very significant adoption. Layer 2s are like Arbidrum are reducing user costs today. but I think it's going to only become more and more and more important over time. And when costs are low on Ethereum layer two, that's the best time to move your app and bridge your funds to, when the costs are low on Ethereum, it's the best time to actually use the bridges and move everything to layer two.

23:18So when the costs of Ethereum start to rise, you're already in that protected zone where the costs are going to stay lower. Hey, everyone, we're going to take another quick break and hear a word from our partners. We'll be right back to the Real Vision Crypto Daily Briefing.