Offchain Labs co-founder Steven Goldfeder is betting his lot with Ethereum.
One of the biggest trends in crypto over the past couple years has been the growth of Ethereum Layer 2 blockchains such as Arbitrum and Optimism. Rather than trying to compete with Ethereum head on like bespoke blockchains such as Solana, Cardano, Avalanche and the like, Layer 2 chains aim to be complementary with Ethereum. Their ultimate goal is to help scale the blockchain, which can only process a few dozen transactions per second and is subject to high fees, without compromising on its security and decentralization.
In this discussion we cover how Arbitrum compares to other Layer 2 and Layer 1 platforms and the ways in which Arbitrum helps scale Ethereum without adding too much centralization, and what is next on the product’s roadmap.
Forbes: What do you see as the big differentiators between Layer 1s and Layer 2s built on top of Ethereum?
Steven Goldfeder: I’m a big believer in Ethereum from a technology perspective, which includes its security and its decentralization, but also from an ecosystem and innovation perspective. That means if we looked at the core technology and the guarantees that it gives, like security and decentralization, Ethereum is the clear winner among alternative Layer 1s. If we look at just the ecosystem, where people are, where interesting things are being built, it takes us right back to Ethereum as well.
When it comes to Arbitrum and Layer 2s versus the alternative Layer 1s, the big differentiator is Layer 2s are part of Ethereum. They’re not competitive with Ethereum. The Layer 1s are generally competitive. I have no issue with technical experimentation. And different Layer 1s make different trade-offs when it comes to some of these core properties like security, decentralization and scalability, and how they try to achieve that.
Forbes: What are some examples of a trade-off?
Goldfeder: Let’s go right into Layer 2s and what they provide. At the end of the day there’s this blockchain trilemma—it is hard to optimize security, decentralization and scalability. We know how to take the Ethereum software and crank up the limits and make it scale much better than Ethereum does—we know how to do that. And some alternative Layer 1s do that. They take literally the same software, and they allow more transactions and what happens is you get a de facto centralization because you make it harder and harder to run the software you make. The machines have to be beefier, and they have to have more compute power, storage and memory, which makes it much more difficult for average people to participate in network-provided security. That’s one of the easiest places to see this trade off.
But Ethereum still needs scalability to be useful. Security and decentralization are great values to have, but if no one can use it, it’s not going to get you anywhere and that’s where Layer 2s come in. And Layer 2s inherit security from Ethereum if they’re designed correctly and are true Layer 2s, but then they go a step further. You might ask the question, how come Layer 2s can crank it up more and process more and the reason is because their security property is fundamentally different. A Layer 1 blockchain like Ethereum, Solana or any other Layer 1 blockchain running a typical consensus protocol, has to generally have a supermajority of nodes that are honestly participating in protocols. That’s not the case of Layer 2s. Layer 2s have a different security threshold, which requires Ethereum to be correct plus one Layer 2 node.
Any one node does the trick. There could be a million nodes in theory and in a Layer 2 world you need Ethereum and one of those millions. In a Layer 1 world you need a supermajority of those million. So that means you can safely turn up the cranks a little bit because the amount of honesty they need is much less; you don’t need a supermajority of network to do the right thing. You just need one needle in the haystack plus Ethereum. There’s no magic going on. So Ethereum can be very decentralized and secure.
Forbes: You said there only has to be one node on Arbitrum that needs to be honest versus however many that could theoretically be dishonest. Can you explain a little bit more of how that works?
Goldfeder: In a protocol like Ethereum the nodes are always actively participating. They’re all trying to make blocks and you can see this running on the network. At Arbitrum the way it works is one node goes ahead and posts something. And then there’s a period where anyone can challenge a transaction before it becomes final.
Forbes: So that is where the word optimism comes into optimistic rollups?
Goldfeder: Exactly. The point is that if there’s nothing fishy going on, you don’t actually know how many people are watching. Because the nodes don’t have an active role. The live Arbitrum chains (there are two) have a white list of who can post these challenges. Anyone can follow along but there’s about a dozen or so players that can post that include some big names like Offchain Labs, the Ethereum Foundation, Google Cloud, ConsenSys and Unit 410, which came out of Coinbase. It also has a broad set of ecosystem participants as well.
Forbes: What are the differences between zk rollups and optimistic rollups? And why did you choose to use optimistic?
Goldfeder: The core differences between zk and optimistic rollups are how they do the proof of a transaction’s validity. What our members do is take all these transactions off chain and then they send the results back to Ethereum in an efficient data structure. That’s how they scale and get the correctness. Zk rollups give us what’s called validity proof. They say, “I’m going to prove to you that those transactions are accurate right here.” Optimistic rollups do what’s called a fraud proof, which is they say, “Okay, these are the transactions, these are the results. I’m not giving you proof, optimistically just accept my work. But if someone challenges me, that’s when the proof mechanism happens.” So that’s the difference between the two.
Optimistic systems have existed in computer science literature for a long time and are often more efficient. The reason is that zk proofs are very expensive and whether there’s fraud or not in the network, you always have to do a proof. In optimistic systems 99.9% of the time you get away with not doing one of the proofs. There are trade-offs, of course. If you look at my academic history, I was using zk proofs well before I was doing any optimistic rollups. Our goal at Offchain Labs was to build the best technology of scale that’s available today. And we didn’t come in with any bias. We looked very closely at zk rollups, but we decided against them for a few reasons, one based on cost—they’re much more expensive—and you can see this today on the live networks at Arbitrum because transactions are much cheaper than on zk chains. It was clear to us that an optimistic system like Arbitrum was a better approach. I still believe that to be the case today. But again, we don’t have a personal connection to this technology. I’m building the best technology available today. And in a year or two or three or four or five, if zk technology is better, it would be up to the DAO, but I would personally advocate for incorporating that at Arbitrum, but I’m not sure that will be the case.
Forbes: Now that the Ethereum merge is completed, the focus for developers of the blockchain are turning more towards scalability. How do you see that impacting Arbitrum?
Goldfeder: At Offchain Labs we are on the front lines of this as well because we have another project called prism. The software came to us about a year ago when we acquired a team called Prysmatic Labs. Our team is very integrated in Layer 1 development. That includes Ethereum Improvement Proposal-4844 (which will help the network scale), and there are some synergies that we can explore. I think that EIP-4844 is going to be very important to lowering our costs even more significantly. I think that we’re going to see a continued migration to Layer 2 technologies. And again, the nice thing—going back to your first question about being an Ethereum Layer 2—is that we’re not competing with Ethereum. Arbitrum chains and other Layer 2 chains are not competing with Ethereum, they are part of Ethereum. And you have a Vitalik [Buterin] who’s writing a blog post that’s also advocating for everyone’s moving to Layer 2s as opposed to the alternative Layer 1s where there’s fierce competition, and that’s not a good place to be, competing with Ethereum considering all the success it has had technologically and ecosystem.
Forbes: You said the Arbitrum Foundation is the most decentralized DAO out there. I think you guys have 44% of the voting power and 56% is in the hands of the community. Is that accurate?
Goldfeder: Forty-four percent was given to insiders generally, which includes everyone, investors, team, future team, advisors. That’s the non-community portion. But there are a few things to mention. Offchain Labs team does not vote by policy. The team is allowed to delegate but we require all employees to submit a form with their delegation to certify that they don’t have any direct control over this person. We are very strict about that.
Forbes: Do you think it’s smart not to vote? Theoretically, if you’re the founding team, you should know the most about the project.
Goldfeder: I think there are a few things to be said here. Number one is that success to the ecosystem will be from a strong community. If you look at who the top delegates are, I think the number one delegate is l2Beat. There are some really good community members such as TreasureDAO, Blockworks and Olimpio, and I think that’s where the success comes from. If the success comes from a specific team, then it’s not going to be the strongest network. And I think that bolstering the independent community and empowering that is more valuable in the long term. What you are saying is true, but if a project can’t find the community that cares, that is going to take up that mantle and put in the hours of the time, then it might not be that successful. I agree with you, but I think the goal is to have an empowered strong community to do that.
Forbes: What do you think is next on the roadmap for Arbitrum? What would you like to see get implemented?
Goldfeder: I mentioned earlier that the challenge protocol today currently has an allow list in which about a dozen industry players are involved. There’s a brand-new challenge protocol that we at Offchain Labs have researched and developed that has been audited and will hopefully be published online later to be available for any chain. It’s a nice new protocol for doing these challenges that will allow anyone in the world to fully challenge and participate in the security of the network, which in my opinion, is a very important step. It will be up to the DAO for how and when it incorporates the challenge protocol.
The second one from a technology perspective, which is really cool, is what’s called Arbitrum Stylus. Today you have chains that are defined by languages. Many chains use Ethereum’s native programming language called Solidity. You have Rust chains. There are the Move chains like Aptos and Sui. Stylus will bring Rust, C and C++ onto the Arbitrum blockchain. This means that developers will also be able to write code in those languages, and this is nice for a few reasons. One is that sometimes we don’t realize how small the ecosystem is today. And when you talk to web2 teams and enterprises, they all have experience in different languages, but not in Solidity. There are great cryptographic libraries that exist in these other languages, but not on Solidity, which will become available.
It also turns out that going through these Stylus contracts allows teams to get up to an order of magnitude of 10x performance increases. This will mean that the capacity of a network you’re using will become significantly higher or you can lower the gas costs for your users of your contract. So I think that’s going to be a game changer.
Forbes: Talk a bit about the trends you’re seeing as far as projects, users, and how that may compare to other projects on Layer 1s, ether and some of the other main ones.
Goldfeder: Arbitrum One has roughly 65% market share when it comes to all the Layer 2s. Arbitrum has two chains, Arbitrum One and Arbitrum Nova. Arbitrum One is the sweet spot for decentralized finance. Remember, I mentioned before, all the innovation happens on Ethereum as opposed to on other chains. Arbitrum One has become that for DeFi. It’s an extension of Ethereum, but almost every interesting DeFi project now happens on Arbitrum One. You see products from other ecosystems coming to join that as well. For example, you have Trader Joe from Avalanche doing their expansion to Arbitrum One, and Polygon’s biggest futures market came to Arbitrum One, and within a week they were doing the majority of their activity. So Arbitrum One definitely has a strong footprint in DeFi, but also has a really strong NFT community. Its metric is a little less than TVL than DeFi, but it’s definitely an extremely strong community. And again, I think the key word here is innovation. You’re seeing DeFi innovation. It’s not copycat projects. When you look at basically every core metric—you mentioned TVL—all other good metrics where Arbitrum is dominating: Arbitrum has over 1.1 billion ether (ETH) on the chain. That’s more ETH than every one of its competitors combined. Another good metric is contracts deployed. Arbitrum is by far leading among Layer 2s. And I think it all points to real organic activity. One thing people thought was the Arbitrum airdrop would happen and then all the activity would go down. And I think the activity has actually gone up since then, and it shows real sustained users and building good projects.
Forbes: I want to get your thoughts on the SEC lawsuits. Although arbitrum (ARB) wasn’t mentioned in either suit, I’m curious if you have any thoughts as to why or if you’re concerned about ARB potentially being named as a security by the SEC in the future?
Goldfeder: The only thing here I would say is Arbitrum is real decentralization. I think it’s the most decentralized DAO that exists today as far as I can tell. I can’t speak specifically on any other project, but Arbitrum is really going after the ethos of real decentralization from every perspective. I don’t have any specific comment on that other than that.
Forbes: Thank you.
Offchain Labs co-founder Steven Goldfeder is betting his lot with Ethereum.