Bankless - Ethereum’s Three Front War | Justin Drake

Episode Date: December 2, 2024

Is the Beam Chain the endgame for Ethereum? At Devcon, Justin Drake presented what he called his most ambitious proposal to date! Beam Chain, a roadmap which fell very flat in public discourse. In this episode we fill the gaps of this reaction and we discuss the rise of Credible Competition to the three major components of Ethereum Roadmap. Execution, which competes with Solana, Data Availability which competes with Celestia and Consensus, which competes with Bitcoin. So what’s Ethereum’s strategy for winning this Three Front War against three legitimate competitors? Stay tuned! ------ 📣SPOTIFY PREMIUM RSS FEED | USE CODE: SPOTIFY24  https://bankless.cc/spotify-premium   ------ BANKLESS SPONSOR TOOLS: 🐙KRAKEN | MOST-TRUSTED CRYPTO EXCHANGE https://k.xyz/bankless-pod-q2    ⁠  🦄UNISWAP | BROWSER EXTENSION https://bankless.cc/uniswap  🐧 CARTESI | LINUX-POWERED ROLLUPS https://bankless.cc/CartesiSimple  🛞MANTLE | MODULAR LAYER 2 NETWORK https://bankless.cc/Mantle 📈 iYield: YOUR FINANCIAL PICTURE, SIMPLIFIED https://go.iyield.com/bankless ------ ✨ Mint the episode on Zora ✨ https://zora.co/collect/zora:0x0c294913a7596b427add7dcbd6d7bbfc7338d53f/105?referrer=0x077Fe9e96Aa9b20Bd36F1C6290f54F8717C5674E   ------ TIMESTAMPS 0:00 Intro 4:29 Beam Chain Reactions 9:43 Is Ethereum Ossifying 18:07 Ethereum’s Three Front War 27:03 Bitcoin Utility Prospects 33:20 The Endgame for Crypto 48:55 Beam Chain 101 54:20 Beam Fork Aftermath 59:51 Pre-Confirmations 1:05:06 L1 Execution Layer 1:12:47 Rollup Design Space 1:26:20 Ethereum’s Midgame 1:32:37 Ethereum 3.0? 1:34:04  Closing & Disclaimers ------ RESOURCES Justin Drake https://x.com/drakefjustin   Beam Chain Announcement  https://youtu.be/rGE_RDumZGg?t=7001   ------ Not financial or tax advice. See our investment disclosures here: https://www.bankless.com/disclosures ⁠ 

Now, my personal take is that the only way to win the money game is to also win the utility game. And the only way to win the utility game is to also win the money game. Like, these two things are fundamentally intertwined. Welcome to Bankless, where we explore the frontier of internet money and internet finance. This is how to get started, how to get better, and how to front run the opportunity. At DevCon, Justin Drake presented what he called his most ambitious proposal today. What emerged from that proposal was the beam chain, which fell flat in public discourse. People wanted Ethereum 3.0. People got not that.

We bring Justin Drake on the podcast to discuss this reaction, fill in some gaps, but more importantly, discuss the rise of credible competition to the three major components of the Ethereum roadmap. Justin Drake, after his beam chain proposal talk at DevCon, gave a talk at the Bankless Summit that categorized all of Ethereum's future roadmap upgrades, of which the beam chain was only one part of, into three different categories. Execution, data availability, and consensus. Each one of these categories directly correlates

with a large ecosystem player in the cryptospace. Execution competes with Solana. Data availability competes with Celestia, and Consensus competes with Bitcoin. So we present Justin with the idea of Ethereum fighting a three-front war against two competitors with centralized teams who can ship fast and break things

and the highly ossified Bitcoin and its network effects. What is Ethereum, strategy for winning this three front war against three legitimate competitors. What's its trick? What does it have that others don't? This is what we get into today on the pod. Let's get right to the episode with Justin. But before we do, we want to thank the sponsors that made this possible, including our recommended exchange for 2024 and picking up some crypto. Do it now before everyone else gets here. That is Cracken. Go create an account.

If you want a crypto trading experience backed by world-class security and award-winning support teams, then head over to Cracken, one of the longest standing and most secure crypto platforms in the world. Cracken is on a journey to build a more accessible, inclusive, and fair financial system, making it simple and secure for everyone, everywhere to trade crypto. Cracken's intuitive trading tools are designed to grow with you, empowering you to make your first or your hundredth trade in just a few clicks. And there's an award-winning client support team available 24-7 to help you along the way, along with a whole range of educational guides, articles, and videos. With products and features like Cracken Pro and Cracken

NFT Marketplace and a seamless app to bring it all together, it's really the perfect place to get your complete crypto experience. So check out the simple, secure, and powerful way for everyone to trade crypto, whether you're a complete beginner or a season pro. Go to crackin.com slash bank lists to see what crypto can be. Not investment advice, crypto trading involves risk of loss. Are you ready to take control over your entire financial life? Crypto, defy, and fiat, all in one place? Me, I yield, the free financial planning tool built for crypto natives. Unlike traditional portfolio,

oil trackers that just show you the value of your assets, I yield goes deeper. It consolidates everything, assets, debts, income, and expenses, offering you a complete financial picture across 16,000 tokens, 40 defy-foyleac protocols, and all fiaturances. With real-time defyield tracking for platforms like AVE, Athena, eigenlayer, and Uniswap, Iyield ensures that you're always on top of your staking rewards, investments, and cash flow. And the best part, it's 100% free with no ads and it's secure, no personal data, no ID requirements, no compromises. So head over to iYield.com. slash bank list to sign up today and get the guesswork out of financial planning. Radically simple ideas always tend to catch on.

That's why Cartese did the hard work of putting Linux on chain so that building DAPs can be radically simple by using Python or JavaScript and their suite of libraries. Simple, like not rebuilding the basics from scratch. Simple, like dedicated, scalable compute for your DAP. Simple, like building DAPs however you want. Web3 should be simple too, like bread and butter. Cartese brings radically simple solutions to Ethereum,

to developers can do what they do best. Build. Go ahead and discover a flexible, modular stack on Cartese and build your most powerful, ambitious project yet. Visit cartesi.io slash simple and simplify your blockchain journey and start building today. Bankless Nation, once again, we have Justin Drake, a researcher at the Ethereum Foundation,

and more recently has been the pillar of the beam chain, a proposed path for making some of the upgrades found in the Ethereum roadmap a reality all in one package. Justin, welcome back to Bankless. Hi, guys. Thanks for having me. Justin, so I was furiously live tweeting your presentation at DevCon as you were giving it. The significance of your talk that you were giving went pretty viral, especially from when you tweeted a tweet that read, announcement of an announcement tomorrow at 5 p.m. on the DevCon main stage, I will unveil my most ambitious initiative to date.

So a lot of people did not know what to make of that tweet, which I think allowed people's imagination to just like run wild. and then because myself and a few others were all live tweeting your presentation in real time, even by the time you had finished your presentation, a lot of people's imaginations had been brought back down to reality. After getting off the stage and like checking in on like Twitter post talk, what was your first reaction? What were the thoughts that you had after seeing some of the Twitter discourse? What's your reaction to the reaction? Sure. So interestingly, I had spent the months prior trying to convince the researchers and the devs that this was not a crazy idea. that it was not too ambitious to bring forward.

And I think the summary of crypto Twitter was that it's not ambitious enough. So I was kind of very surprised and a little bit shocked by this juxtaposition of takes. I think one of the things that all of the researchers and the devs had as context was that I was only focusing on one single layer of the stack

and there was these two other layers, the execution and the data layer, which I didn't talk about. And there's a lot of ambitious stuff. also happening there. And then maybe another thing that the devs and the researchers had in the back of their mind is that they knew that there was kind of a short and medium term roadmap with the next coming forks where some big changes were coming in.

And maybe that's something that I didn't highlight enough in my presentation because people thought kind of incorrectly that this was a kind of a massive one-time sprint over the next few years. And then there would be no incremental upgrades between. now and the fork. But actually, it's kind of the opposite. There's plenty of upgrades that would happen in between. And then another like minor thing that I would highlight is that I had this slide which showed kind of a five-year roadmap, but actually the first year was kind of already complete. It was the pitting phase. And so people were kind of very

underwhelmed by the five-year timeline, which actually is a four-year timeline. And the four-year timelines itself kind of this conservative timeline to try and get all of the devs on board and not put too much pressure on them. And so I think that maybe there's an opportunity to shrink the timeline a little bit, but ultimately I do think it is a realistic one given how slow and how conservative the IFRML1 is. Now, one of the things that you did mention

is that a lot of the imaginations went wild in the few hours between my tweet and the actual presentation. And this was a fantastic social experiment, right? Because it's allowed for the community to express their wildest, dreams and desires. And one of the things that I was very pleased about is that a lot of the community shared this vision around putting ZK front and center in the roadmap. And this is exactly what the beam chain proposal is all about, is about leveraging this fantastic technology that

has matured in the last few years, and that is, you know, poised to mature even further and be ready for production at L1. And then the other thing that I was also very pleased to see is that for some reason or never, like now was the time where the notion of native roll-ups hit mainstream consciousness in a very, very big way. Now, I've been kind of shilling native roll-ups for years at this point, and, you know, it always felt like this futuristic kind of science fiction concept that had never really received the excitement that it deserved. And now finally, you know, when I allowed people's imagination to go, well,

the notion of native roll-ups came in. Now, one of the, I guess we'll talk about, one of the big things I'm trying to do with the beam chain is to accelerate the social layer, the coordination part of things. And the good news for native roll-ups is that there's fairly little coordination to do, right? It's mostly just technology.

It's extremely hardcore engineering. And this is engineering that is already being done by the wider community with all the different ZK vendors. And so in some sense, it's just a matter of sitting back and relaxing and waiting for this technology to mature. And then there's a tiny little bit of coordination to expose the power of these ZKVMs into the R1. But on the other hand, the beam chain is maybe 80% coordination and 20% technology. And so this is where I feel that I have a big impact to make.

Okay, Justin, it was really interesting for us to sort of observe these reactions to the take as well. And so, like, I want to almost divide this into like sort of three separate groups. There's the devs, which you mentioned, who, interestingly enough, you said they saw the beam chain propose and they were like, this is impossible. This is too ambitious. And they were really surprised at it. Then there's the Ethereum natives, like I would say David and myself that have been following the roadmap for years and we had a certain set of expectations. And then there's just like the rest of crypto. Okay. And I think each of these camps had kind of a different reaction to what you proposed. I will say for the middle group, like the group of Ethereum natives of which David and I are probably

part of. I think when I saw that announcement, I was like, okay, here we go. Here it is. You know, native ZK roll-ups on Ethereum. And so when I saw the beam proposal, I was just like, okay, this is great. It still feels like part one to a sequel. And I'm like, I'm missing like the part two. I'm missing like, you know, the fellowship of the ring. I'm missing the two towers and the return of the king. Okay? It's like it's only part of the movie here. And I feel like I got some of that sequel in the Bankless Summit talk that you went through, which was like laying out not just the consensus layer, which is the beam chain, but also the data layer and also the execution layer. And I think David and I want to spend some time talking about that roadmap because to me,

for an Ethereum enthusiast, that was like, almost like, okay, this is it. This is the full picture of what we can achieve in a four-year timeline. But before we get there, I want to address that group of just not Ethereum enthusiasts who've been following the roadmap, everybody else in crypto. Because connecting this to something you said earlier, I think when they saw the beam chain and they saw a five-year timeline, which was the headline, right, it wasn't the four-year, it was the five-year headline. And then hearing kind of your reaction where it's like the Ethereum Devs, Ethereum culture, thought even that was too ambitious. They point to that and they say, yeah, that's part of the problem. Part of the problem with Ethereum

culture right now is that it's not ambitious enough. It has slowed down. and they're looking at things that other alternative layer one communities are doing, and they're just ship, ship, ship, that's what it feels like. They're all shipping. And so even if Ethereum researchers or a certain cohort in Ethereum are proposing, like, big things, big sounding things like the beam chain, if the all-core devs reaction to that is like, whoa, whoa, whoa, slow down. They would point to that and they'd say, yeah, Justin, that's part of the problem with Ethereum right now.

It's got too many breaks, and it's not ambitious enough. And I want you to address that because I think that's something that you have maybe seen and why you're like, you got three different stakeholders here. And the dev kind of stakeholders is like, whoa, this is a lot. Okay, Justin. The beam chain is a lot. And like, are you kidding me? Four or five years? That's really fast. Do we have a problem here? Do we have an endemic problem inside of Ethereum where we're just like going to slow these days? Things have kind of stultified and ossified. And I'll just put one other case out there. That's what we've seen in Bitcoin. Bitcoin definitely did that. Go try to get like Eric Wall's OP cat in. This is like a 10-year, decade-long project to get one single op code into Bitcoin. So communities have a tendency to ossify over time. And are we hitting up against that wall here on Ethereum? I think that's what the wider crypto community wants to know. That is a fantastic question. And I think the criticism from the wider crypto community comes from a misunderstanding of the strategy. of Ethereum.

Ethereum, it has two constituent parts. It has the L1 and it has the L2s. The L1 is competing with Bitcoin. It's competing for maximum robustness, maximum security,

maximum quality, maximum health. It's competing for the moniness in some sense. And then you have the L2s. They're competing with Solana. They're competing on maximizing utility

and UX and quantity. and quantity and performance and volume. And asking the L1 to compete with Solana is just a misunderstanding of the separation of concerns. Now, I agree that as a community of L2s, we need to ship, ship, ship, ship, every week, every month, every day on a continuous basis. And this is exactly what has been happening. Look at base, for example. Like every week, they raise the gas limit. Look at all of the L2s.

There's a lot of effort, for example, on providing pre-confirmations and improving the UX. And actually, if you look at Solana, what are the two key things that it prized itself the most? Latency and throughput. And I think that L-2s are going to vastly outperform Solana on both of these two metrics. Latency is going to go down to zero, just one ping time, thanks to pre-confirmations. You can think of this as being an abstraction of the notion of the slot duration, right? Users don't have to worry about what is the exact slot duration in a chain. They just want the UX to be instant.

And then you have the gas limit, which every single day is being raised and raised and raised. And I invite you to go to this website, rollup.wtf, which shows you that the total throughput, which is measured in megagas per second has grown to 100x versus the L1. So all of the L2s combined have 100x the throughput of the L1. And what is going to happen over the next year is that this 100x is going to become 1,000x and then 10,000 X and then 100,000 X. Can we just pause and make sure we understand that, Justin? So I've got rollup.wop.wtf up.

And what you're referring to is, are you referring to megagas per second? metric or TPS, total transactions per second? So if you look at the very top here in the header, there's the totals. And then the middle column is megagas per second. Okay. So we're at 67 right now. Right. And if you look at the bracket, that's 68x.

Right. So it depends on the time of the day and what. 68x Ethereum. Mainnet. L1. Yes. Okay. And then total transactions per second, it's 23X.

Yes. But the thing is that you want to be looking at the gas per second because there's some transactions that are more complicated, and so you'll do fewer transactions, but those are richer transactions, for example. So the middle column, I think, is the one to focus on. Okay, cool. And I interrupted you. Sorry, so you're going on a flow. This is 60x right now, the middle column, megagast per second, and you're saying something about that, like increasing somewhat logarithmically? Yeah. So right now, I mean, it fluctuates every second, but right now is roughly 60x. What I expected to be next year is 600x, and then the year after that is 6,000x. And the reason is

that L2s can just scale in an unlimited fashion. They can scale horizontally, but they can also scale like base vertically by just increasing the gas limit. And this is something that Solana is incapable of doing. There is no way that Solana is going to increase its throughput 10x per year for the next few years. Like it's already completely maxed out

in terms of the utilization of the hardware with the existing clients. Sure, maybe there could be new clients that provide more resources and they allow for the so-called compute unit limit. So in Solana, we don't use the term gas. They use the term compute limit. But essentially, what I think will happen is that Solana is going to hit a brick wall, right, where it's maxed out its scalability.

Whereas on the other hand, Ethereum has a sustainable scalability story thanks to the L2s. So, Justin, I'm now sharing on the screen a slide from your Bankless Summit talk. And the Bankless Summit talk happened three or four days after your DevCon BeamChance. talk and it was kind of a reaction to the reaction to the beam chain talk. And so it was like, here's what everyone said about the beam chain talk and now let me kind of like fill in the gaps. And you gave this Strawman layer one timeline slide, which I think filled in the parts that people were missing if they just thought that the beam chain was the only thing that the Ethereum was shipping. And the way that you oriented this slide sparked something in my brain. You

color-coded all of the upgrades as they relate to the three components that compose Ethereum. Execution is in purple. So all the upgrades that go into Ethereum over the next five years are in purple. Data is in orange. So all the data blobs, the data availability, all those upgrades are in orange, and then consensus, which is green. And so all of these different like EIP upgrades with Ethereum will kind of fit into these one of three categories. And the connection that this made for me is each one of these kind of correlates to another very big blockchain in the crypto space. Execution is Solana, data availability is Celestia, and consensus is Bitcoin, like very hardened consensus, and then also the monetary policy for ETH is contained in here.

So that caused me to tweet, this one tweet. The Ethereum roadmap is more blobs than Celestia, faster execution than Solana, and better money than Bitcoin. And of course, this really kind of illustrates, I think, the tension that Ethereum has been under over the last two years. It is fighting a three-way war against three different ecosystems that all have very strong merits to them. Like, the money and this of Bitcoin, very hard to fight. It's like all fiat currencies, slowly losing versus Bitcoin. Solana execution, also very hard to fight, not only because they have good execution, but people know that they have good execution. And then blobs, more blobs than Celestia, I actually think that's the easiest

one to really fight against, but nonetheless, like, Celessia does have more data availability than Ethereum. John Charbonneau put this differently. I'm the Ethereum guy. John is more of just like the equal-weighted interpretation of the same idea. He put out this tweet, which is Ethereum's strengths is also its weaknesses. Its ambitious vision means it's competing with everyone a little bit. Bitcoin is better sound money than ETH. Solana is a more perform. form an execution environment than the Ethereum layer one, Celestia is going to have bigger, faster blocks sooner. So there's this vision of like Ethereum being like drawn and quartered, except not quartered, thirded by Bitcoin, Celestia and Solana. And there's this tension here where like can

this Ethereum body pull in and win this tug of war against three different fronts or is it getting stretched by these three different competitors? Bitcoin is better money than Ethereum. Salana's got better execution and Celestia's got more data availability. And right now, Justin, the current market pricing, the market is saying that Ethereum is being stretched and it does not have the strength to win a three-way tug of war. And so this is kind of asking, we did a episode with Fatal a couple years ago called the end game. And I think this is something that the Ethereum research community has really attracted is end game researchers, end game protocol designers who think in very very logical ways how this whole game ends and then building towards that, which ends up being

a five-year-long roadmap. But what this sacrifices just in, which Celestia is doing and Solana is doing, and fortunately for Bitcoin doesn't have to do, is the middle game. And the middle game is very important. The middle game seems to be being lost, at least by the validation of the market, in this present part of Ethereum's history. So I kind of want to throw all of that at you. How do you react to the idea that like Ethereum is actually losing the middle game? And the middle game might be a very important thing to not lose because the middle game dictates how the end game begins.

Yeah, fantastic question. So I think you're right that the easiest to address is probably the DA with Celestia. The way that I see it is that Celestia in some sense is a collaborative player with Ethereum because it provides overflow DA when there isn't sufficient capacity on the L1. And so what I expect will happen is that the use cases of Herb that requires the highest grade security and the highest grade composability, we will use L1DA. But that's not enough for all of the world. And so we're going to need overflow in the form of eigenDA, for example, or Celessia,

which you can think of as being collaborative technologies with Ephemium. Now, you are right that there is competition. in the form of Bitcoin, which is trying to be the money, and Solana, which is trying to provide maximum utility. Now, my personal take is that the only way

to win the money game is to also win the utility game, and the only way to win the utility game is to also win the money game. These two things are fundamentally intertwined. And there's actually this quote that I have here from Alan Kay, who's this computer scientist, who won the Turing Award, he wrote,

people who are really serious about software should make their own hardware, right? And this is an explanation in my mind as to why Apple is so successful, right? It's because they are able to make use of the synergies of simultaneously optimizing for software and the hardware, and that's what makes them special. And I think Ethereum has an amazing opportunity

in front of itself. And the reason is that Bitcoin is not even trying to compete with utility, and Solana is not even trying to compete for money, right? Ethereum is the only system out there that has the guts, the ambition, the vision to compete for both money and utility. And it turns out that there are these synergies and network effects that may be worth flashing out. So when you have large amounts of monetary premium, what do you have on the other? rent. You have large amounts of economic security, large amounts of economic bandwidth, right? And that is fundamentally the driver of utility. If you want to have trillions of dollars

of stable coins, decentralized, trustless stable coins that use a pristine money as collateral, the only way to get there is to accrue monetary premium. If you want to have a blockchain which secures tens of trillions of dollars, hundreds of trillions of dollars, a blockchain that is unquestionably secure from attacks from nation states or even a cabal of nation states all coming together to try and attack the system,

you're going to need very, very large amounts of economic security. And the only way to get that is through monetary premium. And then the other thing happens where in order to get very large amounts of monetary premium, you need to have large amounts of utility. And that is in the form, for example,

of the burn, right? Like the more activity there is, the more utility you provide, the more scarcity you have because you have income for your chain, which you can feed into the token and increase monetary premium. But you also have network effects in the form of liquidity and composability and using the token as a unit of account. If the only thing that you optimize for is being money, what you end up with is a dumb rock, right, the pet rock, like gold. or Bitcoin. And if the only thing that you optimize for is utility, you end up with, you know, a casino like Solana, like shallow and is only trading, you know, meme coins. What we're trying to do here is to build the future of finance, right, the internet of value. And in order to achieve

that, we need to have both the utility and the money. And this is a unique opportunity for Ethereum and we should just keep on going and have the conviction to continue. continue in that direction. I want to see if I can re-articulate what I think you're saying. Once again, showing the graphic that John Charbonneau tweeted out, which is for the listeners, it is a three-way Venn diagram. And you have this orange Bitcoin circle overlapping with this purple Celestia circle, overlapping with this like cyan, Solana circle, right? Like execution, data availability, money. And then where those three circles overlap at the very center is Ethereum. And so this is, I think, the design structure that I think nerdsnipes a lot of Ethereum believers is like, wow, there's this

unified blockchain architecture that does it all and is like the singularity of all logical conclusions of blockchain designs. And then, of course, going back to like the alternative vision, which is like, well, what if three competitors just like unbundle this thing? And I think what you're saying, Justin, is that, well, Ethereum has the benefit of tapping into these synergies that happens when you combine the money of Bitcoin with the execution of Solana and the roll-up sovereignty of Celestia. And so each one kind of has its unique synergies. You know, the utility of Solana plus the money of Bitcoin is like your very censorship-resistant, internet-based currency and financial rails, like going bankless, defy, defy.

And that's very cool and unique. And nation-state resistant. That's cypherpunk values. That's what we're going for. Then you have, you know, the combination of money and. And layer twos, like sovereign blockchains, app chains that can customize themselves into like whatever they see fit. Businesses that are chains that have the funding because of the money.

And then you can combine all of these synergies. And only Ethereum can really tap into these synergies is the design space that Ethereum is going for. And so there's each one of these overlapping parts of the Venn diagram have their like benefits that I think Ethereum thinks that only it can tap into. but also just playing devil's advocate, Starkware has always been interested on settling on Bitcoin. And Bitcoin layer twos like Alpin Labs, the BitVM, is like you actually are starting to see the moneyiness of Bitcoin and sent layer twos. And so you are seeing Bitcoin and Celestia overlap in their

model and their design space in ways that Ethereum is like less relevant to. And so I think kind of the bet is like, is it actually possible to unbundle Ethereum and see the Ethereum vision happen emergently in other ecosystems? And how does Ethereum kind of like defend against that? And how does Ethereum like control the moat of these synergies of these overlapping Venn diagrams? Yeah, fantastic question. So it is totally possible for Bitcoin that currently has a massive lead in terms of monetary premium to also start playing the utility game. And one of the recent innovations is called BitVM. And I think you've had an episode on BitVM. And if BitVM were to materialize in a successful way, then we could see roll-ups on top of Bitcoin.

There are unfortunately several problems with that approach. The first problem is that it is technologically very, very difficult to pull off. So yes, there's progress to do it, but it might take several years. Okay, fine, maybe that's not a big deal. The other big problem is that Bitcoin is fundamentally limited to 4 megabytes per 10 minutes of DA. So even if you have those roll-ups, there's very, very little opportunity to provide high-grade utility just because there isn't much throughput. And the only way to provide more of the, you know,

maximally secure DA on Bitcoin is to make a hard fork. And the community is just unwilling to do these hard forks. And then there's a kind of a third thing, which is that BidVM in some sense is a massive hack, and it comes with a long list of caveats. One of them is that you can only do optimistic roll-ups. You can't do ZK roll-ups. I mean, they use ZK in the ford-proof game, but still, fundamentally, it's an optimistic roll-up. And the assumptions are not as good as on Ethereum, where you need to have one of N participants being honest, and when you do the bridging in and out of the roll-up, you can only use, you know, for example, fixed amounts of,

of Bitcoin. So it has to always be, you know, a 10 Bitcoin deposit, for example, a 10 Bitcoin withdrawals. There's all sorts of complications that due to the fact that Bitcoin is not programmable. Now, you're right that stockware is specifically looking to build on top of Bitcoin if there's this new upcode called Upcat that would happen on Bitcoin, because then a lot of the hacks would go away and you could, you know, natively do ZK roll-ups and things like that. And I guess the issue goes back to what I said initially is that it's going to take a very, very long time for Bitcoin to add this upcode, if ever, just because the social layer is so ossified and in some sense against the notion of adding programmability. But even if Bitcoin does get

Upcat, again, we're in a position where Bitcoin is fundamentally limited from a DA standpoint. So the bottleneck no longer becomes execution, but it becomes DA. And that is something that, as far as I can Intel goes against the DNA of Bitcoin. We all remember the block size was, right? We wanted to increase the block size to something like 8 megabytes per 10 minutes or 16 megabytes per 10 minutes. This is a big part of the history of Bitcoin. I don't see them dramatically increasing the amount of DA at layer one. I have some observations on the conversation thus far, but before I get to those, I want to just ask a follow-up question on Bitcoin because it could be that the end state that we've just defined is just like our crypto-native version of the end state.

I will tell you a lot of the newcomers in this cycle seem pretty satisfied just to have Bitcoin with, you know, 10 transactions per second, and the rest kind of custody by BlackRock, Michael Saylor, you know, the U.S. Bitcoin Reserve. Like, they're not into the whole bankless thesis that David and I have been operating under for the last four years. They're kind of fine. Saylor says private keys is for extremists. As long as number goes up, right?

It's just like, hey, so what if this Bitcoin is kind of custody? As long as, you know, 100 or 1,000 kind of like whales have the ability and can afford to interact with the Bitcoin blockchain, maybe that's enough. And so that's one side of kind of the barbell. And the other side of the barbell is just execution with stable coins. Like, what if you just have something that's not decentralized finance and fully bankless, but a whole bunch of tether and USC in kind of like. a open finance type chain that it's not maximally decentralized, it's not censorship resistant,

but it's just like permissionless at least. Tether doesn't settle on Ethereum either. It settles on Tether. Exactly. And so before we get into kind of the full vision, I just want to mention that this is the bankless vision, if you will. It's kind of the crypto-native vision of the world. It's kind of hardcore. It's pretty hardcore. Like it's very ambitious and it might not come to past. Like the actual practical version of this might just be Bitcoin, digital gold, pet rock type thing. And sure it's got a little bit in kind of a layer two type state, but most of it's custodied by like big whales, essentially, in institutions. And then on the other side of the spectrum, you have kind of open finance. And like it is NASDAQ on chain, right? But like,

it's kind of permissionless, but it's all, there's nothing to centralize really about it. It's all settled in meat space. And I just want to ask. you, Justin, if, I know that's not what we're working towards, but do you sometimes wake up in the middle of the night? That's where the incentives point. And like, worry that that's actually going to be the outcome, that we're not going to be able to achieve. And when I say, we're not going to be able to achieve, I mean, just like this entire crypto movement thing, whether it's Ethereum or something else, that we just won't end up with the end state of actual censorship resistant bankless monetary financial property rights system for the world. Okay, very interesting

question. So I think, you know, it is totally possible for Bitcoin to just do one thing and be a pet rock and be immensely successful. And, you know, the proof that this, you know, is a possibility, I guess is gold, right? Gold is a $17 trillion asset and Bitcoin is only about 10% of that. And so if Bitcoin would to become as successful as gold, which I think it could totally be, then there is a 10x opportunity for Bitcoin. But ultimately, you know, $17 trillion is a successful. small tam. You know, I'm saying this, you know, somewhat provocatively, but the vision that we have is so much bigger where we're talking about decentralized, you know, finance for the whole world that would encompass, you know, bonds and stocks and fiat and store value and derivatives and digital

property and everything else. We're talking about, you know, a vision that is potentially in the hundreds of trillions of dollars. So again, you know, in some sense, Ephraim is dreaming, much bigger dreams than what Bitcoin is. Now, the other thing that I'll say in terms of using trusted money Legos is that they are not scalable when it comes to complexity and composability. So let me try and explain that. So if you're trying to do something very, very simple, then it's actually okay for you to have the decentralized core and then the service on top of it just be totally centralized.

And we see that with Bitcoin. So we see the decentralized money, and then we see all the apparatus around it, which is like the exchanges, you know, being totally centralized. And the reason that works out is because as a user that wants to sell or buy Bitcoin, I'm only trusting a single trusted entity.

And so there is no coordination across trusted entities. And I think the superpower of the more ambitious vision that we have is that we're going to be building super structures that involve multiple money Legos, and sure, like the final wrapper at the very end, the wallet or the broker, the exchange, or the UI can be centralized, and that's fine. But if you zoom in and you look at the money Legos, as soon as there's a trusted entity, which is something other than the wrapper UI, then now this is where things break down. because you start having platform risk, you start having counterparty risk.

One of the classic examples is the Facebook API endpoints. If you were to build a complex application built on top of Facebook, while now Facebook has the ability to rug you, and so this is not scalable in time and space. But if you want to build a money Lego, which has very, very strong guarantees, meaning that other money LEGOs feel comfortable interacting with it, then you need this strong composability, which only comes from strong trustlessness.

And sure, you can take these 10 trustless money that goes and wrap everything in something centralized, and that's probably what the users want, right? The users want a button which says, I forgot my password, please give it back to me. But again, like when you zoom in and you look at the building blocks, those need to be trustless in order for the complexity to scale. Just a recent example of this is just, you know, Bitcoin has been kind of scale. scaling its DFI open finance kind of use case through tokenized Bitcoin like Rapt BTC. And I just noticed earlier this week, like Coinbase decided to delist Rapt BTC, right?

And it's like I haven't dug into the details around this, but, you know, Justin's son has now some participation in Bikos. There's some question as to like what's going on with RAPTC and the market gets kind of scared. And you can't audit that on chain. Like you have no idea. This is not a scalable money Lego for Bitcoin. would be much more scalable is kind of a Bitcoin tokenized on a trustless bridge inside of a true BitVM style layer two. That would be actually a bankless money Lego that you could trust and you could build a civilization and an institution around, like not something like cussied, wrapped Bitcoin. So that's a small example of this. Now, I just want to get to some of my observations on the

commentary so far, this three-way war, right, between other different chains. Like, I guess a few thoughts I have at the top of my head. One, this is why everyone hates Eith, isn't it? Because Heath is like, oh, no, we're a better money than you, Bitcoin. And Bitcoin's like, shut up. Like, sit down, like, you know, Little Eath. And then Ethereum's tried to out blob space Celestia, and it's trying to out-execute Solana. And so if it feels like Ethereum is being tacked all the time in the social layer by other communities, it's because it is. It's because it's kind of trying to compete against all of them and effectively competing against all of them. So that's one observation. But the second is this. I think this is important.

Isn't it the case that everyone is actually playing this game? So everyone is playing the game. If you're Bitcoin, you're playing the game for money, for execution, and for DA. If you're Celestia, the one thing that you want more than anything is for Tia to be money. And you also want to have an execution layer built up around you. If you're Solana, and they may not acknowledge this or realize it yet, they actually need sold to be money. Or else it has no value accrual properties over the long term because all of the value

accrual will leak into the app layer and away from kind of MEV for block builders in the fullness of time. And they actually need some blob space to scale. Like they actually need a salana roll-ups, even though this is unacknowledged so far this cycle. It seems to be the case that all of these chains are actually playing the exact same game as Ethereum, whether they know it or not. And your point earlier is. Everyone's trying to move into the center. Yeah, exactly.

And they need to. I think they need to. Like even thinking about, you know, Bitcoin's scaling strategy, it looks like a much less mature version of the Ethereum like roll-up roadmap, doesn't it? I mean, they need to deal with blob space and they only have four megabytes of blob space. So how are they going to scale that? How are they going to deal with the MEV? Like, they're going to have to go down the Ethereum roadmap if they actually want to scale DA and execution. Anyway, so that's another observation. And I want to throw that by you guys like do you think that everyone is just effectively playing the same game here it's just

they're not acknowledging it whereas the theorem is like yeah we're competing for all of these things and and so it doesn't have as clean and narrative this cycle yeah i mean you could try and zoom out and look at the patterns across the different players and maybe you know everyone's trying to aim for the same end game whether or not they realize it as you said because that's where the network effects lie and the only difference is you know the strategy to go about it One of the strategies that Bitcoin has taken is maximum ossification at the social layer. And unfortunately, it's trapped itself. You know, on paper, it sounds like a very interesting strategy,

but it ultimately means that it will be incapable of changing its security policy, which means that, you know, the long-term security of Bitcoin is jeopardized. And it's also incapable of changing DA. And so unless there's some sort of radical change at the social layer, then I don't see how Bitcoin could go for this bigger vision. I think it's interesting what you say about Celestia, right? Because initially they started as a pure DA, and then they recently added more execution,

or at least planning to add more execution. I don't know the details around being able to verify snarks on Celestia. And once you have these two components and Tia the token, then yes, you have all of the ingredients to potentially win this massive endgame. But if I were to look at the main, you know, competitors in the race, I only see, you know, Bitcoin, Ethereum and Solana as being, you know, the reasonably likely candidates to win out. And as far as I can tell, Bitcoin is ruled out just because of its inability to change and has to change. And I think that Ethereum is also in a position to win out because it has a lot of the existing network effects.

You know, recently I had a look, for example, in terms of how much TVL there is from Tevere on Ethereum versus Solana. And the Delta is striking. There's 75 times more Tevere on Ethereum versus Solana. It's 65 billion versus 862 million. And going back to my previous point, ultimately it's for the L2s to compete with Solana. And the L2s on many metrics are crushing Solana.

If you take Arbitrum, for example, Arbitrum has 3.5 times more tether than all of Solana. If you look at Circle, Circle has more USDC than all of Solana. And again, what I think will happen is that the slot times are going to be abstracted. So the pride and joy of Solana, which is its 400 milliseconds slot times, is not going to be a competitive advantage. And we're already seeing that with Arbitrum. Arbitrum today has faster U.X than Solana.

It has 250 millisecond block times. and this 250 millisecond is effectively going to go down to one millisecond with pre-confirmations. And going back to one of my previous comments, which is that the second pride and joy of Solana is the ability to have large amounts of execution. But there's only so much you can push the execution vertically. At some point, you have to grow horizontally. And Solana is just obstinate that it doesn't want to grow horizontally. It just pushes, pushes everything vertically.

and the problem is that you hit the laws of physics, you hit the limits of computation, and you can't scale past a certain point. And on the other hand, Ephiram has taken the sustainable strategy of scaling horizontally, and what is today, you know, a 100x increase over the L1 throughput

will become 1,000x and then 10,000 X in the year to come. New projects are coming online to the mantle layer 2 every single week. Why is this happening? Maybe it's because Mantle has been on the frontier of layer two design architecture since it first started building Mantle DA, powered by technology from EigenDA. Maybe it's because users are coming onto the mantle layer two

to capture some of the highest yields available in Defy and to automatically receive the points and tokens being accrued by the $3 billion mantle treasury in the Mantle reward station. Maybe it's because the Mantle team is one of the most helpful teams to build with giving you grants, liquidity support, and venture partners to help bootstrap your mantle application. Maybe it's all of these reasons all put together. So if you're a dev and you want to build on one of the best foundations in crypto or your user looking to claim some ownership on Mantle's Defi apps,

click the link in the show notes so getting started with Mantle. Uniswap wallet is officially the preferred wallet of bank lists and it's the one we use any time when we want to transact on chain. Whether you're on your browser or on the go, uniswap wallet makes it easier than ever to swap anytime, anywhere.

Use your wallet to transfer funds directly from a top centralized exchange and taping thousands of tokens across Ethereum and over 10 other chains like base, arbitram, and optimism. Uniswap wallet delivers deep liquidity, fast execution, reliable quotes with zero gas swaps through Uniswap X. And when it comes to security,

you can rest easy knowing it's backed by Uniswop Labs, one of the most trusted teams in Defy. Their code is open source and independently reviewed so you know it's protected. So why wait? Download the Uniswap wallet today on Chrome, iOS, and Android. And don't forget to claim your free uni.combe.combe. Start swapping smarter with Uniswap. Justin, I want to tap back into the Ethereum roadmap and unpack once again this slide that you shared at the Bankless Summit, which I thought was immensely useful to kind of like understanding and like categorizing parts of the Ethereum roadmap. I know Vitalik has his roadmap, the merge, splurge, urge, verge, verge, verge, or something. The three swim lanes that I see and resonate with much more is like these three alternative swim lanes, which is execution, data, consensus, competing with Bitcoin, competing with Celina. And in order to provide just like stronger comprehension by like, broader audience. I think people much more resonate with that kind of understanding of the Ethereum roadmap. I want you to start with the beam chain, which is where this whole entire

conversation started with, which is the consensus swim lane of the Ethereum roadmap, which is the presentation you gave at DevCon. And then the presentation you gave at the Bankless Summit was the other two sides of this, execution and data. But maybe let's talk about the consensus swim lane, the part of the Ethereum roadmap that competes with Bitcoin, the money. And how would you describe the beam chain in a single sentence? There's many different upgrades here, Maxib, Fossil, APS, but if you package it up as a whole, which is what you're doing with beam chain,

how do you describe what the beam chain upgrade does to Ethereum? So the beam chain upgrade is trying to take the slowest parts of the roadmap and accelerate them so that we can complete the full Ethereum roadmap in a timely fashion. And you can think of the beam chain as being this memetic social coordination tool

to ultimately accelerate the roadmap and try and avoid Ethereum ossifying prematurely. Vitalik called the beam chain a mimetic wrapper around the roadmap. And in the short term, because it's just a proposal,

you can think of it as being an invitation to think strategically about the Ephemim roadmap and to widen the discourse and the overturn window around it. In some sense, the beam chain is not about changing the content,

the fundamental of what we want to change. It's about changing metadata around it, about how we would ship the roadmap. So, for example, I want to see more discussion around the opportunity or even the necessity to do ambitious upgrades, to embrace ZK technology, to reject technical debt, to optimize our approach to Ethereum governance.

And as you said, David, as potentially an opportunity to re-skin and simplify the roadmap, because when you look at Vitag's roadmap, it is extremely complicated. It's for researchers. Exactly. It is for researchers. And what I'm trying to do here is to digest the roadmap and present it in a slightly different way. Now, above and beyond, you know, the short-term discussion that I'm hoping will happen with the wider community and the researchers and the devs, ultimately, if the beam chain meme is going to be successful, it's going to have, you know, real impact in.

terms of accelerating the roadmap. So for example, what I'm hoping will happen is that it will nerdsnipe, motivate and unite this wide, you know, diverse group of devs around a common shutting point. And one of the things here that I'm extremely, you know, excited to share is that we now have 12 different teams that want to build a beam client. So that's the five existing teams. And then in the last seven days, since I've given my talk, there's been like five new teams. And then when I made the announcement, there were already two. So that's a total of 12. And there's about 50 different individuals that have approached me on Telegram and that have emailed the beam.com at Ephraim.org, basically asking, you know, how can I help? I want to contribute. And so I think this first part

around nut sniping and motivating the devs has been a success so far. But ultimately, what I want to do is pour this accelerationist fuel over the long term. I want to have all of the devs kind of converge towards this shared mission. And I want to avoid the premature ossification of the R1. Because one of my fears is that Ethereum ossifies prematurely, not because there is a lack of good research ideas or a lack of collective desire to innovate, but because of an ability to coordinate fast enough to match the ambition of the roadmap as well as the rapid growth of the ecosystem. And so again, to summarize, the beam chain is about taking these ambitious kind of long-tail updates that are very, very important to maximize the long-term health of Ethereum and make it

competitive as a super-robust consensus layer and to basically provide the foundations for Ethereum to be successful over the decades and centuries to come. When I look at everything in green in the beam chain layer and I try to summarize it, I feel like as a user as an upgrade, like the things that we're getting from this. And again, the theme here on the consensus layer is better money, better settlement, better consensus, almost like competing against Bitcoin on that angle. What seems like we get are these things called pre-confirmations. And you said somewhere within this red map, we get 90% of transactions with pre-confirmations. So I think that does something for roll-ups, perhaps. we can talk about that. We also get stake capping and one-eath validators. So this decreases the barrier to entry to actually run a node from an economic perspective

and it's kind of like a final solidification of the monetary strategy. Again, competitive against Bitcoin. And then at the end of this, we also get fossil, which I think is a censorship-resistant improvement type of fork. At the end of this, the beam fork, you said this a whole bunch of times, like, we're

snarkifying the Ethereum consensus client. So I think that gives us the ability to basically run a client on like some minimum set of hardware. I mean, there was this vision for Ethereum to kind of like run a ETH validator on your watch. So I guess what I would love to do is could you summarize this in kind of like layman's terms? What do we get on the consensus layer at the end of the beamfork in like four years plus time? So to a very large extent, the consensus layer is all about improving the health of the system more so than the performance. So if you look at fossil, for example, that's about censorship resistance.

For the vast majority of users today, they're not being censored, but still, this is something that we care very much about and is deeply rooted in our cyphepunk values. Another one is attest a proposal separation. Here, the idea is to segregate the validators away from the centralization forces of MEV. And actually, to a very large extent, what the consensus layer, grades are all about is maximizing the viability of being a solar validator. And I have these three different memes. I have Zen staking, which is that as a staker, the only thing that I'm really responsible for is attestations. And attestations are trivial. It's just looking as to whether or not

the previous block arrived on time. If it arrived on time, attest for it. If it didn't arrive on time, I test for the parent. That's as simple as it gets. You don't need to be at all sophisticated. And that's in comparison to what we have today, where you have to pick a relay, you have to play timing games, you have to deal with MEV spikes, and soon you'll have to deal with the pre-confirmations. And then the second meme that I have is what I call fish staking. So today, Ephemstaking is for whales. You need to have 32 eph, that's $100,000, and that is the number one barrier to entry to becoming a validator. And we can reduce that to just one if. and one of the seeds that we're planting in the next fork is called MaxEB,

which allow for the validators to consolidate and ultimately allow for us to unlock the ability for just one eve to be sufficient to make a deposit. And then as you highlighted, there is this vision that ultimately, once we snarkified everything, you'll be able to validate on a smart watch. And I call that feather staking. And once you combine all three, Zen staking, fish staking, and feather staking, and fever-staking, what I'm hoping will happen

is that we'll have this renaissance, this golden age of solar validating, and that will undeniably help make Ethereum be the most robust chain out there. Now, the reason why I included the pre-confirmations in the green section is that there's this whole discussion around, let's make slots fast.

So, for example, Max Resnick, you know, was laughing at the proposal that, you know, we'll have four seconds slots in four years time, that is only a 3x improvement over a huge amount of time. What are these researchers and developers doing? Are they sleeping on the wheel? Well, the thing is that these short slot times are extremely unhealthy from the point of view of decentralization. We want to have the ability for a validator in Australia with a high latency home internet connection to still be able to participate. And the good news here is that we can keep the slot times relatively high,

but bring the UX down to effectively just one ping time thanks to pre-confirmations. And my thesis is that the slot time is not going to matter very, very soon from a UX perspective. And so we should kind of plan for that and make use of it. And in some sense, what Solana is trying to do to reduce its slot times is a trap, right, because it's unnecessarily centralizing its consensus layer when there is a much more pragmatic, ironically solution available today. You've got somewhere in this green layer, there's a milestone at which 90% of transactions are pre-confirmed.

And I feel like we could do an entire episode on pre-confirmation, so that's like an esoteric topic. I've just seen a lot of, like, debate about it. I mean, some people are saying pre-confirmations in and of them themselves are like centralization vehicles or like others are saying they don't really exist yet. And I know that at DevCon, you showed the first main net pre-confirmation, but maybe just very briefly, what gives you confidence that will get to like 90% of transactions pre-confirmed? Like, when does that happen and how do we go about doing that?

This does not require an upgrade. Is it more of like a social coordination type of work to be done here? Well, what's going to happen is that the vast majority of transactions are going to move to L2s, and all of the L2s are going to have pre-confirmations. Like even today, Arbitrum has pre-confirmations, 250 milliseconds. All of the super chain roll-ups, they have pre-confirmations on the order of two seconds. And these are reputational pre-confirmations, basically a pinky promise that the proposer is going to include your transaction in the next block.

And what I think is going to happen is that the technology around pre-confirmations is going to evolve. So, for example, we have Unichain that announced that they're going to have 250 millisecond second T.E. pre-confirmation. So they're going to use trusted hardware to provide. the pre-confirmations. But we can do even better than reputation and trust. We can have crypto-economics. So what I'm suggesting in the context where there is no reputation for the sequencer, especially in the context of base roll-ups, is to use just good old crypto-economics.

So the idea is that the sequencer would pledge collateral, some if or some other collateral. And when they make those promises, they're backed by that collateral. And if the promise gets reneged, then the collateral gets slashed. And sure, like there is a bunch of infrastructure that need to be built. For example, there needs to be a registration contract where the sequences can go and deposit collateral. There needs to be a fraud-proof game.

So basically, there needs to be the ability for a user that was cheated on by a sequencer to provide cryptographic proof that they were cheated on and there needs to be the slashing mechanism. But all of this is, like, totally doable. And actually, it has been done as a proof of concepts,

on Mainnet just a few days ago. And I think this is the future. And there's this kind of common misconception that pre-confirmations are centralizing force. It's actually the opposite. So what happens is that as a user who requires a pre-confirmation, I need to go talk directly to the sequencer. So it's all about cutting the middlemen

between the user and the sequencer. Who are these middlemen that I'm referring to? Well, it's the builders and the relays. In a world with pre-confirmations, we no longer have builders and we no longer have relays. And the vast majority of centralized choke points comes from these entities, the builders and the relays. Instead, with pre-confirmations, at least before we have APS, which should happen soon. But in the short term, we're going to have the need for what we call gateways. So unfortunately, for the base pre-confirmations, the validators themselves are not powerful enough to provide.

them directly. They're assumed to be on a Raspberry Pi with a home internet connection. They can't provide these sophisticated pre-confirmations. And so instead, they will delegate their sequencing rights to what's called a gateway, which is the equivalent of a relay for pre-confirmations. And the important thing to stress here is that the relay is sophisticated from a DevOps perspective, but it is not sophisticated from a financial perspective. So a lot of the centralization forces of block building comes from the fact that you have to be very financially sophisticated. You need to be very well connected to the exchanges, connected to searches. You need to have very fancy algorithms for market making, for arbitrage, for this and that.

And this is the reason why there's only two of the builders that build the vast, vast majority of the blocks on Ethereum, because that's where the centralization comes in. On the other hand, the gateways, which are just relays for pre-confirmations, only require DevOps as a barrier to entry. And DevOps is something that thousands of companies around the world have the skills to provide. This is not some sort of black magic that is required for block building. And so in that sense, I actually believe that pre-confirmations is less centralizing that the status quo, where you have these two middlemen, the builders and the relays, in between the user. and the sequencer. Okay, so that's the consensus layer, aka the beam chain, which is basically

the presentation you gave at DevCon. And this, again, is kind of the technical network side of, you know, competing against Bitcoin. Of course, we've left unaddressed the whole memetic narrative side of things, which is like, how do you like, you know, propagate and get everyone else to sort of like meme into existence, eat this money the way maybe Bitcoin has done? We'll leave that for another episode. In fact, we've done episodes on that, of course. I'm tempted to blow right past the DA layer, which is basically it looks like this is an orange in the diagram if folks are following, which is basically like scaling up blob space essentially. And it just feels like it's pretty straightforward. We're just getting more and more blob space as the years go ahead. And like

incrementally with multiple forks, we're just scaling that part of the roadmap. So I feel like we don't have to camp on there. What would be really interesting from my perspective, Justin, and the story that has been told less often is the execution layer on the layer one. And this has been sort of a chant that has happened recently. Indeed, when we started this episode, that's sort of what I was hoping for from your announcement at DevCon is kind of like, what are we going to do to make the Ethereum layer one great again and to improve that economic zone. And I do see at the end a box that says native roll-ups. But maybe before we get to the end state here, can you just like, summarize, what is the vision for, in this layer one roadmap, Ethereum's layer one execution layer?

Is this a strategy to make the Ethereum layer one great again? Or like, what does this even look like in your mind? So at the execution layer, we're very much constrained by the way that the EVM is defined, right? The EVM is directly consumed by application, and there is a need for forward compatibility. So the way that we change the EVM is extremely slowly and cautiously so that we don't break the existing applications. And so if we're going to make big, massive changes to the AVM, we need to think out-of-the-box. And the out-of-the-box solution basically boils down to snarkifying the whole EVM. In some sense, snarkification is something that is done outside of the EVM, as opposed to changing the internals and the guts of of the EVM.

Now, one of the big performance metrics of the EVM is gas per second. How much execution can we push through the EVM? And the main bottleneck that we have right now is on the validator side of things, they have to verify that these EVM blocks are valid. And the way they do it today is the naive brute force way, in the sense that they have to redownload the entire block, and then they have to re-execute the entire block. And so now you see that the bottleneck is basically the slowest validator on the network

and their ability to validate the block. But what we can do with ZK is that we can create a massive asymmetry between the performance of the builders, on the one hand, that produce the blocks and those that verify the validity of the block, they only have to verify a snok. So the primary reason why we have a gas limit is to prevent a massive block from coming in and then it takes many, many seconds to verify and that's not enough time for the validators to come in and make sure that it's valid.

So ultimately, the strategy here is to grow the gas limit by 10x, by 100x, just by removing this possibility for denial-of-service attack with these very, very big blocks. and the reason, again, is that the validators only have to verify a snock which takes roughly one millisecond, regardless of how much computation was done. So once we've snockified the whole EVM, which is a process that happens off-chain, one of the very first things that we can do is start increasing the gas limit and do so safely without compromising decentralization at all on the validated side of things. But then comes the really cool thing, which you mentioned,

the native roll-ups. What the native roll-ups allow us to do is to scale the EVM horizontally. It allows any developer who wishes to launch a new instance of the EVM to make basically a copy of the EVM, and each copy will have its own gas limit. So if we're able to have, you know, let's say a gas limit that is 100 times larger than today, so that would be three giga-gagas per block, well, you know, now we can have N times 3 gigas per block, where N is the number of copies of these native roll-ups that have been programmed by developers. Now, one of the important things to stress here is that these are programmable native roll-ups, not the, you know, top-down, centrally controlled type of roll-ups known as execution sourcing. So in the previous Ephraim roadmap, we had this notion where there would be, you know, be a fixed number, for example, 64 shots or 1,024 shots, and they would be exact copies of

the EVM. And my personal belief is that this goes against the ethos of Ethereum, because the ethos of Ethereum is all about programmability. And so we should be able to program several things. One, the number of roll-ups shouldn't be some sort of hard-coded and dictated number from the top down. It should just be an emergent property of the market. And then the second aspect is that if someone wants to customize the roll-up, for example, have their own gas token, or have their own sequencer, or have their own governance system around it, then they should have the option to do so. And what that would allow is for all of the existing EVM roll-ups to upgrade to become native roll-ups. And in that sense, what will happen is that all of the roll-ups would become one step

closer to becoming Ethereum itself. Right? Ethereum has many things to offer it. It can offer settlement. It can offer DA. It can offer sequencing. And it can offer the virtual machine. And there's terms that we use for each of those.

If you consume settlement, we say that you're an L2. If you consume DA, we say that you're a roll-up. If you consume the sequencing, we say that you're based. And if you consume the virtual machine, we say that you're native. And so if you do all of these things, and your based native roll-up, then the term that I use is you're an ultrasound roll-up. The maximally aligned thing possible.

But of course, it's up to you to determine how aligned you want to be with Ethereum. I want to put a pin on this subject because I really like this illustration. You say how aligned you are with Ethereum. The way that I would illustrate this is like how close to the core are you of Ethereum. The current roll-up model,

the roll-up that people know about today, Trom, Optimism, World Chain, Base, Inc. These kind of like, these are all kind of businesses. Like, base is a, that's like a corporation. That's a Coinbase. That's like a less aligned version of a roll-up. And then there's even like a layer two, which just uses settlement, doesn't use as DA, even less, like close to the core of Ethereum. And the progression that you just illustrated, like layer two, roll-up, base roll-up, native roll-up. It's like a set of concentric circles, and each one you illustrated is like, it consumes one more part of Ethereum layer one resources. And one of the big conversations going on lately is like,

well, layer twos are unaligned with Ethereum. They're parasitic to Ethereum. They take all the execution fees. They give Ethereum layer one meager amounts of data availability fees, and then they keep all the execution fees for themselves. And they are also highly inoperable with each other. That's another big problem of the Ethereum current structure. As you were illustrating the versions of roll-ups that are a little bit closer to the core of Ethereum, base roll-ups that consume sequencing and native roll-ups that consume execution, these flavors of roll-ups consume much more layer-one resource costs. And they also get much more layer-one composability and composability with roll-ups of their

same ilk. And so a mental model, Justin, that I'd like to check with you, is that, there are new flavors of roll-ups that people are just learning about, we are just learning about, base roll-ups and even closer to Ethereum native roll-ups, that when they are composable with each other, are consuming far more layer-one resources than the roll-ups that are like the optimisms, the bases, the arbitrams. And so the check is that this is like kind of like fixing the economic relationship between roll-ups and the Ethereum layer one. Composable roll-ups need to consume much more layer one resources.

And the mental model that I'm beginning to understand is that layer one resources, if you want to be interoperable with a broader roll-up ecosystem, need to be consumed. And this ultimately shows up as both layer one execution fees in the case of native roll-ups and even further blob space fees for base roll-ups. So I want you to check that understanding that I have. It's like if you want to be composable with broader Ethereum, you need to consume more layer-one resources in order to do that. Yeah, great question. I think directionally, your mental model is correct, but I just want to

highlight a couple aspects of it. The first one is that it's not concentric circles. You can choose to be one, but not the other. So you can choose to be based, but not native. You can choose to be native, but not based. You could be a based validium that doesn't consume the DA, or you could be a base roll-up that does consume the DA. So all of this is up for you to decide. Now, in terms of advantages, there's different types of advantages that are being provided from consuming the different things. And I don't want to necessarily go through all of the combinations, but one of the big ones is this notion of shared security. And then another big one is around shared composability and synchrony. So if you want to have the maximum security,

then what you would want is to consume the DA and the execution. And the reason why I highlight the execution is because oftentimes today, especially if you want to be EVM equivalent, you need to introduce governance for two reasons. One is that if your copy of the EVM has a bug, then you need governance to step in to fix the bug. And if the L1 EVM changes, then you need governance in order to change your copy to mirror the L1. And unfortunately, governance is an attack vector, like fundamentally. And so wouldn't it be nice if you could have a virtual machine, which is completely governed free and therefore has a chance to inherit the full security of the R1? Now, if you want composability, then what you want is to consume DA and sequencing.

And the reason is that you want to have shared sequencing, and the only way to do that is to reuse the L1 sequencer so that you can synchronously compose with the L1 and other base roll-ups. And you also want to use FMDA so that you don't reintroduce a new sequencer. So this is a problem with Celestia. Celestia is itself a sequence chain of blobs. And if you were to consume it, then now suddenly you have this other sequencer which breaks the composability with the R1. Now, one other like subtle thing, if you want, like really, really good composability is that you also want to be native. And the reason is that there's this little trick for the native roll-ups,

which is that the proof of validity for the execution can be delayed by one slot. So this is actually a proposal by Max Resnick and Dan Robinson to delay the execution state route by one slot. And what that would mean for the native roll-ups is that the proof has a full slot to arrive. And so if you want to do synchronous composability, all you need is subslot latency. on the proving side of things, as opposed to like hardcore real-time proving, which would be on the order of 100 milliseconds, let's say. And so the opportunity to do synchronous compatibility would be much, much simpler across these native roll-ups.

If I were to summarize, like, the execution layer, there's a lot of detail here, right? But like, it's basically Ethereum's execution layer strategy is roll-ups in three flavors, first the roll-ups that we know today, just roll-ups, and then based roll-ups, and then native roll-ups. And hopefully there's a progression chain where there's like economic incentives, coordination incentives, to move large roll-ups down that stack

and closer towards native roll-ups. Now, I want to ask a question, Justin, about native roll-ups because others have proposed native roll-ups. I think Martin Copelman from NOS gave a presentation at DevCon about native roll-ups as well that is similar in some ways.

he was using the same term, but I also think different. I think maybe what he was proposing was something closer to like, you know, execution sharding, right? Basically, like we have, you know, a set of Ethereum governance defined roll-ups. It's less pluralistic. Anyway, what are the differences between Martin's proposal and term for native roll-ups and like what you're talking about here? Or differences in similarities, really. Yep, fantastic question.

So I think you're right. Like the term that is usually used is execution sharding and, you know, somewhat confusingly, he used the term a native roll-up for something that is not a native roll-up. So there's two main differences. I think the first one is around the programmability, which I've already expressed, which is that with the native roll-ups, you can have as many as you want, and you can configure them by choosing the sequencer, the governance token, the gas token, all of those

things. But then there's another very key difference, which is around the incentives. It turns out that the native roll-ups are much, much, much, much-frofenes. friendlier to the existing roll-ups. And the reason is that the existing rule-ups can upgrade to being native roll-ups, and they can do so for free. They don't have to pay anything. Like, it's just a raw, commoditized technology that is given for free to the existing roll-ups.

On the other hand, the execution shards would have EF as the enshrined unit of gas, and they would have their only IP-159, and all of that would be, Fiat rule for Ethereum. And unfortunately, these two visions are not compatible. You have to choose one or the other. Now, my personal, I guess, thesis is that we've already lost the opportunity to capture congestion fees at the roll-up layer. This is not something that we can go back. In some sense, we've made a gambit, right? We've told the roll-ups, hey, there's this amazing source of revenue for you, which is the execution fees. That should be a great incentive for you to go build the Ethereum ecosystem. And remember what I said around the separation of concerns between L1 and

L2. The L1 is all about providing maximum robustness and credible neutrality. The L2s is all about scaling. So they are responsible for doing all of the BD and getting all of the users and all of that. And this is the reason why, for example, Coinbase is building on Ethereum, right? Because Coinbase has all of the users and it has an opportunity to bring them and make money in the form of execution fees. Now, one of the things that, you know, is still under discussion is opting into becoming a base roll-up free as well, right? Because from a nominal standpoint, you don't have to pay any gas to use the R1 sequencer. But what some people will say is that by opting into the

R1 sequencer, you lose the M-EV. And so you're indirectly paying for the right to use this sequencer. Now, the good news here is that I believe that MEV is going to be captured by the applications and as such won't be a source of revenue. And so if we accept this premise, then these two pieces of infrastructure that Ephem would provide the sequencing and the execution would be provided for free. There would just be pure technology that would ultimately help the roll-ups do their job, which is to acquire as many users as possible, and provide a complementary job to the R1. If we could replay this back and do it all over again, is there a part of you that wishes we started with the based roll-ups

or even native roll-ups? Or is this the way it had to be, you know, play out? Like, if we had started with based roll-ups, like, wouldn't everything be better now? It's an interesting question, and a difficult one to answer. In some sense, we don't have a choice, right? Because we don't have the technology to go directly

with these execution shots, which you can think of as being ultrasound roll-ups or based and native roleups. And I think had we gone down that path, we wouldn't be as aligned with the programmability ethos of Ephem. And also, there would be less of an incentive for ecosystem players outside of Ethereum to integrate with Ethereum. Like right now, what we're seeing is many exchanges

kind of coming in and building on top of the firm. We're seeing Cracken, we're seeing Coinbase. And then there's, you know, project like Walt Coin, that are coming in. And, you know, unit chain, for example, building its own chain. Like there's been commentators that have said, hey, uniswap, they have the website. They originate a lot of the flow, meaning that the users, you know, come to the website. In some sense, why don't they build their own L1 and capture all of the revenue directly there?

But the incentives are such that you want to be building as an L2 because you get a lot of the security and composability that ultimately leads to more revenue, all while at the same time providing this symbiotic relationship with the L1 because you're growing the users and providing to the L1 a source of revenue, which is on the one hand congestion fees at the DA level. And on the other hand, kind of this memetic premium of making use of Ephemeyerum layer 1 of its token and growing the network effects around it. You called this a gambit earlier, Justin. I think that's a great take. And we're in this like part of the gambit where it's hard for the market to tell. It's even sometimes hard for bankless listeners to tell

myself to tell whether this was like a brilliant move or if we just effed it up. It's like we don't know yet until the story totally plays out. I have long to believe that this is actually a brilliant move to like bring in all of the roll-ups, bringing in all of the ecosystems and have them kind of build out the execution layer. And it's like, I look at this roadmap and I see like, wow, this is mega bullish, right? To me, I look at this and I'm listening to the plan. I'm hearing kind of the tradeoffs and the balance of concerns and what Ethereum is trying to do. And I get bullish. A whole bunch of other people look at this roadmap and they say, this is too complicated. Like, you can't sum it up in a tweet. You know, Ethereum's going to be like pulled apart.

And I think the market is closer to believing that right now, which makes Eath holders, you know, have to be in the conviction zone right now. And I guess maybe one of my questions as we draw this episode to a close is, okay, so like, if we're thinking long term, right, back to David's original premise here, and we see this at the long term vision. But if we're not winning enough of the middle game here, right, this cycle, let's say, we don't win enough market penetration, like traction, we get, like pull the part this cycle, and we kind of like lose this middle game. If you lose,

lose momentum, yeah. If you lose enough middle games and if you lose enough momentum, are you at risk of losing kind of the end state game? And so part of my attraction to Ethereum is like it thinks in terms of years and decades rather than kind of like month to month cycles. But I'm worried at the same time that like month to month is the thing that actually like begins to, I mean, what's a year, if not 12 months, right? It all kind of like, it's all middle game, all the way down, is another way of looking at that. And I'm wondering what your thoughts are on this meta problem. Now as like a whole bunch of ETH holders are kind of like looking at prices and they're saying, ah, I don't know if I fully understand this roadmap, nor do I know if I fully believe it.

Do you have any takes on this? So I do agree with you that, you know, we shouldn't be complacent. And one of the things that would be, you know, very bad in my opinion, is if Solan, were to flip Ethereum. And so I think in the short and medium term, we should think very, very pragmatically and we should put aside some of the ideology, at least for the short and medium term.

And so the two things that I would encourage the ecosystem to do, especially the L2s, is to be very pragmatic and almost fake it till you make it and be ready to fake it even harder than Solana's willing to fake it. So far, what we've done is we've launched L2s, without fraud proofs. We've launched L2s with centralized sequences. And I think we can go even further and have L2s where you consume AltDA as opposed to consuming Ephem DA. One of the potential

outcomes in the next few months is that Ephemda is going to become extremely, extremely expensive. You know, think cents per transaction or tens of cents per transaction. And I think it should be totally socially acceptable for some of those roll-ups to at least temporarily move away from this expensive IFRMDA and use ALDA like eigenDA or Celestia up until to the point where they have a very, very good reason to come back, which is to have first-class security and first-class composability. The other thing that I would try and encourage the ecosystem to do is to band together, right? together we are stronger. And the good news is that we have a sense of shared mission.

We're all trying to make Ethereum the best platform possible. And there are actually incentives to band together within Ethereum. One of the things that I believe in is that there are network effects at the roll-up level. And so if these network effects are strong, then what could happen is that one dominant L2 or roll-up starts eating up the rest of the ecosystem. And the silver lining here is that now there is an incentive for all of the other roll-ups to come together and fight the dominant one. So you could kind of imagine the dark force, right, which has no incentive to try and be

composable with the rest of the ecosystem because they're trying to go for the whole pie and then kind of this grassroots movements of lots of small guys all banding together. and the way that we band together, I think, is through shared standards, shared infrastructure like sequencing, shed deposits and shared bridging, shared block packing and data compression, shared proof verification. Like basically all of this sharing of infrastructure whereby everyone who opts in is contributing to the network effects and to the amortization of costs. So it is quite expensive, you know, to verify a snark on chain,

especially you have to do this in a regular basis, for example, once per minute or once per slot even. But once you've shed that cost and you amortize it across and different roll-ups, it becomes virtually noise. And so what I would encourage the developers and the ecosystems to do in 2025 is to focus on sharing because sharing is caring. and ultimately is what's going to allow us to fight competitors like Solana. Justin Drake, this has been great.

So everything we just talked about including that roadmap, are we allowed to call that Ethereum 3.0? The theorem 3.0. Can we not use that term? The final Ethereum 3.0 is for sure Ethereum 3.0. All the layers. Is this Ethereum 3.0? So I try to stay away from the term, Ethereum 3.0 for various reasons.

The first one is that I think we already have consensus on what the roadmap is, what are the items of the roadmap, the meat, I guess, the flesh that we would put on the bones. The real question is, like, how do we package it? How do we prioritize it? How do we solve the social layer coordination problem? The other reason why I would try and avoid the term, Firm 3.0, is because we already have terms for all of the existing layers, right?

We have the notion of bank sharding for the data layer. We have the native roll-ups for the execution layer. And now we have this new term beam chain for the consensus layer. And I don't think it's necessary to add yet another term. And then the other thing I'll mention is that the term Affirm 2.0 caused a lot of confusion in the space. Like people thought it was a new chain, a new token. Even the regulators got confused and the critics used it as a way to criticize Ethereum. I don't think it's necessary to go down this route of Ephraim 3.0.

All right. We'll hold off the meme layer as long as we can, but no promises there, Justin. Justin Drake, as always, thank you so much for joining us in Bankless today. Thank you. Got to let you know. Of course, crypto is risky. red map stuff is risky as well. This is a gambit, I guess. You could lose what you put in, but we're headed west. This is the frontier. It's not for everyone, but we're glad you're with us

in the bankless journey. Thanks a lot.