Bankless - Do Based Rollups Solve Everything? | Justin Drake

Starting point is 00:00:00 From my perspective, like once you appreciate base sequencing and your base pill, there's no going back. I don't know of a single person who has gone the other direction. It's kind of up only, up to the right in terms of the converts. Welcome to Bankless, where we explore the frontier of internet money and internet finance. And today on Bankless, we explore the frontier of based roll-ups. Today on the show, we have Ethereum researcher Justin Drake to give us the reasoning why he thinks the future of Ethereum is. based. So what are based roll-ups? How do they differ from normal roll-ups? Why the Ethereum layer 1 is the best sequencer for Ethereum's layer 2 roll-ups? How does it fix Ethereum's fragmentation?

Starting point is 00:00:47 And what does this all mean for Ether, the asset? There's already a handful of teams publicly working on the based roll-ups future, Tycho, puffer, spire, but also Justin alludes to at least a few of Ethereum's biggest roll-ups already planning to go based in the not-so-distant future. So with his episode today and perhaps a few more soon, we are front-running all of that by getting you downloaded on based roll-ups. So when Ethereum inevitably does go based, you'll have already seen it coming. I think both David and I, you'll find in this episode, are very bullish on the concept of based roll-ups because it really solves a lot of the challenges that Ethereum is facing with its current roll-up-centric roadmap. So this is the most comprehensive episode I think we've ever done

Starting point is 00:01:28 on based roll-ups. And Justin toward the end said that he thinks that based roll-ups have really reached escape velocity. I know some people listening to that are like, what? They just came on the scene, but realize a lot of work has been going on behind the scenes to sort of like propagate the idea and design architecture for based roll-ups. I think it's getting ready to take off. So this is maybe a few months in advance of front running, but 2025 could be a big year for based roll-ups, David. And I want to talk to you about all of that in the debrief that we do right after this episode. If you are a bankless citizen, you're on bankless premium. That is an extra podcast, no commercials, It's got the bonus content David and I put in every single week.

Starting point is 00:02:06 You have that in your feed. That's the episode that we're about to record where we give our thoughts on the episode. It's called the debrief. So go check that out as well. Speaking of extra content, Justin is giving a talk at the bankless summit. So if you are going to DevCon and want to hear Justin talk, the bankless summit is happening the day after DevCon. It's like TED Talks for Ethereum. I've hand-selected some of the best speakers in Ethereum and charged them with the responsibility of producing the most interesting talk possible.

Starting point is 00:02:32 I asked Justin if he could leak anything about his talk at the end of the episode. He doesn't give me too much, but he did say in a chat that it's painfully bullish, so I'm excited to see what all of that means. So let's go ahead and get right into this episode all about base rolls with Justin Drake. But first, a moment to talk about some of these fantastic sponsors that make the show possible, especially Cracken, our preferred exchange for crypto in 2024. If you do not have an account with Cracken, consider clicking the links in the show notes getting started with Crackin today. If you want a crypto trading experience backed by world-class security and award

Starting point is 00:03:02 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 hundreds 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 and Cracken NFT Marketplace and a seamless app to bring it all together,

Starting point is 00:03:35 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. The Arbitrum portal is your one-stop hub to entering the Ethereum ecosystem. With over 800 apps, Arbitrum offers something for everyone. Dive into the epicenter of DFI,

Starting point is 00:03:59 where advanced trading, lending, and staking platforms are redefining how we interact with money. Explore Arbitrum's rapidly growing gaming hub from immersed role-playing games, fast-paced fantasy MMOs to casual luck-battle mobile games. Move assets effortlessly between chains and access the ecosystem with ease via Arbitrum's expansive network of bridges and onrifts.

Starting point is 00:04:20 Step into Arbitrum's flourishing NFT and creators-based, where artists, collectors, and social converge and support your favorite streamers all on-chain. Find new and trending apps and learn how to earn rewards across the Arbitrum ecosystem with limited time campaigns from your favorite projects. Empower your future with Arbitrum.com. Visit portal.arbitrum.com. To find out what's next on your Web3 journey. Uniswap wallet is officially the preferred wallet of bankless, and it's the one we use any time when we want to transact on chain.

Starting point is 00:04:50 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 tap in business. thousands of tokens across Ethereum and over 10 other chains like base, arbitram, and optimism. Uniswap wallet delivers deep liquidity, fast execution, and reliable quotes with zero gas swaps through Uniswap X. And when it comes through 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.eith username directly in the mobile wall. Start swapping

Starting point is 00:05:27 smarter with Uniswap. Bankless Nation, we are once again here with Justin Drake, which means that we are going to learn something today. Justin Drake, of course, is a reacher at the Ethereum Foundation. He has taught us many things on the Bankless podcast. And today we are hoping to learn about based roll-ups. Justin, welcome back to Bankless. Hi, David and Ryan. Thanks for having me. So before we get into the idea of based roll-ups, first talking about roll-ups generally, how would you say the roll-up-centric roadmap is going? The Ethereum effort defined and then moved to towards the roll-up-centric roadmap in 2019, so five years ago. Five years later, how would you evaluate our progress, our direction,

Starting point is 00:06:05 in the vision that Ethereum set out for itself in 2019? Yeah, I think we're doing exceptionally well on certain dimensions. One of them is that there is a lot of experimentation and a lot of teams that have taken it to build an L2 or roll-up or Validium. Another thing that I'd say we've been successful at is dramatically reducing the fees. And so in some sense, scaling feels like a soul's point. problem. We have a whole roadmap to increase the amount of blob space, and we've seen tremendous progress with the fraud proofs and with the validity proofs. And so from the perspective of scaling

Starting point is 00:06:40 and having relatively secure roll-ups, I feel like we're doing exceptionally well. What I think that we're starting to transition into is basically the interrupt phase of the roll-up centric roadmap, where we realize, we acknowledge that we have a fragmentation problem, we have a siloism problem, and we'd like to bring back some of the network effects that previously existed at layer one and that we have lost. The network effects at layer one were unsustainable because we had anti-network effects, namely transaction fees, right?

Starting point is 00:07:15 It's not sustainable to have a $100 transaction fee, and that's something that we've solved. But now we have another anti-network effects, which arguably, I'd say, is decentralized sequences, which prevents this optimal form of composability, which I call synchronous composability. When you say network effects, Justin, what do you mean by network effects and why are they important?

Starting point is 00:07:36 So network effects is basically when you have a network and every additional node brings in more synergies than just its raw weight. So you can think of it as just 1 plus 1 equals 3. you have positive sum games, you have win-win situations. And one example of network effect would be shared liquidity. Every roll-up kind of adds to the liquidity pool, and now Ethereum as a whole can become this defy hub

Starting point is 00:08:07 with a best-in-class execution and whatnot. But right now we're in a position where liquidity is fragmented, and so that particular network effect has started to break down a little bit. So David asked the question, like, how's the roll-up-centric roadmap going? And you talked about some of its strengths. You also mentioned toward the end some of its, you know, weaknesses maybe short term or medium term or long term. I'm not sure what your perspective is. But could we dwell a bit more on some of the, I guess, you know, tradeoffs or some of the weak points, some of the areas that Ethereum needs to figure out? So one is fragmentation maybe and a loss of some network effect. Are there any others that pop into mind here? Associated with the defragmentation, there's a little bit also of memetic confusion and also a loss of unity maybe at the social layer. There's competitive forces, there's maybe onaligned incentives and things like that. But nothing that fundamentally cannot be solved. So I remain optimistic. And interoperability is a big word under which there's lots of many different constituent parts. So one of the low-hanging fruit, for example, is just having basic shared standards at the wallet level, the equivalent of ERC20, but in a cross-roll-up

Starting point is 00:09:19 model. The idea of having shared governance, this is something that's being experimented on, shared sequencing, shared snark aggregation, shared data compression and blob packing, shared deposits, the idea of having shed fungibility. All of these things are being worked on in parallel, and directly we're making tremendous progress towards fixing this fragmentation. When you say memetic confusion at the social layer, a few things that come to mind for me. And I just want to sort of name them explicitly to make sure we get kind of our problem statement, right, on the sort of the social layer side of things. Like one memetic confusion area to me seems to be this idea that L2s are parasitic. Well, are they parasitic?

Starting point is 00:10:03 Like that's one area of question, which didn't exist prior to the roll up centric roadmap. There was like only one kind of Ethereum shared state. So it's very obvious that every app that deployed on Ethereum was like net good for Ethereum. The social era would sort of hands clap cheer it on. So maybe that's one area. Another area of memetic confusion, I would say, is around ether the asset. And this question of, okay, what are the value accrual mechanisms of ether in this rollup-centric world? We've seen fee revenue go down, at least in the short term, you know, post-Protank sharding and the deployment there.

Starting point is 00:10:38 And so now there's a question of, well, what about this ultrasound money thing? There's not as much fee burn at the base layer for ether the asset. So these are two areas of memetic confusion that I've seen, this idea that L2s are parasitic, this idea that there's no value accrual from a cash flow statement for ether the asset. Is that what you were explicitly referring to when you said memetic confusion? Or if not that, what else? So I think you summarized the perspective very well of the ifholder,

Starting point is 00:11:05 who's a little confused, but there's other entities that are also confused. As a developer, you know, I used to, have it very, very easy for me, I knew that launching on Ethereum just meant deploying on layer one. But now I have to choose across a whole ecosystem, and it's a little bit more confusing. And similar thing as a user, now I need to make an explicit choice as to which ecosystem I need to put my assets on. And then, you know, there's like 20 different flavors of ether with every single deployment of a roll-up. So I think we're in a position of max awkwardness in the puberty phase, and hopefully from from now onwards, it's downhill.

Starting point is 00:11:43 I want to zoom out a little bit and kind of look at the progress of Ethereum in a sense of order of operations. We once upon a time had the very close-knit nucleus of the Ethereum layer one. And in 2019, after we kind of solved plasma, which turned into roll-ups with this idea of fraud-proofs, we had these roll-up teams that extended outwards from Ethereum,

Starting point is 00:12:04 like these vertical integrations, these things that would grow away from the Ethereum layer one. You know, optimism, arbitram, Polygon, all of these networks that would scale away. Like you would move your assets from the Ethereum layer one, you know, away to these layer twos. And this is what would create the fragmentation. And now that you illustrate us being in the interoperability phase of the roll-up-centric

Starting point is 00:12:24 roadmap, now it's like we're filling in the space that has been created as these roll-ups have extended outwards from Ethereum. And I think there's an order of operations right. Like, first we needed to create the fragmentation problem before we could actually solve the fragmentation problem. I don't know if we could have even gone back in time in the Ethereum roadmap and tried to solve this fragmentation problem before it was even created. Maybe you can kind of just talk about the benefits of the fact that we first created all of these different roll-ups, and now we're kind of just like filling in the gap. How do you feel about the order of operations of how Ethereum progress has gone?

Starting point is 00:13:01 Yeah, I very much agree with you that it's a logical progression. I mean, there could have been an alternative past and alternative ways for Ethereum to tackle things. one way would have been to go down the Solana route. Obviously, we don't want to be taking these kinds of shortcuts. Another thing that we could have done is to scale Ethereum layer one using basically randomly sampled committees, which you can think of as sharding. And so here, the safety of each shard is not done through fraud proofs or through validity proofs. It's done through naive re-execution from committees.

Starting point is 00:13:37 I guess what we did is that we took a very pure path where we said that we don't like honest majority assumptions. And we know how to do these things without an honest majority assumption. And so we kind of opened it up to the broader market to build these roll-ups that are, you know, in some sense, fundamentally more robust because they only rely on pure cryptography or, you know, these cryptocurrencies games, which are ford-proofs. reflecting on the past and the different strategy that we could have taken, I guess one advantage of this sharding approach that we didn't take would have been that we would have had more homogeneity across the different roll-ups. But the downside is that we would have had way, way less experimentation. And I'm actually quite pleased that we don't go down the sharding route

Starting point is 00:14:24 because there's a new way to do execution sharding, which is so, so much more powerful. So first of all, it uses SNOCs, ZK roll-ups, to remove the honest majority assumption that we knew how to do, but now the technology has caught up that we have line of sight towards having data layer one. But the other interesting aspect is that we can take this ZK EVM and create a pre-compile within the EVM to programmatically deploy instances of the EVM. So with the vision that I described that we didn't go down, we would have had a fixed number of shots.

Starting point is 00:15:02 something like 64 or 1,024. And that is kind of an awkward position to be. It's just an artificial number that we had to pick. Whereas with this pre-compile, we can have as many as we want. And then the other aspect of programmability is that we allow room for creativity. So we can, for example, have ZKVMs with different sequences, with different governance tokens, with different fee mechanisms. Everything around the virtual machine is fully programmable.

Starting point is 00:15:31 I'm wondering, Justin, if you have an opinion, on the following question. In addition to kind of fragmenting the Ethereum network, because we have all of these different roll-ups, we've also kind of fragmented DeFi apps, right? There's many different implementations of Uniswap across all these different networks. There's many different implementations of AVE across all of these different networks. And DeFi is something that is strictly better when everything is kind of contained in the same environment. Composability and liquidity capital, it's just more efficient in effect when everything is in the same spot. So in a future of Ethereum that we're hoping for, do you see that there are fewer implementations of these Aves, uniswaps, of these applications? And is that kind of something that we should be striving for or hope to be striving for? Is that like a success scenario when maybe instead of like 27 different implementations of uniswap across all of these different chains? There's actually just like maybe just one. Maybe there's just one implementation of uniswap or all the liquidity has aggregated. Maybe there's just one major

Starting point is 00:16:30 implementation of Ave? Is that a goal? Is that something to aspire to or is that maybe something too prescriptive? I mean, I'm personally a believer of the app chain thesis. And I think it does make sense once the barrier to entry to deploying an app chain is low enough that there is one unit chain, one Ava chain, et cetera, et cetera. The issue that we have right now is that each individual app chain comes with overhead. So, for example, if it's a ZK app chain, each one of these ZK app chains need to verify a ZK proof. And if you have a thousand ZK app chains, then you need to verify a thousand proofs every once in a while,

Starting point is 00:17:06 and that's just very, very expensive. So we need infrastructure to, for example, aggregate the snock proof. So you have one snock proof to verify a thousand app chains. And it's a similar thing with the blobs. The blobs are like fairly large chunks of data, and it's possible that a single app chain only consumes, you know, part of it, let's say, one quarter.

Starting point is 00:17:27 And so you want to take all of these data, different app chains and do the block packing together so you have optimal utilization of resources. A similar thing to block packing is this idea of data compression. The more data that you're compressing, the better the compression ratio. And so what's happening right now is that each individual roll-up is doing its own little isolated compression when really what we should be doing, and this is a network effect, it's a win-win thing, where we take all of the data across all of the roll-up, all of the app chains and we do one fell swoop data compression. Justin, you were talking earlier about, you know, like you're happy, you're satisfied that we took the roll-up-centric roadmap

Starting point is 00:18:08 because ZK has made pre-compiles on the L1 possible. And so we have more potential for, you know, like rolling out ZK-based like VMs and incorporating that into the layer one. I want to isolate that for our conversation because I don't think that is the based roll-up conversation, right? This is something that we're talking about in the medium to long-term future and is something that is tangential to based roll-ups. Am I correct in that? That's exactly right. It is orthogonal to base roll-ups, and it's called native roll-ups. So when you look at an L-2, there's three different important criteria. Criterion number one is the data aspect. Do you put data on-chain or do you put it off-chain? The former is called a roll-up, the latter is called a validium. And then the other layer you can look at

Starting point is 00:18:58 is the sequencing layer, like how is the ordering of transactions done? And here you have base roll-ups, which use the L-1 for sequencing, and you have non-base roll-ups that don't use the L-1 for sequencing. And then the final aspect is the virtual machine, the execution. If you are exactly EVM equivalent, because you're using this pre-compile, which is basically Ethereum exposing the EVM to the application layer with this pre-compile, then you are a native roll-up, as opposed to being a custom roll-up with a different virtual machine than the exact copy of the EVM.

Starting point is 00:19:37 I mean, we should do an entire separate episode at some point in the future, probably on native roll-ups, because it sounds like that is kind of the almost the final evolution of bringing some of this execution back to the layer one validators. The reason we're not talking about it for the bulk of today's podcast is because it's a more distant future, right? Think of that as the medium to long term, whereas base rollups, my understanding, Justin, that's the here and now. Like, we're ready for base rollups essentially at this moment. This is not a distant roadmap sort of EIP in the future that we're going to incorporate in some hard fork, though there might be some upgrades necessary to fully optimized based rollups.

Starting point is 00:20:15 But is it right to think of based rollups as maybe the next evolution of rollups in some sense, but the final evolution of rollups? And I don't even know if this is the end game, but maybe I'll just say the end game for roll-ups is native roll-up technology? Is that how you think about it? Yeah, so the very nice thing about base roll-ups is that they don't require hard fork. And so you have dozens of teams now that are racing to go build all the infrastructure and deploy them. To say that they're like ready for prime time might be a little bit of an exaggeration in the sense that we only have like one base roll-up in production at Tyco. And, you know, there's adjacent infrastructure that is being built out. but I do agree with you that the progress in base roll-ups is definitely happening,

Starting point is 00:20:57 and I think 2025 and 2026 should be very big years for base sequencing. As for native roll-ups, in order to get to that very exciting endgame, we need ZKEVMs, but not only that, we need real-time-proven ZKEVMs. And the good news is that we're definitely getting there. We know in theory how to parallelize snock-proving, so that you can get the latency down effectively as low as you want. And historically, we've had snockification of the EVM where a single block would take multiple hours, tens of hours,

Starting point is 00:21:35 and then it go down to single-digit hours, and then tens of minutes, single-digit minutes. And now we're at a point where it takes tens of seconds to prove a layer 1 EVM block. And I wouldn't be surprised if in 2025 we have single-digit at second, proving latency for the layer 1 EVM. Oh, very exciting. You can expect Bankless Nation at like a future episode, I'm sure, on native roll-ups

Starting point is 00:21:59 when that comes closer to fruition. But now let's get to maybe the main episode topic and the bulk of where we want to spend our time because all of what we just said is sort of a context. It's a preamble. And maybe, to your point, Justin, we are not ready for prime time for based rule-ups. But let's maybe call this the dawn of based roll-off because we are off zero. We have at least one. deployed in production, and you think, and I think we share this belief, that 2025 is going to be a

Starting point is 00:22:27 big year for these things we call based roll-ups. I feel somewhat that we should start with a definition. But before we get there, I just want to say the promise of base roll-ups. There's somebody on Twitter who made the comment that, hey, if we had based roll-ups when we started, you know, rolling out the roll-up-centric roadmap in, you know, 2020, 2021, then we wouldn't have the fragmentation problems or value-acrual questions that you were mentioned. these memetic problems in the social layer. If we just started with based roll-ups, and it's almost like, it's a shame we couldn't just start

Starting point is 00:22:58 with based roll-ups to begin with, because the promised of these things, as far as I understand it, and this is, again, the memetic narrative that we'll have to dig into the validity of this, that it solves fragmentation. You have shared liquidity across all of the base roll-ups, and indeed, also with the layer one.

Starting point is 00:23:13 So isn't that nice? We don't have the fragmentation problem. And two, it also solves the value-accrual problem to ETH because we are using the validators as the sequencers. And so we get to preserve some of the security around, you know, censorship resistance at a deeper level than maybe a traditional roll-up. And also, eith validators get some fees from this transaction, a greater share of the fees from this transaction. So if we just say those words, it almost sounds like, oh, base roll-up solves everything. And so coming into the episode, can we title this episode, Base roll-ups solve everything, Justin?

Starting point is 00:23:46 give us a definition for based roll-ups and tell us the promise here. Right. So just to set expectations, base roll-ups don't solve everything. They do try and tackle one specifically of the stack, which is the sequencing layer. And basically, when you build a roll-up, you have a choice as to what is the sequencing mechanism. Who are the actors who will determine the order of transactions? Now, since the dawn of blockchains, you know, Bitcoin 15 years ago, we've had essentially just one thing, which is called base sequencing, where the L1 is doing the sequencing, the validators, the consensus participants, whether it's minors, whether it's validators, all doing the sequencing.

Starting point is 00:24:28 And if you look at every other blockchain, whether it's Solana, the BNB, TESOS, XRP, you name it, they all have base sequencing. And for the vast majority of the existence of Ethereum, everything was just base sequence. That is the natural, normal, default, canonical way of doing sequencing. But then something happened a few years ago where we saw the emergence of these centralized sequences specifically in the context of the roll-up-centric roadmap.

Starting point is 00:25:03 And I think I attributed to three or four different reasons, extremely pragmatic reasons for those different roll-ups. One of them is that they could launch extremely quickly, and it was very, very straightforward as to what needed to be done. Just spin up a server, and that server is responsible for all of the sequencing. But there's other things that centralized sequencers provide. One of them is that they are a training wheel for security. So a lot of the virtual machines of the roll-ups are secured by fraud proofs or validated proofs. And these are extremely complex pieces of code,

Starting point is 00:25:47 potentially thousands and thousands of lines of code. And it just takes one critical bug to jeopardize the whole roll-up. So you can imagine draining all of arbitram in one fell swoop. That would be absolutely terrible. And so what the roll-up sequences are doing is making sure that the blocks that end up being produced are created using your default software as a, opposed to software that is specifically crafted to go exploit a vulnerability. And so as a black hat, even if I identify a bug in, for example, the arbitrar and for

Starting point is 00:26:24 proof, I'm not able to exploit it because I'm not able to convince the Shed Sequencer to go build a block that will go exploit this bug. The other aspect, you have a service, I guess you could say, that centralized sequences provide is MEV protection for users. So traditionally, you'd send your transaction in a public mempool, you gossip it, and now you're exposed to front-running and sandwiching, and that's a terrible user experience. With centralized sequences, wallets can directly connect to the centralized sequencer. It's an end-to-end encrypted connection.

Starting point is 00:27:01 So the only entity that is able to see the transactions before they land on chain is the centralized sequencer. and that entity is kind of trusted to not front-front and not sandwich their users. And so that leads to a great user experience. And then another final service that the centralized sequences provide is that of pre-confirm or extremely low latency user experience, you know, on the order of either two seconds for the super chain or even much faster in the context of arbitram on the order of 250 milliseconds. And you can think of the centralized sequencer as being a cheat code, as taking some sort of a massive shortcut, which is very strategically meaningful to do because it's very pragmatic, but it's not long-term sustainable.

Starting point is 00:27:51 And there's all sorts of reasons for that. One is that it's not credibly neutral, and we'll talk about that, and that means that it makes it very, very difficult to convince your competitors to join force. in shared sequencing. Like, would the optimism superchain want to share a sequencer with off-chain laps and arbitram? It's very difficult to pick one server for all of Ethereum. Another issue is that it is a regulatory central point of failure. It's also a vector for accidental liveliness failures.

Starting point is 00:28:30 This is something that we've seen in the past where roll-ups just randomly go offline for, sometimes several hours, and that is not what we want to base the future of finance on. And then finally, there's a very, very big trust component to centralized sequences. We haven't seen it yet, but we could imagine, for example, a centralized sequencer getting hacked and then millions of dollars of front-running happening because the centralized sequencer has been compromised. And then another aspect is that because centralized sequencing fundamentally leverages trust, as the technology, it's unscatable.

Starting point is 00:29:07 And so if I'm a random Joe, you know, a 16-year-old programmer in my mom's basement, and I want to go deploy a roll-lap, well, I don't have any trust. And so I can't go spin up my own centralized sequencer. And so I think what the ecosystem has realized is that really we want to go down this route of decentralized sequencing. That is, I guess, the end game, the proper way of doing things, the long-term sustainable path, but not only that, we also want shared sequencing. And this goes back to this idea of network effects and universal synchronous composability.

Starting point is 00:29:46 And in order to achieve this shared sequencing in an Ethereum-wide capacity, you know, going above and beyond the clusters that are being formed with Arbitrum orbit and the ZK-Sync elastic chain, we need credible neutrality. And so I think we're going to spend a lot of time in the rest of the episode. talking about one thing, credible neutrality. I want to keep on unpacking the sequencing thing because I think you alluded when we asked, what's the definition of a base roll-up? You alluded to the fact that, like, well, base roll-ups are the same as all other roll-ups

Starting point is 00:30:15 except for this one thing that's different, which is how we sequence things. That is the difference between base roll-ups and, like, arbitral optimism, is that it's only the sequencing layer, which is critically different in a base roll-up versus the roll-ups that we know today. But I do want to put a pin on the typical roll-ups in how they sequence things, the the centralized sequencer. Because I think in the crypto industry, we are all kind of like allergic to this word centralization. And sometimes overly so, decentralization, that's not decentralized. That's not decentralized. Sometimes it's used as like a cudgel in ways that it shouldn't be.

Starting point is 00:30:48 Leaning into like what Ethereum does and what Ethereum provides its roll-ups, the whole point of the roll-up-centric roadmap is that we can actually have centralized layer two's because of the checks and balances that the layer one provides its layer two. So, like the layer twos are in a state of can't be evil, even when they have centralized components. And that actually has been articulated in the Ethereum roll-up-centric roadmap, that that's actually a strength of roll-ups in their current form. Like, Arbitrum can be allowed to have a centralized sequencer and lean into the properties of a centralized sequencer because of the checks and balances that's provided by the

Starting point is 00:31:25 Ethereum layer one. So even with the centralization of the sequencer for optimism, Arbitram-based, whatever, there's only so much evil that can really be done here because of what the Ethereum Layer 1 provides. But maybe you're saying that even if that is true, we have still kind of like tapped out the juice there is to squeeze on the end of the spectrum of there being centralized sequencers. So like we've leaned into the benefits of centralized sequencers as far as we can. And we've hit the limits of the growth of the roll of centric roadmap because like, you know, Arbitrum as a centralized sequencer, optimism as a centralized sequencer, that is kind of the same thing. of play here. And now in order to get the next 10x of scale, we need to actually kind of go to the other end of the spectrum, which is the decentralized sequencer, which is where we

Starting point is 00:32:09 return to credible neutrality and refragmentation. I mean, one way to phrase it is that roll-ups significantly reduce the potential downside. We should be in a position where user funds are safe. And so security is a huge part of being a roll-up. And in some sense, that is what happens first before you can think of anything else. And if you look at the criteria of layer to beat, there's the stage zero roll up, stage one, stage two. Essentially, it's all about security, just making sure that if I have an asset on a roll-up, we can't have some operator that just steals these assets from me. We can't have some sort of security committee that steals them, or we can't have these assets be frozen. I always have the option to force a transaction

Starting point is 00:32:55 on-chain to, at the very least, withdraw my app. assets, and that is, you know, table stakes. But I think once we've all reached, you know, stage two and we have like these ultra-secure roll-ups, there's going to be a first for more. And really, we want strong network effects, like Ethereum-wide network effects. And so, yes, we have dramatically capped the downside, but we've also dramatically capped the upside. And so just to maybe illustrate for listeners here, we have the centralized sequencer paradigm of layer two's today, the one-of-one sequencers, where we get all the scale, we get the MEV protection, we've got a bunch of nice things. Would you say that base roll-ups are on the opposite end of that

Starting point is 00:33:39 spectrum, where you have a single centralized sequencer on one end, but then base roll-ups use Ethereum as a sequencer. We're going to get into that. But base roll-ups are kind of like basically on the farthest other end of the same spectrum of the decentralization of sequencing. Yes, that's exactly right. And in some sense, you know, the two ends of the spectrum meet and then you can go full circle. So as I mentioned, we started with base sequencing. That was the only thing available for 10 years of blockchains. And then we went down this path of centralized sequences. And I think what's going to happen is that we're going to incrementally move towards the centralization. And the kind of steps that are possible is one, what I call a federated sequencer. So you have a

Starting point is 00:34:20 multi-sig instead of trusting a single entity. And this is what Arbitrum is looking to do fairly soon. they have a 14 out of 20 multi-sig, and they're going down this path of a federated sequencer. The path after that is basically a permissionless and decentralized sequencer, where anyone, you know, for example, by locking some stake, can become a sequencer at some of the slots. So you have random selection across a permissionless set of consensus participants. And I believe, but I'm not 100% sure that ZK Sync is going down that path.

Starting point is 00:34:54 And so they're skipping the federated state and going directly for decentralized sequencer. And then there's some roll-ups out there. They say, okay, we're going to go straight for the end game. And we're going to go back to base sequencing. Oh, interesting. Okay, so there's different stages of decentralizing your sequencer that you can get to the federated and then the permission list and then all the way to the end we'll be talking about, which is based roll-ups.

Starting point is 00:35:16 And some roll-ups may just kind of like go all the way to the end. I want to set some more context for this because every time we've said centralized sequencer or decentralized sequencing, you know, listeners might be asking a question. Okay, like, refresh me again. What are we sequencing here? And what we're sequencing is the ordering of block transactions, right? So that, as we've illustrated in a many former bankless podcast, is incredibly important. It's kind of a godlike power that gives you, you know, abilities to say front run, you know, sandwich attack, all sorts of things. It is one of the things that those that are consuming block space actually pay for. So the sequencing, the ordering of

Starting point is 00:35:55 transactions in a given block is a very important power. And that's why we're talking about, you know, decentralizing it rather than keeping it centralized. But you illustrate another point in what you were saying, Justin, was none of this was sort of like roll-ups wanting to be evil. You know, we want to get our cut. We want to, you know, extract rent from users. Therefore, we're going to centralize the sequencer. And some of that framing is out there, I would say, on the memetic level that all roll-ups are doing this because they're greedy and evil assholes or something like this. And like what you're talking about is, no, actually, there's incredibly good reasons for this. From a user experience perspective, from a security perspective, from a like latency

Starting point is 00:36:34 perspective in order to be competitive and make this whole crypto thing useful, there's very good reasons to centralize the sequencer. So you're like you're kind of putting that framing out there. One thing that's become clear to me, I think, is clear to me. And I want to get you to verify with the roll-ups that we have today with some level of centralized sequencers, even if we get those to stage two, which by the way, stage two does not mean decentralizing the sequencer and necessarily, it's in my understanding. So you can still be stage two as a roll-up and still have a centralized sequencer. What you get from that is user protection against the cannot steal mandate. So a roll-up can never just like yoint your asset, can't steal

Starting point is 00:37:14 your assets. They can't sandbank been freed you. They can't go FTCs. They can't Alex Mishinsky, you know, all of your ETH, all right? So that's great, and that's something we've preserved. My understanding, though, Justin, correct me if I'm wrong, is in a stage two roll-up with a centralized sequencer, we may not have preserved another virtue that we have on the layer one, which is real-time censorship resistance. So let's say you want a transaction that is, you know, credibly neutral. And so let's say it's sanctioned by some third-party nation state and you want to preserve the censorship resistance of that transaction. Well, I believe some stage two roll-ups have kind of a forced inclusion where you can kind of like after the fact, like push that

Starting point is 00:37:58 transaction in. But it's not like as real time as the Ethereum layer one. And so even in a stage two roll-up with a centralized sequencer, if they, you know, blacklist some set of transactions, which they may have to do based on their jurisdiction and whatever, you know, regulatory regime they fall under, they may legally be obligated to do this, and you kind of lose that property that's inherit on the layer one, which is like more real-time censorship resistance. This is something I believe I've learned over the past, you know, kind of a set of months going through this as kind of a degraded capability as opposed to the layer one. So we preserve the cannot steal, but we may not have preserved the cannot censor, at least in a real-time fashion. Can you fact-checked

Starting point is 00:38:41 me? Is that kind of true with a stage two centralized sequence and roll-up? Yes, that is correct. So you could even imagine like an extreme example, a Wells Fargo chain, which is only sequencing transactions from 9 to 5 and is closed on the weekends and is filtering OFAC transactions. You could still have a Wells Fargo chain, which is at stage 2. And really, because, you know, there could be a very long delay between you initiating a forced transaction and the forced transaction actually landing on the chain, it would be useless to you. Like if you want to make a payment, you know, you'll pay for coffee. it would take 24 hours.

Starting point is 00:39:17 It would just not be workable. And so your only option, if you're being censored or if you just happen to want to make a transaction on a weekend in some sense, it's just to withdraw and exit completely. The chain is not useful to you if you are actively being censored. And that is pretty good, but I think we can do so, so much better so that we can make sure that the short-term censorship resistance, the short-term liveliness, is as good as the Ethereum layer one.

Starting point is 00:39:43 But in practice, what I think we've observed is that the liveliness is actually pretty good. And again, as you said, the reason why decentralized sequences exist is not because there's some sort of evil plan to the great liveliness. Like every time there's an accidental liveness failure, it's accidental. It was not intended. But really, what I see decentralized sequencing as unlocking is a credibly neutral shared sequencer. And that unlocks synchronous composability, and then that unlocks network effects and brings the theorem much, much closer together and ultimately yield an ecosystem that is, you know, an order of magnitude more powerful.

Starting point is 00:40:24 There's something that the base relaps do, which I think fits into the theme that we see across Ethereum at large, is that base relaps learn to repurpose the Ethereum layer one in a way that benefits them. And this time, it's using Ethereum as the sequencer. Can we go into what that actually means technically? So as a roll-up, what does it actually mean to use the Ethereum layer one as a sequencer? When we pop open the hood, what's going on under the hood? Yeah, so you're right.

Starting point is 00:40:48 In some sense, Ethereum was not designed to provide sequencing as a service. And you could say the same thing for data availability. So in the early days of data availability, we had call data, which was meant to be a place where you put bytes with basically inputs to function calls that you would have. and it was kind of abused in some sense by the roll-ups to put like arbitrary user data. It's kind of hacked in there.

Starting point is 00:41:16 It's hacked in there. And then, you know, we came out with a very clean kind of custom design solution, blobs that were, you know, purpose-built for the roll-ups. And there's kind of a similar story, I guess, going on with sequencing, where initially the base sequencer

Starting point is 00:41:31 was meant to be for the layer one exclusively. And Vitalik kind of had this fourth experiment. Well, if we use the base sequencer for L2, and the term that he used was total anarchy because basically you'd be in a position where all the users would submit their transactions in the MMP pool and there would be a conflict in transactions and some of the transactions would revert,

Starting point is 00:41:52 it would be a complete mess and it would not work. Until we had proposer-builder separation. So, proposer-builder separation basically says that even though the sequencing rights are given to the L-1 consensus participants, the heavy lifting of building optimal blocks is done by the builders. And so what it means in practice, what does base sequencing mean is that from the perspective on L1 validator, nothing changes. I'm connected to MF Boost. I'm given the most valuable block

Starting point is 00:42:25 to sign, and I sign it, and I don't even see the contents of that block. All I sign is the block header. And instead, all of the heavy duty sequencing is done by, specialized searches and specialized builders that will do the heavy lifting on behalf of the L1 validator. So not only are you reusing the proposing rights at the L1, but you're also reusing the full PBS pipeline. You're reusing builders, searches, relays, private mempools, all of that stuff. With the Ethereum roll-up-centric roadmap, anyone can deploy a roll-up though, which means anyone can deploy a based roll-up, but Ethereum, the layer one, isn't actually aware of based roll-ups. It doesn't know that there's a network elsewhere that is a based roll-up. So how do the block builders,

Starting point is 00:43:17 the people that are supposed to be sequencing this roll-up, how do they actually become aware and actually opt into this duty of sequencing the base roll-up if anyone can deploy a roll-up and, you know, how do these things actually, like, integrate? Right. So one of the mental models for base roll-ups is that it's the same thing as a layer one contract, except that fees are 100 times cheaper. So just like anyone could go deploy Uniswap or Avey or whatever it is, you know, Ephraim is not immediately aware. It doesn't know, it has no semantic understanding of what Uniswap is.

Starting point is 00:43:49 It's just, you know, bytecode and instructions that are being blindly run. And ultimately, what is the mechanism to get these transactions on-chain and have them executed is transaction fees, its incentives. And so users are willing to pay transaction fees, then that is an incentive for them to get picked up by the searches and by the builders and ultimately go on chain. It's the exact same situation with base roll-ups. Users, pay transactions.

Starting point is 00:44:12 Is a based roll-up mempool the same thing as the Ethereum Layer 1 mempool? Or is there a separate mempool? This is an implementation detail. You can use whatever you want. One of the trends that we're seeing is that the public mempool is dying. What's happening is that a lot of the top wallets, like MetaMask and Rabbi wallet, have their own private mempool where instead of gossiping transactions

Starting point is 00:44:36 to the public mempool, they have private deals with builders and they privately share it with the builders. And so that has one great advantage is that because you're relying on the trust of the builders, there's no more front running. And what we're seeing is that these wallets are actually incentivized to operate these private memos

Starting point is 00:44:56 because they can extract the back running. So it's kind of this ethical way in some sense to make money from the older flow that you have all while improving the UX for your users. And this is the kind of infrastructure that is very, very much reusable for base sequencing. So one of the things that I said initially is that centralized sequences

Starting point is 00:45:17 provide as a service MEP protection. But what we're seeing is that now MEP protection is done one layer out closer to the edge by the centralized wallet operators. And because of that, It's much easier. It's one less thing that you have to worry about if you want to decentralize your sequencer. Justin, this mental model of a based roll-up being like a smart contract on the Ethereum layer one, except a smart contract where gas fees are, you know, 100x cheaper is really interesting.

Starting point is 00:45:51 And I think a lot falls out from that idea. Can we go over what the main benefits are of a based roll-up, maybe from the different stakeholders involved? So let's talk about from a user perspective, let's talk about from a developer perspective, let's talk about Ethereum as a whole. I don't know where you want to start with this, but when we contrast based roll-ups from traditional roll-ups with centralized sequencers, what are the core advantages here? It sounds a little strange, but the core advantage is to not degrade the network effects that we have with layer one. It's just preserving what we already had with layer one.

Starting point is 00:46:34 So remember how I said that we had these amazing network effects, and that brought ether the asset to half a trillion dollars. And we thought we were all going to the moon. And then there was this big anti-network effects, which was the transaction fees. We hit a wall. We couldn't grow these network effects anymore. and then we go and down the path of roll-ups to scale, but then that introduced a different kind of more subtle type of anti-network effects, which is the breakdown of composability.

Starting point is 00:47:04 And so what I'm hoping can happen is that we reintroduce the ability for any app chain, any contract, to call any other app chain synchronously. So that means at will within a single block. So one canonical example here is AVE, when it makes a liquidation, it's in a single block or even a single transaction will liquidate its position to uniswap.

Starting point is 00:47:36 But what we're going to do is go up one level abstraction and have AVE chain in one single block liquidate its positions on unichain all synchronously. If it's two separate chains that don't share a CERCAN. sequencer and don't have real-time proving and real-time settlement, you can't do this kind of liquidation. And so the network effects that we have at layer one just break down. In summary, all I'm trying to do is just preserve the network effects of layer one while removing the fees as the main blocker that we've seen historically. Okay. And then just to emphasize, those network

Starting point is 00:48:15 effects are basically synchronous composability, atomic composability, where you feel like you're on the same chain is how users will feel and shared liquidity. So there's not this kind of fragmentation of like, oh, this chain has better liquidity for this base pair versus that chain. I got to go navigate and bridge my way and figure out what chain to be on. It's basically one single unified shared state. But we do have that additional really cheap transaction fee capability that has been at the core and the entire purpose of the roll-up-centric roadmap. So it'll all feel like one chain, one-shared liquidity pool, except fees will be incredibly cheap. I mean, earlier in the episode, you know, I said, hey, Justin, do base roll-ups, you know, solve everything? And you said,

Starting point is 00:49:02 well, no, hold on. It doesn't solve everything. But, like, what I just said feels like it solves a lot. Everything? A lot. I mean, it feels like it solves a lot. Let's just say that. But, you know, is this a realistic possibility here if all of these roll-ups become base roll-ups? I think it is. And we can talk about incentive alignment. We can talk about the technology that we need to go build. We can talk about some of the downsides of base roll-ups and how we patch them. But fundamentally, in order to have the luxurious kind of first-class composability that the L-1 enjoys,

Starting point is 00:49:36 we need two key fundamental ingredients. Ingredient number one, shared sequencing. Ingredient number two is real-time proving. Real-time proving is just an injecting. problem. And SNOCs are getting so, so good that we should have them relatively soon. And then another interesting thing with real-time proving is that we have a training wheel called TEEs. T-E's are capable of providing real-time proving today. And so this is an opportunity for someone who wants to build these synchronously composable roll-ups to use T-E's as a training wheel.

Starting point is 00:50:09 But then the other much more difficult problem, in my opinion, is this trillion-dollar dance, this coordination problem, this social layer problem around which sequencer are we going to decide upon to be the canonical ephium-wide shared sequencer? One question I have, Justin, is kind of how this might change, like, user behavior in a based roll-up world. Like, right now, I have, like, most of my activity consolidated to, like, a single layer two, like the layer two that I prefer. And, you know, my assets are there, most of my transactions are there, and I use the apps that are deployed. on that layer two the most.

Starting point is 00:50:48 How does that change? How does my relationship as a user change with the network I'm on in a based roll-up world? Because some of the benefits of based roll-ups is that, well, maybe there's an application on the Ethereum Layer 1 that I want to use, but the assets that I want to use in that Ethereum-Layer 1 are on this based roll-up,

Starting point is 00:51:05 or maybe they're on a different base roll-up. How does my relationship with the network I'm on change if everything is based? So there's a couple of things here. One of them is that to a very large extent, the detail based or not based is relevant to developers. There's been a lot of networks that have tried to go down the async route. There's cosmos, polka dot, the internet computer, Nair, and Tone network. And every time it's kind of a similar story. Like when you have asynchrony,

Starting point is 00:51:39 in theory you can have like massive, massive scalability. But the downside is that the developer experience becomes 10x harder. And so, you know, developers want simplicity, and they'll rather go with Solana than PolkaDot or Cosmos. And really what we're trying to do here is like thread the needle extremely delicately so that we can have the synchrony that the developers want while also preserving all of the values and requirements that the FML1 has imposed itself. Since we had the rollocentric roadmap, the first things and the Donner, prominent things that have come out of that are these very big generalized layer twos. Optimism, main net, arbitram, base. These are very big generalized layer twos. Does a based world lean more

Starting point is 00:52:27 towards the app chain model of things where in a world there's more chains, but they're more application, they're catered to specific applications rather than they're being generalized based roll-ups? There's actually just applications that are based roll-ups. Do you think in a based paradigm that we lean more towards app chains rather than generalize layer two's? Well, in the world where we've perfectly solved interoperability, and that includes many other things above and beyond just shared sequencing. So I mentioned this barrier to entry to becoming an app chain. We need to have optimal proof aggregation, optimal data packing. Then I think, yes, the app chain model makes a lot of sense because it removes the dependency

Starting point is 00:53:08 of having to opt in to someone else's governance, right? Why would Uniswop, for example, want to opt in to arbitrams governance? They're two different things. Uniswap only have its own, you know, sovereign and independent governance. And so it makes more sense for it to launch as a separate unit chain. Radically simple ideas always tend to catch on.

Starting point is 00:53:35 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. Web 3 should be simple, too, like bread and butter. Cartesey brings radically simple solutions to Ethereum, so developers can do what they do best. Build. Go ahead and discover a flexible, modular stack on Cartesey and build your most powerful, ambitious project yet. Visit cartesi.io

Starting point is 00:54:08 slash simple and simplify your blockchain journey and start building today. 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 2 design architecture since it first started building Mantle DA, powered by

Starting point is 00:54:24 technology from EigenDA. Maybe it's because users are coming onto the Mantle Layer 2 to capture some of the highest yields available in Defi 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

Starting point is 00:54:40 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 defy apps, click the link in the show notes. So getting

Starting point is 00:54:56 started with Mantle. It's fascinating here because I think what you're describing, Justin, is Ethereum is trying to attract both types of developers, maybe. So developers through which, you know, asynchronous composability is fine, we'll still have the existing, you know, layer twos with centralized sequencers. But developers who want, you know, synchronous composability and that devX, well, now Ethereum will have this other thing called kind of the based roll-up type of infrastructure and they could, you know, just deploy there. I want to ask the question because

Starting point is 00:55:27 you mentioned sort of two things that were imperative for this vision foundation. for this vision coming about. One is shared sequencers and the second is real-time proving. So when you say shared sequencers, I'm not sure it's clear to bankless listeners, the difference between a shared sequencer and a based roll-up. So my understanding of a based roll-up is that that's where we're using the Ethereum validators as our sequencer. So of course it's shared. It's like definitionally shared across kind of like, you know, all of the validators. And you're saying that maybe becomes the shelling point because that is the most credibly neutral, you know, cohort. of sequencers that you could do. Why? Because you're basically using the L1 validators. But I think by

Starting point is 00:56:08 implying you're using the phrase a shared sequencers, maybe you're implying there's also a step before that. So is it possible for an existing roll-up with a centralized sequencer? Say, base. We'll take them as an example. The base roll-up, centralized sequencer, okay? They're going to continue doing their thing because why? There's all of the other benefits we just talked about with centralized sequencers. They don't want to give those up, you know, like too early, right? They're building kind of incrementally. They want to preserve those things. Then there's going to be this other cohort of straight all the way to the end game of based roll-ups, you know, the tycoes of the world, the spires of the world.

Starting point is 00:56:43 They're building all of these based roll-up app chains on that side of things. Is it possible for those two worlds to connect via route of this interim, you know, set of shared sequencer? Or maybe just more generally, what's the difference between a shared sequencer and the end state of a based roll-up? Great question. So base sequencing is a special type of shared sequencing, but not all shared sequences are based sequences. So what I expect will happen, as you said, is that there's going to be an intermediate step. And this is something that we're already seeing. You could call it clustered sequencing, where within individual ecosystems, whether that's arbitram orbit or the elastic chain or the super chain,

Starting point is 00:57:27 there will be a shared sequencer, which is considered to be credibly neutral within the ecosystem. So you could imagine, for example, ZK Sync to use the ZK token to come up, with a decentralized sequencer, and within the context of this ecosystem, which is ZKSync era and all of the friends of ZKSync, then yes, that is a great place to have Shed sequencing and therefore synchronous composability. And the great thing here is that we're going to see experimentation, and we're going to see the value of synchrony and shed sequencing come up. We're going to see, for example, this data compression that I talked about, the block packing, we're going to see flash loans, just like we see on L1,

Starting point is 00:58:13 kind of do like crazy, you know, kung fu, you know, moves in order to squeeze as much amoe as you can and get the best execution. We're going to see latency go down dramatically because right now, let's say you have two chains within the ZKSync ecosystem. And in order for your application to work, it kind of straddles both chains. And so you need to do a transaction on chain one, and then you need to wait for the sequencer to, pick up the next move and do another transaction on chain two. And so that is a multi-slot process which increases latency. It also removes fee certainty. So as a user today, if I want to make a transaction on L1, I know exactly what I'm paying. But if I have to split my transaction into multiple asynchronous sub-transactions, then I know how much I'm paying for the first one, but I don't

Starting point is 00:59:04 know how much I'm paying for the next ones because the gas prices are always moving. And this is where you can end up with stock transactions. So, for example, you just haven't provided enough gas, and then there's a fee spike, and then now your transaction never executes. Or for some weird reason, the full transaction is unable to complete. So one example here would be that you have some funds on chain A, you want to go buy an NFT on chain B.

Starting point is 00:59:34 You initiate a transfer of funds from A to B, and then someone buys the NFT before you. You're no longer in a position to buy it. And so now, you know, in some sense, you have a partially executed transaction that you need to go revert and things become quite complicated. And so, again, like,

Starting point is 00:59:52 no single asynchronous ecosystem has thrived in the past. And the reason is that there's this long tail of benefits that synchrony gives us. And as you said, the key to achieving Ethereum-wide shared sequencing is credible neutrality, the key to going above and beyond this clustered shared sequencing. I think we've kind of just stated without explaining how

Starting point is 01:00:18 base roll-ups provide synchronicity. I don't think we've actually defined it head on. Maybe we could take a moment and actually illustrate how do base roll-ups actually provide full synchrony across Ethereum? So anytime you have shared sequencing, whether or not it's based, you have an opportunity for synchronous composability if you also have this second ingredient, which is real-time proving. So imagine that you have uniswap on unichain, which is trying to liquidate a position on AvaChain. You can have within a single block a first transaction that is going to go from Ava chain to read the price on Ava chain and tell unit chain exactly what the price is. And because you have the real-time proving, this is an instant

Starting point is 01:01:10 message that can happen within a single block. And then if the liquidation price is favorable, then you can have a unit chain, basically, send assets that need to be liquidated. You have a withdrawal which happens within a single block. So you don't have to wait seven days, in the case of an optimistic roll-up. You don't have to wait several minutes. In the case of today's ZK rules, it just happens within a single block. Assets go from average chain to unit chain, they get liquidated. And then the proceeds of the sale in a very similar way get immediately withdrawn within a single block and go back to Ava chain. I think the key point here is everything is happening inside of a single Ethereum layer one block that is the sequencer of these

Starting point is 01:01:52 transactions that are being made across these based roll-ups. In the generalized roll-up world that we kind of live in today, if I'm on optimism and I want to do something on Arbitrom, I take a bridge across, I take across, I take some sort of bridge aggregator, and it hops me over the layer one from chain to chain. And that's where a lot of the fragmentation happens is because the layer one isn't really aware of what I'm doing until later. And so I think the base roll-ups, the important key factor here is that it's actually the layer one that is doing the bridging of states or the bridging of transactions across the base roll-ups. It's the layer one that's doing the sequencing of both base chain A and base chain B at the same time. And rather than routing my order above the Ethereum layer one across two roll-ups above the Ethereum layer one, is actually going down into the layer one.

Starting point is 01:02:45 And the layer one is able to process that logic. Yes, that's exactly right. So today we have bridges. And the way that bridges are built is that you have the two pillars, and then you have a deck that connects the two pillars. What the real-time proving does is that it brings. brings the deck, you know, it makes it, you know, inferences, like, extremely small.

Starting point is 01:03:04 And then what Shad sequencing does is that it collapses the two pillars into a single pillar. And so now you've taken your whole bridge and you just collapsed it, and everything can happen within a single block. So in a world of base roll-ups, there are many, many, many pillars, but they're all, like, touching each other. And they're touching each other and coordinated by the Ethereum Layer 1. And so there's many vertical chains going out from the... Ethereum, but the facades of each are all touching one another. And it's just like one

Starting point is 01:03:32 defragmented system because they're all completely touching each other. That's exactly right. Yes. And one of the initial critiques of shared sequencing is that people will say this is an unscatable model. And the reason is that you can't have every piece of state touch every other piece of state. You know, you have too much locking. It just doesn't scale. But the answer to that is that synchronous composability is better thought of as optional, opt-in synchronous composability. The vast majority of transactions will stay within a single execution zone. They will only touch one roll-up, one app chain. But then there's going to be a few super-valurable transactions where it makes sense for them

Starting point is 01:04:18 to be touching three or four or five or as many as necessary different app chains. and they will be willing to pay the cost of basically locking the state for that period of time when they're touching the state. It's fascinating, Justin, use that term execution zones. Another term that Dave and I have been thinking of and just observing how it seems like the rule-up union of chains is kind of evolving is this term economic zones. And I want to ask the question of how practically you think we go from a more fragmented world of different. role-ups to a more unified set of roll-ups from a composability perspective, from a shared liquidity and shared state perspective. And one observation that we've made is we're starting to see these economic zones almost congealed together at kind of the hyperchain or super chain type of level.

Starting point is 01:05:13 I mean, you see, you know, the super chain, which is like the OP stack, and you have base now, you have unichane, you have Zora. And they're talking about shared composability. shared liquidity, shared interoperability between all of those chains inside of their economic union. You have the same thing going on with the ZK stack and the elastic chain. They have ZK Synchronos lens, ag layer and the polygon ecosystem doing the same set of things. Polygon mutable there, Ronin on there. You have Arbitrum and doing their orbits thing. And so there seem to be these economic unions of chains kind of forming that are composable within their sort of economic unions.

Starting point is 01:05:52 but you know, like they have loose bridges to the other unions. And then you also see kind of a growing and the dawn of the base chain economic union, which, of course, is sort of the Ethereum Layer 1 plus whatever Tyco does, plus whatever, you know, the future spire does, plus whatever puffer does, these base chains. So you've all of these economic unions that are starting to congeal. And what we want to get to eventually is where we have the right synchronous composability and all of these unions kind of become one super union, super structure. So we have shared like composability across all of them.

Starting point is 01:06:30 Anyway, we're starting to see hints of there being enough advantage, comparative advantage in kind of like creating mini super chain style economic unions. And I'm just wondering if you can kind of extrapolate how all of this might happen. Because now we're not just dealing with technology. We're dealing with incentives. We're dealing with like economies. We're dealing with all sorts of coordination factors that happen socially. So how do you see this really evolving once we are in a based roll-up era, once this technology,

Starting point is 01:07:03 this structure is available generally? Will some of these roll-ups kind of convert to based roll-ups? Will all of the different economic unions eventually migrate and become based roll-ups as well? I think you put it very well, that there's these different unions that are happening. and the based union is in and of itself an important union. And what I think will happen is that in the early days, there's going to be a few pioneers, a few visionaries, who can kind of see the longer term future

Starting point is 01:07:33 and are willing to make a bet and make a name for themselves with this based sequencing union. But you're right, as soon as the network effects become bigger and bigger, like one by one, what I expect will happen, is that these clustered sequences will convert, and that will have to do, as you said, with incentives. So one of the, I guess, observations is that some applications won't have any choice, but deploying from day one

Starting point is 01:08:05 as leveraging the base sequencer. And that has to do with neutrality. So I think a great potential case study here is ENS. N.S is not like uniswap where it can be deployed in many different places at the same time. It has to be deployed in one place. There's like one route of trust for EMS. And ENS is the Switzerland of domain names. I joke that the N in ENS stands for neutrality. And so it can't go deploy on top of optimism or deploy on top of arbitram. That would break the neutrality. And so it essentially has two options as I see it.

Starting point is 01:08:46 Option number one is that there is an ENS-specific decentralized sequencer, so they could use the ENS token as the basis for a new proof of stake, and that is going to be a decentralized sequencer that would power ENSV2. Or option

Starting point is 01:09:01 two is that they use base sequencing. Now, I personally don't see any advantages for them to go down the first route. It's just strictly more complex, and it doesn't buy them anything. Like, Two of the stated reasons for having your own custom sequencer is that you can have shorter slot times. And two is that you can capture MEV for yourself.

Starting point is 01:09:25 But, you know, ENS does not generate any MEV. ENS makes fees through like subscriptions. And ENS doesn't need like super short slot times. If it takes 12 seconds or one second to register a domain, it doesn't make any difference. And so I think, you know, app chains like ENS, makes sense to be deployed, leveraging the base sequencer. I guess something similar could be said for fine art and like credibly neutral NFTs, things like Cryptopunk. So let's say that you want to build Cryptopunk 2.0, where it's the same concept, but instead of 10,000 punks,

Starting point is 01:10:01 you want 10 million punks. And so there's no way that you can deploy it on R1. You have to deploy it as an app chain. And the only thing that makes sense is that you deploy it as a based app chain. Now, if we go back to the incentives, at least in the naive setup, naive-based sequencing, as a roll-up, I'm giving away my MEV to the sequences. And that could be seen as a bad thing and a disincentive to become a base roll-up. But the counter-argument to this is that the shared sequencing will increase your overall footfroll and traffic and increase your execution. fees that you go collect. And so the question becomes, are my gaining more in execution fees than

Starting point is 01:10:48 I am losing in MEV? Now, my personal thesis, and this is a bit of an extreme, you know, caricature is that MEV is going to zero. The amount of MEV is relative to execution fees going to converge to zero. And that's for two reasons. Reason number one is that we have the wallet, you know, protecting users from front running and sandwiching. And we're all. And we're also going to have better applications that are designed so that MEV doesn't leak to the sequencer in the first place. It's recaptured. And so really from a fundamentalist perspective, I think each roll-up needs to focus on execution fees. What are execution fees is basically a toll for the right to use the virtual machine. Any virtual machine has a limit in terms of how

Starting point is 01:11:39 much footfall it can process. There is the famous gas limit, which is 30 million gas on Ethereum. And the way that we charge for execution is that we have mechanisms like EIP-1559. And every roll-up today, all the top roll-ups, they have the equivalent

Starting point is 01:11:55 of EAP-1559, and they're making money through congestion and execution that way. And that is what they should be focusing on. Now, one of the things that you said previously, Ryan, is that one of the big reasons to be excited about base sequencing is value accrual, right, because MEV goes to the

Starting point is 01:12:16 validators. I actually think that's not the case, partly because MEV is going to zero and so it's a rounding error, but also if it's not zero, this is a bad thing because it becomes a disincentive for the roll-ups to opt into a shared sequencer. And with the applications capturing the MEV. This is something that we're starting to see out of the world of Sorella and also Unichain. Unichain just got announced. So we have our kind of our first big application-specific roll-up. And then Sorrella Labs is working on this thing called application-specific sequencing, which explicitly gives applications the ordering of their own internal transactions, more or less with the intent to do exactly what you said, which is allow the uniswap application to retain the MEV that it creates

Starting point is 01:13:02 in order to give that back to LPs and traders. So this isn't just theory. We're starting to actually see this played out. And so the business model then, if this logic continues, is that MEV does drop to zero because we get more efficient, applications can retain it, and MEV doesn't go to the underlying chain, it goes to the application. Well, then the value here is really kind of the GDP of Ethereum. Like what total aggregate transaction volume can we produce?

Starting point is 01:13:31 And I think within the idea of base roll-ups, sharing your apps, sharing your businesses, sharing the things that generate value across other chains is probably the most logical way forward where the value of one-based roll-up does actually create more value for other-based roll-ups. And this kind of gets back into the whole idea of just network effects. When the value of one application on one-based roll-up increases the GDP of a different base roll-up of a neighbor-based roll-up, that is what network effects are. That's kind of the vision that I see. Is that what you see? Yeah, that's exactly right.

Starting point is 01:14:03 So if you have a based unit chain, every time you have another based app chain that requires unit chain, that is additional footfall. So AVE is an example. You know, Maker might be an example. And a significant part of uniswap's footfall today comes from these L1 interactions. And so one of the bearish takes, I guess, on unit chain being non-based,

Starting point is 01:14:30 is that for the most part, it will capture the flow from Uniswop X, basically from the Uniswap labs frontend. So if you go on Uniswap.org, then yes, your flow might be routed to Unichain, but like the flow from Avey and all of the other ecosystem players, that will continue going to L1. And so I could totally see, in addition to the unit chain that has been deployed with an application-specific sequencer to also have a base version of Unichain. What I could also see is actually a hybrid and a merging of the two. So really what the Unitrain is trying to enforce is, first of all, that there's no censorship. So anyone who wants to do arbitrage can do arbitrage.

Starting point is 01:15:21 And number two, that the transactions are ordered decreasing by priority fee. and what you could imagine is an application or chain like unichain basically saying, I'm willing to be based, but I want these two constraints to hold. So you could imagine, for example, a suave kind of network of TEEs that are going to sign off on a set of unisop transactions that have to be included and have to be included in this order, but the base sequencer is allowed to insert additional transactions. So you can kind of think of it like a constrained, even more constrained inclusion list where the relative ordering needs to be preserved,

Starting point is 01:16:07 but you are allowed to insert transactions. And the fact that the Shed Sequencer is allowed to insert transactions, this is what creates the synergies. This is what creates, you know, it allows you to fulfill the coincidence of wants and kind of extract the network effects that are otherwise impossible if the unit chain says, this is the list of transactions, end of story, no synergies possible.

Starting point is 01:16:29 Yeah, I was going to ask the question, you know, Justin, your personal take, if you think the unit chain would have been better off as a based roll-ups or would be in the future better off as a based roll-up

Starting point is 01:16:38 rather than the architecture it's selected. But, you know, that's still maybe a valid question, but maybe the more general question is like, you know, let's get into the real tactics here. How do you think that this could happen?

Starting point is 01:16:49 So a dream scenario for what you're talking about is basically the entire super chain. Let's just pick on kind of the OP stack and you have some pretty large chains there. You have base and you have OP mainnet and you have like unit chain and apologies to the other dozens of chains that I've left out in this. But they're all on the super chain and they kind of fix interoperability amongst each other's. And then they sort of opt into some sort of shared sequencer or are on the path to making their entire super chain based or at least having that option so that

Starting point is 01:17:21 chains inside of their network could flip that off or on and preferably on for the good of kind of like the network for the good of shared liquidity for the good of fixing fragmentation. Do you think that's like realistic? I mean, a naive podcaster is like, hey, like, why can't we just do that? Let's just like convert the super chain into a based super chain and all of the chains within them now are hooked into all of the shared liquidity of the L1. Is it that simple? I mean, you mentioned there's some other designs here that might like kind of like bring the best of all worlds into play. How do you think something like this happens for Unichane or the broader super chain? Right. So I've had, you know, various conversations with various founders and like various degrees of

Starting point is 01:18:03 being base-pilled. I think, you know, Hayden, you know, understands the value of Shed sequencing and understands the value of base sequencing. I don't want to put words in his mouth, but I do think it is a plausible future that Unichain becomes this, you know, base roll-up where you have, you these additional constraints, which makes sure that MEV ultimately goes back to the users and doesn't leak to the sequencer. I think one likely way forward is that there's this bottom-zap grassroot movement of the base cluster that just incrementally grows bigger and bigger every year. But there is also a possibility for incumbents to one day from another just flip a switch and become base. and I don't want to leak too much alpha, but

Starting point is 01:18:52 basically two of the top roll-ups by TVL have founders that are base-pilled. And so it's either arbitram optimist or a base. I don't know if there's many more than that. No, more than the top-spice. Let's say out of the top 10.

Starting point is 01:19:07 Top 10, okay. Justin, I want to bring in the conversation of the Ethereum Layer 1. What kind of capacity does the Ethereum Layer 1 need to have in order to kind of fulfill this base roll-up vision? because the layer one execution is meaningfully constrained.

Starting point is 01:19:22 But I think if entire chains are going to use the layer one as a sequencer, we have to start to talk about what kind of execution powers the layer one has in order to actually manage all of that. How does this impact the role of the Ethereum layer one in terms of just like its capability here? Right. So the Ethereum layer one consumption in terms of execution doesn't really change. We don't need to go above and beyond the 30 million gas. like the EVM as it is, is sufficient to verify the fraud proofs and the snark proofs, and that's fine as is.

Starting point is 01:19:52 What is a much more important factor is the slot duration. So right now we have this 12 second slot times, and one of the discussions during 2024 has been doing need to, or can we, reduce the slot times. And I think my opinion as of today is that there is a lot of low-hanging fruit that could allow us to significantly reduce the slot time for example, to four seconds, all while preserving the decentralization requirements that we have, namely that you can be a validator on a home internet connection and a Raspberry Pi. And so there's basically juice to squeeze in the future for improving base roll-ups in that way. But there's another kind of trick that has been developed in 2024, which is this notion of pre-confirmation, trustless pre-confirmations.

Starting point is 01:20:41 Because today, we have reputation-based pre-confirmations, where the centralized sequencer's, you know, pinky promise that the transactions that they have sequenced off-chain will be indeed settled on-chain in the way that they've been promised off-chain. And the idea of, you know, trustless collateral-based pre-confirmations is that the L1 proposes, put forward collateral, and whenever they make a promise, which is nothing more than a signed message saying, you know, your transaction will be included in position X and will execute in such and such a way. and if you're doing a unisop trade, you know, this is the price you're getting, these are the fees you're paying, you know everything up front. And if this promise is not held, is reneged, then you stand to lose the collateral that you've put forward.

Starting point is 01:21:28 And what this allows us to do is basically, from a UX perspective, mimic 100 milliseconds slot times. Why 100 milliseconds? Millisseconds, because that's the amount of time it takes for information to travel on the internet. That's internet latencies. It's ping times. And so what we're going to see is, you know, base roll-ups like Tycho experiment very, very soon with these pre-confirmations on Mainnet. And then if this can be deployed to Mainnet and be really robust, then what is the advantage of having a non-based centralized sequencer? Because even a two-second slot time, which is what the Super Chain has, that would be 20 times lower than pre-confirmations.

Starting point is 01:22:09 That's fascinating. You brought up earlier in the conversation. and used this analog to, you know, data availability. So in the early days of the roll-up-centric roadmap, you said, Ethereum really, as a network, wasn't built for roll-up data availability. You know, there's kind of like this hackish way that roll-ups were using kind of the call data function and just like stuffing their data inside of that. And so the Ethereum roadmap kind of adapted and said, no, we're going to make this fast lane for

Starting point is 01:22:34 roll-ups. So they could put all of their data in one place. We'll call these blobs. We'll make them super efficient. We'll have a path for scaling them. fast lane increases in terms of capacity, and we have a home for, like, all roll-ups for, you know, their data availability. Is there an analog to, because you also said, the current Ethereum network is not really built for shared sequencing, right? So we have, you know, 12-second slot times,

Starting point is 01:23:00 you know, we need to get our slot times down, that sort of thing. You're also talking about trustless pre-confirmation. It's not clear to me whether that's kind of extra outside of the Ethereum roadmap, outside of the protocol, or whether that could be embodied. But, like, I guess, maybe just the general question of do we need a kind of a proto shared sequencing, you know, roadmap item to get us to an EIP where we deploy this and now, now Ethereum is meant for, it's built for, you know, based roll-ups and shared sequencing? What sort of changes might we need to make to the roadmap in order to facilitate that or prioritizations would you recommend if this indeed is the future? Yes. So the answer is yes, we do have such a roadmap item and it's called

Starting point is 01:23:42 APS, attester, proposer, separation. And the idea of APS is that as a validator, my main responsibility is to attest to the correctness of the chain and to produce inclusion lists. But the ultimate proposing of an execution block, which involves a lot of sophistication

Starting point is 01:24:06 through the block building process, but also sophistication in the pre-confirmation process, that would be offloaded to the broader market. Because right now what we're trying to do, and it's kind of awkward, is that we're trying to shoehorn sophisticated pre-confirmations to L1 validators running on the home internet connection on a Raspberry Pi. And so the solution that has been proposed in the short and medium term

Starting point is 01:24:33 is the equivalent of relays, but for pre-confirmations. And we call them gateways. So as an L1 proposal, I will delegate my sequencing rights to a gateway, and the gateway does all of the heavy lifting on my behalf. And just like we have enshrined PBS that promises to remove the need for relays, APS, attested proposal of separation, also promises to remove the need for these gateways. And so that is the cleanest and most aligned path forward in the roadmap to really make Ethereum custom design. for pre-confirmations. Okay, and just a quick question of when on that? Are we talking about way in the distant future, or is this short to medium term?

Starting point is 01:25:18 I'd say medium term. I'd say years, but less than half a decade. Less than half a decade. I love how that's in the medium term in Ethereum world. Justin, how does a base rule of paradigm, how would that impact the economics of ether the asset, if it does at all? I think any impact would be indirect. Basically, we would have a very competitive, a very strong network effect, and that would give more credence to the notion that, you know, ether is money and it's a powerful money for the internet of value. What I think we potentially, you know, stand to lose if we go down this

Starting point is 01:25:51 asynchronous route is basically lose to a competitor that has decided to fully embrace synchrony. And, you know, here Solana comes to mind. And so the optimistic take here is that, you know, we can have the U.X and the DevX of Solana with shorter. latency. So instead of 100 milliseconds slot times, we have these 100 millisecond pre-confirmations. And we have the strong decentralization because validators can be on home internet connections running on Raspberry Pi's. I think in recent times there's been a debate about the role of centric roadmap, right? Like, is the role of centric roadmap right? Like all of these kinds of debates, especially with the rise of Solana, kind of like shifting the Overton window of the

Starting point is 01:26:34 conversation sphere. And I think a part of the framing of that debate could be illustrated as, where does defy live and the Ethereum layer one with 12 second block times and kind of constrained block space has been losing ground in like the home of defy. It's lost ground to its own layer two's. Arbitrum and base have accrued a lot of defy value. Solana has accrued some defy value. And I think really the more accurate way to illustrate this is like where is the home of defy going to be? Because defy is kind of the golden goose. It has the most valuable transactions. It has the demand for very low latency transactions. In a based roll-up world where many chains are all base roll-ups, how do you think about the question of where does Defi live? Where does the home of

Starting point is 01:27:21 Defi? How does that even work in a world of base roll-ups? Right. So in a world where base roll-ups win, all of the TVL would be under the umbrella of the base sequencer, but there would still be a migration process from L1, where the gas use is extremely high to a CETLLB. system like L2s where the execution is done off-chain and so you can afford to have a hundred times less gas. So going back to the idea of network effects and Malkaf's law, right now the path that we've gone down is that roughly 10, 15% call it, of the L1 TVL has migrated to the L2s. And that's like a relatively, you know, small hit to network effects. Like the L1 is still strong and thriving. But the more we go down this path, the more we're carving out a network effects

Starting point is 01:28:14 and weakening them. And so the better plan, in my opinion, is to have a migration of this TVL that preserves network effects. So we want to go to DAPs, to app, from apps to app chains, and that is going to save on gas, but we don't want to move from base sequencing to centralize sequencing because that breaks the network effects. And so there's a little bit of path dependence here, which is like how quickly can we put the train on track to being in the direction we don't, you know, systematically break down network effects. And instead, we can preserve them under this umbrella of base sequencing. How does a based world impact the users' choosing of the chain that they're on? Like right now, I got to go to a drop-down menu. I select Arbitrum. I select optimism. I select ZKSink.

Starting point is 01:29:05 And there's all these choices that come with that. Like I got to kind of kick the tires of these chains that I'm using. I kind of have to know what chain that I'm using. When we have always assumed in the crypto world that eventually chains will just be completely abstracted. But we've kind of like alluded to that North Star without really knowing the path there. When it comes to actually being on a chain or choosing a chain, what would that look like in a base world? Right. So from the perspective of UX, the number one thing that we can do is have shared fungibility.

Starting point is 01:29:35 And it turns out that fungibility is an orthogonal thing to sequencing. So what we'd like is something like Ag Layer or, you know, many different ecosystems are working on the equivalent of Ag Layer where there's a shared deposit contract where all the different roll-ups basically deposit their assets there. And you get two benefits. One is that this is a, you know, homogenization layer whereby you can always go from one, one roll up to another, and because you're going through this deposit contract, the assets are fungible. But there's also another massive benefit, which is that if you want to go from an

Starting point is 01:30:15 L2 to another layer two, you don't have to withdraw back to the L1, and you don't have to pay the extremely high L1 gas costs. And so it's a direct L2 to L2 withdrawal process. And that is going to remove the drop-down. And this is what Hayden was talking about in his episode, where once you have fungibility, you can abstract away which chain the asset lives on. The bigger impact of base sequencing is around DevX. As a developer, like today, I have to choose an ecosystem. Do I launch on arbitram or do I launch on base or optimism? If I can just launch as my own independent based app chain, then I don't have to make a choice. That is the natural move for me. And I can still get all of the benefits of being proximate to other applications because the boundaries, you know, between these

Starting point is 01:31:07 app chains, as you said, is the touching, they're kissing, there's no space in between. By the way, just for an update on that kind of that shared fungible layer for L2 to L2 kind of communication, we've done episodes on Ag Layer. It sounds like there are some other, you know, technologies that would facilitate that. That is also orthogonal to everything we've been talking about today on base rolls. But how's that progressing? Are you optimistic that 2025 will see some real progress in that space? kind of an aggregated layer for L2 to L2 interoperability,

Starting point is 01:31:35 fungibility between kind of like tokens? 100%. I'm extremely optimistic, and I believe that pretty much every ecosystem is working on the shared deposit system, whether that's Polygon or the super chain or others as well. And the good news is that we're going to see experimentation with these shed deposit contracts, and we're going to be able to extract the best practices.

Starting point is 01:31:55 And eventually, I'm hoping that we're going to see a standard like ERC 20, but for shared deposits. Now, the tricky thing is that in order to achieve credible neutrality for shared deposits, we need to completely remove governance. It needs to be a dump pipe, you know, smart contract. But in order to achieve that and to do it in a safe way, we're going to need multiple redundant implementations of snock verifiers. Because like this, you know, amazing, you know, shared deposit infrastructure depends on snarks.

Starting point is 01:32:29 and snarks are extremely complex and buggy. And so I think the ideal outcome is that you have three different ecosystems, each come up with their own implementation, and then they all agree together to deploy a shared deposit contract where they've all contributed one of the three verifiers. So it's that idea back to we'll have these economic zones kind of creating their own shared layer basically, and then hopefully we get an economic super zone

Starting point is 01:32:57 where it's all pulled together in the future. future too. I want to return as we get to a close here, Justin, to the question of ETH value accrual that David was asking. And so I think there are almost like two camps. Actually, let me broaden this and just say, as long as I've been in crypto, there's always been this perennial question of like, okay, but like how do we price? How do we value layer one assets, right? And the Bitcoin community has a very like noted take on this, which is like, as money, as a store of value, it's digital gold, stupid. That's the whole purpose, right? Other ecosystems. I mean, alternative layer ones, let's call them.

Starting point is 01:33:32 The salon is Napos and Suez of the world. They're not as bold. They do not think that their layer one asset, at least the organizers of this. They don't think their layer one asset is money. They think of it as kind of a capital asset, and it's going to throw off cash flows from execution fees or MEV. I have come full circle to the idea,

Starting point is 01:33:50 and I guess this has been the genesis of the bankless podcast, too, that the bitcoiners were right. Okay, if a layer one asset is going to have value, it's going to be primarily as a monetary asset. And I know, Justin, you've been a proponent of the memetic idea of ETH as money, but more specifically, an ultrasound money, which implies that there's some sort of cash flow burn on the asset that supply goes down over time. We've also called within Ethereum, ETH a programmable money.

Starting point is 01:34:17 There are also, like, other terms, maybe it's a roll-up money, maybe it's money for the on-chain economy. There's all sorts of prefixes you might put to kind of like the idea of ETH's money. but it comes back to at least for me, what is ether? How do you value it? As a monetary unit, as a store of value with some sort of productive yield. I want to ask your take on this. So being kind of an architect partially of the ultrasound money, ETH, meme, how do you think ETH is valued? Is it still money? What is kind of the memetic narrative that you're taking going into 2025 and beyond? Right. So I do agree with you that the value-work role story for EF is mostly money. And one way to

Starting point is 01:34:56 frame the ultrasound money meme is as a selling point to achieve this moneyness. It's about scarcity. If Bitcoin has the 21 million Bitcoin meme, we have an even more powerful meme and we have more economic security. And so surely these are the two most important shilling points, security and scarcity. And so surely we should be able to win the memetic warfare with Bitcoin. having said that, I am still very optimistic on the idea that Ephraim will accrue extremely significant flows on the order of billions of dollars per day. And that sounds completely insane. And by that, you mean you're talking about burn, basically, or capital flows.

Starting point is 01:35:40 Capital flows, yes, in the form of data availability. And, you know, right now we're in this kind of depression phase where like data availability just accrues almost nothing because we're. we went from zero to something pretty big and there's just not enough demand. What I expect will happen is that very, very soon in the matter of months, we're going to see this fee market develop. And I do agree with David's take, which is that we don't want to be too aggressive and celebrate too early. We want to grow the number of blobs as quickly as we can so that we basically have this growth phase of Ethereum before we start accruing value.

Starting point is 01:36:23 basically the Amazon strategy, right? Like grow to be like the world like Titan, and then that ultimately makes you a very, very valuable company. But there's basically two stories to the flows. There's supply and demand. Let's look at supply first. Ethereum is constraining itself to home internet connections. Now, what folding sharding gives us is roughly speaking 100x over

Starting point is 01:36:52 a single validator. So from a supply perspective, we have the equivalent of a hundred home internet connections for the whole world, for like the whole internet, all of finance, just a hundred internet connections. Now, in terms of demand, what are the network effects of data availability? And I see three of them. The first one is this notion of shared security. when you deploy an app chain,

Starting point is 01:37:24 you can either deploy it as a Validium or as a roll-up. But if you want to really tap into these network effects, you want to be a credible money Lego. You want to be a money Lego that other applications feel comfortable, you know, calling into and can know that, you know, in 100 years, they'll be working just as well as they are working today. And so if you're building, you know, a Uniswap, for example, or unichain, there's no way that uniswap can be a money Lego as a Validium.

Starting point is 01:37:56 It has to be a roll-up. The second aspect actually has to do with synchronous composability. If you have an external data available provider, for example, Celestia, well, Celestia introduces its own sequencer. Celestia is a blockchain with blocks coming one after the other, and now you've lost the single sequencer that has monopoly power over everything so that they can extract all of the synergies. And then there's a third network effect, which is something that I only realized late,

Starting point is 01:38:34 but I think it's an extremely powerful one. And that has to do with the native roll-ups. So remember how I said that we can z-kify the EVM and we can expose a pre-compile within the EVM so that roll-ups can become native. Now, it turns out that this pre-compile for subtle technical reasons only works for roll-ups. It does not work for validiums. And so if you want to have all of the benefits that come from native wall-ups,

Starting point is 01:39:06 you have no choice. You have to be a roll-up. And if you want the quick kind of technical explanation, the reason is that the snock-proofs, are generated off-chain and gossiped off-chain. And in order for anyone to be able to produce these proofs, the underlying data, the so-called witness data, needs to be made available.

Starting point is 01:39:30 All of this is technical jargon, but the long story short is that if you want to be a native roll-up and not have to worry about bugs in your EVM implementation, and not have to worry about governance to stay EVM equivalent every time the EVM changes at L1, you have to be a roll-up. And so here we are. We have 100 home internet connections for the whole world,

Starting point is 01:39:55 and then what I expect will happen is there's going to be a massive demand from all of these roll-ups that want shed security, synchronous composability, and also be native. And the rough back-of-the-envelope calculation is that we can scale to 10 million transactions per second. This is this crazy thing where you start with 100 transactions per second, which is just a firm L1, you get 100x from roll-up scaling, you get another 100x from dank sharding, and then you get another 100x from Nielsen's law, which is the exponential growth of bandwidth for a period of 10 years.

Starting point is 01:40:36 And that gives you your 10 million transactions per second. And then once you have that, there's sufficient capacity for everyone in the world. many, many transactions per day, per human. And even if every transaction charges a tiny amount, let's say, a tenth of a cent, that's still a billion dollars per day of income. So what I'm seeing, what I think is a plausible future, is that we basically grow the blob space as aggressively as we can. And then in the period of 10 years, we have this gargantuan network effect.

Starting point is 01:41:12 and because we're fundamentally limited, we only have 100 home internet connections. The fee market is naturally going to do its thing, and even if we're talking about absolute dust per transaction on an aggregated basis, because there's so many transactions that are being processed, we're talking about billions of dollars per day of income. And the reason why I mentioned that flows are important

Starting point is 01:41:36 is because there are a way for us to rationalize the monetary process. premium. And I think the best mental model that I have is gold. Gold is this $18 trillion asset, and some septet of it, you know, the number that's usually quoted is 10%. So about, you know, 1.8 trillion of the whole gold market cap comes from its use as an industrial metal. And that's something that we can understand fairly well just by looking at flows. It's an understood science. We understand, for example, the cost of mining gold from the ground, and we understand how much gold is being lost to electronics every year. There's this crazy statistic that tens of billions of dollars of gold is lost every year in dumps,

Starting point is 01:42:25 right? Because you have your piece of electronic, it has a few, whatever it is, micrograms of gold, and then you just throw it in the dump, and that's effectively a burn. It's a burn, you're burning the gold. And so you can do these flows analysis, and you can look at other metals, and you know, you would have a market cap of roughly 10% of what it is. And then the rest is the monetary premium that comes from the slow money, where you lock up the money in volts, and you want to look at the ratio that is locked.

Starting point is 01:42:53 And so if 90% of all the gold is locked, then you'd expect a one divided by one minus 90%, which is 10x, one divided by 10%. And I think we can be in a similar situation with ether, the assets, where 90% of the supply, is basically locked in these virtual volts, either in the form of staking or in the form of smart contracts that basically provide the economic bandwidth for stable coins or whatever it is. And then only 10% of the economy is really circulating, and that 10% is itself valuable based on flows alone. Justin Drake, that end statement is definitely worth the price of admission

Starting point is 01:43:34 for this entire podcast, very comprehensive. You know, one of my prime questions has been, well, is Ethereum DA a commodity? Can it be replaced with some other DA from another ecosystem? That I got to say is kind of the best argument for why Ethereum DA is not a commodity. Anyways, just so much there. So thank you for that. I guess as we close this out and end it, I anticipate that, you know, because we've been through the blockchain roadmap cycle many times is you create a new solution for a category of problems. And then you also spawn another set of problems that you have to then go chase down and solve. And so my final question, we don't have time for a comprehensive answer. But I'm sure there will be some who listen to this episode and they'll say, but wait, this one thing about based roll-ups. You didn't acknowledge, you know, this set of trade-offs. Roughly before coming to this, I saw a tweet from Torgle from Scroll.

Starting point is 01:44:28 He said, this, base roll-ups do not inherit real-time censorship resistance and liveliness guarantees from Ethereum. There's a lot of myths. He's going to write an article dispelling them. I know you haven't read that article, you know, so I'm sure you can. There's also takes that I've heard that base roll-ups will increase block builder centralization, and that's like a new challenge problem for Ethereum. In general, what do you think that the trade-offs or the problems that might fall out of a max-based

Starting point is 01:44:53 roll-up, you know, Ethereum? What are the next dragons we have to go slay after this? So one of the things that I've observed with base sequencing is that there's a new reason why it doesn't work, like, every month. And so it's a game of Wackamel where, you know, we just have to go through them one by one by one. And I think the best way to prove that this thing actually works is to just stop talking about it and actually deploy it in production. And, you know, we could, you know, have a conversation or debate and go through all of these technical points one by one. as it stands, like, I haven't seen a really solid criticism from someone who's really, really gone down the rabbit hole and taken the time to go through all the 15 sequencing calls and pre-confirmation calls that we've had. From my perspective, like, once you appreciate base sequencing and your base pill, there's no going back.

Starting point is 01:45:54 I don't know of a single person who has gone the other direction. It's kind of up only up to the right in terms of the converts. Now, it is possible that there's unknown unknowns and that we end up hitting roadblocks. My personal thesis is that a lot of the problems that we will encounter are actually universal to any decentralized sequencer and any shared sequencer. And so really, base sequencing is just the final layer of veneer at the very end. It's just a small detail. Is it this server or this Raspberry Pi or this other entity that's doing the work? But ultimately, it's very much the same kind of technology.

Starting point is 01:46:37 And so I'm not at all worried from a technology standpoint. And I would love to have these technical discussions with TogRoll and others. If I may, just ending this episode, I prepared a little meme and I wanted to see how it lands and it's relevant to base sequencing. Let's go. The meme goes as followed. If Bitcoin is digital gold, then Ethereum is Digital Switzerland, a credibly neutral jurisdiction that can be the basis for internet finance. Now, I would liken Ethereum settlement to the Swiss legal system.

Starting point is 01:47:14 It is the final and authoritative arbiter of truth. And I would also liken Ether the asset to this respect, a credibly neutral, unit of account to denominate financial contracts. And this is where a firm sequencing comes in. Ephem sequencing is like Swiss democracy. It is this credibly neutral layer of mediation that can unlock synergies and network effects that would otherwise be trapped by competitive and adversarial tensions. And so I think we shouldn't be focusing so much on the technical details of base sequencing

Starting point is 01:47:55 because those are to a very large extent shared with decentralized sequences and shared sequences. But we should be focusing on what makes face sequencing special, which is, in my opinion, credible neutrality. Well, maybe Ethereum can be Switzerland but with America's military. Something like that. We want some security there, for sure, as well. $100 billion of economic security. There you go. Justin, I've learned quite a lot in this episode. This has been great. We have you speaking at the Bankless Summit at the day after DevCon November.

Starting point is 01:48:25 16th. Now, you've described your talk that you are producing as painfully bullish. Now, I don't really want to give anything away. I also don't know anything. Is there anything that we've covered here on this episode today that's going to be related to or overlapping with your content, your talk at the Bankless Summit? Or is that all novel material, a novel talk that you're giving at the Bankless Summit? I think it's mostly going to be novel stuff. I think that I will be reusing some of the material, but not much. Just as a teaser, in 2024, I have spent most of my time focused on sequencing. And it is a true pleasure for me to say that I think base sequencing and pre-confirmation

Starting point is 01:49:06 has reached escape velocity in a sense that my job as a researcher and coordinator is largely done. There's now dozens of teams, very, very strong teams that are building all this infrastructure. And I think at this point it's hopefully an unstoppable force. And so that means that in 2025 I can start working on new and exciting stuff. And I have all sorts of plans that will be disclosed in due time. Ethereum stuff, Justin? Oh, yeah, Ethereum stuff. Amazing, cannot wait.

Starting point is 01:49:39 David, what are the dates of the Bankless Summit? That's the day after DevCon, right? We'll probably include a link to that in the show notes. But give us the date again. Yeah, it's November 16th. DevCon ends November 15th on Friday. The Bankless Summit is the day after November 16th. unofficially the eth is money conference, just to make sure that at DevCon, there's enough

Starting point is 01:49:57 content around Ether, the asset and its future in the Ethereum world. Please inject some of that painfully bullish material straight into my veins. That's going to be an exciting set of talks. Bankless Nation, this has been Justin Drake. We've talked about based roll-ups here today. We've also talked about the fud out there about Ethereum's future and a very bullish episode here, Justin. Thank you so much for joining us.

Starting point is 01:50:17 Absolutely. Thanks for having me, guys. Got to let you know, of course, roadmaps are risky. So is crypto. you could lose what you put in. But we are headed west. This is the frontier. It's not for everyone, but we're glad you're with us on the bankless journey. Thanks a lot.

Bankless - Do Based Rollups Solve Everything? | Justin Drake

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.