Guide to the Ethereum Roadmap | Jon Charbonneau of Delphi Digital

Jon Charbonneau, Research Analyst at Delphi Digital joins David to talk about his recent essay titled, "The Hitchhiker's Guide to Ethereum." Proposer / Builder Separation (PBS) Danksharding and ProtoDanksharding (DS + PDS) Data Availability Sampling (DAS) What the absolute HELL are these things? What do they mean for Ethereum? What's the common theme behind all of them? And what do they mean for ETH? All of these questions answered and so much more in the livestream. ------ 📣 ALCHEMIX | Get a self-repaying loan today! https://bankless.cc/Alchemix ------ 🚀 SUBSCRIBE TO NEWSLETTER: https://newsletter.banklesshq.com/ 🎙️ SUBSCRIBE TO PODCAST: http://podcast.banklesshq.com/ ------ BANKLESS SPONSOR TOOLS: ⚖️ ARBITRUM | SCALED ETHEREUM https://bankless.cc/Arbitrum ❎ ACROSS | BRIDGE TO LAYER 2 https://bankless.cc/Across 🏦 ALTO IRA | TAX-FREE CRYPTO https://bankless.cc/AltoIRA 👻 AAVE V3 | LEND & BORROW CRYPTO https://bankless.cc/aave ⚡️ LIDO | LIQUID ETH STAKING https://bankless.cc/lido 🔐 LEDGER | NANO S PLUS WALLET https://bankless.cc/Ledger ------ Timestamps: 0:00 Intro 6:00 The Vibe of the Ethereum Roadmap 8:50 The Hitchhiker's Guide to Ethereum 17:06 The End State of Ethereum 23:05 Data Availability Sampling 34:49 Proto to Full Danksharding 48:15 Proposer Builder Separation 54:01 Checks & Balances 56:10 Incentives in Being a Builder 1:01:50 Centralization 1:03:50 Comparing Ethereum’s Scaling Strategies 1:05:54 Value Capture 1:07:20 How Jon Understands Everything ------ Resources: Jon Charbonneau https://twitter.com/jon_charb Delphi Digital https://delphidigital.io/ The Hitchhiker's Guide to Ethereum https://members.delphidigital.io/reports/the-hitchhikers-guide-to-ethereum Valuing Layer 1s - Memes, Money, or More? https://members.delphidigital.io/reports/valuing-layer-1s-memes-money-or-more ----- Not financial or tax advice. This channel is strictly educational and is not investment advice or a solicitation to buy or sell any assets or to make any financial decisions. This video is not tax advice. Talk to your accountant. Do your own research. Disclosure. From time-to-time I may add links in this newsletter to products I use. I may receive commission if you make a purchase through one of these links. Additionally, the Bankless writers hold crypto assets. See our investment disclosures here: https://newsletter.banklesshq.com/p/bankless-disclosures

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:08David Hoffman?

  Welcome, Bankless Nation, to a very special episode. I'm super excited to bring you on today's State of the Nation. State of the Nation is where it comes out every Tuesday on the Bankless livestream and then every Wednesday morning on the RSS feed, where we relate it to big-picture action items. We drop some insights and action items, and I'm happy to bring you some awesome, awesome alpha coming out of a research analyst out of Delphi Digital. We're talking about the Ethereum roadmap beyond the merge. In some ways, the merge feels like the finish line for much of Ethereum's history, but there is still so much left to do. You might have heard of things like Dankscharding, or data availability sampling, or proposer 4.0 builder separation, and you might have thought, wow, that's really complex. I might not ever understand that.

• 0:53David Hoffman?

  And you might be right about that. And I'm sure that we are not totally, we are not going into the full math about some of these things, but we are going into some of the shared themes and shared strategies that each of these complex mechanisms have for the future of Ethereum beyond the merge. There's this common structure to all of these things, and it has to do with harnessing complexity and the separation of powers to make Ethereum scale computation while remaining decentralized. So today on the show, we're bringing John from Delphi Digital, who wrote a fantastic research report praised by some of the leading Ethereum researchers, Tim Baker, Paulina and even Vitalik himself, called "The Hitchhiker's Guide to the Ethereum Roadmap."

• 1:38David Hoffman?

  And so we're bringing him on the show to talk about not the math, but the meaning behind all of these things, and what it means for Ethereum, what it means for you as a potential Ethereum validator, and what it means for you as a potential ETH holder. You guys might be aware that somebody is missing from this livestream. Ryan is in the middle of travel, got rugged by some travel plans, so I'm taking this one solo today, and our fearless leader will be back with us next week for the next week's State of the Nation. But in the meantime, we have to talk about a message from one of our sponsors, Alchemix. Alchemix, you guys know Alchemix. It's the crypto-powered savings paradigm where you can save money either in ETH or DAI or your other stablecoins. You can deposit it into the Alchemix DeFi yield farming account.

• 2:22David Hoffman?

  And for all of the money that you deposit into Alchemix, you can withdraw up to 50% of your deposits. While that deposit grows and grows and grows from your yield farming in DeFi, which Alchemix does for you. So it allows you to save your money and spend it too. It allows you to get your interest payments paid to you upfront through the Alchemix web app. And so you can check them out. There is a link in the show notes, bankless.cc/Alchemix, so the capital A. You can see on screen all the assets that you can leverage, including staked ETH from Lido and Rocketpool, as well as your preferred stablecoin. And if you are an Alchemix token holder, there is a coming tokenomics upgrade. And there's a link in the show notes for you to explore that as well.

• 3:07David Hoffman?

  At this point in time, this is when Ryan would ask me what the state of the nation is. And so I just have to ask myself what the state of the nation is. And today, the state of the nation is checking and balancing. And that, I think, is going to be a theme of the episode. All of these very complex things, like I mentioned earlier, dank sharding, proposer builder separation, data availability sampling, all fits under the theme of checking and balancing. And we've heard the phrase checking and balancing before. This comes from like your basic US history class. And why it's so powerful is it allows for not one part of the power structures around Ethereum to outsize the others. And so today, the state of the nation is we are checking and balancing the state of Ethereum. So we're going ahead and get right into the show with John from Delphi Digital

• 3:51David Hoffman?

  to talk about all the different ways that Ethereum checks and balances itself. Right after we get to some of these fantastic sponsors that make the show possible. The era of proof of stake is upon us and Lido is bringing proof of stake to everyone. Lido is a decentralized staking protocol that allows users to stake their proof of stake assets using Lido's distributed network of nodes. Don't choose between staking your assets or using them as collateral in DeFi. With Lido, you can have both. Using Lido, you can stake any amount of your ETH to the Lido validating network and receive ST ETH in return. ST ETH can be traded, used as collateral for lending and borrowing, or leveraged on your favorite DeFi protocols. All this without giving up your ETH to centralized staking services or exchanges. Lido now supports Terra, Solana, Kusama, and Polygon staking.

• 4:33David Hoffman?

  Whatever your preferred proof of stake asset is, Lido is here to take away the complexities of staking while enabling you to get liquidity on your stake. If you want to stake your ETH, Terra, Sol, or Matic and get liquidity on your stake, go to Lido.FI to get started. That's L-I-D-O dot F-I to get started. The Layer 2 era is upon us. Ethereum's Layer 2 ecosystem is growing every day, and we need bridges to be fast and efficient in order to live a Layer 2 life. A cross is the fastest, cheapest, and most secure cross-chain bridge. With A cross, you don't have to worry about the long wait times or high fees to get your assets to the chain of your choice. Assets are bridged and available for use almost instantaneously. Across bridges are powered by Uma's Optimistic Oracle to securely transfer tokens from Layer 2 back to Ethereum. A token

• 5:16David Hoffman?

  proposal is being deliberated as we speak in the Across forum, where community members will decide on the token distribution. You can have your part of Across's story by joining the Discord and becoming a co-founder and helping to design the fair, fair launch of Across. If you want to bridge your assets quickly and securely, go to across.to to bridge your assets between Ethereum, Optimism, Arbitrum, or Boba networks. If you're trying to grow and preserve your crypto wealth, optimizing your taxes is just as lucrative as trying to find the next hidden gem. Alto IRA can help you invest in crypto in tax-advantaged ways to help you preserve your hard-earned money. Alto Crypto IRA lets you invest in more than 150 coins and tokens with all the same tax advantages of an IRA. They make it easy

• 5:58David Hoffman?

  to fund your alternative IRA or crypto IRA via your 401k or by contributing directly from your bank account. There is no setup or account fees, and it's all you need to do to invest in crypto tax-free. Let me repeat that again. You can invest in crypto tax-free. Diversify like the pros and trade without tax headaches. Open an Alto Crypto IRA to invest in crypto tax-free. Just go to altoira.com/banklist. That's A-L-T-O-I-R-A dot com slash banklist and start investing in crypto today.

• 6:31David Hoffman?

  All right, Bankless Nation, we are here with John Charbonneau from Delphi Digital. John, welcome to the show, man.

• 6:37Speaker 01

  What's up? Happy to be on. First podcast for me.

• 6:40David Hoffman?

  Yeah, congratulations. And we were talking a little bit backstage. You have a very recent history with Ethereum and crypto, which is pretty impressive that you wrote an article praised by Vitalik himself as being extremely accurate and extremely well researched. And so, like I mentioned, I teased in the intro, some of these things are extremely complicated. There is like crazy math, like you gotta go back to algebra and polynomials to understand some of these things. But that's not here what I want to talk about today, because I kind of want to talk more about just the vibes of all of these things, how all of these things have a shared structure, a shared pattern. Before we get into some of the more complicated stuff, can you just summarize the vibe of the Ethereum roadmap post-merge? Like, what should we expect?

• 7:26Speaker 01

  Yeah. So the big thing, obviously, is the merge is not going to be scaling Ethereum. So the priority after the merge is going to be all of these different steps that we need to scale all of the computation on Ethereum through the rollup-centric roadmap. So basically, trying to make Ethereum a really good scalable base layer for rollups, while at the same time scaling that throughput, keeping it really, really decentralized and easy to validate. Because that's what keeps everything in check ultimately. True scaling isn't just what's your TPS, it's throughput relative to what is the cost to validate. And that second part is what a lot of other typical monolithic ones that just try to like jam throughput through on beefier

• 8:08Speaker 01

  hardware, they ignore that second part. So that's not true scalability in mind, it's keeping that second part in mind at the exact same time that you need to do.

• 8:17David Hoffman?

  Right. One version of scalability is we just do away with the whole blockchain thing, and we just go back to a database, and then we have just a scalable system on the planet, but then we lose all of that trustlessness. So to summarize what you're saying, post-merge, post, once we get to proof of stake, it's all about finding these different ways to scale computation, to scale throughput, without losing all of the cool properties that make a trustless blockchain, a trustless blockchain, all right?

• 8:46Speaker 01

  Yep.

• 8:47David Hoffman?

  And so in the intro to your piece, and just to dive into this a little bit more, you wrote,

• 8:54David Hoffman?

  scaling computation without sacrificing decentralized validation. Let's dive into the validation aspect of this. What does it mean to scale computation without sacrificing validation? From the user who might be like thinking about staking Ethereum or Ether in the future to become part of the network, that validation aspect I think is really important. Can you talk about and elaborate on that part?

• 9:18Speaker 01

  Yeah, so the big theme with a lot of what Ethereum is going to be doing going forward, both at the L1 level and the rollup level, is that there's probably going to be some specialization and centralization in tasks, primarily block production. The realization of that is that you just need to really focus the most on decentralizing the validation of that, because that's what keeps everything in check. Like Vitalik's endgame post is what put it the best, that most roads tend to lead to that scenario. Whether it's in rollups, you see that you're probably going to have somewhat specialized block production. If you want to have the highest throughput on them, what's important is to keep regular

• 10:00Speaker 01

  users behind essentially full node security without the resource requirements of what typical like high throughput L1s would require you to need today. So you want to give people that level of security by fully validating the chain. So if you have a malicious block producer, you just reject the transaction. You could just see right away. If you're just running a light client of a monolithic L1, that because it just has too high a hardware requirement to run a full node, they can pretty much do whatever they want to, and you're just not going to notice it because you're trusting the honest majority. That's not the case with a full node. A full node won't be accepting invalid transactions. And if you're working on a rollup, then you're hiding behind the safety of fraud proofs or ZK proofs.

• 10:44Speaker 01

  So keeping that validation for regular users is what really, really matters a lot.

• 10:48David Hoffman?

  And so we, I mentioned three components in the intro. Data availability, sampling, proposer builder separation, and dank sharding slash proto dank sharding. These are all different strategies to get to the same end result, correct? And so different ways to scale Ethereum that scale different parts of Ethereum, but ultimately produce that scaled Ethereum with decentralized validation. Is that all correct?

• 11:13Speaker 01

  Yeah, that's right, because the overall vision of Ethereum right now is obviously the rollup-centric roadmap. The problem with Ethereum right now is it's not built to actually host rollups. It's kind of just makeshift right now, and we're making do with what the L1 can handle. So a lot of these are primarily geared towards scaling the data availability throughput, because that is a big bottleneck for the rollups right now, that they need to post their data to the L1. And the layer one is not optimized for that today, so it's being able to scale that, while at the same time making it really, really easy for regular people to validate that.

• 11:46David Hoffman?

  So people who were familiar with the older Ethereum roadmap, we talked about sharding as a way to like scale computation on the Ethereum at layer one, but we've shifted away from that, where computation scaling has been thought to now move into the rollups. And computation scaling, that's what we're really talking about, like transaction throughput, like how fast transactions can go. And so, once we, now that we have this rollup-centric roadmap of Ethereum, a lot of that computation is happening on the rollups, that's what rollups are do. And but in order to allow those rollups to really become unleashed and go up to their maximum throughput potential, we need to enable data to have make data available to them,

• 12:29David Hoffman?

  because the resource that rollups need the most is data. Can you talk about this transition from like this older version of the Ethereum roadmap, where we were going to actually scale the Ethereum layer one in terms of transactional throughput, to what we've kind of pivoted to now, which we've actually focused on scaling the data availability?

• 12:50Speaker 01

  Yeah, sure. So the old roadmap was, yeah, just jammed through the execution on the L1 through execution sharding. We've since realized essentially that it's better in nearly all respects to probably put that on rollups and then just optimize the base layer for data availability. So that is what the old sharding design was doing, the the previous one after execution shards. It shifted to the previous data sharding design, where there was going to be these 64 different data shards, and you just post data to them as the rollups and they're Committees that are checking all of them. And so as a validator, you're testing that all the data was made available because the data availability is important for the security of those rollups.

• 13:32Speaker 01

  For example, in the case of an optimistic rollup, you need the data available to successfully be able to arbitrate a fraud proof. You also need it in the case of a ZK proof for ZK rollups to be able to recompute the state and be able to exit it safely. So the key thing that we've realized is, okay, we can shift that block production and execution off chain to rollups. While now just focusing really on making a really scalable data layer, while also keeping the regular execution part for settlement of rollups. Like they post the proofs to the L1, they can bridge through it, keeping all of those trust assumptions together. But focusing on now scaling, okay, how do we put a punch of data for the L1 without jacking up resource requirements?

• 14:15David Hoffman?

  And correct me if I'm wrong. And again, listeners, we are going to go into the three things in detail. Data availability, sampling, proposer builder separation and dank sharding. But John, before we get there, just again, correct me if I'm wrong, but it's data availability, sampling and dank sharding that do scale up. The effective data of the Ethereum layer one effectively increasing throughput on the rollups. And proposer builder separation is something different. Is this all correct?

• 14:43Speaker 01

  Yeah. So proposer builder separation is one of the really important, pretty much necessary things to enable dank sharding compared to the previous sharding design. So you could do data sharding without proposer builder separation. But proposer builder separation makes the new sharding design possible, which wasn't previously. Because otherwise it would have been too high of a resource requirement for regular validators. Data availability sampling is part of that making it really easy for validators part. That I can very securely check that all of the data was made available and know with 100% certainty or near 100% certainty that it was all available.

• 15:22Speaker 01

  And that means that, okay, if it was all available and I'm working on a rollup, if the data was available and no fraud proof has been posted. I can safely assume if there is one person who would have posted that fraud proof, that I'm good to go. Or if a ZK proof was posted and I know all the data was available, then I know I'm good to go.

• 15:40David Hoffman?

  And what that does is that allows effectively more data to become useful to the Ethereum validators. Well, because we only have to verify a subset of that data, we get to prove that all of the data is available without having to check all of the data. And so we effectively get a data throughput increase because our validators only have to check a minority subset of the data, but as a whole entire ecosystem, we get to leverage the full expression of all of the data. And this is what we mean by scaling Ethereum without scaling computational requirements. Am I tracking here?

• 16:16Speaker 01

  Yeah, exactly. Today is the opposite of that where it's not built as a data availability layer. So when you post call data, the L1, it's just every node has to fully download it and then you hold on to it. That's a very resource intensive way that you can't scale that up to massive throughput.

• 16:31David Hoffman?

  Okay. And then overall, I would say resource intensive. I would say that these three things, again, which we're going to go into, all fit under that vibe of how do we scale computation without scaling resource requirement as like literally using cool cryptography, using cool math tricks. How do we scale computation, i.e. throughput, i.e. transaction speed and costs, without increasing the resource requirements of these tricks, because that is what preserves Ethereum and keeps it decentralized?

• 17:05Speaker 01

  Yep. Yeah, that's right.

• 17:06David Hoffman?

  Okay. And one last question as we stay high level, because again, super complicated subject. So I want to make sure all the listeners and all the viewers get the right vibe. What does this end state look like? Post the post inclusion of all of these different mechanisms, again, data availability, sampling, proposer builder separation, and thank sharding slash proto thank sharding. What's the end state of Ethereum look like? Can you kind of just walk us through a holistic visualization or interpretation of Ethereum in this end state?

• 17:36Speaker 01

  Yeah, the yeah, the really high level of it will be that you basically introduce this new entity who's going to be a specialized builder. So they're going to be responsible for making this really big block together that has all the data, the beacon chain block, everything put together. And then you have this very decentralized, very low resource requirement set of validators. Who don't have to build the whole block, they just take it and they have to say, okay, is this a, is this block good to go? And was all the data made available and you could pretty easily do that data check with data availability sampling. Which is very different from today. Today you would need to download all of that data. So that's why we can't scale it up. When you introduce the PBS proposal builder separation as that new role.

• 18:21Speaker 01

  You get rid of the high resource requirement parts and then you add data availability sampling to make it really easy to do that proposer job of checking the data was available.

• 18:30David Hoffman?

  Okay, and the the last subject I want to want to get to is the interactions between all these components. And so I want, I want to put my devil's advocate hat on my Ethereum crit- critiquer hat on and say, well, Ethereum is complicated. And it's just like solving its problems with more complication, right? It's just adding complication on top of complication. And then eventually, like, once all of these three things that David keeps on listing off by name. Once those get included, then we're going to have like eight more things to solve the complexity there. And then we're going to have like 18 more things to solve the complexity there. So that's the, that's the devil's advocate version of Ethereum where like it's already complicated and we're just making it more complicated to do all these things.

• 19:10David Hoffman?

  And then there's the opposite interpretation where actually these three things fit together really, really well. And have this sort of like elegance between the interactions between these three components. What, what side, what side of that, that take would you say that you're on?

• 19:26Speaker 01

  It's all a mess.

• 19:29Speaker 01

  Every, everything does fit together really well. Even beyond just these three components, when you zoom out and look at almost every major component of Ethereum's future roadmap. Almost everything does boil down to those two points of scaling computation while making it really, really Really easy to validate. I mean, taking these as an example and how they are necessary and kind of interweave with each other. Dang sharding, which is what we're going for, proposer builder separation is necessary to do that. Helps it scale easily and it makes it easier for us to validate it because you make the proposer job really easy. These, all of these different pieces are kind of just running in parallel and they're all chipping away at different things.

• 20:12Speaker 01

  Whether you look at, even looking at a bunch of stuff on the Ethereum roadmap, it's so much of it is just how do we lower the resource requirements to validate? Whether you look at things like history, like history expiry, that's okay. You don't need to have as big a hard drive to run a full node. If you look at statelessness, that helps like reduce your SSD requirements. You don't have to hold your state on hand. Like they're all chipping away at different pieces. I mean, like data availability sampling, which is one that we'll go into more, like it helps you lower your bandwidth requirements and that you don't have to download all this data. All of these things ultimately do really come into this very cohesive view when you zoom out and you like start to understand how they all work with each other.

• 20:50David Hoffman?

  So I think with one of these components, like take data availability sampling, for example, like we get a bandwidth reduction, which is nice because then blocks propagate faster. We can have more throughput that way.

• 21:02David Hoffman?

  That might be like one single like sized upgrade. I don't know how you want to side this thing, like maybe one order of magnitude. But then when we layer on like the next thing, like proposer builder separation, like we actually get another order of magnitude, but then these things interact. Right? These enter, these orders of magnitude compound on each other. And then we throw in the third thing and we have like three different ways we are decreasing resources requirements and they're not, they don't, they're not linear. It's exponentially scaling. It's exponential reduction of resource requirement, which turns into an exponential scaling of the actual throughput of the holistic Ethereum ecosystem. Is that a fair take?

• 21:42Speaker 01

  Yeah, yeah, that's generally about right. And that's why you see like people will have heard of and will go into this some more depth proto dank sharding, which is a halfway step to dank sharding. And that's exactly what you see where you have a certain amount of data availability throughput today. If you go to proto dank sharding, which implements part of the steps of dank sharding, you get some orders of magnitude scaling. And then dank sharding layers on even more of the exciting changes that are coming. And then you get even more orders of magnitude scaling for data availability on top of that.

• 22:11David Hoffman?

  So just to answer a question, which I think might have popped up in some listeners minds. Dank sharding slash proto dank sharding. Can you talk about how these names came to be?

• 22:21Speaker 01

  Yes. So dank sharding, which is a great name. Play on took the old just sharding design and Don Kraut is the one who came up with that. And then the halfway one proto dank sharding is a play on taking that and then also proto lambda pitched in to add that like halfway measure.

• 22:41David Hoffman?

  Right. Okay. So proto lambda, he's with the, he can't remember where he was, but he was working maybe at the EF before this. Yes, he was working at the EF. Then he joined the optimism team. He's called proto lambda. I'm sure it's not his actual name, but that's what we call him. And then there's dank red feist also at the EF. And so these, these things are fondly named after them. And so Ethereum sticking true to its name of kind of picking weird names for its upgrades. So, okay, so the, here's where I get a little bit confused, John, and I kind of want you to pick the next path forward. We have three different things to talk about. Data availability, sampling, proposer builder separation, dank sharding slash proto dank sharding. Since they're all interrelated, like, is there a starting place? Is there one that we should pick out first to talk about first?

• 23:26David Hoffman?

  Or is it kind of a chicken and an egg problem? Since they're all interrelated, there is no starting point for this whole entire conversation.

• 23:34Speaker 01

  Yeah, they're all pretty interweaved. Like, you're gonna see data availability sampling, proposer-builder separation, and dank sharding all come together. And then proto-dank sharding is the halfway step that implements like some of the transaction formatting. So they all do very much interlock, those three.

• 23:50David Hoffman?

  So we're just gonna have to pick one?

• 23:52Speaker 01

  Yes.

• 23:52David Hoffman?

  All right, let's go with that data availability sampling. Can you explain data availability sampling? And how it scales computation while keeping Ethereum decentralized?

• 24:02Speaker 01

  Yeah, so back to that message of we're trying to scale the data availability throughput of the L1. And just to remember, there's a difference between data availability and data retrievability. What we need for the security guarantees of the consensus layer isn't that this data is retrievable forever. What we need to know is was this data published? Did we attest that it was available such that it's available for some period of time? That any interested party could have downloaded it and could have reasonably submitted or fraud-proof? Or done anything like that? Like, that's what we need for our safety. The history, like, retrievability of it is a separate, much weaker assumption that, like, is dealt with separately.

• 24:45Speaker 01

  So that's what the data availability is. And then...

• 24:49David Hoffman?

  Well, let's actually dive into that a little bit more. So data availability, if I'm recalling my Ethereum knowledge, it makes it availability for an amount of time. And that amount of time is just assumed to be long enough to know that the whole world could have downloaded and had that data available to them. But it doesn't, but the Ethereum protocol does not commit to embedding that into the blockchain perpetually. So it will not be saved inside of the Ethereum blockchain forever. But the Ethereum protocol generates assurances that there was enough time for the whole world to have downloaded that data at some point in time. Well, can you just, let's elaborate and unpack the difference between these two things?

• 25:32David Hoffman?

  And why we wouldn't just save it in the Ethereum protocol forever? And how do we know that it was actually available to the whole world for a sufficient amount of time? And why are we, why do we feel secure pruning it from the blockchain at all?

• 25:45Speaker 01

  Exactly, yeah. So the problem with if you require it stay on chain forever, that every like full node has to hold on to this data forever. You just can't feasibly keep this thing decentralized. That's just way too high of a resource requirement to ask all of these nodes that you want to be validating. To be able to hold on to this like ever growing massively growing thing that's going to have terabytes of data that are running through it when we like implement sharding in these things. So right off the bat, we know that that just can't keep it like decentralized validation if we do that.

• 26:16David Hoffman?

  And the resource bandwidth that we're talking about here is hard drive space, right? Like if we if we let it get out of hand. All Ethereum nodes that like all all of us that plan on staking our ETH and running our nodes. We would have like a normal sized computer, except for like this massive, massive, just like rack of hard drive space. And so that that's the resource requirement that we are minimizing here with this data availability sampling. Is that correct?

• 26:44Speaker 01

  So that would be minimized by the fact that we are going to start pruning. These data blobs when we introduce the new transaction format with proto dank sharding. This this history issue is going to be is going to be exacerbated by the fact that we start posting a bunch of data to the L1, but it's already an issue. Which is why apart from all of this, there is separately another EIP like called EIP 44s. That is going to start allowing nodes to prune. All historical data after a year and stop serving that to the network. The the difference today is that call data, which is what rollups use today. It just persists on chain. It just stays there, but that's not the security guarantee that we really need for rollups.

• 27:28Speaker 01

  So what proto dank sharding is going to do, one of its big halfway steps to going to full dank sharding. Is it introduces the new transaction format that rollups will use. So today they post call data. Once proto dank sharding is enabled, they'll start posting data blobs. And the thing with data blobs is they can get pruned after about a month. So that's more than enough time that we guarantee that it's available, that people could have posted fraud proofs or done whatever. Anyone could have downloaded it in a reasonable period of time. But then full nodes just don't have to hang on to it. It's not their problem. Because history is just a one of n, like trust assumption. As long as someone out there is holding the history, it's fine. I could just go ask them for it. You don't need everyone to be holding that on hands. Like, it's just not necessary.

• 28:12David Hoffman?

  So one of n assumption, as in if there's one Ethereum node that's out there that has this data? Or like, what is the actual process of why, why would I would want to go retrieve very old data? And then how do I actually do go about finding that data?

• 28:31Speaker 01

  So, yeah, I like, it could be anyone. I mean, like, the easiest example is BlockExplorers. If you have some kind of business that depends on, I need to be able to serve this history whenever like anyone asks for it. They're a perfect example of, as long as they're storing it, you're fine. There are other decentralized and like more innovative options to make sure that we are better incentivizing history. Retrieval as it starts to become more of an issue. It's not like a big issue today. Like another thing is, it's really not an issue for the Ethereum protocol. It's only an issue for apps, basically. If they use, if they lose like their old history and they're like, oh, I can't prove that this thing happens.

• 29:12Speaker 01

  So one thing that, for example, rollups could do is they could just mandate that they have to hold on to their own, like, relevant parts of their history. There's lots of innovative solutions where as long as someone is holding on to it and I could just ask them, I can always prove that what they gave me is valid. So as long as someone still has it, it's not lost forever.

• 29:30David Hoffman?

  Okay, so if a rollup makes a layer one transaction with a specific amount of interactions with a part of the Ethereum layer one, like it talked to these tokens and talked to those tokens, the rollup can make sure that the data for those relevant transactions are always stored by the rollup providers, right? The rollup nodes. And so that would make that that part of Ethereum persistently and perpetually available for, for those, that particular rollup. And so that part of Ethereum is secured, and then we could probably repeat this process like 50, 100, 1,000 times. And then eventually we'll get to the full entire, full archive state of Ethereum, but it's no one single node is responsible for storing all of it. Is this, is this correct?

• 30:12Speaker 01

  Yeah, it's never going to be mandated in protocol that Ethereum needs it. You could also try to do decentralized uh different like incentivization where you can have like different networks. Um where you are paying people to like hold your data, but it it's just fundamentally not the Ethereum L1 like consensus layer's problem. Um because it's not like breaking anything if you can't like resurrect your history. It's a it's a problem for that app like, hey, I can't do this transaction or whatever. Now I can't prove this thing. Um but that fundamentally is just a different task that's much easier to outsource and it makes sense to to like keep everything very decentralized.

• 30:47David Hoffman?

  Okay, it's not breaking anything if you can't uh. I can't remember how you exactly said it, but I think it led into my my next question where if I put on my Bitcoiner hat. And they're like, you guys are pruning the data from your chain like I would get triggered, right? Like Bitcoiners are extremely hyper focused on backwards compatibility and making sure that if somebody mined a block in in 2010 and then they didn't move those coins from 2010 onwards into 2050. Like the the emphasis on Bitcoin is that you will always be able to retrieve your coins and the data and you also get the the self sovereignty from always being able to run a node and spin up a node from Genesis and verify every single transaction.

• 31:30David Hoffman?

  So we're we're not violating that principle by doing this?

• 31:35Speaker 01

  Yeah, because the the reality is is you won't yourself have the data, but as long as you can get from one single person out there, like it's really just not an issue and it fundamentally isn't a big concern because there are always going to be these things and you can incentivize them in different ways. You can use like they're like third party like indexing things like you'd have like the graph. Um like there there's work with like all of these different things that like as long as someone's holding on to it like it it's not a major concern.

• 32:03David Hoffman?

  And the uh and the idea is that even if one person, if again, putting on my Bitcoiner hat. Okay, but you're asking me to trust one single person to serve me the data like what if they serve me incorrect data that uh serves their needs instead of mine? Uh what what's the response to that?

• 32:21Speaker 01

  So you only need one out of all of these different candidates who should be storing it, whether it's at, whether it's rollups storing theirs, their own data, whether it's individuals, block explorers, if people start using like the graph or all these different solutions, you just need one of them to provide it to you.

• 32:38David Hoffman?

  But how do I know that they're giving me the correct data?

• 32:41Speaker 01

  Exactly. Like they can't give you fraudulent data. You could always still prove that, hey, this thing happened. Like you can, like that, that's not like, they can't just serve you invalid data and then you'll take it. You can always prove on chain that like, you can go back and look that, hey, is, does this match? Like they can't trick you into doing something. The worst thing that could happen is just, they don't give you the data and then you just can't retrieve what happens.

• 33:03David Hoffman?

  Okay. Now, is this still, this is where some of these interaction between these things get a little hairy. Is this what we're discussing now about the ultimate pruning of the, of the blockchain? Is this data availability sampling or is this dank sharding or proto dank sharding? Or is it both?

• 33:19Speaker 01

  Uh, this will be implemented with proto dank sharding.

• 33:23David Hoffman?

  Okay.

• 33:23Speaker 01

  And then the other steps will be implemented with full dank sharding. So this is the main reason that proto dank sharding is able to scale is because it introduces the new transaction format that are going to be carrying these data blogs. And these data blogs will be pruned after about a month or so. Once we know that they've safely been available, anyone could have downloaded them, anyone could have checked them. You can prune them. And then because of that, we can like send through orders of magnitude, more data throughput, because we're not requiring people to hold on to these things forever. So we don't have to worry about blowing up the history with them.

• 33:56David Hoffman?

  Okay. So with, with dank sharding and proto dank sharding, we can juice up the throughput of Ethereum because we are only requiring the nodes to really store the more recent parts of Ethereum and not the long-term parts of Ethereum. Therefore, we're, we're more okay with juicing up the data requirements because it's all in the short term, not in the long term. Correct?

• 34:18Speaker 01

  Yeah, and the reason that we could then take another step up from proto-dang sharding to full dang sharding is because in proto-dang sharding, we're not going to have data availability sampling or any of that. We are still going to require the validating nodes to fully download the data. That's how they'll have to check that the data is available is they will have to fully download everything and say, hey, was this data available? When we go to full dang sharding, after that, you'll only be checking a subset of the data. So then the data will be sharded. So then that gives you the extra orders of magnitude scalability.

• 34:49David Hoffman?

  Okay. And then the, the going from proto-dang sharding to full dank sharding, that's when that crazy like polynomial math comes into play where like we can allow for more data to come in because the validators only have to check parts of it. And because of crazy polynomial math, we have the assurances that to like the 99.999% likelihood that all of that data is there. And so we can, do you know like the order of magnitude increase in data from going from proto-dang sharding to dang sharding? Do you know what that is?

• 35:24Speaker 01

  From, sorry, from proto to a full dang sharding?

• 35:26David Hoffman?

  Yeah, uh huh. Like what's the scalability increase on that one?

• 35:30Speaker 01

  I want to say it's about a 30x if I remember correctly.

• 35:33David Hoffman?

  Wow. Okay. So we, we take proto-dang sharding and then we add this like polynomial math to get to full dang sharding. And that gives us a 30x increase in, in data availability?

• 35:46Speaker 01

  Uh, yes, or I want to say 16. I don't remember the multiple on it, but it's, yeah, it's, it's in the, it's in the low orders of magnitude number. Um, the target block size is going to be around a megabyte um, uh, for data availability. For, um, for proto-dang sharding, and then it'll go up another couple orders of magnitude or so. From proto to full. So because we have data availability sampling, because you don't have to download everything anymore.

• 36:11David Hoffman?

  Okay. So data availability sampling, is that the polynomial math thing that I was just talking about?

• 36:17Speaker 01

  Um, that is part of it. Um, you, yeah, the KZG commitments are like how you um commit to the data and that like includes um that that's the like polynomial part of it. Um, that's where like some of the trickier math comes in. Um, that would be the KZG commitments would be partially introduced um by proto dank sharding, but not the full like 2D KZG scheme with data availability sampling, like some of the like the like nitty gritty stuff.

• 36:46David Hoffman?

  Okay, so what components of data availability sampling have we not already discussed that we need to dive into? And and or maybe we should start from start from scratch because I'm literally, I think the listeners can like figure this out by now. I'm literally learning as we go. Um, okay, let's let's go back to data availability sampling. Uh, what what is the resource requirement that that is trying to mitigate and how does that work?

• 37:08Speaker 01

  Yeah, so yeah, first I'll explain how it works. Um, so the basic idea is, like I said, with proto dank sharing, it's not super scalable to just download all the data. So what we want to do is make it such that you only have to download a piece of the data. Um, a really small portion of it, but how do you like the naive way to do that is just to like take this data that was posted, I check a bunch of little pieces, and if they're all there, I'd say it's good. Um, the problem with that is is you could have missed the one transaction that like prints a million ETH or like whatever, and then you're screwed. Um, and like all your money's gone or whatever. Um, so that doesn't give you any guarantee.

• 37:41David Hoffman?

  Every single transaction and we

• 37:43Speaker 01

  Exactly. So that would be the naive way to do it. Um, the better way to do it is what you you do what's called erasure coding. So that initial data, the original like actual data gets extended. So you have this like parity data. And this is like used in like error correction. It's the same like type of technology that's used in like CDs that if they get scratched, you can like reconstruct it with only like part of the data. It's the same like concept as that. Um. Where now you have this original data and you have the extension data and you only actually need 50% of that data and then you could reconstruct the whole thing and it could be any 50% of the data. It could be a little bit of this, a little bit of this, like it could be

• 38:25Speaker 01

  anywhere. So then it makes it really easy to do that sampling math because as long as I have 50% of the data, I could recreate the whole thing. So that makes it really easy now because I could do the samples and for you to trick me into like a testing to this block, you have to be hiding more than 50% of the block, not, not just like one transaction, you have to be hiding more than 50% of like this data that was given to me. So now let's say you were trying to do that and I checked once and I was successful. I got the data. So I still have a 50 50 shot that like the block isn't fully available. But if I check it again, now I have a 75% chance and then you have an 87.5%

• 39:06Speaker 01

  chance. Like you keep doing that. And if you do that like a fixed number of times where you have a 50% chance of being tricked each time and you do that like 30 times or so, your odds of being successfully tricked like 30 times in a row at 50% odds is like essentially zero. So that's what we do is we extend the data using Erasure coding such that you only need 50% of it to be available and then you could safely sample the different pieces. And then if those were successful, you can say with pretty much statistical certainty that like the block is available to me.

• 39:44David Hoffman?

  Okay, so let me, I'm going to repeat this back to you to, because to, to hope that I got it. So there's this, there's this amount of data, there's this data that I got to download and verify and I need to make sure that 100% of all transactions in this bit of data is completely valid. It's not enough to do all transactions, but one, we can't cut that shortcut because if one single transaction gets in, if this is coming from an attacker, they might do something like print a billion ether and give it to themselves and all of a sudden we have like a huge problem with Ethereum security if somebody is able to print a billion ether because now they probably control the network. They definitely do control the network. So we need to make sure that 100% of all transactions are completely valid. But we do, what we want to do.

• 40:28Speaker 01

  We don't, we, we don't have to make sure that they're valid. Sorry. That, that's one point to, that, that is really important. Is the, the layer one says nothing about the validity of those transactions that are posted to the L1 from Rolos. It just says that they're available. And then you rely on proofs to tell you, was it valid or not? But the L1 guy who's just checking the data doesn't say it's valid. It just says it was available and someone else will check it.

• 40:52David Hoffman?

  Okay, okay, cool. Again, leaning into the whole separation of checks and balances. This one person says this data is available, somebody else go check if it's valid. But right now we're talking about, okay, all transactions are available to be checked by somebody else. Uh, and so this one person's job is to say, all of this data, every single transaction that, uh, is, uh, exists can be checked by someone. And again, we need to make sure that every single transaction can be checked. Uh, so we have this amount of this thing of data and we do this thing, this weird thing, this complex math thing called a razor, a razor data. Uh, razor, uh, it extends the data and my version of that is it, it makes, uh, my interpretation of that is like, it makes like a,

• 41:34David Hoffman?

  kind of like a, like a square root of negative one version of it. I don't know. It takes this data and it makes this like random, uh, uh, manipulation to it to extend the data. And so, and what that does, again, with crazy math is that, uh, it takes the same blob of data, but then it makes this like clone of it. That is in a particular, uh, out of a particular derivation. And then this person, as a result of that, uh, I might be skipping this up here. As a result of that, uh, you, I need to check that. This is when we get into the, the 50% odds thing, right? Because of this particular way of this expression of this, uh, creation of extra data, uh, and also with polynomial math.

• 42:15David Hoffman?

  Yeah.

• 42:15Speaker 01

  Yeah, so the, yeah. Yeah, so the way the polynomial. Yeah, so there's two parts of it. You basically need to know that, one, the data was made available. So that's what data availability sampling gives you, that you can check all these random pieces, say it was available and I'm good to go. The other part of it is you need to make sure that the data was extended properly.

• 42:34David Hoffman?

  Yes.

• 42:34Speaker 01

  Because if that extra 50% they gave you was just like dummy data, then I can't actually reconstruct it because that was just like useless data. So that's where the like polynomial math comes in.

• 42:44David Hoffman?

  Okay.

• 42:44Speaker 01

  These things called KZG commitments. Those will prove to you that the data was extended correctly.

• 42:51David Hoffman?

  Okay.

• 42:51Speaker 01

  So then you have the KZG commitments telling you, proving that the data was extended correctly. And then I could data availability sample saying it was extended correctly and it was also available.

• 43:00David Hoffman?

  Okay, so once we get to that point, we have, we have those assurances. And then we just start taking actual samples, right? So the first step is to prove that all of the data is, is extended correctly. And that extending correctly, once we get to that point of extending the data correctly, that's when we get to sample it with the increase, increase assurances that there's nothing hidden. So the extension allows us to have more effective sampling. And then, once we get to the extension, then we start sampling this thing. And sampling is like flipping a coin, except once we get to like 30, 30 coin flips, like you will never, ever, anyone in anyone's life, flip a coin 30 times and have it be heads every single time. That will never, ever happen in your life.

• 43:45David Hoffman?

  And so that's how we get to the assurances that we feel secure about, like, there is, it's literally impossible to hide a malicious transaction in this like fake bit of data. And so the first step is to extend the data to allow for more efficient and more effective sampling. And then we sample it about 30 times, because that's where like the level that we deem

• 44:07David Hoffman?

  safely, safe, to assume that all the rest of the data is also valid, or not valid, but available. Is this all correct?

• 44:16Speaker 01

  Yeah. In the real number, it's going to be 75 for some like complicated 2D KZG scheme stuff, but it's probably not like necessary to go into the details of that.

• 44:28David Hoffman?

  Okay, okay. All right. So as you can see, my brain is starting to break. John, can you just walk us through one more time? Now that we've gone through this, can you just walk us through the data availability process one more time from the high level?

• 44:40Speaker 01

  Yeah, so in proto dank sharding, it's going to be simple. You still just fully download the data and then that's it. If you fully downloaded it, you got it, then you're good to go. You sign off on it. The reason that you could do more throughput for the data in full dank sharding is because now you're only going to be, the validators will only be downloading parts of the data. They'll be doing data availability sampling. They'll be downloading like certain rows and columns. And then they'll test you if theirs were available. Then collectively, you're good to go as long as there's enough people who are sampling it. So that like easier requirement allowed because you don't have to download the full thing makes it easier for you to like,

• 45:24Speaker 01

  kind of like juice up the data availability throughput at another level.

• 45:27David Hoffman?

  Fantastic. Okay, so this is where we're about to get into a very fun part of this whole entire story, which is proposer builder separation. And this is where we start to talk about getting paid in Ether. And so this is where the Ether economics of being an ETH validator come into play. And overall, where we'll actually start to be able to differentiate Ethereum scaling strategy from the rest of the alt layer ones. And so, John, I'm going to pick your brain on that because I know you've also written another article comparing like the overall, the different strategies of alt layer ones and the relations to their L1 assets. So definitely a very fun part of the conversation, which I definitely understand a little bit more than the first part of the conversation. So we're going to get there right after we talk about some of these fantastic sponsors that make the show possible.

• 46:10David Hoffman?

  AVE is the leading decentralized liquidity protocol. And now AVE V3 is here. AVE V3 has powerful new features to enable you to get the most out of DeFi, including isolation mode, which allows for many more markets to be launched with more exotic collateral types. And also efficiency mode, which allows for a higher loan to value ratios. And of course portals, allowing users to port their AVE position across all of the networks that AVE operates on, like Polygon, Phantom, Avalanche, Arbitrum, Optimism, and Harmony. The beautiful thing about AVE is that it's completely open source, decentralized, and governed by its community, enabling a truly bankless future for us all. To get your first crypto collateralized loan, get started at ave.com. That's A A B E dot com.

• 46:52David Hoffman?

  And also check out the Aave Protocol Governance Forums to see what more than 100,000 DAO members are all robbing about at governance dot Aave dot com. Arbitrum is an Ethereum Layer 2 scaling solution that's going to completely change how we use DeFi and NFTs. Over 300 projects have already deployed to Arbitrum and the DeFi and NFT ecosystems are growing rapidly. Some of the coolest and newest NFT collections have chosen Arbitrum as their home. All the while, DeFi protocols continue to see increased usage and liquidity. Using Arbitrum has never been easier, especially with the ability to deposit directly into Arbitrum through all the exchanges, including Binance, FTX, Huobi, and Crypto.com. Once inside, you'll notice Arbitrum increases Ethereum's speed by orders of magnitude for a fraction of the cost of the average gas fee. If you're a developer who wants low gas fees and instant transactions for your users,

• 47:37David Hoffman?

  visit arbitrum.io slash developer to start building your DApp on Arbitrum. If you're a Dgen, many of your favorite DApps on Ethereum are already on Arbitrum, with many moving over every day. Go to bridge dot arbitrum dot I O now to start bridging over your ETH and other tokens in order to experience DeFi and NFTs in the way it was always meant to be. Fast, cheap, secure, and friction-free. Living a bankless life requires taking control over your own private keys. And that's why so many in the bankless nation already have their Ledger hardware wallet. And brand new to the Ledger lineup of hardware wallets is the Ledger Nano S Plus, a huge upgrade to the world's most popular hardware wallet. With more memory and a larger screen, the Nano S Plus makes it easy to navigate and verify your transactions.

• 48:17David Hoffman?

  And the paired Ledger Live desktop app gets you increased transparency as to what is about to happen with your NFT. What you see is what you sign. The Nano S Plus gives you the smoothest possible user experience while you're doing all of your crypto things. So go to the Ledger website to check out the features of the new Ledger Nano S Plus and join the waitlist to get yours. And don't forget about the Crypto Life Card, also powered by Ledger. The CL Card is a crypto debit card that hooks right into the Ledger Live app, right next to all the DeFi apps and services that you're already used to doing, like swapping tokens and staking. So if you don't have a Ledger hardware wallet, go to ledger.com, grab a Ledger, and take control over your crypto. All right, Bank Station, we are back with John from Delphi. We're going to talk about the last of the three strategies to get Ethereum scaled computation without scaling

• 49:02David Hoffman?

  trustlessness and permissionlessness. And that comes to proposer-builder separation, which thankfully, it's actually baked into the title about what this is. But John, what's a proposer? What's a builder? And what are we separating?

• 49:17Speaker 01

  Yeah, so this is one of the key things that makes dank sharding possible from what the old sharding design was. But it was actually initially designed as like an MEV fighting strategy that like fights the centralizing forces of it. So the way that it works generally is today, you have, if you're like a validator, like after the merge or a miner today, you have to do two things. You have to like actually build the block. And you also have to say that like this thing is valid and sign off on it. The realization is that like those don't have to be like the same person who's doing those. And like, it makes a lot of sense to split those out. Because that person who builds the block today, it's like a mining

• 49:57Speaker 01

  pool operator. But like, the point is, it's a very difficult role to do that effectively. Because like, one of the biggest like advantages of being a block producer is MEV, like maximum extractable value. So ordering these transactions in a certain way that I can extract the most, like whether it's sandwich attacks, or like, whatever. The problem is, is that like, like, I don't know how to do that on my computer. Like, I'm not going to be able to effectively like, get MEV out of this block. So I'm just going to say, screw it. Like, give my money to someone else. Like, you validate for me. And then just like, it's fine. We don't want that to happen. We want it to be really easy that like a validator doesn't have to worry about MEV. So we need to just like, outsource that building task.

• 50:43David Hoffman?

  And just to put listener, the mind of the listener in the right spot, I think there's a lot of listeners here, a lot of viewers on YouTube, who are planning on staking their ETH. And hopefully, they plan on staking their ETH from the comfort of their own home with their own computer. And then they're probably hearing about like, how lucrative MEV is. And they're probably really excited to get some of that MEV when it comes to the merge. Problem is like, yo, listener, yo, YouTube viewer, do you actually know how to get MEV? Like, are you technical? Or, and because if you're not, then you're just going to leak MEV in favor of those who do know how to capture it. Uh, and so if you are thinking about like, oh, I can't wait to get my hands on some of those juicy MEV rewards. Uh, what you need is proposer builder separation. So John, I'll throw it back to you.

• 51:29Speaker 01

  Yeah, and exactly. So this won't be implemented until dank sharding, which is why at the merge, um, flashbots is going to come to the rescue again. Um, and basically create the, which is this thing, which is effectively out of protocol, proposer builder separation. They're going to introduce something called MEV boost, which is like a plugin to your consensus client, which basically replaces like your execution aspects of your client. And like, they will take the bids from searchers and builders who are like, they want to bid. It will like build this optimal block and it'll just hand it to your consensus client. Um, the problem with this is you now have trust in that like relayer.

• 52:12Speaker 01

  Um, like through MV boost who's sending you this thing that like, that all this is valid and like all of it checks out. And like, obviously you don't want to like have that trust and like all these different things. Um, so PBS will then actually bring all of that, like building and auction, like fully in protocol. And that's what, um, allows for dank sharding as opposed to the old design. Because in the old design, um, we had these 64 like individual separate shard blocks. So it wasn't like that hard to produce a single shard block. The big advantage of dank sharding, like the big difference between then and now is that we just put everything in one big block. It's not sharded in the sense that like we normally think of sharding where it's like these different blockchains that all have different data.

• 52:57Speaker 01

  Everything is put into one big block. The problem is that's really resource intensive, like you can't do that on like a consumer laptop or anything. So we realized that like, hey, we have this specialized role now that we designed because of MEV. Let's just take advantage of that and make the blocks bigger. Because like they're the ones who are building the box anyway. We don't need it to be like super easy to build the blocks. We just need to validate them. Um, so that's what it takes advantage of is now we have this higher resource builder who's going to build the block. The whole thing with like the beacon chain block, all of the shard data, like, put it all together. And there, all the builders are going to send like their block headers along with their bids. And all the proposer has to do is just take the block that has the highest bid,

• 53:41Speaker 01

  propose it to the network and people can take that block. And they'll be able to do data availability sampling and say, hey, if my part was available, like, and the execution on like the execution chain was valid, just sign off on it. So it makes that proposer role like super, super easy where that builder role is going to be a very high resource, like relatively specialized role.

• 54:02David Hoffman?

  Okay. And so going back to the theme of this particular episode, which is checks and balances. This is separating Ethereum block production into two separate groups of people. Where you have the consumers, consumer people, people that plan on staking their ETH, don't know how to capture MEV, they probably have like a Mac laptop or some like home-built PC that they just want to stake their ETH on, but they want to capture MEV. And so the proposer, the proposer builder separation is it separates this consumer group into the proposers. And then it can separates the builder group into the builders and the builders take all of the transactions and bundle it up into a, a, a block that's ready to go.

• 54:44David Hoffman?

  All then they already did all that computation and that block is ready to go. It just needs to be approved by the proposer who is staking their ETH. And then the proposer says like, okay, I will take this block and embed it into the blockchain. How does the proposer choose to select which builder's block to include?

• 55:04Speaker 01

  So they should take the one with the highest bid is what they should do. The, what you, you obviously want to make sure that they are like acting honestly. One of the issues with this is now you give the builder more power because they can censor transactions. Like maybe they bid the highest, but in like, they keep bidding the highest for every block. But like, they want to censor me because like, they don't like me or whatever. So the builders would never include my thing. That's where something called CR list comes in, censorship resistance list. So the proposer will put out like this list of these are all of the transactions I see in the memple.

• 55:47Speaker 01

  And so when the builder is like submitting their block and like it goes through, they have to either prove that the block was full. So I just couldn't include everyone. So you don't obviously don't have a guarantee in that scenario or that I included everyone in there. So it helps to like offset that power. But yeah, I mean, like all the proposer really has to do is just take the one with the highest bid.

• 56:10David Hoffman?

  And so it's outsourced all of the responsibility of collecting, capturing as much MEV as possible to this other group of people who have to compete with each other to bid the highest. What's the incentive for the builders to be builders? Can you just walk us through at the basic level? Like, why would a builder, what's the value in being a builder for the builder types of, for the builder role?

• 56:34Speaker 01

  Yeah, so they're the ones who will directly get the MEV and the transaction fees out of this block.

• 56:42Speaker 01

  So that gets paid to them, like searchers will bid to the builders, like, hey, I'll pay you this much to include my bundle. Another search will bid to them, I'll pay you this much to include my bundle. And the builder will get the transaction fees out of that block. And then that's why we want to have at minimum enough builders such that there's an efficient market. Because if there's an efficient market, every builder should build, should bid up to like the maximum value of whatever they can get out of that block. And then so indirectly, all of the value of that block then flows over to the validators. Because obviously if you have an inefficient market and there's just like one super builder, well, I could just, I could take all the money and then I could just bid whatever. But as long as you have an efficient market with like enough builders,

• 57:25Speaker 01

  then they should be able to bid up to the full value of what like that block is. So that all the, all the value flows indirectly to the validators, which keeps like super decentralized.

• 57:34David Hoffman?

  So once we get to this point in Ethereum's history where we have these builders who are building blocks, what, what group of people today does that correlate to? Like, who are, who are the, the people in the Ethereum ecosystem that will be likely candidates for being builders later on?

• 57:51Speaker 01

  So that is going to be like actually a new role because the person who's playing that role today is the mining pool operators.

• 57:59David Hoffman?

  Okay.

• 58:00Speaker 01

  Um, the actual miners themselves like don't actually like build the blocks or do anything. The mining pool operator, basically they're the one who does it. And then they send it to like the winning miner and like they put their proof of work and like prove that the block is good. Um, what's going to happen with this MEV boost, which is what will happen like initially after the merge. Um, which is like this flashbots program, flashbots is going to run, um, the builder and the relayer to start. Um, so they'll be the ones taking on that specialized role and then the searchers are the ones who just submit bundles. So they like, they run a specific strategy and just say include these transactions like a little piece. The building is like a separate process because it's not just figuring out like specific strategies.

• 58:44Speaker 01

  It's how to take all of the bids that you were given and like optimize like with some algorithm, like what's the best way to put them together? Um, so that's going to be actually like a pretty new role. Um, flashbots is going to be taking that initially.

• 58:56David Hoffman?

  Okay.

• 58:56Speaker 01

  And then looking to like kind of progressively decentralize that and allow like other builders in. And then when you get to full PBS, it'll be like totally permissionless, like anyone can do it. And like hopefully by then, especially because that may be boost will decentralize like enough people know how to do this, like there should be parties doing it at that point.

• 59:13David Hoffman?

  Okay. So the, the people that are running MEV bots today, are they the likely candidates to become searchers in the future?

• 59:20Speaker 01

  The people who are running MEV bots today? Yeah, like pretty much nothing's going to change for them.

• 59:24David Hoffman?

  Okay.

• 59:24Speaker 01

  If you're running, if you're like an MEV searcher today, right now you're just submitting, um, like to miners. It like after the merge, you'll be submitting through MEV boost like to their builders. And then like after a proposer builder separation, like you'll just be submitting your bundles to like the builders and that. Um, but all like MEV searchers, like they for the most part run like a single or like a couple optimized strategies of like, I'm just doing like arbitrage on this. Or I'm just doing sandwich attacks here. Like they're not able to build like what is all of the MEV out there most efficiently? They're running like two or three strategies. So the builder is going to be that new role of taking all of those things and like putting in them like into a block, which the operator is doing today.

• 1:00:07David Hoffman?

  Cool. So it's like a meta searcher, right? And so we have like, some searchers are doing, are making sure like the, there's a balance between balancer, uniswap and sushi swap. And so they are having, they're taking the best arbitrage opportunities there. And that is a highly competitive market that will be many, many searchers for. And then there's like liquidation searchers. So people that liquidate people on compound and AVE and other borrowing money markets. And those people have a highly complex, highly refined strategy and they compete in that realm. And then there's like 17 other different sources of MEV in the, in the ecosystem and every single source likely has its own searcher.

• 1:00:47David Hoffman?

  And the searchers produces bundles, which bundles up all the MEV that they can capture out of the, the MEV opportunities in Ethereum. And then the builders take all the searchers bundles and they bundle it up into, into one single block. And then that, then they propose that to the proposers and the proposers accept the highest bid, which likely comes from the most efficient collection of MEV from all the searchers, coming from the most competitive builder, going to the people that are just staking Ether. And so all of this work that goes on behind the scenes from all the searchers, collecting all the MEV and all the block builders competing to build the best blocks ultimately gets captured by the, the

• 1:01:29David Hoffman?

  single solo staker who's just staking ETH on their Ethereum node and they didn't have to worry about any of it. That's pretty cool. That's pretty cool.

• 1:01:38Speaker 01

  Yeah, it's, there, there's a lot of like steps to all these specialized parties, but like, that's the beauty of all this is it just completely abstracts all of that away. I as a proposer need to say, that's the highest bid. Like, good. Like, that's, that's pretty much it.

• 1:01:51David Hoffman?

  Simple. And something that Vitalik said in his end game, which you cited in your article is some central, uh, some centralization is needed to scale, which is like, you know, centralization, that's like a bad word in, in this industry. Uh, and additionally with this whole proposer builder separation, we're going to have people on their tiny little consumer laptops. And then we're going to have other people with their supercomputers that are competing against other people with supercomputers. And so we do have this stratification of like, we do kind of have this alt layer one super node type setup in Ethereum. But it's okay in this version because of the checks and balances by this proposer 4.0 builder separation. Would you say that's a fair take?

• 1:02:34Speaker 01

  Yeah, you still want the builder to not be like literally one guy at like Google who needs like this ridiculous computer to do it like you want it to be enough that it's decentralized enough that like you have an efficient market. And that you don't have to worry about like liveness concerns that like, oh, it's five builders in the same place and like some regulator is just going to go shut them down. Um, and the hardware requirements that we're talking about for these builders, like it's like running a GPU and like having like, like, I think it's like two and a half gigabits, like at least bandwidth, like Those are like, those are like high resource requirements for a regular person, but like, like just on your consumer laptop, you can't do, but like, they're not like these crazy requirements that like it's decentralized enough that like

• 1:03:17Speaker 01

  We shouldn't have to worry about with all these checks in place about like censorship or like liveness from them. And then we're just relying on the decentralized validators at the end of the day who like they're the ones who are like ultimately still going to like sign off on everything.

• 1:03:28David Hoffman?

  Would you say that it goes from a proposer you can do on your regular like consumer laptop and then a builder, you kind of need a high powered gaming computer? Is that a fair take?

• 1:03:38Speaker 01

  Yeah. Yeah. You would need like a GPU or like a really high end like CPU and like a very good internet connection.

• 1:03:45David Hoffman?

  Okay.

• 1:03:45Speaker 01

  And to be a proposer, like I could do it on my laptop, like, like it won't be crazy.

• 1:03:50David Hoffman?

  Well, okay. So those are the three things that that finishes off proposer builder separation, which I do think is like the most fun and interesting part of this conversation. Cause that's where we start to talk about who gets paid ether. Um, but I do want to like, kind of compare and contrast this to other L1 scaling strategies. And so at the highest of levels, how does Ethereum's like roadmap with all this stuff differ from the strategies of like the generalized alternative layer one? How, how would you classify that?

• 1:04:16Speaker 01

  Yeah. So there's a few different approaches. One approach is I would like simplify as like the sauna approach, which is For the most part, like you make some optimizations to the execution environment, like parallelism, et cetera. Um, but mostly you get like a lot of your scaling out of like entire hardware requirements, like you need higher hardware requirements to run a full node and that like lets you like process. More transactions downside of that, obviously, like I can't like check like I can't like run like a validating node or like be a full node of Solana like on my laptop like it's like it's not a reasonable constraint for me. And so that's trying to like, all right, let's jam everything through one chain. The other vision would be more along the lines of like Avalanche or Cosmos,

• 1:04:57Speaker 01

  which use like subnets and zones respectively, saying, okay, one chain is not enough. We'll split it across a bunch of chains. Um, problem with that obviously is a fragmented security and you probably end up with like pretty like small, like not super secure, like validator sets for a lot of these things. And you're still just like trusting the honest majority of them. Like you're not going to have people like running full nodes and validating like all these different subnets and zones. Like you're trusting the honest majority of them and you're fragmented security across all of them. And then there's like the modular vision approach, um, which is like Ethereum, Celestia, Polygon, Avail, where, hey, we're going to be this base layer. And then we're just going to let roll up slow on top of us and they can inherit like our security

• 1:05:40Speaker 01

  and run it like really low resource requirements for users who like just need to be like checking proofs, basically. And from that, like they will have like the full security and like the higher throughput all in this like interconnected network.

• 1:05:53David Hoffman?

  And talking about a line from one of your other reports and you alluded to this, but with a component that I want to lean into. Uh, you say many monolithic ecosystems such as Avalanche and Cosmos can achieve meaningful scale. However, this fragmented security approach inherits far less value capture back to the base asset. It will instead disproportionately accrue to subnets and zones respectively. Can you talk about the value capture part of this thing and how that fits into this conversation?

• 1:06:22Speaker 01

  Yeah, yeah. So that's a big difference of this is like, it's like effectively, when you look at either Avalanche or Cosmos, like they're not inheriting security from like the primary network or like the hub or anything. Um, so there's no reason to like pay like that main network. And so you don't like they run that like each zone and each subnet, like they run their own validators and they get paid the fees for that chain, like anything that gets. Paid or burnt or anything like that goes to them. Um, that's obviously not the case with rollups. Um, and that's where you get like a real staking yield and all of these things. Um, like out of them, not just like a made up like staking yield that like it's inflationary and you're giving out money. Um, like rollups will continue to pay back.

• 1:07:06Speaker 01

  Um, to the L one like fees. And at the same time, like Ethereum will continue to have like its native execution environment that also has like. That is ultra secure and that you're getting like settlement to the L one. Um, like everything gets paid back to it still.

• 1:07:22David Hoffman?

  Okay, John, this is a lot. This is super complicated. How long did it take you to actually go through and like understand all these things?

• 1:07:33Speaker 01

  Um, so I started working at Delphi two months ago. Um, and prior, so these were my first two reports. Um, prior to that, I got interested in crypto last year, um, like beginning of last year.

• 1:07:49Speaker 01

  Basically just started like buying Bitcoin when it was going up because I was like, this is going up. Uh, I got really interested in D five around like April, May, like just in, just in time to get wrecked by the crash there. Um, and then over like summer, it's like end of the year was actually when I found like you guys. And like, that's when I started to get like red pulled on like the whole like modular like vision.

• 1:08:10Speaker 01

  Like the first time that I ever first like understood like, okay, what is the modular approach? Like, how do these roles like actually scale? Um, was the like ultra scalable Ethereum article that you put out like end of October. And like, that's when I like had the like, oh shit moment when I, I remember when I got to the part of that where. Like you start to realize like how these modular things scale is they actually scale by getting more decentralized. Which is like a crazy realization because as you have more of these like decentralized validators saying the data is available. You can safely increase throughput even more. And then like you just keep adding more and then you can add more data availability throughput, which means more rollups, which means more transactions. And it's just like this virtuous cycle. Um.

• 1:08:51Speaker 01

  So yeah, I mean, like I first got into like looking at modular like six months ago and started full time like two months ago. So it's all, it's all out there on the internet.

• 1:09:00David Hoffman?

  Yeah. Can you just explain for the listeners, what's it, what it's like to, people talk, some people go down the rabbit hole pretty damn fast. But I would say you've gone down the rabbit hole faster than I've ever seen anyone go down the rabbit hole. Can you just talk a little bit about like what that's been like for the last few months for people who are like perhaps sitting on the edge of the rabbit hole? Thinking about like, do I dive in or not? And like, do I get a job in crypto? Like, how much do I commit myself to this? Can you talk a little bit about just like how you got versed in these things that are like, like super niche? Like, polynomials are one thing, but then we can get into like erasure coding and all that stuff like, I don't even understand this. Uh, and I, and from what I've gathered from you, you didn't understand it either more than like a couple of months ago. So like, can you explain for the listeners, like what that's been like?

• 1:09:46Speaker 01

  Yeah. Uh, like I would definitely encourage people to like, if you are really interested, like dive in. Um, like I was like slow on that for a while, like the first like eight months or whatever the longer that I was in crypto. Like I was spending like a little bit of time on it on the side, but I was like, like, I wasn't thinking like, I was probably going to go do this as a job kind of thing.

• 1:10:07Speaker 01

  Once I got into like the whole modular, like, like from like, from you guys, I then found Paulina. And that's when I just like, I went through like all of his stuff over like end of last year, like beginning of this year. And that's when I hit the point of like, okay, I'm going to go do this as a job, like full time. Like I enjoy doing this, like after my job in banking, like way too much. Like I need to just like spend a couple of months, like really learn all this stuff. Um, and then just like immediately start like go interviewing. Um, so like, that's what I ended up doing was like, so like, if you just like bust your ass on it, like, like, like literally everything is out there. And like, that's what I did. Like I like basically read like a ton of Paulina for like a couple of months. And then like I put out like a modular related post and then like started interviewing from there.

• 1:10:48Speaker 01

  And like it moves really quickly. Like that is one of the really cool things about like crypto that is like very, very different. Like no one really cares what your background is. Like if you're smart and like, you know, your shit, like you'll get hired like, and all of it is out there on the internet. Like it's, it like, it was a lot of fun.

• 1:11:03David Hoffman?

  Amazing. Well, John, fantastic work on this report. I'm gonna have to go listen back to this podcast again. I'm gonna have to read your report again to fully understand it. But again, for the listeners who brains feel a little bit broken, I'm with you, but all you have to understand is that there's a separation of powers that puts power and keeps power in the hands of the individual, the people that hold and stake ether at home. And so the cool thing about Ethereum is if you are just doing that, that is enough. John, thank you so much for diving into the very technical parts of the Ethereum roadmapping, explaining that to us here on the Bankless State of the Nation.

• 1:11:39Speaker 01

  It's fun.

• 1:11:41David Hoffman?

  Awesome, Bankless Nation. You guys know what to do next. There's a link, going to be links in the show notes, linking to John's report, both the one that we were talking about, the majority of this podcast, as well as his other report comparing Ethereum to other Layer Ones. As you all know, ETH is risky, crypto is risky, DeFi is risky. You can lose what you put in, but we are headed west. We're on the frontier. It's not for everyone, but we are glad you are with us on the Bankless journey. Thanks a lot. Hey, we hope you enjoyed the video. If you did, head over to Bankless HQ right now to develop your crypto investing skills and learn how to free yourself from banks and gain your financial independence. We recommend joining our daily newsletter, podcast, and community as a Bankless Premium subscriber to get the most out of your Bankless experience.

• 1:12:26David Hoffman?

  You'll get access to our market analysis, our alpha leaks, and exclusive content, and even the Bankless token for airdrops, raffles, and unlocks. If you're interested in crypto, the Bankless community is where you want to be. Click the link in the description to become a Bankless Premium subscriber today. Also, don't forget to subscribe to the channel for in-depth interviews with industry leaders, Ask Me Anythings, and weekly rollups, where we summarize the week in crypto and other fantastic content. Thanks, everyone, for watching and being