Bankless - Celestia’s Building the Multi-Chain Universe with Nick White | Alpha Leak

Episode Date: March 23, 2022

Nick White is the COO of Celestia Labs. Celestia is the first modular blockchain network. If you have no idea what that is, you’re in luck! Celestia wants it to be as easy to deploy blockchains as it is smart contracts. Nick and David cover all things modular blockchains, the network effect that happens as a result of this multi-chain universe, what the far away distant future might look like, and so much more! ------ 📣 ZERION | Trade Across 7 Networks and 500+ protocols https://bankless.cc/Zerion  ------ 🚀 SUBSCRIBE TO NEWSLETTER: https://newsletter.banklesshq.com/  🎙️ SUBSCRIBE TO PODCAST: http://podcast.banklesshq.com/  ------ BANKLESS SPONSOR TOOLS: 👀 POLYGON | LAYER 2 DEFI https://bankless.cc/Polygon  ❎ ACROSS | BRIDGE TO LAYER 2 https://bankless.cc/Across  🦊 METAMASK | THE CRYPTO WALLET https://bankless.cc/metamask  💳 LEDGER | THE CRYPTO LIFE CARD https://bankless.cc/Ledger  🧙‍♂️ ALCHEMIX | SELF REPAYING LOANS https://bankless.cc/Alchemix  🦄 UNISWAP | DECENTRALIZED FUNDING https://bankless.cc/UniGrants  ------ Topics Covered: 0:00 Intro 4:00 What is Celestia 5:30 Types of Blockchains 9:30 How Celestia Modularizes 16:15 Rule Checking 19:40 How Rollups Work 21:10 Different Kinds of Layers 27:31 Prices & Data Availability Sampling 40:00 Celestia’s Blockspace 42:01 Fees & Nodes 45:35 Goals of Celestia 51:43 Celestia Vampire Attack?! 53:20 The Future of Modular Blockchains ------ Resources: Nick’s Twitter https://twitter.com/nickwh8te  Celestia’s Website https://t.co/oaUV9NvYwH  Celestia’s Twitter https://twitter.com/CelestiaOrg  John Adler Delphi Podcast https://podcasts.apple.com/au/podcast/celestia-the-worlds-first-modular-blockchain-network/id1438148082?i=1000552936346  Celestia’s Discord & Telegram https://discord.com/invite/YsnTPcSfWQ  https://t.me/CelestiaCommunity  ----- 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 

What Celestia enables, again, is experiment more experimentation. Like, it'll be, it'll increase the speed at which people can spin up new blockchains and try new things and experiment, which I think will overall increase the amount of innovation in the space. Hey, Bankless Nation, welcome to this edition of Alpha Leak. Alpha Link is actually just Meet the Nation, rebranded into a little bit more of an exciting name. So this is where we'd go and dive deep into various projects that catch my attention around in the Cryptosphere. Today we are talking to Nick White from Celestia, and Celestia is a layer one blockchain that's living inside of the modular blockchain paradigm.

So Celestia and all the people behind it, I think, are just very committed to this whole modular blockchain thesis, which both me and Ryan are definitely aligned on. We think the long-term conclusion of this cryptocurrency space, however it may be, will be modular, Ethereum or a different chain, like perhaps Celestia. So let it be known that we are not ETH Maxis. We're actually modular blockchain maxis, which is why we're talking to Nick from the Celestia team here today. I've already kind of given the pitch for Celestia.

It's even more modular than Ethereum is in its current shape. And so Nick does a fantastic job walking us all the way through the design architecture of Celestia, its differences and similarities to Ethereum, what it's optimized for, how it's going to do cross-chain, multi-chain stuff, how consensus works, all of that deals. So it's relatively technical of a show.

But if you are here to learn about Celestia, this is going to be the perfect show for you. And so let's go ahead and get right into this Alpha League episode with Nick White of Celestia. Right after we talk about some of these fantastic sponsors that make the show possible. 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 NanoS Plus, a huge upgrade to the world's most popular hardware wallet. With more memory and a larger screen, the NanoSplus makes it easy to navigate and verify your transactions, 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 NanoSplus 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 NanoS Plus and join the wait list to get yours. And don't forget about the Crypto Life card, also powered by Leisure.

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. If you're going bankless, you need Metamask. This is your tool to unlock the world of Defi without giving up custody over your private keys. Metamask is both a secure in-browser wallet and also a secure bridge for your hardware wallet.

You can now trade tokens on any decks or aggregator. Metamask swap gathers real-time pricing information across all the defy exchanges, allowing you to select your best price while getting all the Metamask benefits of self-custody, lower gas costs, and increased transaction success rates. Metamask also has a fantastic mobile wallet that I use when I'm out and about, which I used to collect PO apps, NFTs, and do all my defy things while I'm away from home. If you haven't downloaded Metamask, you got to try it out. Web3 wouldn't be the same without it. Download Metamask for desktop and mobile at Metamask.io and load up your Treasurer, ledger, lattice, or Keystone hardware wallets so that they too can get into the world of Web 3. Polygon is Ethereum's largest

and most vibrant scaling solution to date. With millions of monthly users and all of the biggest defy-apps, the Polygon ecosystem has turned into a blossoming metropolis of defy-activity. Transactions on Polygon are quick and cheap, allowing users the freedom to achieve their defy-goals, all while being economically

anchored to Ethereum. But Polygon isn't just the proof-of-stakesychein. The Polygon team is building a suite of scaling solutions, including Polygon Hermes, Midin, Nightfall, and Zero, all with different design choices in order to be optimized for all possible crypto use cases. If you're a developer who wants to build on the Polygon ecosystem, go to the link in the show notes to check out their fantastic documentation. And if you're a user who just wants to experience fast and cheap defy-de-fi, you can bridge over your eth or other tokens and start playing around with any of the

thousands of applications that are available on Polygon. All right, Bankless Nation, I'm here with Nick White, who is the C-O-O of Celestia Lab. Nick, welcome to the show. I'm excited to get into everything about Celestia today. Thanks, David. Excited to be here. So let's go ahead and dive right in. I think bankless listeners will already be familiar with the modular blockchain thesis. And so this is kind of where that conversation starts. Nick, what is Celestia at the highest level? Celestia is the first modular blockchain network. So it kind of was the origin of this modular blockchain paradigm in which you separate the two core functions of what typical monolithic blockchains, that's Bitcoin, Ethereum,

well, the original Ethereum, and Solon and all the other L1s do, which is consensus and execution. So basically, Celestia's core insight is that you can split those two functions apart into separate layers. And so Celestia focuses on the consensus part. So it's the consensus and data availability layer. So data availability is another topic that we can dive into, but basically it's another core primitive that is crucial for building, you know, decentralized applications. Okay, so I think we should probably start with like definitions and mental models and just to allow the listeners to get up to speed, splitting the blockchain between execution and consensus. Let's dive into each one of those and we'll take them apart separately.

Consensus, what are we talking about? This is proof of work, proofs of stake. What are we talking about with the consensus side of a blockchain? Yeah, so consensus is basically the mechanism by which you, you, all the participants in the network agree on the order of transactions. So when like a block, for example, has consensus on it, that means that it's now placed in a certain order among all the other blocks. So once a block has consensus, in theory it should never be reorganized.

Like it should always be in that place, like that block height, essentially. And all the transactions in that block are now ordered. So with that, like once you have consensus, everyone network has the same view of reality. Everyone has the same view of history of events. So that's how you prevent things like double spend. That's how you know basically what has happened. One single shared source of truth that doesn't have the blockchain.

It's not forked. There are two competing versions of the truth. There's everyone come to terms on the same protocol, the same data, and everyone has strong assurances that they are on the same page as everyone else. exactly everyone's looking at the same thing everyone's in agreement like this is the reality this is the canonical chain cool all right okay so that's consensus and then execution let's get if you're just divine execution for the listeners yeah so execution is perhaps the more interesting part of it right execution is where those transactions that that history gets interpreted right so there's there's

something that you call a state transition function which basically defines you know this is like how you interpret a transaction and this is like how what the effect of that transaction is. So for example, in Bitcoin, you know, the state machine is pretty simple. It's just basically like, okay, well, this, is this person, this, that's signing this actually have the right to spend this UTXO. And if so, then like we'll move the balance to this new address essentially. But in, you know, Ethereum obviously got way more interesting where we have a full, you know, Turing complete sort of like VM as the state transition function. So that's like how applications are born is basically like, okay, now this transaction interacts

with a smart contract and executes this trade or, you know, it votes on this like governance proposal or whatever it is basically. Execution is where, you know, the activity is. Like without execution, there's really no point. You just basically have this ledger, but there's no like, you know, effect of that ledger. So execution is like applications, smart contracts, and all the stuff that end users end up interacting with. Correct me if I'm wrong, but execution, another way to explain execution is just like, execution is what goes from like block one to block one plus or block N as in any number to N plus one, right? It's the plus, it's the plus side of that part because we have to actually make changes.

And when we say we're making changes to those ledgers, we actually have to execute changes. and kind of like how you said with Bitcoin, the math for Bitcoin is very simple. It's just basically checks the balance of the sending address, and then if the balance is larger than the amount that's being sent, then it says yes, execute that. And then, like you said with Ethereum, we're getting into code. The VM is a virtual machine.

But basically, it's just computation at this point, and it's just like a bunch of smart contracts or just a bunch of rules to execute upon, and that's, and if you follow the rules and it gets verified by all the notes, node operators, you're allowed to go from block N to N plus one. And that's execution. And so previously, and mostly still actually, we're in this paradigm where both consensus

and execution are done by the same parties. And I think the modular blockchain thesis is that we actually can separate out those two parties into two different groups and allow these resources to be managed separately. How does, where does the conversation with Celestia begin? How does Celestia actually modularize these things? Yeah, that's a really good question. But I want to, I guess, double tap into this concept of separating, like, the execution and the state being updated. Because in a typical blockchain, what happens is every time there's a new block, right, it doesn't just contain the transactions in that block.

There's also what you call a state route, which is basically a commitment to these transactions that I'm including. This is the new state that they create. And so both of those things, every time a new block is mined in a monolithic chain, both the transactions and the new state are included in that block. So in Celestia, like basically all it is, well, the original paper that kind of inspired the project was called Lazy Ledger. And the reason it was called Lazy Ledger is because Celesteia is a lazy blockchain. So it does less. It actually doesn't even interpret, it doesn't even look at the transaction data that people are dumping into it. it just mercilizes it and it has a special kind of scheme of building the block that enables

something called data availability sampling, which makes it really, really scalable. Because one of the reasons that you, there are a few different reasons why you want to separate consensus from execution. So the first one is that, you know, in a blockchain where it's monolithic, there is only really one state transition function that you can implement. So you can either, you know, stuck with Bitcoin. Like you can't run the EVM on the Bitcoin blockchain, right? But also, like, you know, on Ethereum, you can only really interact with Ethereum through the EVM, right?

Ethereum blockchain doesn't run other forms of execution. So by separating those two things, you can actually have a blockchain where the execution, like you can define different kinds of execution environments. So you're not just locked into the EVM or, let's say, Salana C level. Like you can define that after the fact on layer two. So that's one of the big ones. The other big one is that you get more scalability. And that's kind of where this whole data availability problem comes in.

Okay, yeah. So we'll take those one at a time. We'll get into the data availability part later. I want to go back to the execution side. So the EVM is Ethereum's execution environment, right? Yeah. So like that's the execution environment is like, this is where the execution happens.

all the transactions are contained inside the EVM. And what you're saying is that Celestia allows that part to be swapped in or out for any other execution environment. Would it be fair to say that every single blockchain must, it has to have an execution environment in it somewhere, like it has to have an execution. And so Celestia allows you to pick and choose what your execution environment is. Is this, am I using the right words? Yeah, exactly.

So, I mean, you could even think of, you know, I mean, I guess I wouldn't call, I wouldn't called Bitcoin's state transition function necessarily an execution environment because that execution environment kind of implies that there's sort of like computation sort of like rich stateful kind of like like the EVM like a turn-complete like programming environment but yeah basically the execution environment is exactly what you said and so yeah Celesteia frees up developers it makes it more flexible right because now let's say either it could be as simple as like actually I do want to build in the EVM, but I actually want to add in a new pre-compile or I want to remove this one function, like for example, like self-destruct, which can cause a lot of problems and make things

more difficult. You could actually use the EVM, but just, you know, make some small modifications for your application. So it could be better for DFIR or NFTs. Or you can just completely experiment. Like I actually believe in the Celestia team believes that innovation in execution environments has been held back in blockchain artificially because of this sort of monolithic paradigm. The fact that if I wanted to build a new experiment with a new execution environment, I had to build an entirely new blockchain, right? And that the bootstrapping costs of doing that is so high that it's kind of prevented people from trying things out.

So in the same way, like before Ethereum, there's actually a lot of parallels here. So before Ethereum, you know, Bitcoin proved that you could build a decentralized application for the first time. So people are like, oh, this is so cool, we could do lots of stuff. But because there was no general purpose blockchain, every time you wanted to build a new decentralized application, you had to spin up an entirely new blockchain. But then when Ethereum came along, I said, hey, we're going to build a general purpose blockchain on which you can build a decentralized application without building a new blockchain. So Celestia takes that one step further and makes it so that you can do the same thing, but for the execution environment itself. So I can build a new blockchain for all intents and purposes without actually.

building a new consensus network. So that's really like the key part there. Okay, I really like that. That tells a great story. So first we have Bitcoin and we have the first decentralized application on the internet. And then like, oh, cool. We have the, it's like on the spectrum of dimensions, we're at like 1D, right? Like we're a dot.

It's like we have like this one thing. Like hello world. Which is said, yeah, hello world. We can send bitcoins around. That's all we got. And then Ethereum was like, all right, well, we can either. make more blockchains and make more dots or we can or we can have a a grow in a whole entire different dimension and just allow any blockchain to have any application and that is the what

and so we added a an evm to bitcoin right and created ethereum right we added an execution environment to bitcoin and then instead of having to make many many many dots we go from a dot to a line and so what you're saying with celesia is that you buy but then but then ethereum's is the EVM. So where Bitcoin's constraint was that it can only move the Bitcoin's around. Ethereum is constrained by it can only do stuff in Defi inside of the EVM. And then Celestia is saying like, okay, we can go from a line to a three-dimensional shape, 3D, and by saying like, well, we can remove the constraint of the EVM and allow for

any execution environment to come and exist on Celestia. Is this right? Exactly. Yeah, exactly. It's almost like, you know, the EVM only speaks one language, and then a modular chain like Celestea is multilingual. Like you can load any language and speak any language that you choose. I love that before.

So how, you said, okay, you said load any language. My next question was, so Ethereum uses the EVM because, like, that's just the language that it uses, and it forces, it checks all of the rules of the blocks using the EVM. Like the EVM is the police. It's the enforcement of the chain, checks all the rules. How does, if Celessia is completely modular with this execution environment, how does it know what rules to check? That's a really, really good question.

And that is where layer two constructions like roll-ups come in. So basically what roll-ups are and what L-2s are is a way in which you enforce. So like, let's take a step back. Like whenever in a monolithic chain, whenever, transactions or a new block or mined, they are made sure that they're valid first and foremost by the miners or the validators, like the proof of stake validators, right? So that's one way. In theory, everyone wants a full node and they re-verify for themselves, but also the miners are the ones enforcing the validity, right? Now, in Celestia, anyone can submit a transaction. And the transaction

doesn't even need to be valid because in Celestia's eyes, there's no such thing as a valid or invalid transaction, it's all just data. But then that, of course, leads to your question. So how do you determine what a valid transaction is or who is enforcing the execution, the validity of the state transitions that are happening on Celestia? And so the answer to that is L2 construction. So roll-ups are a really new and interesting technology that basically make it possible for someone to trust, not without having to re-execute,

transactions or re-execute a block to know that that block is valid. So in short, it's either a validity proof, which is where zero knowledge roll-ups sort of how they enforce validity, or it's a fraud proof, which is basically how optimistic roll-ups enforce the validity of blocks and transactions. So when we say that Celestia is modular in terms of its like execution, like you can put any execution environment on top of Celestia, are we really talking about, like, the layer two, are the actual modular component of the execution environment, as in like this layer two uses the EVM, and this layer two uses Starkware's Cairo or whatever.

And so that's the actual part that is modular. Yeah, exactly. So you can kind of think, you know, what we're doing is we're decomposing the typical monolithic blockchain into components. So Celestia is the sort of base level component, sort of like the L1 component, which is consensus, which orders the data, and data availability.

And then the other component, the execution layer component, is the roll-ups. So when I say any execution, you can run any execution environment, what I mean is that there's no, like the roll-ups themselves define their own execution that they run on top of Celestia. Okay. And then when you said that Celestia doesn't, isn't aware of anything like an invalid or invalid transaction. That's because it just is a place for the data of the roll-up to be deposited into, like the data, Celestia just receives the data from the roll-up.

Yeah, exactly. So maybe we should go a little bit deeper into how roll-ups work. So basically, a roll-up is, for most intensive purposes, just, like, actually a blockchain. So it's almost analogous to a normal blockchain where instead of having a validator set, and instead of like doing their own consensus, there are roll up operators. And they, so they collect transactions from all the users. They, they put, roll them up into a block. They basically create a block and they, they, they compute what the next state is. So they do the like the validation of that block. And then they post that block onto another blockchain that is actually doing the consensus and also data availability. So, um, I guess like the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the,

roll up data is is itself like its own block within a block. Does that make sense? And so, but Celestia just allows anyone to post data there. So we don't, you know, we don't, we're not aware. I mean, we're aware of the fact that there, the people are putting data there, but we don't actually interpret it or look at it. We don't say like, okay, this, you know, in a monolithic chain, if I, if I try to send an invalid transaction to Ethereum, it never ends up on the chain because it's, it's the miners discarded. They're like, oh, well, you don't have the right to, you know, withdraw this money or to spend this money or whatever.

So it never goes into the chain. What I'm saying is in the Celestius case, there is no interpretation, there is no valid or invalid. So any, all transactions get posted there. So how do all the roll-ups then have shared state and assets, right? So like with Ethereum, because you do have like the invalid or valid transactions, right? Like it also has like native tokens, the eth, native tokens, native tokens on the layer one and it goes from one execution environment on a roll-up down to the

L-1 and then across the L-1 and then up to a different roll-up. How can you, how does Celestia manage that if it doesn't have any rules at the L-1? How do you have shared state between all the roll-ups? Really, really good question. So actually, so this gets into, so we already talked about the consensus and data availability layer, which is, let's say, Celestia, and the execution layer, which is a roll-up. There's also another component that I think will be very important. in the modular blockchain stack, and that is what we call a settlement layer.

So in the model of Ethereum, an Ethereum-based roll-up, Ethereum acts as the consensus layer, the data availability layer, and a settlement layer, and then the Ethereum roll-ups act as execution layers. And so to explain a bit more about what a settlement layer is, it's basically a coordination mechanism between different roll-ups. So like the, it's like a bridge essentially. So the, when you, when you deposit an asset from Ethereum into a rollup and you're bonding it into like a contract at the L1. And then that contract listens for the state updates from the roll up.

But I guess like the good, the good thing about a settlement layer is that if it's kind of like in a multi-chain world, right, like let's talk about like Cosmos as an example. There is no, you know, people talked about the. Cosmos Hub being sort of the central spoke, right, in which all liquidity can pass. So rather than having all every chain, talk to every other chain in a very, very, like, you know, chaotic kind of like topology, there would be like this ordered system where there's like one canonical spot that everyone kind of relays to, to get between each other. So that's kind of what a settlement layer is. It's sort of like this bridge.

It also, in a sense, custody's assets. And it helps to make it easier for those roll-up. to talk to each other, but you don't actually need, like, the settlement layer is not fundamental to the security of these roll-ups. In fact, you know, these, like, roll-ups can actually have bridges with each other directly. They don't have to go through, for example, Ethereum as a settlement layer. It's like a nice to have, if that makes sense. Okay, okay. And so the plan for, like, an asset on Roll-up A to go to Roll-Up B, if both of these roll-ups settle on Celestia is to, it's up to their own,

it's up to those two roll-ups to establish communication between them rather than having to rely on Celestia as the settlement layer? Yeah, basically. So, well, I think the way to think about it is that Celestia doesn't enshrine a settlement layer.

So like an Ethereum, I mean, it's debatable if you would call like the Ethereum, like L1, EVM, an enshrined settlement layer. Because roll-ups on Ethereum can also just connect to each other. on their own independently of that. But in Celestia, there is no enshrined settlement layer,

but you can build a settlement layer on top of Celestia as a roll-up. So like settlement, a settlement layer is basically just a special case of an execution layer. It's an execution layer that's just, it's not really meant to do any fancy kind of execution. It's just meant to verify state transitions of other roll-ups and kind of like, you know, facilitate the movement of funds between places. What would you say is the advantage of that design structure? The advantage is basically that to me, you kind of have, well, there's a few. So one is, as we talked about, like, it adds order.

Like we imagine a world where, you know, it's not just, you know, hundreds of blockchains, not just thousands, but in the future, long-term future, millions of blockchains, right? And when you have that level of complexity and emergence, you're going to need some more, some organization. And so I think settlement layers act as like a way in which, like, coordination mechanisms between all these different state spaces so that they can easily route funds to each other. Like rather than, you know, me having to create, you know, run like a hundred clients to all the different other chains that I want to talk to, I just run one client to the settlement layer. and every other chain that is connected to that settlement layer is also just one hop away.

So it's almost like a kind of like a, I don't know, like a network kind of thing. But like you you're able to like the topography of the network gets, it's more organized. The other thing that's very important is that fungibility in a multi-chain world becomes really challenging because depending, like the fungibility of an asset is path dependent. So it depends on how you bridged. like let's say I took a USDA from chain A and bridge it to chain B and then to chain C

that USC is actually different than if I took it from chain A to chain D and then to chain C. So like, but if you routed it through a common settlement layer, that USC then can, if it originates at the same settlement layer, then it becomes fungible across that sort of like universe. So I think that's a really important feature for settlement. And also actually the last point about why settlement's important is that if I am running a client,

the settlement layer verifies the state transitions of all these other roll-ups. So if I'm a user and I don't, like, I want to have, I want to be trust minimized. I want to like make sure that all the state of all the roll-ups that I'm using are is legit. Like no one's like pulling a fast one on me, which is key to decentralization. then I, instead of having to run 100 clients of every different roll-up, I could just run a client of that one settlement layer, and that settlement layer will tell me the state of all these other roll-ups. So it also kind of reduces the amount of overhead that I need to verify other chains.

So this is all talking about the advantages of having just like a single shared source of truth, right? Like this is like, you have this one place where everyone can look at me like, all right, here's where the, here's what the state is. but I'm having a hard time differentiating that between like that and the Ethereum L1, because isn't that also what the Ethereum L1 does? Yeah, so I mean, the Ethereum L1, I think in this model is a settlement layer in a sense. But the thing is, so why, okay, so I think maybe you're touching on two things. Why not have settlement on the L1 itself? Well, one of the main reasons is that the settlement layer is this, it's, it, it,

actually reduces, it makes it harder to make that layer one trust minimized. So like in Celestia, we have made the L1, there is execution on the L1, but it is minimal, minimal, minimal. It is like Bitcoin. It's basically just sending and receiving funds and staking them. That's it. And the reason that we did that is that it incurs, like, okay, so if I'm a roll up and I want to use an L1 to post my data there, I actually have to take an interest in the, you know,

the validity of that chain for my own security. So with Celestia with minimal execution, it incurs minimal overhead on the roll-ups that use it, if that makes sense. So that's, I think, a really key advantage. So that's why we have chosen to remove as much execution as possible from L1. And then, there was another point on this that I wanted to talk about, but I'm kind of blanking on what it was. Maybe it'll come back to me. Yeah, it probably probably will. When you said it'll the roll up will have the minimum overhead when it posts to Celestia because Celestia has reduced,

it minimized the amount of execution it needs to do at the L1. Is this related to like gas fees? Because I know like, I know that the Ethereum of gas fees actually does impact a layer two gas fees. And a large part of that is because there actually is computation that happens at the L1. That's why we have, that's why

I have high gas prices. Is that the same conversation? That's part of it. Although if you, you know, I know that like, for example, in the new Ethereum roadmap, as I understand, they're going to actually separate fees for data availability space and execution on like gas, basically, on that one. So it's possible to do that. But you're right. But it doesn't, and that's why Ethereum is doing that.

They don't want to conflate these two resources. So this is like this topic of resource pricing, which John Adler gave a talk on at ECC last year, which basically is like, you know, and this is part of the, the modules of blockchain paradigm is the like pricing there's different resources that need to be consumed to build a decentralized application one resource is execution like you know i want i want to be in the chain of this order like i want to have priority right of my transaction that's a that's a big one and then another one is just straight up block space like i want to be i want i need block space i need data i need data to be made available so those are two different kind of resources and in celestia

It's purely data availability. That is the resource basically that we are selling or providing as a service. And then the layer two's, the rollups, we'll have probably have their own token or their own fee model, which is all about execution, which is about like priority in this queue to like, I want to be execute. I want my transaction to get executed,

if that makes sense. But, yeah, we can go deeper into that. Okay, yeah. So if it's an L1 that doesn't, doesn't have an execution layer baked into it in the same way that Ethereum does. Does it have its own native token then? It does.

Yeah. So it's a proof of stake network. And so we need that for the crypto economic security of, you know, ordering, guaranteeing that the blocks will not be reorganized, they won't be reverted, which is critical to all the L2's building on top. So it's critical for that. And then the token is also used in the fee market. So there's, you know, there's a scarce resource, which is block space. So like, and every roll-up will basically, instead of an Ethereum where you pay gas to like, you know, have space on the Ethereum computer, you just pay, you pay in the Celestia token to have space in the Celestia block, right?

Because like, I want my block for my roll-up to be included. Well, I have to pay. And it basically will be a fee market apportioned by the like the number of bytes of. of block space that you want to take up. Okay. And so that doesn't seem all that different from Ethereum in the sense that there's just like there's demand for block space when demand goes up, prices go up. Yeah. Well, so that there's a key difference.

And this is maybe a good, good time to transition into data availability sampling. So one of the key reasons why you want to remove execution from L1 is that one of the one of the bottlenecks for scaling any chain is that. the full nodes of the network have to re-verify or like recompute every transaction. They have to re-compute every state transition and make sure that it's legit before they accept a block. So what that creates is this dynamic where you can only be as fast, you only do as many transactions as the slowest person, right? The slowest full node, right? So all of a sudden you can only, you reach this maximum amount of scaling and you can't go past that unless you basically do what Salon did and says. hey, everyone just has to upgrade their machine.

But then you run into centralization because then all of a sudden, hey, I just have a laptop, all of a sudden I can't run a full node. I'm no longer like a first class citizen of this network. Now I have to delegate my trust to, I don't know, some someone with more money than me, basically. So we want to avoid that situation. So when you remove execution from the layer one, all of a sudden you remove that constraint because what you can do is now, instead of re-executing transactions, you're just verifying that the block data is available.

And it's a very nuanced thing. But basically for all L2 security, you need to make sure that the transactions were published to the network. That someone out there, if they wanted to, could download them at that point in time. So, and you can verify that. Normally, you would have to download the whole block yourself to make sure that all the data is actually there.

but with these data availability sampling schemes, you can verify that a massive block was available by only downloading a very, very small subset of the entire data. So all of a sudden, you remove this constraint where, like, you know, it used to be at every node and the network had to download the full block. Now every node in the network just has to sample, you know, maybe a few kilobytes of data,

and all of a sudden they have security that the whole block data is available. So it increases, like, orders of magnitude, the amount of, like, the size of the blocks that you can create in the network, still in a trust-minimized way. And this sounds a lot like sharding. It is, it is like sharding, actually.

It's a very good point. So in sharding, you try, you build a system where there's a bunch of different blockchains running in parallel, but they're all sort of secured by, like, this sort of like shuffling a validator. So like all the, like the security is uniform. So and the point is that each node,

rather than having to verify the whole network in every single chain, they just verify their one chain. So it is similar to that. But in sharding, you A, the, like you can corrupt each like sub shard committee, which is like I think very,

very feasible attack that you can do. The other thing is that you're still monolithic in the sense that each shard is still executing, uh, every transaction. So in Celestia, it is similar in that, like, you reduce the resource requirements for the nodes and the network. But it is actually quite different in the end. You do end up, like, you can think of roll-ups in a modular blockchain as sort of like the shards in a sharded model, except the difference is, again, in a sharded blockchain, you can only pretty much run that one execution environment.

And you can't just spin up shards or spin down shards. and in a modular blockchain, you can have, there is no limit to the number of shards that you can have, you can have any kind of execution that's like way, way, way more flexible. The layer two era is upon us. Ethereum's layer two ecosystem is growing every day and we need L2 bridges to be fast and efficient in order to live a layer two life. A cross is the fastest and 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 back to the layer one.

assets are bridged and available for use almost instantaneously. Across's bridges are powered by UMA's optimistic Oracle to securely transfer tokens from layer 2 back to Ethereum. Across is critical ecosystem infrastructure and ownership is being handed over to the community. You can be a part of this story of Across 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.tto to bridge your assets between ETH, Optimism, Arbitrum, or boba networks. Bankless is proud to be sponsored by Uniswap. Uniswap is a new paradigm in asset exchange infrastructure. Instead of a cumbersome order book system where trades are matched with other humans, uniswap is an autonomous piece of software on Ethereum that lets you trade any token at the

current market price. No human counterparties or centralized intermediaries, just autonomous code on Ethereum. Input the token you want to sell and receive the token you want to buy. The Uniswop Grants Program is accepting applications for grants. Do you have something of value that you think you want to contribute to the Uniswap ecosystem? No matter how big or small your idea is, you can apply for a unique grant at Uniswopgrant.org and help steer Uniswap in the direction that you think it should go. Thank you, Uniswap, for sponsoring bankless. So, you've got some money, but how are you going to use it?

You want to spend. You, me shopping now, bro. When you know you should be saving. You'll never buy a house at this rate. But what if you could spend and save at the same? time for the enlightened kind with inquiring minds a new world awaits set yourself free with completely flexible self-repaying loan technology supported on desktop and mobile seize the power of Alchemics allowing you to spend and save

at the same time leverage your wealth without risk of liquidation take out a loan that repays itself by using yield from your deposit to pay off your balance Your only debt is time. What was once inconceivable is now within your grasp. Are you winning some? I think perhaps the more correct word was parallelization. With Celestia, what you guys are paralyzing is data availability by not having to have every single node download every single block,

but just some of the data. And because there's many, many nodes, and they're all doing the same sort of thing, there's sufficient overlap and the data that's being downloaded so you have redundancy and assurances that all the data is actually there. And the way that you have these assurances is through sampling. You sample like random bits of data, make sure it's all there. If it's all there, you have assurances that all the rest of the data is there.

But all that data is actually split among many, many, many, many, many nodes, allowing lighter work for each individual node, but a heavier total summation of all the nodes altogether. Yeah, absolutely. I mean, we can go into the exact mechanism if you want. It's pretty technical, but you can think of it as like strength and numbers. It's like everyone's pitching in, everyone's sampling. Yeah, exactly. And then by that sort of like group effort, everyone becomes secure. Okay, so does the Celessia blockchain, in terms of like total available block space, does it have, like, is all these innovations that we're talking about, especially with data availability sampling, does it give Celestia the ability to have more block space? Because, like, Ethereum's biggest constraint is that it's block space, L and the L1 is really, really limited.

I would imagine, like, Celessia doesn't have, like, it's not like the Salonah-type level of scale, where it's basically, like, infinite level of block space, but does it get tailwinds or just, like, perks from this model to have more available block space at the L-1? That's a really good question. In fact, this is why I think I forgot to like follow up, but we were talking about fees, right? And you were saying that in Ethereum, the more demand there is for blockchain space, the more that fees go up.

And so the really key thing about data availability sampling is that there is no limit. Well, okay, there is probably a theoretical limit. But for all intents and purposes, for the current scale of like blockchain utilization, like there is almost, there is no limit to the box size. Like you can keep increasing it with the number of nodes in the network. So more nodes sampling means you can safely increase the block size. So and if you assume that the more people that are using the network, the more nodes there are in the network, that means that the more users there are, the bigger the block size gets. So as there's more demand, the block size increases. So the fees actually stay constant.

So like there's an aggregate more demand for the total. being spent to pay for that block space, but the fee per user or per byte of data to use the block is the same. So it's like a fundamentally new sort of paradigm in terms of like the fees. Every monolithic chain, as more users come on inherently because of the design, the fees have to increase. Yeah, but we can talk about the limits of scaling or sort of the scalability properties if that's interesting. Yeah. the cool thing is that when more users are paying more fees to the chain, the block size increases at some function related to the level of new fees being paid. So the fees, is it hard-coded

in that like the user just always pays the same exact fee or is it like flexible like Ethereum's gas markets? Yeah, so I want to be clear. So like it is just a fee market. And they will have to actually decide to. to increase the block size. So like, it's over long terms. It's helpful. Yeah, over long terms. So it's not like, oh, well, from one block to the next,

it's gonna scale up or scale down or whatever. It's more like, okay, well, now we're realizing we have more users and most of the block space is filled. So now we're gonna increase the block size. So that's kind of the way to work. And you have the capacity to increase the block size because you have extra fees being collected by the use. So you have extra incentives for people to spend up more

nodes? No, so it's a little bit different. So there's the validator nodes who are the ones who are earning block rewards. But then they're the, and they are also sampling and downloading the data, so they're securing the network. But there's also just the end users who are sampling because they're running what we call a light node. So the more people using the network, the more sampling is going on, and therefore the bigger block size you can get. So the fee, the fee part doesn't actually incentivize the increase in in block size. It doesn't increase the number of people sampling per se. Okay.

Okay. Then what is the incentive to spin up more nodes to increase the block size? So there's a few different things. So first of all, the more roll-ups that are being run, they're running nodes, right? And they're sampling and they're doing everything. And then all the clients of those roll-ups, if they want to be trust-mini-ins. So we assume that users of chains want to have full security and be trust minimized, which I think they do.

That's the world I want to live in is where I'm not just like pinging in Fuera to know what the state of Ethereum is. I want to know myself directly. Then every user will be running. And it will be feasible because of validity proofs, fraud proofs, and data availability sampling for every user on their phone, for example, to be running a full node. Then more users of these products equals more sampling. aggregate. So that's kind of the base assumption there. The base assumption is that it just becomes more and more feasible to contribute resources to the network. Well, the base assumption is that end users want to have, want to be secure. If they want to be secure, then they will run the,

like a sampling node. And the reason, so I think one of the reasons why people don't run, I mean, how many people do we actually know that actually run an Ethereum full node? Not very many because it's A, quite technical. B, it's kind of. expensive, right? But in the Celestia model, on a modular blockchain model, one of the things that we want to do is democratize the ability for people to run full nodes. So like, for example, on Ethereum, to run a full node in theory, it's like a laptop. I don't know if that's really feasible anymore, but that was sort of like the target, right? In Salana, it's like, you know, a megapower, like cloud machine with a gigabit internet. But in Celestia, we want to target in like modular

of blockchain, we can target something like much more broadly accessible like a smartphone because it's actually feasible because of the way that the technology works. So zooming out and looking at Celestia holistically, is this particular construction like designed for something? Like Ethereum, I think in my mind is very much designed for defy, like hard money assets, tokens, high security because of high value transfers. Does this change in how Celestia's design change what is kind of optimized for? Or is it the goal definitely something like defy, NFTs, kind of all the stuff that we know of today?

It's basically like as general purpose as any layer one blockchain can get. So it can be used for any application. I would say that the real, the raison d'atch for Celestia is to create plentiful block. plentiful trust-minimized block space that any developer or community, whoever, can leverage to build their own blockchain or their own decentralization. So I wouldn't say that it's specific for any particular thing. I think what Celestia enables, again, is more experimentation. It'll be, it'll increase the speed at which people can spin up new blockchains and try new things and experiment, which I think will overall increase the amount of innovation in the space. And then I also think that, you know, Celestia is really this place that makes it easier for people to have access to their own blockchain computer.

So like before, like, let's say, I don't know, I'm someone who just, like, it's a huge overhead to deploy my own blockchain. And if I wanted to have that level of sovereignty and control over my own decentralized computer, they have to pay a high price. And so I think Celestia brings that price down significantly. And so, yeah, we just want to democratize access to block space. Maybe that's the best way to put it. Are there any other components or aspects of Celessia that we haven't touched on yet? That's a good question. I would talk about two things.

One is Celestiums. So we released a blog post maybe a month ago that talks about this idea where we want to help Ethereum roll-ups. scale. So as you might be aware, like as of now, the call data on Ethereum is still quite expensive. So even though these roll-ups have massive increases in scalability, the fact that they post their data back to Ethereum main chain makes it kind of expensive, like still more expensive than I think these roll-ups would like to be. And if adoption continues, then that price will only go up, right, because Ethereum's block space is still limited. So Celestia is trying to build a solution for that problem. So what we do, what we've done is we're building a bridge that

it will allow Ethereum roll-ups. They still settle to Ethereum, so they post their state updates to the Ethereum EVM, but they will post their transaction data that needs to be made available for their security to Celestia. And then the Celestia validators will attest to that data being available. And if they lie, it's a slashable offense. So there's actually crypto-economic guarantees. And then that attestation gets relayed to Ethereum so that the contracts, the roll-up contracts, can actually verify them. So it's a way of basically being this sort of like off-Ethereum can offload some of this data availability need to Celestia so we can help the Ethereum community scale in the time being.

And for, this is called a validium, right? On the Ethereum. Yeah. Well, yeah. So validium, yeah. So that's why we call it Celestium, because basically it's like a portmanteau of Celestia and validium. So validium is a zero knowledge rollup where rather than posting the data on chain,

you post it off chain. But the problem is that a lot of these validiums today is, they're really not decentralized. They use what often are called data availability committees. And so, and that's just basically like a multi-sig. And we all know multisigs are really not that secure. So Celestia is like a much more decentralized version of that with actual crypto-economic guarantees. Like it's a slashable offense. So we think of them as different, like an upgrade above volidiums.

So that's why we call them Celestiums. So with a validium, like in theory, you could just place that data on AWS, right? And because you still have the execution of the assets that get settled on a thing, Ethereum, the assets are safe, but then there's data that's relevant, that's needed that could go in theory just on like AWS, right? But what Celestia is doing, correct me if I'm wrong, but what Celestia is doing is like, well, you don't have to put it on AWS. You could just put it on Celestia instead and retain some of those crypto-economic

assurances that we all know and love in this industry. Is that all right? Kind of, yeah. So what you need, right, for the security of all these layer two constructions is data availability. And so what that is is not, it's different than storage. This, like, it's something that, you know, a lot of people get hung up on when they're first learning about data availability. It's very understandable because it's a nuanced concept. And the term data availability sounds a lot like, oh, well, you got to make it available. You just, like, store it somewhere, right? But it's different. It's not that you need to store the data, is that you need a guarantee that the data is public information that anyone can download. So, you know, just the validator telling me, hey, well, don't worry, I uploaded the data to AWS, doesn't give me that guarantee because they can just delete it, like, or they didn't even necessarily release it. They're the only ones who know it. It's about information dispersal.

It's like multiple people know what this data is. That's what you need. And so Celestia actually can give you like a guarantee of that, whereas other off-chain data availability mechanisms don't really give you very strong. security on that. That's the key difference. Okay, so this next question comes from the the ETH Maxi camp of the world. What about the idea that Celestia is just trying to vampire attack all the Ethereum roll-ups, trying to get all the Ethereum rolls to settle their data on Celestia, and then eventually tries and just convince the roll-ups to come over to Celestia? That's a really good question. So, first of all, ironically,

actually, we never thought that this would be a use case. We were very dismissive of it because we didn't believe, because what happens when you build a validity, or when you settle, if you're a roll-up and you're putting your data on Celestia, now your minimum threshold of security is actually now the Celestia blockchain, right? So like you're no longer really a roll-up in the sense that you're no longer that's the same security level as Ethereum, right? Because that's the whole point of a roll-up.

So that's why we were always kind of like, really, are people going to want to do this? But then when we looked into it and we saw how desperate the need was and the fact that people were using these data availability committees were like, actually we can do better than that. But frankly, like if I were an Ethereum roll up and I were settling to Ethereum and Ethereum actually had plentiful block space, then I wouldn't use Celestia. Like there is no reason to do that. Like, you know, if all your assets and your users are coming from the Ethereum chain and the Ethereum ecosystem, then you should stay there. So it's not, I don't think it works as a vampire attack for that reason. And that's really not our intention. Okay, cool.

I don't know, I don't know much about the technical details to be able to go further with that question. But another question I have for you is just like, I mean, I'm already totally sold on like the modular blockchain thesis. how in if we zoom forward five, 10, 50 years, how does the general crypto ecosystem landscape topology look like under under like a Celestia paradigm or just a modular blockchain paradigm like who's doing what what applications are using Celestia how just like how does this thing mature and develop and expand over time? That's a really good question. So I also this reminded me of the fact that we never,

when we were talking about scalability, you were saying like oh well you know Celestia wouldn't be as scale. as Salon or something like that. Actually, the answer is no, it's the opposite. Let'saosia will be far more scalable than Solana or any monolithic chain ever possibly could be. Because basically we get the best of both worlds. We can have these like mega-powered nodes

where like the block production is, I'm going to call it centralized, but very high resource, but actually doesn't cause centralization. So that the, because the end users are still verifying everything. And so I foresee a future where Celestia has maybe orders of magnitude more data throughput than any monolithic chain ever did. And so I think we'll really reach this. Like if we want to transition to mass adoption, to me, the only viable path is via modular of bond chain. So but beyond that, what is the future? I mean, I see a proliferation of execution layers.

I think, and I'm already seeing this sort of anecdotally within our. our own ecosystem and within the blockchain space, like a lot of people are getting excited about roll-ups because they're realizing how they're really a new paradigm. And I also, frankly, see a proliferation in the long term of other data availability schemes and other data availability layers. Like I think, and I think that's healthy, right? Like the, like we have this saying in Celestia, which is modularism, not maximalism. And so what that means is basically that, you know, I think one of the reasons why the blockchain space has had this crazy tribalism and maximalism is that every monolithic chain has to be its own island and is really locked into its one design paradigm

and has to do everything on its own, right? So it's very much like dog eat dog, kind of every man for themselves. And in a modular blockchain paradigm, it's more about like, hey, I'm building this one chain to focus on this one function. and then the end users and the developers can plug in and mix and match and do what they want. And there let the market sort of decide, right? Like there will still be a competition between different execution layers or different data availability layers.

But the point is that we're all sort of like on an even playing field and we're all trying to build the most trust minimized, most useful, most like plugable and modular components as possible. I think to put that differently is that we're all very tribal because our our blockchains are all siloed away from each other but then when you put them onto a substrate that allows cross-chain communication cross-chain assets cross-chain whatever all of a sudden that those tribes perhaps melt away just because like we're all on the same protocol for the first time yeah exactly and a key part of this vision we didn't really touch on this but one of the

other reasons why this modular blockchain paradigm is so important and I think revolutionary and is the future for blockchain infrastructure is that the roll-ups or these chains when they share a common data availability layer, they can interoperate with the highest security. So they can have truly trust-minimized securities. It's almost as if, so it's it's almost as if they will be applications running on the EVM. Like, you know, we're used to the fact that applications of the EVM, they all share the Ethereum as their security layers. You don't, for an application to talk to, each other, they don't have to worry, like, oh, is this application secure or not?

So you get that same thing. What you lose, though, is atomicity. So, like, you know, different roll-ups will still, you know, have this sort of cross-chain thing where you can't guarantee that a transaction here is going to be automically execute with another, like, contract here. So you do lose that. But I think it will supercharge this multi-chain, the evolution of this multi-chain universe and really, like, create, like, an internet of blockchains where there's, you know,

this common security layer that underpins everything. And then you can proliferate. You can like deploy chains at will. Like we want to make it possible for people to deploy a blockchain as easily as it is right now to deploy a smart contract on Ethereum. So like if you imagine that world, we'll have the same amount of proliferation of contracts or of chains as there are contracts and applications on Ethereum in the long term. Like that's that's really what gets me jazzed up. is like just the future sort of explosion in Cambrian sort of like explosion of like the multi-chain universe. Like we're so early still in this whole exploration of what blockchains are capable of.

And I think this is a fundamental evolution and how we'll be able to build these systems and how we'll be able to scale them. Well, Nick, I can certainly see your excitement. So it gets me excited as well. If any listeners or viewers on the YouTube are excited and want to learn more about Celestia, where should they go? So our website is a good place to start, Celestia.org, and also a Twitter account,

Celestiaorg, one word. You know, we have a bunch of, there's a resources page, is the FAQ page. You can go on to YouTube, a bunch of other podcasts and talks that we've given. I really enjoy, there was a recent interview that John Adler did on the Delphi podcast, which I think is a great starting point.

But yeah, like reach out, and join our Discord, of course, and Telegram. You'll find those links on our website. And we will also put them into the show notes as well. Nick, thank you so much for coming on to Bankless and telling us all about Celestia. David, it's my pleasure. Thank you again. And let's stay in touch.

Thanks to Bankless community.