Is $TIA Modular Money? | Nick White

Point of build whatever is there's no optimal point. It's all like a trade-off curve, right? And and Celestia is optimizing for a certain thing, Ethereum optimizes for a different thing. And again, also to tie it back, and that's that's why it's not they're not commodities because they're they're they're just gonna be very different. It's like Ethereum block space is gonna be different than Celestia Block Space because they make different choices, they're optimized for different things. I think that like Celestia just has a different philosophy and that results in a different design.

Welcome to Bankless, where we explore the frontier of internet money and internet finance. And today we explore the frontier of Celestia and the roll-up landscape that it wants to bring into the crypto universe.

Bankless Nation, there is a paper-thin wall between the design of Ethereum and the design of Celestia. These two networks are both designing for very similar outcomes. It's often discussed that Ethereum and Solana are the biggest competitors out there, but I actually think the reason why these two networks are positioned as big competitors is because the design philosophies of Ethereum and Solana are just completely opposed to each other.

I suggest to you, Bankless Listener, that it's actually Celestia that is the real competition to Ethereum, because it's made the same choices, optimizing for the same outcome, but with just some few key differences that we're going to unpack here on the show today. Perhaps this is also why there is a much more friendly relationship between these two ecosystems, between Celestia and Ethereum, because despite being more meaningful competitors, our values are the same.

So, personally speaking, if Celestia won the Game of Thrones in crypto, I would actually still consider that a positive outcome for the preservation of crypto values in the bankless thesis to which I hold dear,

and to which other networks I could not say the same for. Today on the show is Nick White, the COO of Celestia Labs, and has been with Celestia from the earliest of days. We had Nick on the show about two years ago, but since then, both Celestia and my understanding of it has grown quite a bit. And also this meme of Tia as modular money has also emerged. And whenever some token presents itself as money in the crypto space, I immediately want to go kick its tires and see where that idea holds up or where it breaks down.

So let's go ahead and get right into the episode with Nick White from Celestia Labs. But first I want to talk about some of these fantastic sponsors that make the show possible. Bakeless Nation, I'm here with Nick White, COO at Celestia Labs. Nick, welcome back to the podcast, my man.

Thank you, David. I think it's been almost exactly two years since I first came on Banklist.

Yeah, and two years ago, let me tell you, I had much less of an understanding of what Celestia is. Um, I think even we were using like Lazy Lazy Ledger as like the name for this whole system, and the concept of DA was just very foreign to me. So I'm hoping to do kind of just like a reprise of that same podcast. But your host this time has uh a lot more knowledge to ask the right questions and take this conversation into the right direction. Uh Celestia has like the the project has grown immensely uh and has really like uh redefined an entire meta. Uh so A, just like congrats on the success, my man. Like, what what's it been like over the last like two years of Celestia? Like uh give us a peek behind the scenes.

Well, it's been uh a grind. There's been a lot of work happening behind the scenes. You know, people like to s you know

Call things overnight success or whatever, but really the reality is it started way back in 2019 and maybe even earlier in 2018 when Mustafa and folks like John Adler, some of the co-founders of Celestia were, you know, researching roll-ups and L2s back when no one was talking about them, you know, forming the foundation of data availability sampling and all the technology that now is live and in production and forms the basis of modularity. So it's been a huge amount of effort and work to get where we are. And I think to us, um, it was kind of inevitable because the these things just make so much sense when you when you get to the bottom of it. It's like this is this is the right way to build blockchains.

Mm-hmm. One thing I appreciate about Celestia is that uh there is a notion of like a logical conclusion of the design construction. Um, like uh from first principles, like you can arrive at the conclusion of Celestia. Uh and so um, maybe we can kind of start there. Like Celestia's motto is build whatever. Um, and I think uh uh going into the design construction of Celestia, people listening to this podcast will be able to understand like why and how that motto emerges. Uh but maybe maybe you can just like kind of walk us through like the first principles nature of Celestia, why Celestia, it's a data availability layer. Maybe you can kind of like walk us through just the base foundations of like the theory, the thesis behind Celestia, and why might might one be interested in building a blockchain that starts using starts with DA first and foremost to build like to be at the bottom of the stack. What does it mean to build whatever? And how did you start here at DA?

So, first,

what is Celestia? Well, you can think of well, Celestia is the first modular blockchain network. And the goal of Celestia is to make it easy for anyone to securely launch their own blockchain.

So you can kind of think of Celestia as an L1 that's been designed from the ground up

to do one thing, which is to provide secure and scalable data availability for roll ups.

And so now to go like a step further, like, well, what does it mean to be a modular blockchain, right? And this is kind of

what ties into the build whatever concept.

Monolithic blockchains, which is pretty much every blockchain up until you know the first roll-ups and Celestia came along, were blockchains where all the functions were combined and coupled into one protocol. And so the nodes in that protocol had to do all the functions of running a decentralized application at one time.

And you when you do things that way, you fundamentally have a bunch of constraints that when you take a modular approach where you decouple those functions and break them into separate protocols that are optimized for that and that can be remixed.

You get all this, you remove those constraints and you have all this freedom as a developer, as a builder, to build new things that weren't possible before. So Build Whatever is touching on this uh aspect of modular blockchains that really removes all the constraints and just blows the doors open in what's possible. So

that's that's really what Build Whatever is about. It's about, so first of all, you know, what what are some of the constraints that get removed? First,

uh scalability

in a modular system.

Becomes uh like is is

significantly better because you have a really high performance data availability network at the bottom, which provides this very cheap, scalable, secure block space. And that's enabled by data availability sampling. Then you have rollups which scale the execution part. So all of a sudden you can have a blockchain that can have really high throughput while still being secure and trust minimized and all the things that we want. And that wasn't really possible before those two innovations. So that's the first thing scalability. You don't have to worry about the throughput and congestion and all those kind of things.

and get dedicated block space basically because you're running your own chain. Um the second thing is that you get to mix and match

The different components so that you can actually choose the kinds of features that you want your chain, your application to have. So you can choose different data availability layers with different properties, different trade-offs. You can choose different execution layers. Like you don't have to be just in the EVM or the Solana VM. You can choose any kind of VM or build your own custom execution environment. You can choose various kinds of fast confirmations like sequencing schemes to have different trade-offs or different kinds of mempools that have MEV tooling or privacy-enabled uh features. And so all these different pieces can be, and you can control the economics of your chain as well, which is really important because you can all of a sudden, instead of giving all the value back to the L1 that you built your application on, you can actually capture some of the value that you're creating as the application builder for yourself and for for your community and your network. And so

It's all those kind of things that that as a builder I like open the the the door wide open

to what what you can build. So build whatever's like whatever you can imagine, you can build. Like let's let's exp and we want to explore and encourage the exploration of this design space because ultimately

How how we get from where we are to where blockchains and web 3 are used on a daily basis worldwide is by by by app builders experimenting and showing us what's possible. Because, you know, uh ultimately a lot of people also say, like, oh, well, modular blockchains are just a bunch of infrastructure without real users or real applications. Well,

and and and so it's like, you know, you guys are just really just, I don't know, it's kind of like a

uh I don't know, you guys aren't aren't actually thinking about that, the end product um or what what like the real users, right? But really, modular blockchains are about that. And it's about

experimenting and unlocking new capabilities that I think are going to be.

The key to figuring out how blockchains become useful,

you know, for for the rest of the world.

Baked into there there are some opinions in the Celestia model. There are opinions in every single layer one. And so I kind of want to like define what the opinions are of the Celestia design architecture. First and foremost, like modularity, of course. But like first and foremost, I think like there the Celestia model is opinions that there will be many, many, many, many blockchains. And in fact, it's trying to encourage the growth of that. And what do all blockchains do is that they need to provide like assurances and verification about their state uh externally so that the world can verify the state of all of these blockchains. And so, like, um, you know, every single blockchain like with that we've had from the beginning of time has some sort of just like node set internally, like Litecoin has nodes, Bitcoin has nodes, Ethereum has nodes, uh, you know, you pick pick your blockchain, it's got some nodes doing some verification. Uh, and that verification is some sort of data, and all these nodes are verifying the data of the state of some particular chain. And so Celestia's opinion is that, like, hey, like we will uh be the place where that data is posted. We call this data availability. That's the what Celestia is providing to the world. And there's some sort of like opinion that like data availability is like at the bottom of the stack because in order to have verification of all of these blockchains, we need to verify the proofs, uh the state proofs of all of these different chains. And so, like, I think one of the opinions of Celestia is that it's a multi-chain world. Uh, there's going to be uh infinity of different chains, and actually, like, not only was there we can actually encourage that, uh, and we can actually support more and more chains by actually doing one of the one of the jobs that all chains have, which is the verification of the state. And like where whether you're an Ethereum layer two.

And you are posting your fraud proofs or your uh blobs to blob space, it's kind of doing the same role or to the Celestia data availability. Like there is proofs, there's uh statements about the state of all of these chains, and you can post them on Celestia in order to like get the security of people able to verify the state of all these things. And so this is something that's like this is just fundamentally true about crypto. This is not something that Celestia invented. This is something that Celestia discovered and is now optimizing for. And so when I hear like kind of the build anything model, uh, or the the excuse me, the build whatever um motto, it's to me, it's like Celestia is providing this like space for blockchains to do something that they all must do, which is allow for the verification of their state. Except the difference is Celestia is just being this one central place to kind of like all of these proofs to be posted, uh, which helps more and more blockchains kind of like manifest. That's kind of like my attempt at explaining this. How would you how would you riff on that?

That was great. There's so many things I want to say uh following that. One is

um

so Celestia is the most minimal blockchain that you can build.

Because when you when you go down to first principles,

all that you need to build these trust minimized applications

is some form of.

Ordering, which is consensus, you need to figure out like what are the order of the inputs to this, you know, state machine, right? Because every every application on a blockchain is a state machine, it's a computer. And you're trying to make sure that everyone can verify and arrive at the same state. So if you want to be able to all arrive at the same state, you need to agree on what are the what's the order of the inputs to the state machine. And you need to know what, first of all, what are those inputs in the first place? And that's data availability. That's data publishing, is maybe a better way to put it. You need to know everyone that's participating needs to be able to see what what's actually happening, right? So, but then the actual state machine and the execution, the verification of those transactions doesn't have to live on on the layer one on that chain. It can be run sort of client side or on a different layer, a layer two, like a roll up. And so that's that's really the insight. So when you think of things that way, what one way to sort of tie this back to build whatever.

is when you're L one when you're layer one, your base layer is that minimal,

You're kind of, you know, you're basically providing the maximum expressivity for the other the people building on top of it. Because you've

It's a tradesman, right? And so we're we're taking some opinion out of the layer one to allow for all of that choice of what that opinion is to be made by all of the layer twos on top of Celestia.

exactly yeah, and so and and like also to to talk about monolithic blockchains, right? Like Solana or whatever, it's sort of a one size fits all model. It's like let us make the choice for you. We know what's best. Yeah, exactly. We we're the experts, we're we have the best execution, we have the best this and that, and just like build on us and we'll take care of it for you. Now that I I think there's there's validity in that, but I think that the the vast majority of the opportunity is in this space of giving expressivity to developers and and app builders because they're smart, they know what they want, right? There's no one size fits all. It's like everyone has a different style, everyone has a different size that fits them. It's anyway, and then

Uh another thing about when you make the block space minimal, one maybe one analogy that could help people is that

uh it's sort of like

r like a a a refined

Sort of uh commodity, we we can also touch on this, like refined oil or something that's like now been processed into something new, like a value added product, versus something that's raw and unrefined, like a pure, just a pure, like just got it out of the ground resource. And so block space that has an execution layer embedded in it to me, like if whether it's on Ethereum based layer or a Solana or a monolithic chain, is a refined, sort of value added, like already transformed good. You can't take it back into the raw form that easily. Yes, you can, like in Ethereum, you can just like post blobs, but that

Layer is not actually built for that. Whereas Celestia is optimized just for

massive scale of the raw block space, unrefined block space. And then you can go and like take however much you want, basically purchase it from the protocol and refine it to do the thing that you want. It could be jet fuel, you know. Maybe you're running like a SVM L2 and you want it to go really fast. So you refine the block space, the the crude oil into jet fuel, or maybe you're just like, hey, I just want a you know everyday EVM chain, turn it into gasoline to put it into your car. I don't know. It's it's that kind of mentality.

Yeah, I I I totally under uh see that pattern. And I I know you read the an article that's gonna come out this week on Bankless. I call like Celestia in this, there's an axis here on size of blocks and sophistication of blocks. Uh sophistication being like basically is there a virtual machine in the blockchain or is there not? Uh and Ethereum and Celestia actually uh occupy opposing corners here, where the Ethereum layer one is a small blockchain. We have like small constrained data with a virtual machine, whereas Celestia is the opposite, where it is uh unconstrained, as in the layer one can support like massive uh size, massive capacity, blobs, but it doesn't have a virtual machine because that's one of the opinions that it allows, it pushes up to the higher layers to the layer twos. Uh, interestingly, even though we occupy like kind of opposing corners here, we've taken like the opposite trade-offs, there's still a lot of like shared values and shared like philosophy here, like easy to verify clients. Like, even though Celestia is like kind of a big, a big blockchain, since all the blocks are like simple because they're all unsophisticated, there's no virtual machine, there's no computation here, it's still trivially easy to verify, which is also one of the goals of Ethereum. Maybe even Nestia is like even more easy to verify than Ethereum because Ethereum, you have to actually redo computation to verify it. Um

And but like also at the same time, both are interested in like producing this layer two design space. May maybe like correct me if I'm wrong, but I think like kind of the main difference between Ethereum and Celestia, especially now that Ethereum has 4844, it has its own blob space, whereas Celestia is like only blob space. So the only difference between Ethereum Left and Celestia is that like Ethereum retains its virtual machine at the layer one. And uh whereas uh Celestia just like has no virtual machine, it's like just post your blobs. Uh is there any other differences that we should parse out, or is that kind of just like the main one?

I I think that I I really love that diagram by the way of like, and I like that way of thinking about a sophisticated, unsophisticated, and big blocks and small blocks. Um, and I think that uh basically Celestia really believes in yeah, big, unsophisticated blocks. Or I like to say big, lazy blocks um that are because when you as you add sophistication.

What happens is it becomes more difficult to verify. If you have state and execution, then the transactions that come through, like all of a sudden, to run a node and verify the chain, you have to have a bigger machine that holds that state that actually processes those transactions. Whereas the point of Celestia is that because of data availability sampling and the fact that it has no execution.

You can run a light node. You don't, it has all you can run on your phone. It requires almost no bandwidth and no no compute and very little storage. And yet you can basically have the assurances of verification of a full node. And when you start to add execution and state, the more state you have and the more execution you add, the the worse and worse that becomes basically. Like you're you lose that ability and it becomes heavier and heavier to run a node that that has like you know fully verifies the chain. So in Ethereum's case,