Nick White - Crossing the Chasm - TOKEN2049 Singapore 2025

For those of you who don't know what Celestia is, Celestia is the first modular layer 1 blockchain that's built for the extreme speed and specialization required to take crypto mainstream. I'm going to talk a lot more about that in a second. But to frame this talk and this conversation, I first want to zoom out and take a look at where the crypto industry is today. And in my opinion, we're at this very critical inflection point of the adoption of crypto as a technology where we're exiting the early market and transitioning to the mainstream market. And this is what business people call crossing the chasm.

And it's basically where crypto ceases to be just for technology enthusiasts and people at the fringes of society and starts to be used for real world businesses and by everyday people. Um, and the reason I say that is for a few different reasons. One is we're finally getting regulatory clarity. So we have the Genius Stablecoin Act. We have the uh market structure bill.

We have streamlined approval of ETFs. This is totally a 180 degree change from a year ago. We have major enterprises entering the chat, starting to build onchain. So Google for example, Stripe building their own layer 1 blockchains. We have Cloudflare announcing a stable coin.

And this trend is only accelerating. We also have some of the first cryptonnative apps that are really reaching product market fit. So things like USC and Tether reaching hundreds of billions in AUM and projected to go to trillions. Hyperlid uh actually for the first time challenging centralized exchanges for market dominance in crypto trading. And then of course apps like Poly Market that are breaking into the mainstream consciousness, influencing politics and uh the mainstream media conversation, bringing a lot of net new users into crypto.

And then last but not least, crypto is also maturing as an asset class. So it's no longer just driven by speculation, but people are starting to pay attention to crypto tokens both for apps and protocols as these revenue generating entities. And that opens up crypto to a whole new class of more sophisticated uh investors. And so while it's really exciting that we have reached this point in the development of crypto, um we're at the the chasm and and seeking to get to the other side, I think, uh we have a problem, which is that the layer 1 technologies that underpin um basically crypto as an industry that have gotten us to this point are not enough actually to get us across into this high growth sort of mainstream uh market adoption phase. And so why do I say that?

Well, I think there's two main things that a layer 1 blockchain needs if it wants to actually support mainstream adoption. The first is speed. Tons and tons of speed. Lots and lots of throughput. So just for some examples um let's say that the uh stable coin that um Cloudflare wants to launch one of the use cases they cited was for bots and AI agents to make microp payments as they're browsing the web and accessing content.

So the fact that like I think over half the internet's traffic is bots and agents could kind of project that there would be hundreds of thousands if not maybe in the future millions of these micro payments happening as all the internet is being crawled by all these bots. That's just one example. Another example would be trading right people want to build the decentralized NASDAQ. Well NASDAQ does a roughly 200,000 orders per second and that's just one venue right in one part of the world. If you scale that up to all venues globally and all assets, uh, global trading is probably in the orders of millions of transactions per second.

So, we need we need that scale of speed. Um, the other thing is specialization. Um, you know, I think we're have F1 on the weekend here in Singapore. So, I thought I'd use a car analogy to make this point, but I think of these smart contract general purpose chains that have gotten us here like Ethereum and Salana uh, as like a like a minivan, right? A minivan is a very functional car.

Um, and it's a good all-around vehicle, but it's not really optimized for any specific thing. Like, if you wanted to go fast, you need a sports car that has a bigger engine, has a more sleek body, or if you wanted to go off-roading, you need a car with four-wheel drive, more clearance, you know, perhaps off-road tires. And um, a minivan is just not going to be suitable for either of those use cases. So similarly, the blockchains that we have today are more like minivans. They're like these all-purpose things, but they're not enough to kind of push the frontier of what's possible to build in crypto.

And so we need to support more specialization. So with that in mind, speed and specialization as these two critical um features for any L1, let's look at the L1 landscape today. So on the one hand, we have Salana. And Salana's very fast. It's an extremely performant chain and they have plans to, you know, make it even more performant with things like Fire Dancer, but it doesn't, it's still a minivan.

It's still a uh general purpose chain. So, no matter how fast it can go, it just can't support all the whole universe of use cases for crypto. And then on the other hand, we have Ethereum that has gone down this rollup ccentric road map, but ultimately is still too slow. If you saturated all of Ethereum's block space, you'd only be able to perform roughly 300 transactions per second. And that's what brings us to Celestia.

And Celestia again is a modular layer 1 that was designed from the ground up. It had the the advantage of being being a new chain to be designed for this future of mainstream adoption knowing everything that we know now. And specifically, how does Celestia achieve this specialization and speed? because of its modular design, it doesn't even support any smart contracts. The way that you build on Celestia is you build a vertically integrated application where you as the developer, the builder can customize all these components.

Um, and so you have full stack control and customizability. So it's a maximum amount of specialization. And then Celestia is also built for speed because by being this very thin, simple, modular layer 1 chain, we can use new cryptographic primitives like data availability sampling that allows us to scale in ways that other layer ones can't. Um, and this is not just in theory, but actually in practice. So, we have uh a an upgrade called Matcha coming to the network this month that will increase Celestia's maximum throughput to 21 megabytes per second, which is equivalent roughly to 64 times the total throughput that Salana processes today uh on on L1.

So with that in mind, the speed and specialization, I just want to highlight a couple of applications building on Celestia that I think really take advantage and are a good example of what speed and specialization on Celestia can look like. So the first is Bullet. Um it's a decentralized purpose exchange. Um and they have built a highly optimized, highly performant rollup uh based in Rust that can actually execute trades at 1 millisecond latency. This is insane.

This is a 100 times faster roughly than even Hyperlquid's uh you know famously fast 100 millisecond uh exchange latency. And they've also because they're a rollup they can control the sequencing the order of transactions. So they can add things like cancel order prioritization which reduces toxic flow and improves the liquidity in the system. And of course their goal eventually is to be a decentralized NASDAQ right. They want to reach uh you know scales of 200,000 orders per second and only Celestia can really support that level of throughput.

So that's why they're building on Celestia. A second app that I'm a big fan of is Pay. So Pay is a stable coin payments app. Amazing. You guys should download it.

And they take care uh they take advantage of specialization to add privacy. So every payment that you make on the platform is totally private. Absolutely essential for mainstream adoption. And they're only able to do that because they can control the actual execution model of their chain. So they've built uh a similar model to Zcash using ZK proofs.

And then they also customize their wallet so that it works natively on your phone. So you never touch a seed phrase. It's backed up in iCloud. You sign with Face ID. It's it's really amazing UX.

And if they actually achieve their goal of wanting to compete with Visa or JP Morgan Chase, they're again going to need the kind of throughput that only Celestia can provide. tens and thousands of transactions per second. Now, you might be wondering, um, well, where are the enterprises, right? And, you know, if you paid attention to Stripe or Google, uh, or Circle, they're all they're not building roll-ups. They're actually building their own layer 1 chains.

So, you might conclude, oh, well, that's, you know, most enterprises are going to choose to launch their own L1s, not roll-ups. I think Mike from Blockworks actually explains why this won't be the case uh going forward necessarily and that is because the work that it takes to actually launch your own L1 is a huge amount of effort. It's a high cost. You have to hire a whole team. You have to launch the network and maintain it continuously and ultimately it's not really a differentiator.

If you can still get all that control and specialization on a rollup, then what's the point of going through the effort of doing all this heavy lifting on the infrastructure side? And I think most enterprises will realize that it's not worth it. And in fact, this is how it plays out in web 2, right? Even some of the biggest web 2 companies like Netflix, Airbnb, and Spotify, they don't they have a huge compute demands. They don't build their own compute infrastructure.

they use uh providers like AWS and that's what allows them to focus on the what's important which is building and improving their business and so that's actually really the cloud model is a good way to understand what Celestia is as a product. Celelesia is like the equivalent of a cloud computing like data center provider but in web 3 where instead of deploying a virtual server as you would in web two cloud in Celestia's case you deploy a virtual blockchain also known as a rollup and um that brings me to the last point of where I think L1's need to mature in order to actually be ready for mainstream adoption and that is on the the revenue and economic model front. So historically chains like Ethereum and Salana have made significant revenue but it's mostly been driven by me and priority fees and the reality is that these fees are not sustainable are not going to be long-term they're not going to belong to L1. They're going to graduate up to the application layer either through application specific sequencing or by people launching their own rollups and controlling the sequencing layer themselves. So long-term me is not a sustainable source of revenue.

So well then what should be the business model for L1's? And I think again if we look at cloud computing there's a lot of parallels. Um, instead of trying to sell very limited, highv value block space like me, like some kind of Wagu steak, we should actually be trying to build a really really really scalable economic model where we're selling more like McDonald's basically selling a massive amount of beef um but at at a global scale with low margins. But as demand overall scales for blockspace, the overall revenue actually becomes very compelling. And so I think this is the real correct revenue model for L1's long term and it's what Celestia is focusing on.

And aside from just where revenue comes from, Celestia is also innovating on the cost side. So we're actually contemplating a proposal to shift to something called proof of governance which would reduce our issuance, right, the cost to run the network from roughly 5% what it is today to 0.25%. So a massive reduction in the overhead and the issuance of new tokens to pay to operate the network. And that would mean that even at relatively reasonable uh levels of throughput, Celestia could actually be a deflationary uh protocol like a profitable protocol.

Um so anyway to you know sort of conclude things here um what I want to leave you with is if you actually believe and take seriously the vision for what crypto could be right that we onboard people or global payments with stable coins all around the world we onboard uh onchain agents you know browsing the web transacting with each other we have all of global trading coming on chain all these things are going to require enormous speed and specialization and that's something that I believe only Celestia's really suited to provide. And so my hope is that Celestia can be the L1 that actually enables crypto as an industry to cross this chasm into sustainable long-term growth and mainstream adoption. So, thank you very much. Um, I'm hanging out if you want to talk afterwards.