Permissionlessly Connecting the Modular Future with Jim Chang

Morning everyone. Thanks for showing up bright and early on a Saturday. Hopefully you guys aren't too tired or too drained from last night's activities. Excited to tell you guys about a project that I've been working on for a few months now called Catalyst. Uh but broadly the talk is going to be about permissionlessly connecting modular blockchains.

So, a quick uh introduction on myself before I get into what uh Catalyst is. Uh a little bit kind of a background for me. Uh my name is Jim. I go by Zerx Jim on Twitter. Um I am the co-founder of Catalyst alongside uh three other individuals in the founding team.

Um I'm also a former product manager at A uh as well as Unstoppable Domains and Ripple. So, I've been in the crypto space for a few years now, coming up on five. uh and uh it's it's been kind of really exciting to explore different parts of the space as it pertains to NFTTS uh building monolithic L1s as well as DeFi. Um and so also contributor to pleaser and gitcoin but not as uh relevant for this talk but uh yeah very very excited to talk to you guys today about defi and the intersection of infrastructure and crosschain that we've been building at catalyst and so kind of level setting the stage that we have in front of us. Um basically uh there's a problem that we have uh around imagining or reimagining rather the infrastructure for the modular future.

And so what is the modular future? Uh essentially I think in the future there will be a kind of world in which there are millions of blockchains. And so for perspective uh coin gecko says that there's about 900 blockchains right now. And so I actually think that within the next 3 to four years there's going to be millions of them. Uh, and why is that the case?

I think it's going to be enabled by the modular blockchain stack. And so I'm sure many folks in the audience are familiar with the modular blockchain stack. So I don't need to rehash it. But the implication of the modular blockchain stack is essentially the ability to create uh new execution environments using a shared settlement layer or shared data availability layer and it becomes very trivial to make a new execution layer. And uh that's a really kind of powerful uh enablement of the space and so that you're able to create new blockchains really easily.

In fact, there's a lot of kind of solutions in the space right now thinking about what does a one-click blockchain look like and that's actually something that's coming out very very soon. Yeah, sure. Try this one because it's blinking. Okay. Oh, here we go.

Cool. Sorry about the technical difficulties. Um, and so in this world of a million blockchains, uh, enabled by the modular blockchain stack, it becomes not only trivial to spin up a new execution environment, it actually becomes very configurable. And so I actually believe that every application will have its own chain. And so whether it's kind of the NFT platforms that folks are using right now or the DeFi primitives that were really popular in 2020 and 2021, I do think that the configuration to be able to optimize for gas prices based on different op codes or have offchain DA or have parallel execution for high throughput low latency environments uh becomes really important because you can have that level of customization solely for your application.

And so I do think that there's going to be millions of apps in the future enabled by the modular blockchain stack and therefore there will be millions of applications in the foreseeable future. Again I say in the next 3 to four years and so what happens in this world and what do we need to do in order to kind of see the fruition of this vision of a million blockchains. And so unfortunately as good it is to have a world of millions of blockchains that are application specific, there are things that going to break uh that we need to be very mindful of. And so one of the things that happens is we're actually going to be losing connectivity. And so kind of the the beauty of having a shared execution environment is that connectivity to other applications.

We actually lose that in a multi-chain/hyperchain environment of millions of chains. The next thing that we're losing and somewhat related to connectivity is composability. And so the ability to create atomic transactions and actually have money legos that became so popular in 2020, we actually lose that as well. And that's kind of a drawback of having a multi-chain world. And then lastly, something that I don't think many people are talking about, we're actually losing the existing infrastructure that we have in place.

So whether that's wallets, whether that's block explorers, data indexers, data analytics, oracles, RPC nodes, you know, um governance forums, you name it. when you create your own application specific environment, when you have a paradigm in which it becomes trivial to make a new blockchain, um all these existing infrastructures that we have just simply don't exist. How do you use a new wallet for a brand new chain that doesn't talk to any other the specifications? How do you use data indexing in this world? And so, I think that's a really strong implication that I don't see anyone in this conference talking about right now.

And I'm kind of raising it because I think it's really important because we're all marching in tandem towards this future of a million chains. And so the question arises based on those kind of problems that we see. How do we actually build for the modular world? How do we build for this world of a million plus blockchains? And so I kind of lay out three different tenants that I think all the new infrastructure in the space needs to have in order to adapt to this new world of millions of blockchains.

So the first thing that I say is that they need to be automatic. And so what does automatic mean in a crosschain sense? It means integrating new chains as they're deployed. And so on the right hand side you kind of see two examples that are kind of doing this but again not exactly as extensible as I would like to see. And so one of them is MetaMask of course if you have a new EM uh EVM compatible environment you can add your new chain using the network URL and the chain ID to MetaMask.

It's a bit manual but at least it's there for us right and so that's why people really like EVM chains and interacting with it using MetaMask. Uh the second thing is a new company that uh is called Modular Cloud now but they used to be called Celestia Scan. So they're basically a block explorer for Celestia roll-ups. And so anytime there's a new Celestia rollup that's posted on the share DA layer of Celestia, it actually pops up in their block explorer. So that's really powerful.

So again, we're moving from the current state where there are kind of siloed experiences that are bespokly created for these individual chains and we're going to a future state in which all these new chains are being created automatically feed into the primitives that we're familiar with whether it be data indexing, wallets, block explorers, oracles, etc. I think the second principle that's super important is permissionlessness. And so the ability to actually deploy developer products with no gatekeeping. And so the ability to create a new governance forum and a new governance kind of uh proposal from scratch like a joked dow or the ability to create a new rollup with no gatekeeping with caldera. I think that's really powerful and you know doesn't allow for gatekeeping that we see currently right now with a multi-chain environment.

And then lastly, it's really extensible. And so extensibility for me means the ability to expand to new environments. And so we're in a multi-chain world where not only are there different kind of specifications to communicate between them. There's different virtual machines, different languages, different consensus mechanisms, different signature schemes. And so you can see in this world in which it becomes really difficult to actually connect all those different pieces together.

And so you need to build primitives and standards that are actually able to um overlay over all those different standards in order for a more seamless and extensible experience. And so Eclipse is something that I think is is kind of marching towards that vision having a shared settlement layer where any rollup of any VM can actually settle on it. Uh and risk zero is another really great example where they're essentially building a shared prover spec from a validity proof perspective so you can actually have a shared kind of um you know shared kind of prover spec that can be settled anywhere and composable. And so this kind of brings me to, you know, now that I've set the stage of of what I see the problem is with, you know, having a a endstate vision of millions of chains and seeing all the existing infrastructure break and I've kind of laid the path into what are the three tenants that I think are incredibly important for us to actually um build towards this modular future. What are we actually building for this modular world?

And u we're building something that I think is super important and super kind of foundational, which is liquidity, which is bridges. How do we actually move value in and out of all these new chains as they're coming live permissionlessly? And so with Catalyst, we have our three tenants that we're building towards that are perfectly aligned towards exactly what I was talking about in terms of how you're actually building infrastructure for a modular future. So the first thing we're doing is an automatic chain connections. So anytime a chain comes live, it's automatically connected to the catalyst liquidity network.

And so there's no reconfiguration that needs to happen. there's no um governance proposal that needs to happen. It's just automatically connected uh which is an incredibly powerful primitive. The second piece is that we have permissionless crosschain pools. And so if you actually want to connect value and you want to move value in between two chains, anyone can spin up a pool between these new chains.

And so you can see a paradigm in which there are brand new chains kind of coming into the network of catalyst. They come online, they're automatically primed to connect to catalyst. Of course, prime doesn't mean they're actually going to be moving value. you need deposit liquidity. What you can do is actually create these permissionless crosschain pools.

Anyone can do it for any asset. And again, now you're able to actually swap value in and out of these two chains. And so the example that I kind of give in the slide is um Ethereum moving ETH from Ethereum onto Madic on some sort of modular chain on Polygon, whether it's their ZK rollup solution or whether it's their supernet solution using their POS chain. Uh and then lastly, what we're building at Catalyst is essentially an extensible protocol design. And so catalyst um first and foremost is a mathematical library and I can talk a little bit more about what actually happens underneath the hood to make catalyst work.

Um but as a result this mathematical library only needs certain approximization kind of um implementations that can span any sort of language in any sort of virtual machine. And so whether it's EVM, fuel VM, move VM, SVM, WASOM, any other VM that people want to throw out at me and there's a business case feel free to DM me. um we can do that. And so that is definitely the extensible design that we've thought of at Catalyst. And so as you could see, everything that we've built at Catalyst is for this modular future is for the future in which we see millions of chains pop up because everything that gets added into afterwards um gets automatically slotted into the Catalyst liquidity network so that people can enjoy transferring value in and out of any chain anywhere.

And so that's a lot of kind of talk about what Catalyst is from a principles perspective. I'm sure many of you actually wondering how it works. And so Catalyst is effectively an AMM design in which we have two assets that live on separate chains. And so it's kind of like unis swap in design, but instead of having two assets live on one smart contract on one chain, we actually break it apart. And so it's two assets living on two separate contracts on two separate chains.

and how they're actually connected together is using an arbitrary messaging bridge, something like an IBC. And so from a user's perspective, after a pool is initiated, um they deposit one ETH into the Ethereum pool. And you can kind of see here in this little box over here that uh you know, the the amount of Ethereum actually increases into the pool. And so if you guys are familiar with how price and variance work in terms of x yals k um on automated market makers, essentially we're moving along the curve um of this price curve. And so we're adding more Ethereum into this price curve.

And actually what happens is when you're moving along this curve, you generate an integral. And so this integral is actually an area that we calculate that we call a liquidity unit. And so in this instance, we calculated X amount of liquidity units that's been added to the ETH pool on Ethereum. Uh and we actually pass that as a message using arbitrary messaging protocol into the destination chain. So in this case, you could see X amount of liquidity units have been deposited withdraw the same amount of liquidity.

And so when that message is received on another destination pool of the destination chain, in this case we're saying Madic on the Polygon modular chain, um you can see now that they're actually taking that unit of liquidity. They're recalculating as a unit in the curve and they're actually moving up along a separate locally set price and variant on a destination pool. Sorry if this is a little bit technical for some folks, but basically you can see here that they're withdrawing the amount of Madic from the Madic pool on the Polygon modular chain and it's actually being held equal by the amount of liquidity that's been moved in the set. And so the entire system has the same amount of liquidity as calculated by our system, but essentially the actual tangible assets in this case ETH, in this case Madic changes, right? And so ETH increases and then Madic decreases.

And so from a user's perspective, again, all they did was swap one ETH on Ethereum. And on a destination chain, they're receiving something that we call the liquidity equivalent amount of Madic in which in a case of very healthy arbitrage and kind of healthy price sets is going to be the market equivalent amount of MADC. And so the user just sold one ETH and they on Ethereum and they are able to receive the liquidity equivalent amount of Madic on this Polygon modular chain. And so that is kind of what catalyst does from a onetoone use case perspective. I think what becomes really powerful is actually thinking about how catalyst works from a one to many perspective.

And so catalyst is not just you know unis swap in the sense of you're swapping two assets back and forth. Catalyst has the ability to actually connect to any number of chains. And so in this instance let's say the user wants to sell their one ETH but they don't want to actually interact with the polygon modular chain. Instead they want to interact with the DIYX app chain. um Catalyst can do that too.

And so in this case, the same exact process happens. The user deposits one ETH. The system calculates the amount of liquidity units that's been added to the system and they withdraw the same amount of liquidity from that system. And so in this case, the user now receives the liquidity equivalent amount of atom on the DYDX chain. And so you could start seeing how this one to one relationship becomes a one to many relationship becomes basically an entire web.

And so when you look at the three principles that Catalyst is building upon, automatic chain connections, permissionless pool creations and extensibility of any sort of virtual machine, consensus mechanism, language, signature scheme, etc. It becomes a very composable and expandable liquidity network for the modular world. And so users are able to create pools, they're able to swap, they're able to um move liquidity and move value in and out of any chain that's connected to the Katus network. And so I think our go to market is is really special because we're not just a DeFi primitive. We're not just kind of a DeFi app, but instead we're building kind of developer tooling and infrastructure for these modular chain builders.

And so if you're building on Polygon's ZK rollup or Polygon Supernets or Caldera or OP stack or arbitrum any trust, I I can start naming more if people have time. um you will have Catalyst be a module in the SDK for that solution. And so when you deploy a new chain, you have Catalyst baked in either as a pre-eployed smart contract or baked in as business logic into the execution environment. And so you can start seeing how yes, we're an application on top of all these modular chains, but we're actually kind of very core to the code of a lot of these modular chains as well. And that is the ability that gives these chains the ability to connect to any other catalyst deployment.

And so essentially catalyst becomes this web that becomes more and more powerful and more kind of enticing from a value proposition perspective to build with uh as there's more chains that come online. And so that's kind of the vision of what we're building at Catalyst. You know, we see the problem of a world of a million chains. We're building from a ground up. what does it mean to create a liquidity network from a crosschain modular world environment?

Uh, and that's a little bit of a sneak preview into how it actually works. Uh, and so if you guys are curious, um, we are on Twitter happy to, uh, share some more updates on there. Uh, but I'm also happy to answer any questions after this talk on stage and then, uh, Christian is one of our developers on our team. He can answer some of the more Gigab brain questions. Thank you.