The Rise of Appchains: Why Control Over Your Stack and Sequencer Matters

Transcript

[Music] Hey, how are you doing? Um, I'm Will Pepper. I'm a co-founder of Syndicate. Um, and today I will talk about app chains and in particular why control over your stack via an app chain and why control over your sequencer especially matters a lot. Some quick background on me. Um, I co-founded Syndicate um with my co-founder Ian in January 2021. Um, before that I studied philosophy, but I'd been coding for a while as well. I got into crypto through Mesh Networks in 2013, another major pillar of decentralized tech. Um, found the Ethereum pre-sale in 2014. Took my $10

Coinbase signup bonus and put that in. Um, if I had put a hundred bucks in, I could have bought a house with it, but you know, hindsight's 2020. Um, I was also on the core team of Constitution DAO where we raised $40 million to try to buy the US Constitution. Um, worked on some projects in the loot ecosystem um, called Adventure Gold and I contribute to nouns as you can tell by my PFP. So, across the ecosystem in a bunch of different ways. First, let's start with the fact that app chains are already big. So we already have examples that are live like WorldChain for example which offers block space for humans and not bots. We have uni chain

by unis swap which centers very heavily around dex trading. Hyperlquid of course um which is a bit of a complicated example. We'll talk about that one a bit later. And then most recently here um Robin Hood which is also launching their own app chain for Robin Hood's tokenized shares. And there's something really important to distinguish about app chains. So we at Syndicate define app chains as a chain centered around an app. That sounds kind of obvious when you say it, but that creates a distinction between app chains and general purpose roll-ups like base where they're not centered around a specific app and community chains um that might

be centered around a memecoin or tipping token or things along those lines that might have a custom gas token and some token mechanics but not a marquee app. App chains really have a marquee app. The chain activity centers around that. We find that that tends to be a very successful model especially compared to the nth general purpose rollup or community chain. Um why are people launching app chains? Why do we have all these examples coming up? Why did Robin Hood just announce an app chain? Well, if you look at the data, it's very interesting what it reveals. So app chains and rollups in general are a $100 million plus pie. there's already

substantial profit, not just revenue but profit being captured by these roll-ups already. Um, if you look at the data, base for example is the most dominant um uh uh beneficiary of this increase in profit. And there's this excellent dashboard by Caunch on Dune where you can see that base alone has made $75 million all time. And then the pie is spread out amongst everyone else. If you even look at um the trends too, they're very strong. Um of course there was a massive runup during the memecoin boom um especially um when there was a ton of memecoin activity across Salana base arbitum and others but the appchain and

rollup revenue is way higher than a year ago and it has stayed relatively strong. I'd note that the majority of this revenue right now is general purpose chains, but um there are a mix of chains um already, but if you look at the data for the last month, it's still very very heavily weighted towards general purpose chains. These general purpose chains are capturing a lot of profits. These are all data on profits, not revenue. They're capturing a lot of profit. Um, and that profit is profit that could be going to token holders, could be going to community members, and could be going to apps. So, every app chain is seeing, why would I pay millions of dollars to someone else when I can have the

benefits of having my own economy? Um, and you might be thinking, okay, this like hundred million plus dollars in profit, where is it coming from? So, 80% of it is coming from ME. Most of that is related to DEX swaps. um and similar trading activity. Um essentially um people are paying for priority fees um so that their transactions can be at the top of the block. Dan Smith has this great um analysis where he basically goes through the revenue sources for base. Um and essentially the vast vast majority of that $und00 million is me and the vast vast majority of that is specifically DEX trading. When you look

at that data, uni chain makes a whole lot more sense. And the other 20% that's not me is mostly fees paid by apps and users. So the apps themselves of that $100 million pie about $20 million of that is paid by apps and users to the chain. Um, and that is money that could be going to the apps. So, let's say you're someone like Robin Hood and you're thinking to yourself, okay, like this is a huge pie and I want to make sure that this can benefit Robin Hood and its users. One thing that you want to do is you want to have a really nice experience with gas subsidy. You want to have a whole managed interface. You

probably want the wallet to feel very, very usable. So, you probably want to subsidize gas. And if you want to subsidize gas, it's important to note that small spikes in traffic actually cost you a lot. So this is some back of the napkin math I did where if you submit roughly 12,200 transactions, you're looking at $1 gas fees on a chain like base. And that of course at Robin Hood scale is not very much. Um it is it is very very easy to hit that. And um I have an article that describes this in more detail of why these dynamics are the case. But the main reason why this is happening is EIP559.

So the EVM itself is limited in the number of transactions it can process when it gets too many transactions. Under EIP559 fees go up. So if any block which um on most OP stack chains for example is 2 seconds um if any block is uh over 50% full um so starts to hit these transaction thresholds gas fees go up by up to 12.5%. You play that out. You look at the chart. Gas fees keep going up.

That's why it's stepwise. That's why you see those jumps. Those jumps are as each block increases fees. and then uh bam, you're paying $1 fees. So, you might be offering, you know, subsidized trading like Robin Hood. You're not going to be able to pay $1 fees to chains when your transactions cost like when your transaction when your trading costs are very low, but your transaction fees are very high. The economics just don't work. Users expect gas subsidies, but these gas subsidies get expensive very, very quickly. And you might be thinking, if you're someone like Robin Hood, why not recycle gas subsidies over yourself?

You have this great ecosystem. One of my favorite memes. Um, it describes a lot of DeFi. Um, but, um, uh, you have these fees flowing into your chain. They'll flow into your wallet. You can then recycle those back and keep subsidizing users indefinitely. If you're on a general purpose chain, those fees are flowing to the chain operator. Maybe that's Optimism, maybe that's Arbitum, maybe that's Coinbase. and then you can't recycle those fees back. You're stuck paying in some cases, especially when you hit scale, very high transaction fees to someone else.

And if you're an app chain, you might be thinking, why leak the revenue to someone else? Why would you pay millions of dollars to another group for the privilege of hosting a chain when you can host one yourself um much much more cheaply than the gas fees would cost you? And more revenue is nice. Gas subsidies are nice. But there's something bigger here too about why groups are pursuing app chains. The second major reason in addition to the revenue available is greater economic control. So you see groups wanting to tune their chain economies to match their apps. World Chain pretty much just last week launched priority block space for humans where um they put humans that are world

verified ahead of bots in their blocks. So humans get priority. That's a really nice example where they're changing the economy and how it works to match the goals of world. Um and if you can customize the sequencer, you can create more value. World becomes more desirable. World chain apps start to have an advantage. there are more reasons to get world verified if you can customize the sequencer. Um, and there's a bunch of different sequencer customization options that you might want to choose. Arbitum, for example, has time boost where it will selectively delay transactions and you can pay to speed them up. There's initiatives like Cicada which are used for private auctions. So maybe you want to run a

totally private auction for block space and then whoever wins that auction um is the one who gets to build the block. you don't need to do like an effective off like you don't need to do public bids. You could do private bids. So if someone's willing to pay a lot more, they they they they're able to do so. You could even sell control of a sequencer. Um Jeff Strong at Stanford um he has a uh initiative called contestable control. It's an idea where basically if someone believes that they can run an initiative more profitably, they could buy the rights to it and run it. And then someone else can buy those rights from them. Um, and maybe you're not, you know, auctioning off every block. Maybe you're auctioning off the

sequencing rights until you're out bid. And that might be days, weeks, months later if someone comes in and sees they can do a better job than you. Um, there's even some really, really funky examples of this. So, one of my favorite smart contract artists is um he goes by Steve P on Twitter. Um, Steve puts together all these weird smart contract designs ranging from negative claims on ETH after a hack to like things that are, you know, a Ponzi scheme that's called, you know, like Ponzi scheme um to like um all these all these cool NFT projects like a collection called free where you could go through a set of um of of challenges and all the NFTs are

free to mint but you need to solve the challenges. um he uh worked with us um to develop some sequencing module ideas as part of a fellowship and he came up with a list of seven really cool ideas. They're all on GitHub. Um I can share the links later um and would encourage you to check them out. One of the favorite designs that I'd call out is a design he made called nuclear war where you essentially instantiate nine sequencers and then those nine sequencers can disable the other sequencers but in the period where the sequencer is being disabled in this 12minut period the other sequencer can disable you back. So it's mutually

assured destruction um just like the current worldwide nuclear doctrine. Um, and uh, if everyone decides to destroy each other, then you end up with a chain with no sequencers left. Um, just like what could happen in actual nuclear war. Um, and these are examples where like sequencer customization is super super deep in terms of where it can go. And it goes from, you know, the like obvious and uh useful and cool like the world chain priority block space for humans um all the way to the really weird and wonky and like experimental. There is a lot of untapped design space here. Um another thing that you can do with

sequencer customization is sequencers can guarantee any requirements. So this ranges from initiatives like for example dream chain that we work with where only so in world you have humans who are allowed to participate in dream chain only AI is allowed to participate and no humans are allowed as an example of a requirement you might want to guarantee that someone is a verified AI agent. Um you can also do things like um guarantee that randomness is included. So we're working on a game right now. Um there's an architecture diagram that you can find on Twitter. It's a little hard to read here. Um but essentially what we can do is we can create transaction bundles where every single transaction

to a game must include randomness in front of it. So the sequencer basically says that you cannot transact with a game unless there is random data right beforehand. And this means that you don't need to set up like a complicated randomness oracle. You don't need to add latency in multiblock randomness, multiblock VRF. The app can rely on an always up-to-date randomness source that is secure and verifiable because it is enforced by the sequencer in a smart contract. Um, this is uh not a hypothetical example. This is an example we're bringing live. You will be able to be the first ones to sign up for it.

We'll talk a little bit about that um later on. And um in short, if you control your own economy, whether that's customizations or new revenue mechanisms, um you can create more value for users and token holders, there's a lot of money being left on the table with the pursuit of priority fees for deck swaps. There's a whole lot of things that could be more valuable that make an app more useful. And the sequencer design space is a huge, huge, huge untapped design space. And this is kind of shocking to me because the sequencer is where all of the value flows. It's where all the transactions are going through. And why are we not customizing it? Why why is this such an

untapped potential? One reason why is because sequencer customization used to be hard. Um, if you look at WorldChain, for example, they announced Priority Blockspace in October and they launched it last week and they're very much a pioneer here, but they're a really well-resourced group and it took them with great partners. Um, and it took them a while to get this done. It's it's not easy right now, but um our hunch is that if you take all of the sequencing activity that lives in off-chain centralized sequencers and moves them onchain um you can make this much much simpler because all customizations

become smart contracts. It's like 50 lines of code instead of entire thousands of lines of client customization. Um, if all sequencing is a smart contract, it becomes transparent or verifiable. You can guarantee ordering. You can inspect the prerequisites like randomness injection. Um, and you can customize with one command. Maybe you want to try a sealed bid auction one day and maybe you want to try contestable control the next day. You easily can do so. Um, and these fees can flow anywhere and including in ways that token holders can benefit. um you don't need to worry about um you know are the fees accountable. If you encode it all into smart contracts which are a

great primitive for moving money around then they're accountable from the start. Um I'd note that there are also centralization risks um in app chains right now. So um with Hyperlid which is very very centralized um you can't export more than 10,000 transactions. the data is just gone which means that if you're trying to do something like your taxes um it becomes very very important to have this data and centralization can have downsides. Um base also um they send the fees to Coinbase to custody them so people don't know where they go. Um Bass says that they keep the fees and I I I I think

that the team I I trust the team. I like the team. I think that um the team I I I I think that the team is portraying this accurately that um you know they are a substantial holder of ETH and a contributor to the ETH ecosystem. But the security motto is always trust but verify and can we verify? Um this centralization makes it quite difficult. Um and that leads to the question of how might we hold roll-ups to a higher standard? How might we demand smart contract level transparency? um and how do we prevent so much of crypto from recentralizing as more of the ecosystem moves to rollups and more of the ecosystem moves to app chains. These are very very important questions to solve.

And I think that if you start to encode more and more of this in smart contracts which are primitive for transparency, for verifiability, for control by token holders, you end up in a spot where sequencers can decentralize. you can have a world that is closer to Vitalik's original vision where he in his guide to roll-ups back in January 2021, way before the roll-ups were even close to being live, he sketched out a bunch of different sequencer designs, ranging from total anarchy, where anyone can sequence at any point to a smart contract to the centralized sequencers we know and maybe don't love um but we know them. um to uh proof of stake and

sequencer auctions and elections and things along those lines. We've only tried one of these methods 4 and a half years later in terms of what he's laid out for roll-ups and there's a huge huge amount more that we can do to make this better. Better world does exist. We really do believe it does and um it's not theoretical. You can try it yourself. Um, so you might have interacted with app chains before, you might not have. Um, if you're here, you can be the first one to interact with a new one. Um, that randomness example that I talked about earlier where there's um, uh, sequencer injected randomness for games. We put

together a game to demo this. Um, and what you can do with the game is you can sign up and uh in the background you essentially have a you have a ship and the ship is lost at sea and you need to keep repairing your ship so that it does not sink. How good your repair is is determined by the sequence of randomness. Maybe your repair lasts you a day, maybe your repair lasts you an hour. You need to make sure that the chip is staying in top shape. Um, and then there will be prizes for the last ship standing. Um, and all of this becomes much much simpler to build and fully gas subsidized, fully embedded.

The app chain experience becomes very easy. This is the very first time we've shared this URL. So, um, all the addresses you see are just the devs on our team. Um, and and everyone here. And, um, I think that this leads to an important point. The more we can make apps better, um, the better the whole ecosystem will be. And remember, wait, it's all about the apps. It always has been. Thank you. [Music]