Fireside Chat with Ben Jones and Robert Miller

[Music] We are ready for our last talk for Pragma New York. And without further ado, I'd like to bring on Ben from Optimism and Robert from Flash Boss to talk about the future of rollups. Please give them a big round of applause.

That's one too. Yo, Scoo, give an applause to us. How about for Karthik and the rest of the ETH Global Pragma crew for an amazing event. Wow.

Hell yeah.

Wow, that was pretty hype.

We got to be hyped. I mean, I think they should have made David's presentation the last one to be honest because that was the most bullish I've seen in a while.

We really needed that about two months ago. We're feeling pretty good now. Yeah. So, Ben and I are going to be having a conversation. I'm going to be asking him a couple questions.

is going to be asking me a little bit. Why don't we just start off? Uh Ben, can you talk about what optimism and the super chain is?

Heck yeah. Um so, you know, David was just here up up here talking about the importance of ETH being a really resilient, you know, grandchildren, great-grandchildren kind of a network. Um but it's also the case that to bring crypto to the masses, we need to scale these chains. What that means is that we need to be able to handle all of the demand for block space for sending transactions on chain that everybody has. So what optimism is doing is building what we call the super chain a techn and a technology stack called the OP stack that allows us to build more chains.

These layer 2 chains that are themselves extensions of Ethereum. So they use this stuff called being a rollup data availability. whatever whether all those details matter for this conversation probably not. But what is important to understand is that they inherit the security of Ethereum. So they allow you to spawn more chains that borrow security from Ethereum but increase the throughput so that we can satisfy the demand of the global onchain economy.

So folks like B world uni chain uh many many others in the superchain are using the OP stack to scale block space.

And what's the superchain? So the superchain is just the term that we use to refer to the network of aligned chains that are participating in this scaling journey together. Um so it's basically like the op stack is the word for the technology the just like open- source MIT licensed like codebase that anyone can go and do something with and then the super chain refers to the set of people that are building using that technology stack that are actually building real chains that are like live on Ethereum in production running. There's a re recent trend of closed source software within crypto. We have a bunch of closed source smart contracts getting traction on Salana.

We have a bunch of closed source sequencers these days. But uh optimism and the super chain have leaned in super hard to open source. Uh why is that a competitive advantage for you or for any of the chains that are using the OP stack? I think in a lot of ways actually it probably echoes what David was just saying about EVM network effects. Um but basically the philosophy that optimism takes to scaling crypto and I think the philosophy that we should be taking as an industry is that we need to build the version the plat as close to the platonic version of the thing that we know should exist.

Not something that we think can exist and make us a lot of money but the thing that should exist and should be the long-term dominant play. And if you look at uh the the precrypto web, if you like look at web 2, a lot of what you see and a lot of things being closed source in web 2 are a result of people looking at some market opportunity, realizing that they can build a closed source product and extract from that closed closed source product to make money. That is not that is a only a local maxima. It is not the global maxima of technology. So, if Blockspace is going to be everywhere and everyone's going to be running these EVM chains, there should be an open- source free do whatever you want with this codebase to make a chain piece of software that should exist.

And, you know, you could argue, oh, you should make it closed source, slap a slap a license on it that allows you to charge some percentage of transaction fees and go off to the races. But, it's our view that the positive network effects that will come out of that outweigh the short-term gains you could make. So the fact that everybody building in the super chain is contributing to the same codebase and making it better. People are coming along that don't want to pay us that just want to use the software are also updating that software and making it better. All of that value can return to the stack and make the OP stack the platonic way to scale blockchains.

And what is the role of the OP stack and roll-ups more broadly in a world where uh increasingly users and roll-ups are kind of bringing their own asset and so the thinking behind the question is that uh originally a really important part of roll-ups was that you had a way to take assets on ETH mainet like USDC like ETH bridge it onto a rollup and then you use those assets in a really secure way in a way that inherited the security of L1. But if you're a fintech company, you have all these customer deposits already. You have all this USDC or this USD already. You can just bring that onto your chain. You don't really need the bridge to ETH mainet.

You don't really need those same security guarantees. So what what's the role of a rollup and the OP stack in this world uh increasingly of fintech companies launching their own chains?

Yeah, definitely you're right that the writing's on the wall. I mean, I loved it. I guess for those in the audience you can't see but we are looking at a beautiful vista of Manhattan across the water right now and like looking at the financial district and I think David was right to say that like we are waking up but great choice caric on this venue space my god you're welcome so you know I think um quite frankly I think as an industry we are learning to undo some of the tight coupling mistakes that we made in our past. And basically what I mean by what I mean to say by that is that you're right. We we built L2s and part of the reason to build an L2 was to make it very amendable to bridging.

And I I think that's actually maybe a slightly different framing than we used to think of it back in the day. It's like an L2 is a bridge, but there are actually two different things going on here. There's the notion of a chain that progresses transactions and like updates state in a decentralized accessible way. And then there's the notion of bridging, which is to take some asset in that state and move it to some other state that is in the same chain. Ultimately, I think what we're going to come to realize is that both of these things are valuable in different contexts.

And the security properties of both of those things can be different and valuable in different ways. So at the end of the day, the reason that all these centralized entities still have an incredible use to be found in chains is because there is value in having a decentralized open ledger that anybody can come and transact in and participate in even if that is a very centralized player doing very centralized things. Its ability to publish that data and make that open is relevant to what they want to do. Additionally like uh there are a lot like the the nature of the EVM for example is so composable like anybody being able to deploy a smart contract. I think fintech and like traditional finance are only really just waking up to like how big of a deal it is going to be.

Like I can't imagine going to those people over there in FIA and being like I've got a pitch for you. you're gonna let any user in the world run arbitrary code on your servers? No shot. Right? That's like a nightmare.

That's like such a foreign concept that that might even be possible um in the world of traditional finance. And so for that reason as well, chains are useful is because they give have a new substrate where anybody can run applications on top of a chain and have a much better um more composable experience and like build apps on top of the primitives that these institutions offer. Anyway, so I'm getting rambly here, but the point is there are value in chains. And it may be the case that the the really decentralized bridging that we set out for L2s is really useful for ETH because we want ETH to be around for our kids, great grandkids, etc., but is not going to be as useful or as relevant to assets that are essentially issued.

Does that mean that there's no value um to be provided by Ethereum and L2s? I don't think so.

And that's interesting. What uh so you're alluding to sort of the value prop for a fintech company of having anyone be able to run arbitrary code, build all these financial primitives. Do you have anything in mind that you feel like uh fintech companies and others deploying chains should be exploring that they shouldn't? Is there like a use case that really excites you or a feature someone should build that no one is

like that it should be but aren't?

Yeah. Yeah. Yeah.

Um I think

because we're doing the obvious like put your assets and a DEX on a chain and let people trade it or whatever. What's the non-obvious thing that you think people should be thinking of?

I think it's a give and take. I think like the beauty of Ethereum is that like Ethereum wasn't really able to perfectly answer that question at the outset either, but it knew that allowing anybody to write an application on top of this platform is going to be a really really really powerful tool. And you know, a lot of things in Vitalik's white paper totally came true in practice. And like we have Polymarket in the prediction markets, we have a bunch of Dows, we have a bunch of finance. But for example, we didn't realize just how crazy powerful that finance was going to get.

And nobody like read the Ethereum white paper and immediately went, "Oh my god, the money Legos are here. Let's start pitching these fintexs." But as time goes on, the those use cases became clear and it worked out. So yeah, I don't know that I have some like golden goose startup opportunity for someone in the audience to immediately run up and grab and like this is the this is the reason to be doing all of this. Um, but I just think as time goes on, as people continue being like given the opportunity to build applications that interface more closely with the block space that these institutions bring to the table, I think we'll they'll pay a lot of dividends and we'll see a lot of really cool stuff that just like with the start of Ethereum was hard to predict.

And you might give me a frustrating answer to this of the same, but uh uh how do you see these chains differentiating and competing? like are we going to have a chain per fintech company that are all just you know the same uh superchain standard config or are they going to be differentiating and competing on interesting dimensions

I definitely don't think in the long run we'll be competing on technology we might compete on the usage of technology but again our platonic goal for the OP stack is that it does everything and it is this perfect open source codebase to do whatever it is you want to go and do and I think you know if we look at the internet that's a pretty likely scenario to uh it's like pretty clear that that's what played out in the internet too, right? Like um no, there's no SAS startup that is selling themselves as differentiated because of the database that they use or the network transport layer that they're going with or whatever, right? Like none of these things are really relevant. They're like building products and they're like, "Yo, we have a sick product that you can pay us monthly for." That's where they get the money.

That's where they differentiate. So, I think we'll see that um in in Web 3 as well. There'll definitely be technological differentiators. um that as we continue to grow like have temporary advantages and there will also be like very useful technologies that plug into differentiators. So like we'll unlock new technical differentiators that unlock new product differentiators and then we'll just end up seeing that technical differentiator that unlock that getting super standardized and at that point I think at the end of the day we're competing on what we should be in in industry in the in the world which is on products that help people and so everyone will compete on providing utility to humans and that is the most important thing of all.

What does the network topology of this end up looking like? Like does every firm have its own chain? Does every app have its own chain? uh do you see a role for places in the long term like an OP mainet? Where does that fit into your vision of kind of how this this develops?

Yeah, I mean the whole like is the future of app chains or one giant single chain the future is like a very interesting question and it's going to be hard. It's going to be answered naturally over time. It's like hard to predict. I think what we can say in broadstrokes is there will probably be both. There will be use cases that are very centralized that are relatively siloed that can get away just running on their own chain and it will make the most sense from an economic perspective to be running on their own chain.

There will also be use cases that want to be applications that compose with and leverage multiple chains that are kind of like independent or like owned by the institutions. And if you want to do something that spans all of them, it makes more sense for that application to be hosted on a shared piece of infrastructure like something like OP mainet that is a more neutral piece of block space that gives a kind of shared, you know, level playing field for certain kinds of applications that want to compose across many other chains um to work on. So yeah, I hope it's not a non-answer, but that's kind of the nature of the beast, I think, is we'll have a bit of both. But wait, if the idea, this is getting a little in the weeds, but if the idea is that you can deploy something that composes across multiple chains and that's the reason to be on OP mainet, then why not just deploy on your own chain and compose across multiple chains? So, what's the value of OP mainet there?

Yeah. So, the I mean this is this is very in the weeds. I think that the real answer comes down to the actually not like the chain that you're deployed on but the governance structure of the chains that your application is running on which is like a little similar but a little bit different. Um, basically what I mean by that is I think ultimately they they're going to be a bunch of chains and a bunch of apps running all over them and the centralized apps that have big central points of control will have centralized components to the governance of those chains and will like allow the people that are interested in making those chains better to like perform upgrades to like make improvements to that chain relatively unilaterally. So it's not about um them creating one chain or the other.

It's about them deploying on some chain of which opinet I think is a good example that has a more neutral governance system that that application developer couldn't create themselves. It's like they want in the same way that an appdev deploys on Ethereum allows them to credibly limit their own control of the application. It's like only the rules of this smart contract can be followed no matter what I as the application developer do which is very unusual. Um, it's it's kind of like that but for crosschain stuff.

And what uh Sorry, were you going to say something?

Well, I want to play it back to you and ask you about me because that's something that we haven't talked about and you are one of the foremost experts in um along with your organization. So, I don't know. ME is a whole other part of the story that I wouldn't want to jump into without knowing what the heck me is. So, Robert, what's MEV, bro?

ME it stands for maximal extractable value. It's the value that you can extract from transaction ordering. Basically, having the right transaction at the right place at the right time. If someone buys a bunch of assets on uh one exchange, that might mean that there's an arbitrage opportunity between that exchange and another one. And so, MEV is the money that you could make by arbiting between two different exchanges.

And um it ends up being kind of a a really deep and important problem in how these chains work, how um you know people who are interacting with them and trading with them, how good of outcomes they get, what uh kind of negative externalities are left by the games that people play to capture me. Uh and it it I think is a really important part of the story for why crypto will end up being successful or not. Yeah, I think one like mental model for MU that's like very oversimplified is just like transaction ordering matters and like if there's a transaction that anybody could make that makes them a little bit of profit like that arbitrage, who ends up being able to do that is like very relevant and the economics of that are super relevant. I guess like my mental model for like the MEV landscape today is like basically there are these validators that are actually producing the blocks and they basically expose an API where a bunch of biders can say hey this is the ordering of transactions you should have another says hey this is ordering transactions you should have and the validator kind of looks at that says okay who's going to pay me the most here I'll choose that ordering um I think the there seem I my suspicion is that there's something deeper and like nowhere in that story was the application developer and this seems like very relevant to like will there be app chains is like me for app chains is now a thing but me for apps kind of is a thing today that we just don't think about anyway that's kind of a ramble but do you have like thoughts on that am I speaking crazy or does that make sense

well so uh a lot of people will deploy apps onchain and then they will discover that their apps create a lot of value that is leaked to these sophisticated parties that run bots or to validators and so uh we've got our link Inc. Marines in in uh the the crowd here somewhere. Uh we work with chain link as an example. Their oracle updates create um opportunities to clear out collateral from lending protocols and these can be worth millions of dollars sometimes. Uh two weeks ago, we had an Oracle update from Chainlink that went through our infrastructure that got them uh paid back $100,000 in a single transaction because there was a lot of money to be made from having the right transaction placed in the right right uh place at the right time.

And Chain Link wants this to go to Chain Link or or they want to redistribute it to their partners in some way. You don't want to just leave this money on the table. But um it is hard and kind of complicated uh uh to recapture this value uh as an application and so people work with uh projects like flashbots or some other people to try to enable applications to capture the value from transaction ordering that they've got. Um one way that you can do that is by using our infrastructure on Ethereum mainet. Like I was saying, chain link had this $100,000 uh repayment of me that otherwise would have been left on the table as an example.

Another way that you could do that is by launching your own chain. Um if you think that you can get a lot of usage and a lot of of users, then uh kind of naturally if a lot of fees are being paid from transaction ordering, you can capture that if you're running the chains sequencer or or its validator. I think it's a very interesting question for the long term in the industry of sort of how as an application developer do you monetize or do you want to monetize? Are we going to force everybody to launch their own chain? Maybe that's the equilibrium in the in the end and and that's how these things work.

Or do you end up um having some some other ways where uh you know just by nature of deploying an app on somewhere you receive all the value that it creates back to you. Uh, and there's four or five different experiments that are ongoing and nobody's sort of cracked the code exactly on how to let apps monetize from the value that they're creating. They're mostly in the history of crypto have been leaking so far. And when I'm saying value, I also mean this is in the order of hundreds of millions, even billions of dollars annually, too. It's we're we're not talking small amounts of money.

Yeah, it's huge. And I mean, I look back to a few years ago and I almost think like there was a period where me had like kind of a bad rap in some sense. So there's like a negative connotation there like everyone was getting sandwiched and my slippage is worse and my trade isn't executing at a better price because these bots are going crazy and paying the validators these little bids to change the ordering. It sounds like I'm hearing but I mean just caller if you agree or disagree but it sounds like you're saying me is not necessarily a bad thing. It's just a I guess you would just call it a property of these systems like like I guess the reason I asked this is because you didn't mention a future where there's no more me and I feel like there was an era of like we need to kill the ME.

It's like this is death to me. Do you agree with that? Me

is just the money that's left on table in having the right transaction at the right place at the right time. It's kind of the economic engine behind these systems and why they work. And it's always going to be there. But I our task is to develop systems that channel it into positive outcomes for users. So there's all sorts of dexes these days that use me to give users a better price.

And I think in in the long run, it's that kind of system that ends up deciding whether or not we outco compete tradfi. Why is anyone going to trade on these chains? Why is anyone going to use them? It's because they give them better outcomes than traditional systems do. And uh in Tradfi, you have these super entrenched market structures.

They're extremely hard to change. It's extremely hard to kind of experiment with new ways of routing value. Our opportunity in crypto is to try a lot of experiments on what can actually deliver better outcomes for users and uh to use ME to do that because ME is the economic incentive to get people to do things. So uh ME certainly has a lot of bad outcomes. The task is just or it has a lot of negative externalities.

It has in the past. The task is to try to harness that in order to make it a positive thing for users because that ends up being an integral part of why this industry will win or fail, I think.

Yeah, that makes sense. It's like um you know, MEV in in Tradfi is like spending 10 billion dollars to run a new fiber optic cable secretly under a mountain so that you can get 0.001 milliseconds better tradeex. It feels like we have an opportunity here to democratize that a little bit. And just because people can extract from financial systems doesn't mean that there's not opportunity to do it in a better way.

Yeah. Exactly. And I I mean you're always going to have this competition to win. You're always going to have this competition to get arbs. The task is how do you channel that into something that gives a user a better outcome.

And so we really like auctions at flashbots. Uh instead of just letting one way this is manifest is users have their trades auctioned off. So, who can give the user the best price to to execute their trade as an example instead of uh uh sort of payment for order flow market structures that aren't as competitive in in Trafy?

Ask not what you can do to extract, but what extraction can do for your users.

Ben is full of bangers. Yeah,

we got to close it out strong. And I mean, I think we're at time, so maybe that's the banger we ended on.

Thanks everybody for listening. Cheers.

Thanks, guys. [Music]