1 Billion Users Onchain | Jesse Pollak of Base

I think that this is the decade of the world moving on chain and I think that we're gonna see it faster than people anticipate. And so I don't know whether it will be in the next one to two years or the next three to five years, but I really do believe that this is the decade where we'll see it happen. And by the end of the decade, the whole world's gonna be on chain.

Welcome to Bankless, where we explore the on-chain frontier with BASE's Jesse Pollock, co-creator of BASE. We're here to explore how BASE is trying to get to 1 billion users on-chain. We also are trying to get to how BASE is trying to increase its scale by 400x to 1 billion gigagas. We'll explain what that is in the show. How Coinbase is using Base to put more of Coinbase on-chain. And where all of this leads to is on-chain summer round two. We've already done this once, but we're about to do it again. And then next summer we'll do it again, and next summer we'll do it again.

I think this is a cool conversation, David, because uh we're talking about um something that's very important in crypto, which is number one, how we're scaling Ethereum itself. So this is a story about layer twos and how that scales, and especially in a world where we're we have blob space. And also, this is a case study for how a centralized company, uh Coinbase, an exchange, is slowly migrating on chain, what that looks like. I think it's a fascinating case study and why folks should tune into this episode and be following in general the base story.

I think the most interesting part of this conversation was at the very, very end, just to tease the end of the conversation, when I asked uh Jesse whether or not uh Bass would ever spin out of Coinbase, which was a question that was a means to an end of something bigger, but I'll leave that up to the imagination of the listeners. So let's go ahead and get right into the conversation with Jesse Pollock from Bass. But first a moment to talk about some of these fantastic sponsors that make the show possible.

Bankless Nation, we have Jesse Polak. He's the create co-creator of Base, which is the Coinbase layer two. Jesse, how you doing?

Doing great. Happy Monday. Hey.

Hey, it's great to uh have you back on Bankless. You know, we were just checking some of the stats on BASE recently, and we thought we we have to have Jesse back on to tell us what's going on. And uh a couple stats that come to mind. Base is now number three in terms of layer twos by uh total asset value, so total locked value at like 5.2 billion. In terms of uh daily or weekly active users, it's trending at 1.3 uh million. You've got um fees that are you know between like five cents or under on base, and I think base launched in August of uh 2023, if I'm recalling correctly.

Give us a summary. So how have things gone so far and how uh far in the journey are we?

Yeah, yeah. Well, we're still at day one. This is what I tell my team every day. It's like it's still day one. We have so much more to build. Uh there's there's so much more work we have to do to bring the whole whole world on chain. That said, I mean, we're about eight and a half months in right now, and I think uh the team and I am are pretty blown away with the amount of progress that we've made. I I kind of see this in three general kind of phases of the last nine months. The first phase was on-chain summer. Um it was so exciting. Uh we saw you know millions of people come on-chain on base, so much energy. I think a really like great way to kick off the the network and the economy and get a lot of that kind of culture that we wanted to make happen kind of thriving. Um, and and there's a big boom in metrics and uh all of the kind of uh outcomes that we were looking for. Then I'd say after on-chain summer in the kind of fall, we basically saw things kind of like tail off, flatline a little bit, like a little bit of trough of sorrow. Like we had this big launch, it was really exciting. Now like we're we're met with reality. Uh our team was also candidly just pretty exhausted. Like we'd pushed so hard for a year to get that thing out, um, worked so hard during on-chain summer, and took like September, October, November uh as a little bit of a reset where it's like, okay, let's get our like equilibrium back. Let's also think a little bit broader of like, okay, we made this first push. Where do we go over the next one to two years? How do we really crystallize our mission, our strategy? Um, I think we did a bunch of good work there.

And then I'd say the third phase was really like Q1 of this year, which we s we entered the year kind of in the top 10 by activity and TVL. And I think if you look at the metrics now, yes, we're third in terms of TVL, which is a very like lagging metric, I'd say. And there's people who've been building for three years, and so it's gonna take a long time for us to catch up there. But if you look at daily actives, DEX volume, uh, you know, developers, at this point, base is the largest layer two in the world. Um, and most days were uh you know competing to be the largest chain in the world uh after Ethereum. And so that I think is pretty remarkable. And there were really two catalysts that I think we we really saw kind of driving this in Q1. One was uh the launch of Farcaster Frames, which I think um you know happened in January. We really leaned in, built a ton of developer tooling, and as a result, saw our market share of like Farcaster activity go from 15% to something like 75%. And I think a lot of the best builders are on Farcaster, and so that was this catalyst that kind of got us uh having more and more of those builders building on base. And then I'd say the second big catalyst was EIP 4844, which we'd uh worked on for two years starting in 2022, even before base was like kind of a twinkle in my eye. Um uh and we landed and were there on day one with support for. And I I feel like there was this kind of moment that I don't know if you guys remember that day, but I don't think our team will ever forget it. It's like 484 landed on Ethereum, base upgraded to support it, and like the internet broke. It's like every single post in my Twitter feed was people being like, oh my god, look at how cheap fees are on base. Um, and that I think kind of like we'd we'd we'd been flat, we saw a little up curve from the Farcaster stuff, and then post 4844 went kind of vertical in all of the metrics that we we care about, and um

That has kind of been the story of Q1. And now I think we've we've established ourselves as the market leader, the place where everyone wants to build. And I think it's mostly about okay, how do we make sure that we're further solidifying that that that kind of place, that role, and invest more heavily in our ecosystem and our economy to ensure it can keep growing? Because kind of going back to the beginning of my answer, it's day one. If you look at how many people are on chain, it's still, you know, hundreds of thousands a day, probably, and we have to get to billions. And so that's like a

ten thousand X. Uh so a lot lot more work ahead.

There's been uh just this iterative cycle that um exists before crypto between infrastructure and apps, right? Where uh infrastructure gets built out, apps consume that infrastructure, then apps get constrained by the scale, so then the the scaling goes back down to the infrastructure. It takes the infrastructure a little bit of a while to scale out, but then it provides more scale to the apps, and then that's this is the iterative cycle. And we've seen this happen with um a handful of protocols, a handful of layer twos and also layer ones. Uh Solana is experiencing the same thing, right? Like demand for Solana block space produced some of the growing pains out of the Solana ecosystem. Not too long after 4844 happened, we actually saw the constraints of base. Uh despite despite all of the abundance of block space, there also seems to be an abundant demand for block space. And we we saw some of those those constraints hit the hit thereafter. So uh you've been on this campaign, Jesse, of the road to one gigagas per second. Maybe you can talk about just like why this goal is important and where where what is a gigagas per second? What is that a measure of? And like why why is this North Star important for the base, uh the base vision?

Yeah, absolutely. And maybe just to kind of refresh folks, so on on like middle of March, we upgraded Ethereum with 4844. What that did was it basically brought down the L1 costs for base and other L2s. So now we have this new place where we can store data that's purpose-built for roll-ups. That makes it so instead of paying a ton of money for all of the data we're storing on L1, we pay a lot less. Really incredible solution that I think is going to enable L2s to scale as part of kind of Ethereum's roll-up-centric scaling strategy. And so post that launch, we saw base fees drop really, really low, but then we also saw a massive increase in demand. I think we 5xed demand over about like a week and a half. And what we started seeing on base as a result is now the L1 costs were very low, but because of uh the increase in demand, we started to see actually like somewhat high fees on the L2. And this is because, like Ethereum, base has uh a fee market, which is uh defined basically based off of uh two things. One is what's called the gas target, uh, which is basically you know, what is the target amount of gas per block?

Or per second that the base network can kind of process. If it's processing above the target, fees start going up. And that's the second part of the mechanism. It's 1559, same as Ethereum. They start going up exponentially. If the amount of gas being processed every block is below the target, they start going down. And so it's kind of this like automatically recalibrating system where if there's more demand on the network than the network can support, you have rising fees. If there's less demand, then you have lowering fees. And so we saw more demand than our target could support, which kind of drove up our fees very aggressively. Now, the the thing to know about the target is uh it's a variable, so you can kind of change the target, but if you change the target, it has a bunch of other downstream effects. Uh and

You can't change the target willy-nilly because those effects can be really problematic. And so instead, you need to kind of apply a very principled approach to how do you change the target such that the network can scale healthily. And so we kind of look at that as three primary problems to solve. One is when you increase the target, you need to have more data availability. That's you can actually put all those transactions in. More gas means more transactions, means more storage that has to put on the L1. So we need to continue scaling the amount of L1 data availability through things like PeerDAS and the whole proto or Dank Sharding Roadmap. So that's

This is an Ethereum layer one roadmap question.

Ethereum Layer One roadmap question, but we're gonna keep investing in that. We're working on pure DAS prototype, like we've been contributing to Ethereum L1, we're gonna continue contributing to Ethereum L1.

The the second kind of challenge is uh like how much gas can the actual L2 node process? Like literally, how fast can it run? If we're running two second blocks and we want to process, you know, 10 million gas a second, can we get all of that gas processed in those two seconds or in less than those two seconds so that all the nodes can keep running? And that's just a lot of kind of like uh EVM optimization. Can we do things optimistically? Uh can we change the database format so it works better? So that's the second thing. And then the third thing is um uh and and there's a kind of actually in the second thing, there's kind of a related point around uh hardware. It's like, can we increase the hardware requirements? Um and the cool thing about L2s is that they don't necessarily have the same hardware requirements as the L1 because of how they're designed. So we can, but we don't want to go too far because it makes it too expensive to run the nodes. So that's the that's the the second big thing. And then the third big thing is state growth, um, which is basically like if you run the chain very fast and now you have it running very fast, you have all the data availability on it, then you get this really, really, really big chain. And it's all this data that people have to store, and that causes a bunch of other downstream problems. And Ethereum's been kind of starting to think about these problems and solving them with things like vertical trees and state expiry or state uh kind of like offloading to other parties. Um, and that's a third kind of open research problem that we have to solve. And so, in order to feel confident increasing the target, which is just this variable, behind the scenes you actually need to solve really, really, really hard engineering challenges. And so when BAESE started seeing these increased fees, I think the the feeling for me was like, okay, we kind of need to change our focus and narrative a little bit. We've been just focused on this L1 thing of increasing data availability and lowering costs for two years.

And now we need to kind of like shake the team and shake the industry and be like, okay, the rules have changed. Like now we have a bunch of L1 data availability. You know, we're going to use it, but we also need to focus on EVM scalability. We also need to start focusing on state growth. And so I kind of came out there and said, hey, we're going to start solving this. We're going to set a really aggressive goal, which is one gigagas a second. And that goal had kind of come from conversations with the Paradigm and Ref team, who we've been working really closely with. Huge shout out to them. And if you look at where we started, which was 2.5 million gas a second, or 2.5 m gas a second, one gigagas is 1000 million gas a second. So it's 400x.

Where we started in this process. So we kind of set that really aggressive North Star, and then we basically started working internally to be like, okay, like how do we 400x? And so what we pretty quickly initially figured out was we had enough head space to do a couple increases. So over the next two weeks, we did about a 2x. We went from 2.5 m gas a second to 5 m gas a second. We're now in the process this quarter of trying to get to 10 m gas a second. We feel like there's a pretty clear line of sight there. And then we're gonna continue just trying to double until we get all the way up to a thousand m gas a second or one gigagas a second. Um, and along the way, we're gonna solve or push forward at least a lot of these really hard challenges that are kind of invisible constraints that are limiting our ability to continue scaling the network.

So this is just what it's like to be uh a base engineer. It's like this is the landscape that you have to work with and and the North Star that you need to get through. I remember post um 4844, I think Vitalik released a blog post where he said that the uh the end of Ethereum zero to one moments uh is upon us. Like now that's in the rearview mirror, and now it's more about getting from one to ten. Uh and what he meant by that is a bunch of uh incremental steps rather than a binary like step function of uh like like the introduction of 4844, right? Like a binary improvement in Ethereum. Now it's all about just like incremental steps in like seven different directions that are all kind of slowly multiplying together. There's not one silver bullet here, but now it's just a bunch of incremental improvements in like five or six different directions, which base has its fingers in all of them because it's a network. But the beautiful thing about this is that like if base solves any sort of um you know frontier problem, this actually gets to be scaled out for the rest of the Ethereum layer two space, right? Like so like you guys are are building for yourselves, but also for for the rest of the industry.

Yeah, and I think this is really like one of the most important things about how we've decided to build base. It's like all of our contributions to Ethereum L1 obviously impact the whole industry, but we also decide to build on an open platform in the OP stack. And so all of our contributions to the OP stack also really help the industry. And if you look at the current OP stack core development process and outcomes,

The base team just led the most recent upgrade. We did almost most of the work. We did the governance proposal. And we're also leading the next upgrade, Fjord, where it's more scaling stuff. We're going to decrease L1 costs 10 to 20%. We have a bunch of other really positive changes going in there. And all of those benefits, they accrue to any other network that's running on the OP stack. And I think that that's been really essential to us from the beginning. One of our values has been decentralized and open source. And it's because we believe that this really is a collective effort. And the more we can contribute to things that accelerate us but also accelerate the industry, the better outcomes we're going to get for everyone building on-chain.

You you said with the uh introduction of 4844 there was just a marked increase in the consumption of of of base transactions, just and this was more or less uh expected. We were all chanting the chant of induced demand ahead of the 4844. But it also seems to be that there was actually more than some induced demand. This is the kind of my my gut take. Do you know where that demand came from? Like why were people doing extra things on base? Like what w what was driving some of that that demand?

Yeah, it was it was definitely more than I think we anticipated. I think if you look at the cross the whole industry though, it it maybe wasn't like it wasn't more for everyone. It was just more for base. Um and that that you know, there's some exceptions to that rule, but like that there's a reason why base is now you know processing twice as many transactions than Arbitrum or other networks. It's because there's been more transactions that have moved to base than elsewhere.

I don't have like a specific answer for you there of like, oh yeah, it was this specific thing. I more think that there was something about like the combination of the developer go-to-market we've been running and the timing for our launch and just the moment of all these pieces fitting together, where when people like kind of saw this 4844 thing happened, they projected it onto base the most. And they were like, oh, this is where this is unlocking us to build more better things. Um I think if you look at the um kind of builder metrics, that's what you see. Like one of the metrics that we look at the the most is um uh what we call like revenue generating deployers, uh, which is basically like uh how many people deploy smart contracts onto base and other networks that go on to generate meaningful revenue. Because it's basically a measure of like.

Substantive high quality developers. And if you look at the graph, it's like we were flat, and then the Farcaster frame stuff happened and we did like a slight upcurve and started growing. And then 4844 happened and we we started going through the roof. And today has more revenue generating deployers than every other L2, than Polygon POS. And I think we're about across Ethereum. And that has just been like up and to the right. And I think a lot of that is basically the moment and the like convergence of a bunch of different factors that caused a lot of developers to come on chain for the first time on base, as well as a lot of developers to refocus their efforts onto deploying things on base.

I I really like that that metric of revenue uh generating deployers. Like you like one framing for this is it's basically entrepreneurs, it's basically small businesses that have decided to uh set up shop in this new area. We've often talked about Ethereum in the context of it's Manhattan, it's busy, it's grown up, it's really expensive, right? And now with these these layer twos, uh, we've got new suburbs that are opening. And and when you're talking about revenue generating deployers, you're just talking about like businesses popping up. There's businesses, there's restaurants, right? Uh there's movie theaters, there's all of these small businesses popping up. That's very cool. I want to ask some kind of like dumb questions, maybe from a user perspective. No dumb questions. Uh for you, Jesse. Okay, so like maybe some obvious questions to you, but you're talking about it from a system perspective. I want to look at this from uh from a user perspective. So uh it cost, I believe, right now, as just looking at the charts coming to this, about one cent to transfer ETH on base. Like that's the cost for a user for a transaction. If I want to send you uh some ETH, Jesse's gonna cost me one cent. What was that prior to EIP 4844?

Yeah, um I I think like our media I was just looking at the metrics. I I feel like our median fee decrease was something like 80 to 90 percent. Um and so that makes it so like the you know, but that actually impacted like higher gas usage things more than the else uh e sends. And so I think that that would have probably been in like the 10 to 30 cent range basically for doing an east send.

And then what do you want it to be? Or is that an impossible uh you know question to answer?

We want it to be as low as possible. I think our kind of like internal North Star across all of Coinbase is one second, one cent, which is we want uh average and median transaction fees to be less than one second, be pro less than one cent processed in less than one second. Um and that's that's just kind of like I think for us a thesis around kind of like the on chain broadband moment, where it's like if you have cheap enough transactions and fast enough transactions, anything is possible on chain. Um and so yeah, we want to scale to probably another 10,000 X in terms of the number of users while keeping those kind of costs and speed down subsecond, subset.

Okay, so you you're talking about the the three uh buckets of of cost that go into a transaction. So of that of that one cent, you know, how much of that is that first bucket that you're talking about, which is uh Ethereum uh layer one versus the the other two buckets?

Yeah. So today most of the cost is actually not Ethereum Layer One because of the uh kind of usage of the L1 gas market. Um like we aren't using all of the blob space. And what that means is it means that blob fees are very, very cheap, essentially free. Um I think that that will change in the next few months. Um we'll start using that, and then we'll have a market based uh fee which will bring it back to probably something like you know five to thirty percent of the L1 fee. Um and then there's kind of the like L2 market based fee, um, which is like depending on how much the target is uh and how much demand there is, there's gonna be higher or lower fees. And so that will make up the rest of the fee.

The way we've described VIP 4844 is it's almost like another lane that's that's uh opened up, like a high occupancy vehicle lane. And right now it's it's cheap because it's not full, right? But but that could change in the future. Let's say if that changes and that HOV lane becomes uh expensive uh yet again. So Ethereum blob space is is full and all of the layer twos are kind of competing for it. Would you ever consider using a uh non Ethereum DA DA solution? Because there are some of those springing up, and that could be a source to reduce fees as well for BASE.

Yeah, I mean, like, look, we think pretty expansively always of like what are the different options for for scaling. I think our current perspective is there's a very clear roadmap for Ethereum data availability scaling that just needs to be executed. Um, like PureDAS is the next milestone. That gets us another 10x potentially of uh data availability. And so our top priority is executing on that, just like our top priority previously was executing on 4844. And we think that that's gonna be more than enough, at least data availability for the for the next phase of BASE's growth. When we think about the like mission and vision and strategy of BASE, um, we're really trying to build a global on-chain economy that increases innovation, creativity, and freedom. And I think one of the things that is required for building a global economy is trust and security and surety. And when I think about the trust and security and surety that you get from depending on Ethereum for data availability alongside uh settlement and kind of all of the L1 um logic that sits there for L2s, it's just so much higher than I think the the trust that you get from starting to kind of like fragment those things and introduce basically new trust systems into your your technology stack. And so I think for base itself, um it would it would it would have to be pretty extreme costs for us to to I I don't even know if there would be a scenario. I think like again, we're we're optimizing for trust and and kind of that that long term outcome. And I think we're gonna be able to solve all of our needs on the Ethereum side.

So Jesse, if layer ones are expensive and layer twos are less expensive, why don't we just scale with a whole bunch of layer threes on base?

I think we I think we well, A, I think there will be a lot of layer threes on base. Um and my thinking on this has somewhat evolved in the last year as I've started to kind of build a better mental model for how these pieces work. Um I I generally think like layer threes can be analogized to just servers. It's it's they're just servers that run uh uh slightly more connected on-chain and let you more easily bring in the on-chain assets into your kind of dedicated compute environment. And there's a lot of people who want to run servers, uh, and they run all kinds of different servers with different trust characteristics and different stacks and languages and tooling. Um, and I think we're gonna see the same proliferation at L3 on BASE and other places. You know, today I think we are tracking like six or seven L3s that are running uh actively on BASE. I expect that to grow to hundreds this year, um potentially more. And um I think it's gonna be a really fertile ground for experimentation because it's gonna be way cheaper because you're gonna have easier on ramps for users.

And an L3 on base is just another chain that uh you know uh it deploys on top of the base L2, and so settlement on base L2 and data availability on base L2 or elsewhere. Well what about another solution to your second and third bucket for low low um trans or like high transaction fees and gigagas, which is just like parallelization, right? So let like uh there are a lot of uh Solana has kind of like pioneered this, but now there are other EVM compatible um parallelized uh execution environments as well. What do you think about that uh in terms of roadmap items for base?

Yeah, absolutely. I mean, I I kind of I bucket this into the like EVM optimization, which was kind of bucket two of the what do we need to do to increase target. Um, there's a lot of stuff you can do with kind of parallel, like optimistic parallelization just in the EVM that doesn't break EVM compatibility. Uh you know, it doesn't get you always all of the gains, but you can still do a lot there, and we're really excited about that. And again, like we've been working closely with Paradigm and Ref team, and you know, they've been thinking a lot about that, and um we're excited to collaborate with them further there. Um, I also think there's some things that could be um like breaking changes. For instance, um, you can do a lot more with parallelization and in general, kind of like state isolation if you require access lists, um, where developers actually have to specify here are the things I'm gonna access in this contract, because it then lets you or in this call, because it then lets you kind of uh do a lot more in terms of uh figuring out how it's gonna execute.

That would be a breaking change. Like, you know, there's all the stuff that's deployed that doesn't have those uh kind of access uh lists defined, and we need to change it for new things being deployed.

I I don't have a strong opinion about is that the right thing to do or not. I do believe that

There's value in thinking expansively about kind of this L2 problem space and how we evolve the EVM at L2 and how that might change versus L1. And I think it's been exciting to see efforts to start to do that in a way that is both kind of thinking expansively and staying kind of aligned and connected. You know, there's this thing called RIP or roll up improvement proposal. And that's like run by Ethereum Core Devs in collaboration with L2s, working to evolve the L2 EVM while still staying connected. And I think we'll see a lot of progress and impact as a result of those sorts of experiments.

So, Jesse, crypto natives are they're used to paying fees for transactions, right? So you take go from Ethereum to you go to base, and you're like, oh man, this is so much cheaper. Isn't this great? Okay, but but normies out there, mainstream, the billions that you're talking about that you want to bring on chain, they're not used to paying transaction fees for doing things at all in web 2. So what's your thought on that? And like why generally? So if someone comes externally and they're like, Jesse, you know, one cent is great, but you know what's better than great is uh what than one cent is free. So how about you go make my transactions free? Why can't you do that?

Yeah, yeah, I I totally agree with that. And I think this is one of our theses that's starting to play out, which is that as costs on the uh L2s and other compute environments go down, you'll actually see developers internalize more of the costs into their own business model and structure. Um so whereas right now, because fees have been so high and because the technology hasn't really been there, like users are paying for those fees. I actually think we're gonna move towards a world um increasingly where developers cover all the kind of gas compute costs, um, and then they make money from their business in some way, and they use that to cover their expenses. And and this is not like some crazy novel idea. This is the way businesses have worked online forever, right? Like you don't pay your AWS costs to use Google Chrome or

Yeah.

To use Riverside or TikTok. That would be weird. But instead, TikTok or whoever pays those costs, they make revenue, they pay those costs. And I just think it hasn't been really possible. But now with lower fees and with things like Paymasters, which are now possible with smart wallets and 4337, we're seeing more and more of that happen. And so one of the ways we're trying to accelerate that is we're actually giving away gas grants to developers all across base. So today if you go to Coinbase Cloud developer or Coinbase Developer Platform or Alchemy or Byconomy or StackUp or Pimblico or pick your favorites account abstraction tooling, you can go get a grant from base to fund transactions for you, to give your users free transactions. And that kind of gives you the opportunity to experiment and learn about how do you kind of evolve your business model and product such that you can continue to provide kind of a gasless experience while generating revenue that you can use to then cover gas. And so I think that this is going to be a really, really fruitful place for innovation and something we're pushing on really hard because it will make the user experience much better.

So this isn't something that you are actively working on as part of the base like protocol. You're this is something you're kind of pushing responsibility towards the app layer devs and then giving them the support and the tools that they need to be able to execute on like a gas free vision.

Yeah, yeah, exactly. We basically think that like

y building infrastructure on top of the chain is the best place for this. Um because I think if you look at um

This sort of compute, like it always is going to have to be priced in some way because that's how you get this kind of like efficient market where people are paying however much it's worth. Um, and I think you run into a bunch of other challenges if you don't have that. And you know, I if you talk to Solana folks, I actually think this is one of the things that they're running into a little bit right now is you don't have an economic model for priority fees in and getting transactions submitted on Solana right now. And so what that's leading to is uh people basically trying to gain the system by submitting you know thousands, millions of transactions, and that then leading to network instability. And so, like economic kind of mechanisms are really, really valuable for designing the sort of highly available, highly decentralized, highly resilient systems that and compute that blockchains can offer. So I think they need to exist, but I think by kind of working up one layer above and then building a bunch of infrastructure there that enables us to build on top of that in a really seamless experience, um, we can get a lot done. And one, I was actually tweeting and casting about this the other day is like one primitive that I think is missing today that I think we really need is better kind of like hedging and market infrastructure for both L1 gas prices and L2 gas prices. Like ideally, we'd be able to create a market on base where we could um efficiently price future gas costs and the gas market such that we could give developers actually more consistency. So they could say, oh, we're gonna spend you know 10 million gas over the next three months of running our app.

Let's commit to buying that up front and have a like known cost that we can understand and then kind of smooth that out with the market so that someone else is kind of taking on the risk and you know willing to trade that. So I I'm talking to a lot of teams who are doing that right now. I think that's gonna be kind of the next big primitive that will unlock a lot more design space here.

Yeah, it sounds like um gas credits. And then of course there's that famous um Ray Dalio story of like he did some like crazy financial engineering with some sort of like chicken futures thing. And at the end of that, at the end of that uh financial engineering, the chicken nuggets product was produced out of McDonald's. Uh so I maybe uh that maybe that's kind of the thing that's gonna be. Thank you, I guess. Thanks, Finance, for the chicken nuggies. I'm gonna open up the conversation to uh Coinbase's relationship with Base. Uh there's a tweet that I want to read from Max Brandsberg. He says, going forward, Coinbase is going to be storing more corporate and customer USDC balances on base. This enables us to manage and secure customer funds with lower fees and faster settlement times with no impact to the Coinbase user experience. We're excited to continue to move our business on-chain and hope other companies will follow our lead. Since this tweet went out, uh the supply of USDC on base is just absolutely parabolic. Uh, I can only imagine because Coinbase followed through on exactly what they would say of actually using base as their its own back end. Uh can you talk about the just this decision? Um I'm assuming it starts with USDC from customer and corporate accounts, but it might increase in scope from there. Um talk about like what this means and where this goes.

Yeah. I mean, I I think if you look back at kind of the mission vision strategy of Coinbase, uh what you see in a blog post, you know, Brian wrote in 2016 is you know, we've had this.

Four phase plan for building the on-chain economy for eight years at this point. You know, he laid it out. He was like, first we're gonna build the protocols, then we're gonna build the exchange, then we're gonna build the consumer interfaces, then we're gonna have millions of apps that billions of people use. And I feel like right now we're kind of between phases three and four, um, where we have the starts of these interfaces in Coinbase and Coinbase wallet and other kind of browsers and um wallets. Um, we're also starting to see kind of the start of these apps where we have lots of apps that are being built on chain on base and other networks that people are starting to use. Um but we're obviously like pretty far away from having millions of apps that billions of people are using. And I think the thesis with Coinbase is um we too are going through this transformation where we started from a place of 2012, you know, Web 2, let you buy Bitcoin, that was all we did. And now here we are 12 years later. And in many ways, we're still a kind of off-chain custodial centralized business with crypto in it. Um, we're letting people do things with crypto, but they're doing it off-chain, you know, in their kind of like web 2 uh products. And I think we see an opportunity for kind of us to lead the way in showing everyone uh what does it look like to take these users, these assets, these products, and move them on-chain. Uh, and what are the actual benefits that we get from that? And I think when we think first principles about the benefits, it starts to become pretty obvious to us why we would do this. First off, it's it's way cheaper. You can build the same products on-chain as off-chain uh at 10x lower cost. And I think Uniswap is a great example of this. You know, it took like two engineers to build the first version of Uniswap, versus building a centralized exchange can take you know tens or hundreds of engineers uh from an engineering perspective. So that's one benefit, uh, way cheaper. The second benefit is

Things are globally available from day one, right? Like if you build something on-chain, it's available to everyone by default. Whereas if you build things off-chain, you run into more challenges in terms of how you make it available to people and what the different kinds of engagements you need in different places in order to unlock that. So that's a that's a really big benefit. The third benefit is composability. If you build something on-chain, it automatically works with everything else that's on-chain. Things just plug in. People can build on top of it, they can leverage it. Versus if you build things off-chain, then everyone needs to integrate with you through legacy APIs. And so when we kind of evaluate this from a business perspective, from a Coinbase perspective, I think folks like Max, who runs our consumer businesses, they look at this and they're like, yes, we need to do this because it's gonna be good for our business, but also because if we do this, we're gonna build a bunch of tooling and playbooks and knowledge that then's gonna let us help everyone else make this transition. And so I think this is really what the USDC kind of announcement from him is indicative of. It's us trying to find kind of what are the steps we can take to move more and more parts of our business on-chain. And I think the the thing that I'm really excited about is if you think about Coinbase today, um uh you kind of think about it like a glacier. It's a massive company with so many things going on. Um, and today most of that on-chain stuff is like underwater. Like it's not happening yet. And people are seeing the tip of the of the iceberg where it's like, oh, USDC balances are starting to come on chain. Awesome. Like Coinbase is starting to let people do things on chain on base and wallet, awesome. But I think what's gonna happen over the next year and change is like that whole iceberg is gonna rise out of the water, and it's like, oh no.

Every single part of Coinbase can come on-chain.

Wow.

All of our products, all of our users, all of our assets, they can move onto these new platforms where they're going to be cheaper, faster to build, globally available, composable with everything else, and better experiences for our customers. And I think that transformation will both blow people's minds. It will lead to a bunch of better products from Coinbase, it will accelerate the base ecosystem, and it will set the playbook for millions of other businesses around the world to make a similar transformation. And so that's what this is about. It's about showing the world that this is possible and showing our customers that it's not only possible, it's better.

USDC on base is definitely like the most approximate thing to put on chain. It's like the f the first step, right? Uh and you get like you you can only imagine where it could go from here. It is the idea here that like Coinbase kind of like unbundles itself as it discovers ways to put more of its own business logic on the actual chain.