Layer 2, rollups, and building onchain (with Base)

Transcript

so when we were thinking about building base probably our largest concern the thing that was like holding us back from building it was will people take us seriously when we say we're going to decentralize because it's coming from a big centralized Corporation I think that a lot of people might look at that and be like okay who are these Jokers what are they doing why are they trying to co-op [Music] this hello and welcome to web 3 with a16z a show about building the next era of the Internet by the team at a16z crypto that includes me host Robert hackit today's episode covers the bustling area of layer 2 rollups a technology for scaling layer one blockchains such as ethereum joining us is Jesse Pollock who previously LED

engineering for coinbase's retail side and who now is the company's head of protocols where he founded and leads the popular Layer Two rollup base we're also joined by Noah Citron an engineer at a16z crypto who works on many open source projects and protocols and who closely tracks developments in this area our conversation digs into the shifting history and future of ethereum the arrival of upgrades like EIP 4844 experiments in futarchy and the difference between leading and innovating inside companies versus within decentralized communities we also discuss the challenges of winning

developer mind share how to refine business metrics and measures understanding the Tangled interactions between rollups and bridges and whether you should ever hyphenate the word onchain as a reminder none of the following should be taken as tax business legal or investment advice see a16z crypto.com disclosures for more important information including a link to a list of our [Music] investments okay so I want to start big picture here which is what is a layer two and what is it about a layer two that enables you to run things cheaper and faster that a layer one like

ethereum or you know many of the other multitudes out there what is it about a layer two that enables that yeah I might think about a layer two just use an analogy kind of like a carpool lane in a highway so ethereum is built a highway you have all these cars that are going on it but what layer tws do is they make it so instead of trying to have every individual transaction be a car you can take a ton of transactions you can put them in one car so maybe it's a bus or maybe it's actually public transit and now all of those transactions can move together at a much faster pace and so the way that that works is layer twos basically take big batches of transactions so thousands of transactions that people submit and they batch and compress them and then publish

them to layer one and what that means it means that all of those transactions that are being published onto layer one through the layer two get the same security they get the same kind of platform benefits they use the the same technology but they're batched and compressed which makes them much much cheaper and so you can get all of these benefits from a speed perspective from a cost perspective that comes from that compression I wanted to also add a little bit on sort of the how and the why we got to this design yeah and to go just a little bit deeper a little bit more technical because it wasn't initially on the road map for ethereum ethereum had a different plan for how they wanted to scale yes ether had a different plan but they kind of wanted the same outcome of you know they had

this realization that they can't scale just one lane so like let's make multiple lanes and the original design behind us with sharding and it turns out like rollups are in a very roundabout way implementing sharding but what the realization that I think ethereum had and a lot of other blockchains had this realization too such as cosmos was that blockchains were very good at taking things putting them in order and then keeping those things around for a little while and it turns out we can do that at pretty high scale at the time we didn't realize we could do that one thing on a very high throughput but we can what blockchains are not so good at though is is re-executing all these transactions you sort of put all these things these transactions in order you kept them around for a little while but every single person who wants to connect to

the network needs to rerun every single transaction and of course you can do that but when you do that you need bigger and bigger and bigger machines and eventually like you run out of machine there aren't computers big enough there aren't data C is big enough there's not enough Network inbound and outbound from your data center to run the entire internet on a blockchain and what you're describing is this issue that if you want to run things off chain and then you end up having to reexecute everything to prove that it's all true and everything is like valid you run into an issue where basically you're just recomputing everything and it gets to be you know even more complicated problem than it was before you tried to offload all the computation exactly and the problem isn't really at first that you can't just have many people re

executing things but around there but you cannot get a high enough stake of each thing and reexecute it so at the time you wanted to build a new thing and you need High throughput you could build new blockchain you can convince people that your token is valuable get a lot of stake on that blockchain a lot of money backing it make it hard to sort of 66% attack for a pro to stake chain and that was sort of the way you did it but now you're under the problem if every blockchain need to do this again and again and again it gets somewhat Capital efficient James like cosmos had this cool idea recently of like oh we could sort of share about Sak around and that is interesting but rups went in this opposite direction of saying you know what we figured out that we can order

things and keep them around for a little while really really well so let's make one blockchain like ethereum that orders things keeps them around for a little while and then let's make lots of rollups that prove what the outcome of the things that the transactions that were ordered and kept around for a little while on ethereum yeah exactly and so like Noah said what was originally a vision of okay we're going to take this one blockchain ethereum and we're going to break it up into a bunch of parts and that's going to let us scale instead said hey actually we're going to take this blockchain like ethereum we're going to make it work really really well for this ordering component that's super important from a decentralization perspective and security perspective and then we're going to create kind of a free market where we say hey we're going

to build a platform that then lets other people do all of the transaction processing on top of us and we're going to give that security we're going to give that ordering as a service and everyone else can kind of innovate in the space of how do we process more transactions and put them into that ordering security that ethereum is providing and I think what we've seen is that like Noah said we're kind of converging back to where we started we're ending up in a world where there are going to be many shards of execution that are happening in parallel that are all scaling ethereum but instead of that having been like topped down where ethereum did it all in place instead we've had being built bottoms up where you now have all of these rollups that are coming together there taking sometimes different approaches to

scaling ethereum and are are making a pretty big impact so more users can come on chain that's really interesting so the initial plan around sharding the blockchain was sort of this top down approach from the community of developers that work on ethereum and this new approach this rollup Centric approach is really more of a Bottoms Up movement where you're kind of like okay well we've got this part the blockchain that can do this sequencing it can tell you the ground Truth for a little while and then really it's up to everybody else to come up with these sort of extracurricular systems that are G to take care of the real load exactly one credit I'll give to vitalic as a leader vitalic buan who created ethereum I think he's done a really great job of creating the like strategic context from

the top down for creating that market like if him and the ethereum foundation other core developers have been like yeah like we don't really know what to do so like you guys just go figure out I don't think that would worked but instead in 20221 they basically said hey we were going down this path of scaling ethereum in place with charting but now we actually want to change our strategy and we want to embrace this Ro Centric strategy and here's how we're going to start evolving ethereum to do that and we want all of you other folks to contribute to this worldcentric strategy and I think that kind of leadership from the center of ethereum where they've been really clear in what the long-term strategy is they've invested in long-term research and Technology Investments and then they've created space with Clarity for a bunch of other

teams to crop up and execute the core parts of that strategy that are decentralized and bottoms up I think it's been very very impressive and I think it's a certain type of leadership that we could all learn something from you know for any of the sort of the theory and historians there who want to see like the defining document of like when that was laid out and like actually see what happens to have that leadership there's a fantastic post written by metalic was post on E research as well as his blog called the rollup Centric road map of ethereum I was written 2020 or 2021 so it's getting somewhat old now but at the time it was the first time in public there was this admission that we're not going to do this sort of execution sharding based approach and set the stage for everything that's

happening right now yeah and mentioning vitalic leadership when I interviewed him like many moons ago back in like I think it was 2016 I was working at Fortune as a journalist and I did a profile of him one thing he told me was he wanted to take inspiration from Linus tals of uh Linux Fame and I think he's been able to do that pretty successfully if less of a surly presence yeah it's like nice Linux yes it's like nice Linux that's not an easy thing to do and I'd say this isn't the only place who's done it I mean I think another place where vitalik's really been an incredible leader is like the road maps that he's published on a yearly basis for ethereum for the last couple years like literally he's diagramed out here's what's happening in

ethereum here's where we're going and I think that that sort of Cl it from kind of the spiritual leader of ethereum has been such a valuable like shelling point and coordinating function for clarifying for all the other teams that are working on ethereum oh this is how I fit it this is where my work contributes to this broader Mission and I think with decentralization sometimes it can be easy for people to say oh like we're going to leave it up to the community and I actually think that's totally backwards like with the centralization you still need leadership you still need people who are willing to come out and say here's our long-term strategy here's where we're going it's just a lot harder for leadership because there isn't the sort of command and control where you can just tell people what to do instead you need to inspire people and you need

to get them to follow you and to trust you and to influence them and I think metallic has done such an incredible job of both articulating those strategies and road maps and building the sort of like trust and Confidence from the broader ethereum developer community that let them execute out the vision that he's doing that he's articulating with many other people in a totally decentralized way I mean just to extend that like Brian Armstrong of coinbase also is managing to tow this line he's obviously leading a centralized business and exchange but with all of his interest in decentralization and just a really interesting Vision on how he's going to take the company yeah and if you look at Brian you know in 2016 Brian published the coinbase secret master plan which talked about four stages where you have a first stage which is

building out these core protocols like Bitcoin ethereum then you have a second stage which is building out these exchanges that let people get into those protocols then a third stat where you build consumer interfaces so people can actually access the protocols and then a fourth three of millions of apps that billions of people use he wrote that in 2016 and I feel like we're on the verge of that fourth stage now you know we have protocols we have the exchanges we're starting to have the wallets and browsers we're starting to have these apps but Brian didn't know exactly how to get there and you know as we built base inside of coinbase I think it was much more emergent in Bottoms Up you know in many ways base was a result of a team of people inside of coinbase kind of looking at the strategy and the

vision that Brian was articulating and saying hey we think we know a path to help and that sort of Bottoms Up Innovation happening inside a large corporation it's very rare and I think it is enabled by Brian's leadership and by his willingness to articulate that Clear Vision then also make space for other leaders and the organization to really step up and make an impact let's talk about base yeah well it's a good segue where did it come from where is it now where is it going yeah so I've now been in the coinbase orbit for seven years I led all of our consumer businesses for five years on the engineering side So coinbase coinbase Pro coinbase wallet and then in about 2021 like early 2021 I decided I wanted to get closer to the metal of crypto and that was kind of for me just like I've

been doing it for five years I had a big organization I am still a hungry kind of founder I come in through an acquisition and it felt like I needed to figure out how to push forward like Global crypto adoption and the best way to do that was by getting into the infrastructure because it felt like there were still a bunch of really hard infrastructure problems that needed to be solved so late 2021 after taking a little bit of a break I came back to coinbase and took on what felt like a kind of clear Mission without a clear strategy which was figure out how to bring coinbase on Chain by building infrastructure and products that let us move our customers our products our assets on chain and that kind of begun what was quite the journey and honestly felt like wandering

in the desert for about a year we iterated through four different products that we tried building it was me and a small team none of them got product Market fit internally we talked to customers didn't really feel like they were to fit there we kind of wound down each one of them and I think as we were going through that process we kept running into a bunch of the same problems in particular like we were starting to build these onchain products but we kept running into a bunch of foundational technology decisions that were platform decisions like where do we build these things how do we make sure that when we build them on chain they work well with coinbase products how do we make sure that the platform we build them on works at the scale that coinbase needs and so I think after you know

literally a year of trying and failing at the product level and running into a bunch of these same infrastructure challenges and starting to kind of do the exploration that the decision- making around okay how do we solve some of these infrastructure challenges the small team that I was working with got together and said hey what if instead of just continuing to operate at the product level we go down to the platform and say let's solve this for ourselves first and foremost let's build a platform that we would have wanted if we were starting to build new products on chain of coinbase let's s that and then let's see if that's useful for coinbase and let's see if it's useful for the world and so that was late 2022 we started with an internal prototype that we called baset which was an internal L2 that was launched on a test net that we

kind of opened up to coinbase developers say hey here's a playground for you to play around with stuff on chain people loved it they had a lot of fun building things on it and so that gave us the confidence to say maybe we should make this a real blockchain we decided in December to do it and then we launched the test net in February like two months later at East Denver and opened it up from a developer perspective and then we launched the main net in August 2023 and then you know here we are after the main net launch which was August 9th and if you look at the number of transactions the number of daily transacting addresses the number of daily transacting and monthly transacting Builders who are deploying contracts the amount of Dex volume basis number one across all of them and I think we're

seeing a ground swell of energy of people starting to to come on chain and so that's the short tldr of Base we have a a lot ahead of us I like to say to our team it's still day one we're just getting started and I think when we think about what a successful outcome for base looks like it's building a global onchain economy that increases Innovation creativity and freedom what I love about this story about it being an internal project you know you guys had a problem and you were trying to come up with a solution for it for yourselves let alone in external parties is that there's a rich product history of companies that do just this and I think like maybe the top example might be uh Amazon coming up with Amazon web services to solve their own sort of web hosting problems and then that becoming

you know just a huge profit driver for that company you mentioned that you like to say that it's still day one which maybe tipped me off to this because that is a famous Jeff Bezos ISM famous Bezos ISM absolutely and that's you know credit to Bezos for that one that's not Jesse Innovation but yeah I think about Bas is the AWS to coinbase this really is the platform it's the developer platform that's going to enable developers to build on chain and it came from coinbase who spent 10 years building the infrastructure to let our teams build on chain internally and now we're opening it up to the rest of the world we're trying to say hey it's time for everyone to come on chain because you can get better outcomes for the products You're Building better outcomes for your customers better outcomes for

the economies that you want to be a part of you talked about Bas as being a top L2 and you listed a bunch of attributes in supportive that one of the measures that a lot of people in this industry like to use is tvl or total value locked and there are plenty of people who take issue with that as being like you know the north star that a lot of people cite when it comes to tvl base is ranked third you got arbitrum up at number one optimism or the op main net at two and then base at three so we got a long way to go there and I think the reason why we look primarily at the daily monthly metrics is because they're much more in our control you know if you look at the tvl which we actually call assets on platform because there's two metrics

that people look at that they call tvl one is the total value of assets bridged and natively issued on a blockchain and that's what l2b reports another which defi Lama reports is the total value locked in D5 protocols so it's like how much is in Unis swap Etc people call both of those tvl internally we call the former assets on platform we call it the ladder tvl and so assets on platform describes all of the value of all of the assets that are secured by base whereas tvl describes the value of the assets that are locked in protocols on both of those we still have some catching up to do to be number one and I think the reason why we're less indexed on those dayto day is they're very time dependent if you look at optimism and arbitrum they've been around for three plus years

at this point and once assets flow into l2s they're very sticky and so you don't see that much reshuffling of assets once they're into an L2 and you know base in the last eight months grew from zero to 5 billion in the last three years arbitrum and optimism has grown from zero to whatever they're at today but the productivity of those assets the size of the economy we think is much more reflected by basically how many users are actually transacting how many people are building what's the volume so that's what we're focus on day-to-day but we think that over the next couple years we'll see the tvl and aop grow as well on base and we're excited for that to clarify let's talk about those measures and metrics that you care about you mentioned a few of them I just want to make sure that we have them so yeah

interested to know how you measure your success and sort of how you're performing on those fronts yeah and I was actually just talking about this with someone who is thinking about joining our team and the base team is still very small but we're we're slowly starting to grow it you know it's not quite a two Pizza team two Pizza team is like 10 people but it's a handful of pizza teams yeah it's like a good measure of like how many people you want in a room to get something done yeah exactly so we're like a few two Pizza teams so we think about base as an economy that we're trying to grow global economy global onchain economy and from a strategic perspective we're focused on four things that make up that economy the first is the developer platform which is where developers actually come to build they write evm code they deploy

it on base this creates apps the metric that we look at for whether we're building a successful developer platform is what we call tmtb or quality monthly transacting Builders so we basically look at all of the people who've deployed smart contracts then we add some quality filters to it to make sure that they're not just Spam or Bots and that's our like qmb metric got it so that's the like supply side our developers being successful building our developer platform and able to build apps the next is the demand side which we think about is are we building an app ecosystem that consumers and businesses actually want to use and the metric that we look at there is qm or quality monthly transacting addresses so this is basically who are

all the people who are on chain doing transactions whether that's a buy or swap or a mint or a you know playing a game or sending usdc to a friend and then we apply some quality filters there as well to filter out boxs and spam and try and get to real Humanity can you say anything about what sort of filters you apply to uh get a quality determination I can't explicitly Define it but what I'll say is we look at activity so like what kinds of transactions are they doing we look at like identity characteristics that are starting to be on chain for instance like social graph from foraster verifications from coinbase or World coin and all of that lets us get a pretty clear picture it's not perfect but it does give us enough Clarity and filter out enough noise and it's appliable across all contexts all

chains so that basically allows us to then do market share and kind of look at how much market share base has versus other people pretty consistently so that's the second metric the third metric and the third norstar is really about growing the largest Capital Market in the world onchain and making basic center of the onchain economy and there we're looking at assets on platform alongside tvl and Dex volume so assets on platform is the total value of assets secured by base tvl is the value of Assets in D5 protocols on base and then volume is how much trading is happening on a daily basis and like you called out on aop assets on platform and TBL we're still lagging but on the Dex volume we're seeing that B has been growing very quickly we have some catching up to do to get to salana and BSC who are

bigger and have been around for a little bit longer that's the third Northstar and then the fourth one is we need to decentralize and scale this platform that we're building on and so the metric here is not just a number instead we're looking at basically a four-pronged goal which is we want to achieve a stage two level of decentralization which is the maximum level of decentralization on a rollup that has at a minimum four nines of availability so 99.99% availability that can process one giga gas a second of gas which is kind of a throughput metric for how much gas it can process at a less than one cent average transaction cost and so those characteristics of the underlying technology platform is the fourth major pillar of what our teams invest in okay

so I want to get Noah into the mix Noah as somebody who is tracking these l2s how are you evaluating them like what metrics are important to you as somebody who's observing all this take shape I guess there are two metrics one is sort of this highly practical right now metric of like how's the chain doing how are we doing on transactions how are we seeing like quality developers built on basically like the traction from an ecosystem perspective and that stuff is is probably the most important and what's unfortunate about a lot of that is you can win on the ecosystem without winning on the tech so the thing that I SP a lot my time focusing on is how do you win on the tech side like you can make a lot of concessions to build

something that seems to be scalable and sort of works and if you win the ecosystem and you do that you might actually end up as the winner so I think the thing that's most important to me is that the one who wins the ecosystem is also the one who wins the tech and we need to make sure that that is actually how it lands so I spent a long time picking a lot more out into the Futures like what are the end games of these rools going to look like sort of what is their ability both to get to this like stage two decentralization thing as well as what stage two decentralization looks like for that rollup because each rollup has a lot of somewhat interesting characteristics where what stage two looks like the security model of it looks ever so slightly different so it's

really about how the choices that they're making now and the the way they're thinking about it right now is going to lead to the optimal end game future and just I'm here crossing my fingers that the one who wins the ecosystem is the one who sort of wins also this long-term technical story explain what you mean by that winning the ecosystem versus winning the technology there's nothing stopping me from having a gazillion users you can build chains that have relatively High throughput eventually you do hit a cap but it can get very very high and you can sort of Clone those chains you can imagine building sort of rollups except each rollup is 12 validators who are sort of a proof of authority right it's you and 11 of your friends are each of the validators and you clone these blockchains run across a bunch of data

centers you have something that kind of looks like this multi road map it scales a lot it's not very secure it's not really scure at all you have very little ability to verify what's going on you have very little trust in you know 12 or it could be 12 or 20 or 100 you want it to be 10,000 there's nothing stopping that world from winning other than we need to have people who are sort of educated as well as the people who are winning the ecosystem to be of the people who care about this very long-term Tech which you know one of the reasons I'm very excited to hear a lot of what Jesse's been saying is it seems that Bas to a certain extent has been winning the ecosystem as well as carriers quite a lot the stack as well as you know really the whole op stack

your point about winning the ecosystem means like you might be the product that everybody's using even if it's not the most Superior most decentralized most secure product in the market and what you want to see is a product that is both the superior tech product and being used by everybody exactly I mean we see this all the time we all still use Visa yeah it's a great Point yeah although I mean I think one of our thesis from the beginning on Bas is that the best way to build the biggest ecosystem is to embrace the values of ethereum and to lean into the decentralization road map and take a path that we think gets us to the best endgame and so when we were thinking about building base probably our largest concern the thing that was like holding

us back from building it was will people take us seriously when we say we're going to build Layer Two that's going to decentralize all the way to stage two because it's coming from a big centralized Corporation I think that a lot of people might look at that and be like okay who are these Jokers what are they doing why are they trying to co-op this and so from the beginning our whole ethos was no we're going to build on open source that's why we built on the op stack so it's not a proprietary stack we're going to lean into multi clients that's why we've worked with Magi from a16z and you knowah and op re and kind of taking the whole stack approach that we've done with optimism stack that's going to let us get to stage two by having multiple clients I think all of

that for us has basically been a bet that yeah we could build a great ecosystem but the best way to build a great ecosystem is to attract incredible developers and the best way to attract incredible developers is to live the values that they want to see which in the ethereum world is about decentralization it's about open source it's about building something that embodies those values and so I think it's been cool to feel like oh yeah we're actually being rewarded for prioritizing decentralization and being rewarded for thinking about this in the right way and I think maybe there's another you know reality where someone can just build a chain or roll up that throws away all those things but still gets the most adoption but I'm skeptical I think that the hardest thing to win is mind share of developers and we kind of see that again and again and again and

that's what you need because those are the people who are going to build all the incredible products in the world and the best way to win that mind share is by being the same and caring about the same things that they care about and prioritizing them and earning their trust I think that you know we're starting to do that on base and I'm excited to see how much further we can go you know I think that's something we're extremely lucky for right like this didn't have to shake out that way you are absolutely right like if you're going to have a big ecosystem you need great developers building great things and fortunately we have this culture among ethereum developers of really really caring about these things yeah it didn't have to be that way it's something very special that land in a the I don't know exactly why it happened

that way I have some guesses but it did and because of that I think you're right that it's likely that the rollups who do do it s the longm right way and say they're going to do it the right way from the GetGo are going to build that ecosystem much more quickly yeah I can't quite describe why it's happened the way it has as well although I will sayess it was the beginning of 2023 we were working on 4844 which is the scaling upgrade to ethereum that coinbase and base started working on about two years ago that just shipped in March and because we were kind of doing a bunch of core development related to it the ethereum foundation put on like a core development Team Summit where they brought all the core development teams which are a bunch of different teams bunch of different companies building a

bunch of different clients all together in Austria for a week and me and someone else on my team Roberto who leads all of our cor we got invited because we were contributing and I went and was there for a week and watched all of these decentralized core development teams come together and it was one of the most profound experiences I've ever had in my entire life Wow because I've seen corporations trying to organize in a central like top down command and control way to get people to accomplish things and seeing the Counterpoint which is oh my God none of these people work together none none of them work for the same Corporation but all of them have the same value set all of them are aligning and all of them are shipping incredibly complex software on like a pretty intense timeline with a really

high level of quality to build a global economy that's almost trillions of dollars value that to me was like oh I don't fully understand this I don't know why but there's something profound here that is happening that I want to be a part of so I'm glad you brought up 4844 I do want to talk about that but be before we do I also want to just clarify for people who are tuning in Noah and Jesse you both have cited various stages of maturity for rollups and have cited stage two as sort of like the fully decentralized stage so maybe you could just elaborate on those three stages quickly yeah rollups bootstrap off of ethereum's decentralization and security so like getting to the optimal state of

oh this is fully decentralized it's really fully like having the same security and decentralization character as ethereum wasn't something that was really possible on day one instead like any system you kind of had to build it incrementally and so that's what people started doing you know three years ago we started building rollups or four years ago the first ones went live in production you know two and a half to three years ago and when those first rollups went live in production they had a lot of training wheels so there was basically a bunch of Protections in place in the rollup so that if there was a bug it could get fixed or if something went wrong someone could react and respond and you could have a human intervention that could override some issue in the way that the program was

designed so we've had more and more of those rollups launch they've all had various stages of training wheels and so about a year and year and a half ago vitalic again vitalic glum proposed this framework for thinking about the level of decentralization of rollups and he articulated three different stages stage zero stage one stage two that have sort of become Canon for evaluating where rollups are in their decentralization journey and the north start for all of them is to get to stage two which you can kind of think about as having the same security and decentralization characteristics as ethereum Base today is at stage zero but we're working to get to stage two and the next stop is stage one there's a couple things that

we need to do in order to get there both of which are on track right now one is changing the security Council which manages the upgrade keys for base the second is rolling out fault proofs the security Council we've designed we've composed we've deployed it on Main we now need to switch base over to use it and then the fault proofs we have live on testet right now we just went through on it and we're in the process of rolling them out to mainnet over the next few months and so I think there's a really clear path to us resolving both of those training wheels this year for sure hopefully much faster than this year and getting to stage one and then beyond stage one the next big challenge is how do you start to remove even more of the human input that goes into roll

up so for instance even having a security Council how do you constrain what the security Council can do how do you give it so that all the actions they take have to play out on a long time Horizon and in order to get there we need to have what's called multic client which is basically having multiple different clients that are all running the same software that enforce the rules of the rollup and so one of the things we've really prioritized as we've built base and built it on the op stack which is the framework that's open source that we build along with the optimism Collective is designing such a way where it's easy to get to multi client or at least relatively straightforward and so we have one client stack which is called up and op node and then in parallel we've actually been working with a16z

and Paradigm and others on a second client stack which is op breth and Magi which is built in Rust and then there's third and fourth client Stacks that are also being explored and built with OP Aragon and up nether mine and so what's going to happen over the next year is once we get to stage one then we're going to be kind of hardening all of these client stacks and getting them to work well together so that sometime in 2025 we can have multiple of these Stacks running in parallel and we can actually get that kind of last Milestone to get to stage two and so it's going to be a journey we don't have exact timelines for any of it but I will say that I think we've made a lot of really good design decisions that are going to set us up to progress from stage one to

stage two in a pretty straightforward way and I'm really excited about the collection of different organizations and people all around the world all from different teams who have come together to build this open source stack in the op stack that can make it happen Noah you got anything to add I do I guess maybe a little bit related to the fact that there's many clients many teams many folks working on just the OB stack in general and we run into like a hard problem I think this has to interplay with how we get stage two is how do we govern this thing when you look at like l1's you get this really nice thing which is social consensus there is no real L1 State You can disagree and a minority or a majority can run a different node software and the chain somewhat peacefully Forks off there's a

little bit of oddness that happens there but for the most part chains can just split in two and you have social pressures keeping it in line when people are acting well this is very very hard to do with rollups so you need much more restrict governance at the end these rollups are going to want to make changes the evm is evolving it's not likely to stop evolving soon so how do we decide together when there's a hard button of go or no go on every single upgrade yeah I think really hard challenge when we were thinking about building base we had a few different choices for how we built it one of the choices would have been okay we're going to go build our own whole stack and just kind of do our own thing another choice would have been we're going to work with

someone else to co-build some of this technology and where we ended up was deciding to build with optimism build on the super chain build on the op stack which is this fully open- Source technology toolkit that lets you run a chain and I think one of the big reasons why we did that was because we saw exactly what you're seeing now which is is that at the end of the day a lot of this is going to come down to governance how do you govern the upgrades to these chains how do you make sure that they remain neutral and open and permissionless in a global context even as you have actors like coinbase who are obviously like us domicile businesses and how do we do all of that in a really thoughtful ridiculous long-term way and when we thought about doing that ourselves or thought about doing with

other folks we really felt like optimism was really the leader in thinking about governance at the level of rigor and scale and Longevity that would actually let us solve those challenges and so I don't think we fully built governance systems that can do that yet but I'm really optimistic and confident that the path that we're on with the op stack with optimism Collective and the governance systems that they're building are going to let us do that in a way that works for base works for the rest of the super chain it hopefully works for the world I'm curious if you ever looked a little bit at some of the governance designs that azc has been proposing they've got sort of a whole bunch of different post on The Forum where they have been putting up rfps and discussing things and they did land on this very interesting idea which I think

no one else has explored and I hope others do start exploring it because it's really a fascinating idea where they made the concession that the bridge does not define the rollup and I think people have argued about this in the past for a while I think Kelvin fiter from alism is the champion of the bridge does not define the rollup and they really lean into this idea and say well if that's the case there's nothing actually stopping us from having many bridges to roll up the only constraint you have is that each Bridge needs to me a unique token on the rollup so you know if you have Alice's bridge and Bob's bridge and they meant usdc there's going to be a usdc and b usdc which is unfortunate but it does mean that you can have many bridges each with their individual governance which starts to

look a little bit more like social consensus I'm curious if you thought anything about these kind of weird maybe impractical designs but maybe there's something we can learn from them and just before you jump in Jesse no maybe you can just Define for our listeners what that means that the bridge does not define the roll up yeah no can you just Define that in like one sentence you just that you just solve this theological problem I can give you a links to 12 separate blog posts Each of which are like 5,000 words well that's great well let let's add that to our show notes this is like the most confusing least intuitive thing about building these chains and I'd say I feel like I've had like many existential crises I'll try and give in 30 seconds or maybe a minute or less an explainer

in that rollups at the end of the day if you go by the theory of Kelvin a rollup is a function it has some inputs with your transactions and the function produces some outputs you can compute this function yourself or have a proof that helps you compute this function or whatever and that's the rollup like what the state of the rollup is is when you apply this function to the inputs now a bridge or any other way of computing this function is simply an instantiation of computing this function the bridge canut it wrong the bridge can disagree on how you compute it and kin makes this argument and I think he give a good talk called rollups aren't real where I think he maybe first made this argument but what he argues is because the bridge is sort of just like an implementation of how you go from input to Output the

bridge doesn't really Define the RO the bridge can break the bridge can disagree with the RO up State at the end of the day like there's some social consensus that defines the function that the rollet computes on the inputs to get the outputs and that function when executed correctly defines the St of the rollup rather than any individual bridge and this is what I was saying is I like really leaning into that saying like Okay just let's have lot of bridges it's maybe a little practical but it's certainly reallyy cool so the idea being that there should be some social consensus that ultimately determines the state here rather than letting the bridges which can do all sorts of funky weird things that get problematic determine it exactly and this is sort of the point I was coming across is that even though we may have social consens

we may all get together and sort of agree what the function should be or how it to change in a normal blockchain you can just run with that when it comes to a rollup the contract doesn't care about your social consensus the contract does what it's supposed to do and we have to Grapple with that now and I think it's going to be probably the most difficult thing we're going to Grapple with as we go from stage one to stage two oh I haven't read as text posts but I do think that this question of what's the relationship between the governance on L1 and the rollups on L2 is a really really important one and it's one that we've been thinking a lot about and working really closely with optimism on and I don't think I have any new insights to share here but I think it's

a hard one and it's one that we'll keep working on as we get to stage two you cited optimism and their governance as a reason that you worked with are you referring to their two house system where they allow token holders to have votes and yet it's also Guided by you know a sort of Council of people who are in this tech for the Long Haul yeah i' say citizens is the way they call it so it's a two house system one that's Market driven where you have the token being like one token one vote which is proportional to ownership of those tokens and then the other which is you know one citizen one boat so they're basically gradually building concentric circles of different citizens who are optimism citizens and I think it's up to like 500 now or something like that and those two houses have different

responsibilities they vote on different things they have checks and balances we're now introducing the security Council which is kind of a third governance component that's starting to have other checks and balances and you know whether or not like the existing exanti of that governance is perfect certainly not because it's just a system that you know it's incrementally evolving the thing that I think really Drew us to work with optimism is the level of thought that they were putting into it the way I think about this is at the end of the day a lot of this is just like human systems at scale that are informed in technology or enshrined in technology and you know this goes back to what we're were just talking about like a lot of what we're even talking about with chain is like what do the people socially have consensus around

when they're running this chain what's their consensus around the function that is a human problem and when you have these complex human problems I think one of the things that BL chain potentially enables is it enables us to design much more flexible systems that can reflect human problems and then solve those human problems with code and whereas we previously would have used like laws that we write in paper and you know lawyers review and the Court's judge we now have the opportunity to write contracts en code that Define programmatic relationships between different people and different entities and enforce things you know permanently Forever on chain and I think that opens up both a huge new design space in terms of what's possible because we can Define

things that you could never have defined in a contract and I think it basically opens up the possibility that we could build things that are way way way better like a hundred times better a thousand times better than any governance system that we've ever had today and that could let us better Express human intent and say hey we have all these people who want to make the world a better place but maybe we're just being held back by the way we can represent their intent in these governance systems that we're designing and if we can 10x or 100x our ability to represent that intent and to build these governance systems that reflect that intent I think we can potentially build a much better place and so that thesis and that intentionality that I think optimism is

bringing to it is one of the big things that Drew us to working with them and I think where we are is still day one like the governance systems that have been built are still very very nent but that way of thinking which is like this is a tool we have a new tool that's a new technolog that can enable us to build better governance systems is going to let us accomplish things I don't think anyone can even think of right now yeah it's a hard challenge it's sort of the fundamental proposition of web 3 is this idea that you can have like networks that are community-owned Chris Dixon talks about this a lot in his book read WR own and yet how do you manage those things you know you don't want direct democracy mod rule but you also don't want plutocracy you don't want autocracy

there are many examples of like authoritarian sort of network systems in web 2 that exist today and yeah so how do you actually navigate that practically it's a really intriguing set of problems if you look at the optimism governance system today I think the like theoretical or like the philosophical governance framework that it most resembles from my perspective is something called futarchy which was defined by this guy Robin Hansen in the late 90s early 2000s which basically posits that the market is very good at certain things but very bad at other things in particular it's very good at finding Solutions if it's given a constraint problem but it's very bad at defining values and like outcomes that a society might want to drive and so his thesis with future Archy was basically what if we set up like a dual system

where you basically had humans defining the outcomes that we wanted to drive and then markets defining the inputs or the solutions that could drive those outcomes and you created these feedback loops between the human defined part of governance we was defining outcomes and the market defined part of governance that was driving the solutions and that could get us to like this future Archy world that could drive better outcomes and when you look at optimism governance today you kind of see that we're starting to have some shape like that right you have the markets the token house where you have people who are voting on specific proposals about what we should be doing and then you have the citizens house which is the human side of things which is defining what are the value system should what's the Constitution what is the impact we actually want to reward and those two

things can come together to get to basically better outcomes than uh pure human system or a pure market system might be able to get to and so I don't know again we're still very early but it is cool to see that we're already being able to use this technology to take things that were previously ideas but didn't really have a way to be actualized and actualizing them and then iterating on them and learning from them yeah futar is absolutely fascinating prediction markets are one of the reasons I actually got into this whole industry to begin with Noah do you have any other questions on optimism on governance or any areas you might want to probe there yeah I mean I guess I have one question I'm not sure how plugged into this you are but you know I've been sort of Loosely following this interesting shift where like ethereum is

sort of spinning off the evm into its own thing that is not really necessar governed by it but is governed by the collection of rollups and the rollups are getting ever improving the evm and upgrading it from what I hear the first upgrade that we think it's going to happen is probably adding the secp to6 R1 I can't remember it's R1 K1 one of those curves the curve that works in the enclaves yeah I'm curious to hear if you're plugged into about how you think that's been going is there anything that's like further down the pipeline you're excited about you want to see the and also like what do you think the end state is like how far off are going to go from what ethereum built evm with yeah so this thing yeah it's been a process that's been forming just in the last six to nine months it's called Rip

rip which is a funny name but stands for rollup Improvement proposal it's not an auspicious name right it's like no you guys but hey sometimes we're stuck with names and I think one of the really cool things about rep and rip is that it's kind of like the ethereum foundation ethereum L1 innovating itself it's rare that you see people be like oh we're going to give up control because we think that that's going to lead to a better long-term outcome and that's kind of exactly what they're doing here they're saying hey what we're seeing emerge with this rollup Centric scaling strategy where we have massive proliferation of l2s is that those l2s are having different constraints than the L1 and they're able to explore more aggressively how to leverage those different constraints to get better

outcomes for developers and users and we want to figure out how do we make it so those people are actually empowered to do that while still keeping them close enough to L1 That You Don't See massive Divergence and so that was the intent I think about six to nine months ago some people from the ethereum foundation on Scar and one other person who I'm totly blanking on started a telegram group and a bi-weekly Cadence where people are now meeting we just you know are passing EIP 7212 which is the first thing that's gone through rip which is adding this signing system to L2 so that people can ver ver ify Pass key signatures and use less gas I expect that to go back to L1 once it's on L2 I think that that's a pretty good proof point that like we're

starting to make some real progress with this I'm generally like very bullish on this I do think that in the medium to long term we will see some Divergence from what makes sense for the L1 evm versus what makes sense for the L2 evm because l2s are able to just push on different constraints but I do think we get the best outcome if those things are not happening willy-nilly but instead we're in the same strategy and execution approach where it's like we're setting long-term strategy and then we're empowering people to execute against it and I think that's the context that RP is starting to create where it's like we can have this long-term strategy which is we want the evm to remain relatively compatible but we also want to empower L2 to start to diverge from L1 and we

want to empower L2 to stay standard across a bunch of different rollups so developers have a consistent strategy now go and execute on that mandate and I think it's going to work I think it's going to work and I think we're going get better outcomes as a result I think it's going to help developers and help users so obviously a lot more proof in the pudding but I'm positive and bullish on where we're head so far and are you worried a little bit that unlike the ethereum core doubs where not only do they have a shared goal but they're the goal is very shared they're working on ethereum the ret core dos are working on their own rollup stacks for the most part their own rollups some of who be each other as being ruthlessly competitive are you worried that rebuilding as you said this fantastic

core Dev community that ethereum has among like the all rollup devs is gonna be possible or yeah I mean look the hardest thing in the world is getting people to trust each other and build together like regardless of whether you're in a decentralized ecosystem or a centralized company like there's no easy way to make it so people work really great together and I think this is a different environment than the L1 environment I mean I'm an optimist so that's something you should know about me and everyone should know about me but I do feel like having the anchor of of us contributing to ethereum and scaling ethereum and having the leadership of metallic and the ethereum core development Community is going to be a pretty strong forcing function to align people to good behavior on this and my sense is that as people step out of line

there will be not like repercussions like there's nothing people can do but there will be social pressure to continue to work well together and so that's definitely the attitude that the base team is showing up with is like we want to work with L1 we want to work with ethereum we want to work with ethereum core developers we want to work with other l2s to find common ground and to figure out what are the things that we can push forward together so we can have more of that standardization and my hope is that if we set that example and L1 sets that example other people will follow the lead and show up with best intentions to drive what we think will be the best outcome for all of L1 and L2 okay great one thing that we haven't talked about is EIP 4844 we had the

denune upgrade go live in March which kicked off ethereum's scaling stage I don't know are people still calling it The Surge is that like a p a term that's what matal calls it in the road map right so yeah you had the merge this shift to proof of stake in September 2022 and then in March we've kicked off The Surge the sort of earliest stages how has life changed for you since then yeah yeah so that happened on March 13th I mean immediately we saw a really big reduction in fees that night you know fees were Subs sent for a long time and you know that was exactly what we've been working towards for two years and so that was super exciting just to feel like oh my God the thing that we've been working on for two years actually is delivering the results that we wanted

now the KnockOn effect of us reducing fees by more than 100x was immediately we started to see a lot more demand uh so in the last you know few weeks we've seen demand more than 10x on base basically that has brought a lot more people onto the chain a lot more developers a lot more users but it also has started to expose other scaling bottleneck because we solved one of them like the data availability on ethereum we drove down those cost a ton but there's others for instance how many transactions can the blockchain process at one time that they then published to that data availability on ethereum and so we actually saw a couple days the weeks after that where fees on base started to go way high again because there was so much demand and we had scaled the chain to support all that

demand from a capacity perspective so we did our first two increases of What's called the gas Target which is basically kind of like how much gas or how many transactions can the blockchain process every block we've about doubled that now from 2.5 million gas to five million gas and we've also set what we think is a very ambitious but also achievable medium-term goal of growing that to one giga gas which is a, million gas so it's 400x where we originally were which is 2.5 million gas and so we're kind of anchoring to this idea of how can we get 400x in 400 Days we like it it's mimetic we think it's really ambitious maybe a

little Out Of Reach but we want to push ourselves to be ambitious and in order to accomplish that in order to scale base you know 400x there's a few different challenges we need to solve one is we need to keep scaling data availability so the data availability we created with 4844 is great but it's still pretty constrained so we're need to grow that really significantly if we are going to support 400x more through prodon base the second is we need to actually support literally the change or run that fast you know it needs to be able to process all the transactions in every block in less than the block time consistently in order to be able to process that much gas and we've made some progress on that but there's a lot more work to do we're really excited about working with gu you know Powers

the core client for Base today we're also really excited about new clients like op breth which is a new client B rust that's really optimized for performance and we've been testing that and seeing really good results and so we think we're going to be able to push the limits really intensely there and and get that up to our goal and then the third thing that we need to solve is something called State growth which is basically as you run the chain faster and faster it creates way more data that now everyone who wants to run a node has to keep track of and that means that the hardware requirements get bigger and bigger and that state sometimes can't fit into memory which means you need new kinds of storage or you need to start charting it and so there's some open research questions there that we need to

solve if we want to push the chain to run as fast as we're talking I'd say the way we're thinking about this now is we've set a very ambitious goal we have a bunch of really important execution work to do and we're just going to burn it down and I think one of the really good things that I feel about this is that me and the whole engineering team that's working on base we spent five years scaling coinbase and obviously we haven't been perfect there there's a lot of challenges from 2017 to 2021 but we learn so much about what it takes to deliver 100x scaling improvements in you know months or years and I think we're going to be able to take all of those learnings and apply them to base and to the op stack and scaling Layer Two and layer one and to build on the legacy of

what we've contributed already with 4844 and make a huge impact over the next couple years so that's what we're laser focused on right now you know I have the gas cost on base in my menu bar right now and you know we're at 90 mu right now which is not low it's it means that we're basically running at our Target which is 5 million gas a second and that means that with more surges we're going to be over Target are going to start getting high again and so every day I'm waking up and talking to our team about what can we do to increase data availability to scale the execution bandwidth and to solve the state growth problem so we can keep bumping those limits and keep driving down fees look I mean it's a good problem to have right

the you come down and now you're seeing induced demand you've got all these people who want to build and transact and now you've got to deal with the repercussions of that totally it's a good problem to have definitely a good problem to have we've been looking at the views too actually because Noah created some ni dashboards so we've been measuring the gas usage on Bas and how the blocks have been how much they've been filling up yeah it's pretty cool and I mean like Bas is the biggest consumer of blobs right now we're seeing just so much demand for people to build on base to transact on base and we're just going to try and keep up and I'm sure it won't be all smooth sailing I'm sure there's G to be a lot of bumpiness and I met with our Builder community on Friday and the thing I said to them is

give us feedback share how what you think we should be doing but also be patient with us like we're going to try and scale this thing 400x and 4 100 days and that's going to be a hell of a journey and it's not going to be perfect and so the more we can all just like come together to figure out how to do that together I think the better off will be how are people using Base today and assuming that you're able to pull off these optimizations that you're able to bring the costs down for transacting on the network and the fees how do you want people to be using it moving forward yeah yeah well I mean like if you go do an eent today it costs a Cent and so even with the demand that we're at right now it's still super cheap and

way way cheaper than it's ever been before to transact on L2 and I think what we're seeing is that people are doing all sorts of things on base I mean obviously usdc for payments both peer-to-peer and commerce has been growing we've seen a ton of trading activity where people have been trading on decentralized exchanges and doing perss and other things a ton of bar and lend activity we've seen games proliferate on base pretty quickly in particular a lot of casual games think you know early Facebook platform in 20072 2008 those people are starting to build on chain and it works really well for like current transaction costs and the way people are engaging we're seeing a lot of like local things start to come on chain so Blackbird for instance which I think a16z as an investor in you know they've been bringing restaurants on

chain which is dramatically improving their margins driving more business through loyalty this summer we're going to have a lot of onchain stuff as a part of onchain summer so I'd say like I said at the beginning we're seeing emerge on bases the next generation of the internet and just like with the App Store and the internet you had things from every single part of you know the economy being transformed we're seeing the exact same thing happen on base I think as fees go down further and we increase capacity more we're just going to see more of everything it's kind of hard to predict what exactly will drive the most growth but I'm very bullish in the next year on the consumer use cases social you know things that can be driven by a feed like farcaster where we actually have the opportunity to take

advantage and see the benefits of that Viral growth that really can only come from sharing and then channel that into onchain Apps and so that's a place we've been really leaning in you know if you look at forecaster usage across Chains It's like 70% base market share today and I think that's going to be a place we continue to lean in because we think that social platform is g to be a huge distribution channel for new onchain use cases it's very cool to be seeing all these new apps coming out now finally you know we've been talking about for years that like the apps are coming you know the web three apps will be here they're going to be really cool consumer experiences and now it's actually happening now it's actually happening yeah it's epic it's really epic and I've been working on chain and crypto for 10

almost 11 years now and I did not think it would take us 11 years to get here but I do think that the industry is a little bit kind of traumatized from the last 15 years where it's like we're so conditioned to kind of flat growth punctuated by these speculated Manas that we aren't that prepared for what's to come which is going to be actually pretty normal which is viral consumer growth and it doesn't look like what we've seen like it's not going to be flat with spective Mayas instead it's going to be we're going to start going viral and then it's going to get steeper and steeper and steeper and steeper until the entire world is on chain well that is a bullish prediction I think we've hit the inflection point already I think it's starting to show up in the metrics for instance if you look at you

know L2 usage month over month I think we're starting to see that kind of Bend up in the curve and I think it's going to play out on the order of years but this cycle of growth is not going to be the same as the last ones it is going to be driven by technology driven by use cases driven by people using onchain to build the next generation of the internet and that's going to bring in billions of people that otherwise just wouldn't be here yeah I liked your mission statement the world onchain I know you like to say onchain is the new online it just reminded me of like the latter day America Online yeah exactly we really do think about this as the next generation of the internet and we think that the transformation and movement of the world from online to onchain is going to be equally as

impactful if not more impactful as the the movement that we did in the early 2000s from offline to online and so it's going to be a sea change but I think it's going to be a really positive sea change for the world so one more note but I'll just say on the team from the editorial perspective we've had discussions about when and whether to hyphenate on Shane um where are you guys at on that right now are you still hying or no hyphen I've been no hyphen since day one thank you Noah we appreciate no hying on onchain no hyping on open source everything one word yep I'm in that camp too Teddy Roosevelt gave a famous speech about like not having hyphenated Americans like you can't be Italian-American or German ameran you just be American all the way but anyway yeah I'm antiy too I

think that's right and I think you know we saw this exact transformation with online right on dash line was what we originally called it it was originally e-mail and then as those Technologies became a part of our Lives The Hyphen disappeared because now this is just email it's just online and it was just a part of our day-to-day experience I think the movement the cultural movement that's happened around onchain and the kind of meme that you know has taken the internet a fair amount about like not having the hyphen so it's not on dash chain but it's instead onchain it's about embracing onchain as a part of our life it's about saying this is something that we all need to get used to and something that we all need to embrace because it's going to be a part of our lives forever and I think you know maybe

we're a little bit ahead of the curve maybe we could wait a few more years to drop the hyphen but I'm someone who loves to live in the future and I think that it feels better and looks better people grock it more easily and I think we're going be able to help the world come on chain faster if we talk about it without the Hy is it working is it catching on or is this going to be like uh like mean girls trying to make fetch a thing I I think it's catching on when we look at the data from consumer preference perspective from a builder preference perspective we see a lot of energy around the word onchain you know historically people have tried a bunch of different things they've talked about Bitcoin they've talked about blockchain they've talked about crypto they' talked about web 3 all of them have had a fair

amount of baggage attached to them they've been pretty technical I don't think we've actually been able to build successful mental models for everyday people around what this thing is and I think the thing that I really love about onchain is that first off it's a very clear analogy you can say onchain like online and everyone immediately gets it immediately they're like oh this is a new technology platform and then the second thing about it is that it is fundamentally about the platform like we're not talking about money we're not talking about speculation we're talking about technology and we're saying hey this is a new version of the internet that's onchain and I think that grounds people in the right way of thinking about what we're doing which is that we are building a new technology that enables people to build incredible new experiences that can of course use money

and of course could include speculative things but that's just one part of it and really what the outcome that we're trying to drive is like in the early days of the internet we're trying to build a better future by leveraging this new technology to build new kinds of applications that bring more people into this global economy that we're building and I think that mental model it removes a lot of the fear the uncertainty the threat that sometimes people can feel when they get into that kind of like money speculation ground place and they can instead think much more expansively about oh this is opportunity this is something that I can benefit from in this new world that we're building I'm all for that the revolution will be on chain not just online that's exactly right cool thank you so much for joining Jesse yeah thanks for having me a great

job yeah I love it excited to keep building with you guys likewise awesome thanks guys [Music] w w w w w w w w w w w [Music]