Unichain and MEV On L2s | Hasu & Hayden

Transcript

but when it comes to UNI chain and what Unis swop means right Unis swop is not just a single smart contract protocol it's like a world of things that interact with it and are built on top of it uh and so when it comes to like a uni chain it's a you know kind of a a chain for a specific Community instead of use cases and you know I think that it will kind of people will self- select into that and uh and we'll figure out what that means over the over the years hey everyone this episode is brought to you by Fuel Labs who's launching fuel ignition a new high performance ethereum L2 powered by the fuel VM now you're going to be hearing all about ignition later in the program but for now what you need to know is they are launching phase two of their points program so

click the link at the bottom of the show notes and head over start experimenting messing around on fuel today thank me later hey everyone quick disclaimer before we get into today's episode nothing said on bell curve is a recommendation to buy or sell Securities or tokens this podcast is for informational purposes only and the views expressed by anyone on the show are solely our opinions not Financial advice our guest and I may hold positions in the companies funds or projects discussed all right everyone welcome back to another episode of el curve today I'm joined by Hayden Adams and hu guys welcome to the show great to be here same yeah this is going to be fun uh we're going to be talking about uni chain and rollup boost a sort of joint announcement that came out of Unis swap

labs and Flash Bots but before we get there I would actually like to start uh by zooming out to kind of get the 10,000 foot View and I think the um just try kind of get your guys view on the State of the Union around ethereum today and the rollup centc road map which is everyone's favorite conversation and I think once we get your guys uh sort of perspective on that that's going to inform more of the technical discussion about what do unit chain look like what is rollup boost look like and what were the different uh decisions that went into the technical architecture there so maybe Hayden can I kick it over to you give us your 10,000 foot view on sort of State of the Union around ethereum and rup Centric red map yeah um you know I actually uh am pretty said about the rooll of centric road map

I think it's pretty cool this it was this you know came out of this I believe vital coined the term a few years ago and really what to me the rollup Centric road map means is that ethereum is trying to scale horizontally uh and you know that there's basically two types of scaling vertical scaling horizontal scaling and vertical scaling is basically like let's get be and B machines be and B nodes um and just try to run everything on on one you know essentially uh you know and blockchains are kind of like servers so run on one server they're like servers but with these nice decentralization properties and then uh the you know role of centric road map and horizontal scaling is well you know at some point the hardware capacity of what you can run on one device caps out and it does it caps out

pretty soon actually right and then so to scale beyond that you need to have multiple you know pieces of hardware and there's sort of the the limit of scaling um horizontally is almost unlimited it's really just like how well can you coordinate across you know many different uh you know machines essentially and you know that's when that's why people talk about things like bandwidth and and all of that it comes down to like you know the the limits of scaling horizontally uh but it's far greater scale and and when I think about like where blockchains are going to be and you know I see blockchains as uh basically the the things that are most applications are going to run on in the future uh because they add these really nice properties like provability transparen transparency auditability and

uh you know thing and decentral ation and stuff like that and you know to scale to the to the needs of the world you really need like a huge amount you you need vertical scaling but that you need a huge amount of horizontal scaling as well and so Ro of centric road map is saying like yeah there's a lot of problem like you know let's do that and then what ethereum is now grappling with is how do we do that in a way that uh creates a really good user experience uh that really just just works when you think about the internet right it it has scaled horizontally there's a million servers everywhere they're all connected to each other and you can use applications and they all seamlessly flow between them and it does that using standard information sharing protocols uh SMTP and https and all all this stuff

and uh you know and there so you know what ethereum needs to get right is its abstraction layers its standards um for how we connect different blockchains to each other uh while preserving the things that we care about um with within ethereum and so that that's what the role of centric Ro map is to me and I believe in it I think it's the right path for ethereum and when I think about you know uni chain and what we're doing it's trying to make sure that we get the abstraction layers right and that we you know build a really solid technical foundation and uh you know the right technical standards also what do you think yeah so I would um agree with all of what you just said Hayden I think um you know the goal that ethereum is driving towards is

not anywhere close like the goal is actually pretty far away the goal is something like you know we want a horizontally scalable system with great user experience and it must be Global and decentralized and you know no blockchain ecosystem is anywhere close to this right so it's a lot of the still the future value lies in what the road map is and I think ethereum's road map is very thoughtful and honestly like thinking the furthest ahead on many of these important challenges and I think you know people often confuse like short-term wins in one ecosystem with like you know when you kind of zoom out you know did you make a meaningful step

towards this like overarching goal you didn't really right because you know you can make a blockchain that's very fast you know and like jacks up the not requirements but did you get closer to solving for example the horizontal scalability system or the like Global and decentralized challenge no like you didn't really right and so for example when people say you know the rollup Cent regroupment was a mistake steak or rollups are parasitic like yeah rollups today are probably as a concept underutilized like a lot of rollups look the same you know they just kind of you know copy the the main ethereum layer one client and like you know Jack

centralized segments are like on top of it right and and then they call that Innovation like I think of course we have a long way to go but we have some rollups that are genuinely innovating and I think like uh you know both both in terms of what's already here but also like upand cominging ones like uni chain um that are you know pioneering like very Innovative technology uh and making trade-offs toward whats here and I think we need more of that so I think we rollups differ from kind of ethereum's original sharting Vision in the sense that they are not all homogeneous right um and that they also not uh you know just all kind of spun out of ethereum but that they are actually created by a

venture backed ecosystem where you can have this kind of bottomup innovation and I think that's a good thing so we should lean more into that of course like right now I mean it was always going to be the pain period like if you went back two three years and you know and you know somebody tells me okay we have you know now the first 20 rollups that have some utilization will we have solved interoperability no we probably won't so of course it will look you know like the ecosystem is drifting towards more fragmentation but this is kind of the I mean almost like the the lowest point um in that and uh we will solve it like we will solve fragmentation and um would we rather not

have you know all of these rollups as part of the ethereum ecosystem that are aligned and that that are like bridging to ethereum and posting data to ethereum like I think we we are very glad that we have them and so we now only got to focus on a getting more value out of the rollups that we have making them more competitive with you know alternative layer ones with centralized exchanges and with traditional server infrastructure and we got to make them interoperable and I think then you know ECM is looking actually pretty incredible on now being a painful moment I'll just Qui like I gave a talk actually internally to the to unisoft Labs um where I was trying to explain the RO Centric road map and where we're at as an industry and like all these like questions people have

about fragmentation and I basically compared it to teenage Penguins um where teenage Penguins go through like a really if you if you just Google te Penguins you'll see that like they look extremely awkward um they're like the baby penguins are incredibly cute the adult penguin of human Teenagers by the way too yeah I I actually that's where my talk started and then I like my talk looked and felt too weird when just showing like awkward human teenagers so I pivoted to Penguins um but penguins look extra weird uh as teenagers and that's sort of like the the phase we're in this like Teenage molting penguin phase of blockchains so you know it's also there's a quote hos I can't remember it's some podcasts in the past you've said this there was there's this feature you could view it as a feature of bug of crypto that it's so

relentlessly procyclical and one of the one of the challenges of that is that there can be long periods of time where it looks like a competitor that has made actually a bad decision it looks like they're winning or that it's a good decision but then you have this cleansing fire so to speak and after a period of time and uh the the wheat gets separated from the chaff and I think if you've been operating in crypto for a number of years you've probably experienced this yourself where it looks like hey I'm making a very thoughtful decision here and my competitors don't seem like they're making thoughtful decisions but it looks like they're winning and you just have to have the I don't know conviction to stick it out for a period of time and ultimately hopefully be proven right and I'm sure Hayden you you're smiling I'm sure you

experienced that at Unis swap a number of times I just remember like five years ago you know when Sushi launched and I was like oh that's not sustainable it's gonna run out of tokens in four years and then four years from that day they ran out of tokens and again I'm not taking shots at at competive I I actually think that like all the forks and competition out of D summer really pushed the industry forward and pushed us to do better so I think it was like a good thing for crypto and the world that all that stuff happened but I just remember it's like sometimes it takes four years to be right and or and crypto Twitter like can't think past one to two weeks and so it's a very interesting uh public forum for us to all be communicate again yeah I mean you see it

in so many things right I mean you've seen like companies uh you know right now kind of a la anouncing announcing like large layoffs Etc and it's kind of this like if you don't manage your risk like you'll have the market manage it for you and I think in a lot of ways like the companies like like the projects were kind of you know were more lucky or like did a better job man like managing for example like their road maps their their treasuries Etc didn't over invest in some hype for example that then didn't materialize then it times like this of course now they stand to benefit right by for example scooping up Talent um that that's let go in other places yeah yeah we've hired 45 people since

April ni yeah that's that literally is what it it's a lot of this is just a talent being acquired at you know kind of ridiculous prices and then it floods the market at some time and you got to be in a position to pick up the pieces when that happens so all right so I've got more questions for you actually on the rollup Centric road map but I want to make sure that we got enough time to flesh out what the two of you just launched in tandem and yeah this where maybe uh I want to get into what exactly some of the the aspects of the vision for unit chain as well as rollup boost because I think they're both super interesting in their own right but especially interesting when paired together so uh maybe Hayden can I kick it over to you again to just can you

give us just an overview of you know what did we just launch here and then I want to hear from the two of you I'm very interested in how you guys divided responsibility in between the flash Bots team and the labs Unis swap Labs team when you were developing this yeah um cool happy to get started and I'd say that you know we were actually very excited to get into chain level development uh and we we first started this in you know with our work on EIP um I don't remember the EIP number but the transing storage one and you know for us uh we think that building full stack really gives you a more complete picture to the industry and so that's one of the reasons we like to span all the way from user facing products to chain level development is that it lets us I think

what happens in crypto a lot is that people push their problems on other parts of the stack all right they're like I'm a modular piece in this portion sliver of the stack and all my problems are going to go up here down here uh and if you live up there or down there then you know Screw you and I think that you know one thing that is exciting to me about just making a fora into chain development is that we actually can't push our problems anywhere in the stack because we have infrastructure at every layer of the stack and so we can't offload our problems we have to like solve them with what we think are the best abstraction layers so there's like this meta reason that we're very excited about getting into chain level development which is that we think we can bring something to the table and

when we look look at ethereum's role of centric road map we think we can be a part of that and we can advance it and uh within that context you know we're the way our approach here and what we're trying to do is uh build a chain that takes all of our learnings and uh tries to meet the needs of a specific you know Community uh and and that's really you know the unop community and defi more broadly and so we're you know we're I I one thing that's interesting is people talk about like app chains and try to compare and contrast like is this an app you know there's been this like a lot of I've seen a lot of debates about whether not uni chain is an app chain and it's kind of in some ways it's complicated you can think of it as an app chain you

can think of it as not an app chain uh when you actually think about what is the best design for an app chain or for UNIS swap specific app chain funny enough you still need general purpose uh use cases right because Unis swap is not this like single protocol it's actually like a protocol with many things being built up and on and around it and you know you know unisoft before is coming out right and it has these Hooks and anyone can build a hook how do you create expressivity for general purpose hooks while not having open you know uh permissionless smart contracts and so uh you know uni chain is a chain for Defi and and you know that takes all our learnings from what Unis swap needs uh to be most successful on an L2 right and that's why we're focused on things like

know significantly lowering the block times uh using this uh this te Builder with with flash Bots uh as well as enhancing the decentralization with this you know uni chain validation Network and so I'd say that like you know I think what uni train ultimately becomes is actually not truly up to us it's what you know it's up to the people that use it and and deploy things to it uh but I think that you know one thing that I think is really under talked about in um in blockchain scaling we have this really funny thing that's happening which is basically so if there's a bunch of competing l2s and they all are trying to be like the general purpose L2 uh but when it comes to like horizontal scaling

and what that actually means you know and each of them has a unisoft deployment and Unis swof or a Unis swof Fork is usually the top uh app on every single chain in terms of usage and that's not an accident and it's not even just like due to Unis sw's brand or whatever like that it's it's actually partially because amms eat block space like no other right they they create constant Arbitrage opportunities that just like eat all like the floor gas price is going to be the lowest value Arbitrage in any unisoft pool on that chain because otherwise you know you'll do the Arbitrage and so to actually scale horizontally we actually need to have more focused chains and you might and when it comes to an application

developer right like an application wants low fees so while we have these like l2s that are competing to be you know these defi hubs or or whatever it is the that need to build on them actually need lower costs and and the am collocating on the same block space as an amm will actually raise your costs so what we're going to start seeing is is apps that don't want to co don't want to be in the same block space pricing as an amm that actually need to trade across chain you know when when you when someone wants to make a swap or if you need to swap from an application it might you know bridge and Swap and Bridge back uh and so I think that you know part of the UN the UN vision is to create a non-fragmented uh source of liquidity and uh you know that can

interoperate really quickly and and uh cheaply against other layer twos uh so that it can sort of you the same role that uniswap right right now you have Unis swap on mainnet and other smart contracts trade through it and you're gonna have uni chain as this chain and then you're gonna have other chains that trade through it and so there's a bunch of components that go into that but that's really like part of the vision here and and it's actually necessary and it kind of gets it like the funny incentives across like people building the chains and people building on top of them again is like people want low cost people want to make more fees Which is higher like the people develing the chains want higher fees and the people building the applications want lower fees and so uh you know a non-fragmented kind of cross chain interrupting uh you

know uni chain can really help that yeah I think in terms uh for me the motivation for building uni chain or like for us collaborator on this is actually I think is actually threefold so on the one hand I mean it's good that we started with the rollup Centric road map because of course this is just an incredibly important I think and pivotal topic right now for ethereum and I strongly believe in the rollup Centric road map and um but at the same time like rollups are not maybe like living to their full potential and so this is something that we want to change with rollup boost like we want to make a

small contribution towards rollups eating all of the Innovation from other chains making like it very easy for them to innovate and adopt new features um and become more decentralized as well and overall like become more competitive but in very like in particular uh like for years I've been like every month I look at this you know Dex versus sex volume chart and um I've just been very like for a very long time been very interested in decentralized es and like I've written like Fair bit about the like micr structure of exchanges and like how that maps to blockchains and this is actually what got me into me on ethereum in the first place because when you like think about dexes there's no way that you don't

stumble over me at some point and realize this like incredibly important topic and the way that a DEX outsources certain tasks for example ordering like like unop for like a very long time like has been Outsourcing uh you know the like the matching you know of of like bits and asks to ethereum right in a sense right this is kind of the beauty of like it's so easy to build um an exchange on ethereum because like you can be really focused on like a small part of the stack and like use the like the blockchain for other parts of it right um but of course like it creates these problems and I think if we want to uh you know get away from coinbase get

away from binance is these like very Central hubs for crypto trading then there are specific things that we can do to make decentralized exchanges more competitive and I think that uh you know lower block times is an important one but manage it in a way that is actually me aware uh I think privacy is another huge one and then I think like expressiveness and overall like you know better programmability of your transactions would be a third one and I think that you know un un Chan is pushing like both rollups and decentralized exchanges a lot into that direction where I think if this you know gains meaningful attaction then it'll

end up being you know a much more competitive experience compared to centralized exchanges then uh you can get anywhere else today yeah and even when you think about things like the 250 millisecond block time and where that comes from it's like you know when someone makes a that I humans perceive um you know when you're like clicking on a website you only really notice up to like 150 to 200 milliseconds and so we're kind of pushing towards a boundary of human perception uh which is really the goal right people want things to feel instant they don't want to wait 12 seconds I think the Mev I'm glad you mentioned Mev Mev is really important you know imagine a world where like H you know someone build like a online poker I'll use poker because yeah H he likes Poker but um someone builds like

an online poker thing on and Amazon like inserts their own like employees into the game and use their privileged ability to like see what's happening in the at the server level to like arbitrarily like favor their poker player over like Mev extraction is not a great like we really do like users are ultimately going to choose chains that don't siphon their Mev to other people um that's like pretty obvious to me and so you know a lot of like like especially for like high Mev generating use cases like Unis swop and so I think that you know what we're able to do with this with this Builder and uh you know we we have this this whole priority

ordering uh version of of trade execution that we're doing on top of unit chain that should really help you know have most Mev return basically go right back to the users and there's a bunch of theories uh Phil at at flash has a whole theory about how extracting the meev and then returning it is actually a more efficient market than just uh than just not extracting in the first place uh basically that there's market efficiency to extracting it but then you have to return it I I don't have an incredibly strong opinion about the the the relative differences there but I think that like the high level zoomed out thing is that like users don't want their Mev siphoned and users are going to choose uh things that you know where most Mev is is return to them uh at the risk of going down a rabbit

hole here because I have so many questions for you I would love hosu for you to respond to that because I find that such an interesting debate actually should we be maximally extractive and then decide what to do with it and then return it or should we try to prevent it from ever happening just I would love to get your response and then I'll go back to my list of questions yeah I think for me the difference um is you know which what mechanism can you design that that can be credibly decentralized because you can like in a decentralized setting you can Design Systems where like a user from their side programs their transactions for example to be only usable in a certain way and then you know even if if many like very kind of strongly rational actors like block

Builders and Searchers who are willing to push like any Edge that they can get have no way to use that transaction in a way that's not intended for example right but let's say you implement a different you know uh like a different method where you just say um you know like we we are just Banning me at the protocol layer for example whatever that may mean right like for example we are like kindly asking U validators to please like run this honest ordering mechanism right then you just end up creating a system where um the parties who don't follow the rules end up making more money than than those that do and that's a problem right because you you

like over time you'll end up centralizing your validate ass set and your Builder and sequencer set like to the parties to end up extracting the me and you push out like all of the quote unquote honest parties right and so I think what Phil is uh is saying most of all is that you know you like you can only basically like we all want me to be returned to users and applications basically to the source who creates it but those parties need to take like control of the of their own ordering basically and of their own data like you can't prevent me at the chain level itself uh not if you have like the strong constraint that for example

ethereum layer one has uh for decentralized ation and this is an area where um this constraint is a bit looser for layer twos because they basically rely on the layer one for safety of their state transition and that is that is actually a pretty big deal um uh because then you can have uh me mechanisms on layer 2 that can be incentive compatible there that could probably not be like or that not probably that could definitely not be incentive compatible on layer one and so this is another area where I think you know we can innovate on rollups um but it's not it's not like one to one transferable to the ler one yeah and

that is where you get into maybe I think Dan Robinson had a great write up on this about priority ordering is all you need and this concept of me taxes which could work on an L2 and never on the L1 and that is to me when I looked at this this team up in between flash Bots and Unis swap I was like oh this makes sense to me because I spent a whole season with Dan on Dex design actually and if I had to do a one sentence summary of that entire season it would be reducing leakage from the system and so that's you know how do you make sure that M goes back to users and also how do we make sure that we aren't leaking Meb to the L1 in all of these various ways and so when I looked at unit chain I was

like okay this this solves a lot of those issues you know so um I and we have an implementation of the priority is all un need thing on top of our unit chain test net which is you know and like the the I could quickly explain what that is yeah yeah that'd be great yeah so priority you know and I think that like the PRI I think that like you know if again to like priorities all you need is like kind of a Dan is good at coming with like good memes I think that like one way of frame it would be like priority is pretty good right now um versus like all you need forever hit the same as the title of doesn't as good you know you gotta say PRI but like uh or maybe even like very good right now yeah relative to to everything else but PRI

what we're essentially doing is you know we have this this Builder that is this collaboration with flash Bots and uh that's going to basically be enforcing a few one it's going to be doing the faster you know subblocks to 50 milliseconds the other thing it's going to be doing is really enforcing a fair priority uh gas auction on everything that is fed into it and so you know I I think the cool thing here is really that like you know if you have a a let's say that you do have like strong properties around the priority ordering of transactions within a block right you have like a pretty you know you you have like a pretty strong belief that there was a fair option there and then the transactions that were output were ordered uh in you know whatever paid the most the highest priority fee

but you what you can actually do is you can start to construct intent based transa uh trades transactions that are aware of the priority fee and aware of the the the the you know uh this auction such that the users uh desire the user is represented within it and what I mean by that is uh let's say that you have an order uh that executes at a fixed price plus some multiple on the priority fee now you can't do this uh if you don't have like you can't do this if there's not a fair priority gas auction happening because you know then the priority fee is less meaningful you can have these like side channels and and different ways that the blocker has been built but if you know that like you know and and that's what the T where the te

comes you and you get a stronger property around the the ordering due to Priority uh you can actually say like you know yeah this trade is going to execute at 3,000 us uh usdc per eth plus eight priority fees or 15 priority fees and you can just like fine- tune this priority fee multiple and what that does is like you know the more Mev you leave on the table the bigger of a of a ARB opportunity that creates people start bidding it up um in this priority gas auction but every increment every unit they bid up the priority fee actually increases the output amount to the user so they can only bid it up up until the point where um that output amount makes the they're no longer profitable uh AR uh Arbitrage opportunity so essentially

let's say that you do like nine priority or 10 priority fees then like 90% of the me is going to go back to the user um so here this is an example of like you leave the Mev on the table uh but you and you you let people compete in an auction uh but it gets bit up to the point where uh you know where the the arbitrager breaks even and then or some portion still goes to the arbitrager uh but ultimately most of it is returned to the user and why do this versus like not have Mev I think one thing that's really interesting is it's hard to know how much Mev will be in a transaction at the time you generate it and this allows us to like uh basically parameterize orders in a way that is like a little flexible because uh today you don't if we leave

too much me on the table you know in the current way that trades execute on ethereum it just all gets eaten up there's no user protection and so we just have this like multiple number we can tune you can even create like a an algorithm or an AI tuning system or control system that basically just if transactions are taking too long to get included uh you lower it if they if they're getting included right away you might experiment with raising it right because users also want their transaction to happen in the next block and that's the problem with Dutch auctions is at a duch auction uh you know you have to Decay a price over time and as time passes you know like the longer and so what you're experimenting with is like how long it takes to order execute but people don't want to wait

and actually waiting also can work against the user if the price is falling the thing they're trying to sell or rising of something they're trying to buy so time is you you don't want to like have a huge amount of time for your order to execute um but you also don't want to leave so much me but so you need like a so this this essentially this order format on top of priority auctions to us and in our early experiments feels like a really really good trade-off space where we can have orders that fill in the next block have that that be very very fast we can optimistically if we can optimistically leave like a little bit of Mev on the table because we know that most of it will be rebated and that ultimately I think will lead to like the right trade-off space of user properties

that where we think that most you know trades can essentially execute through this on through this design on uni chain and create really really desirable properties um yeah yeah I think I mean it's hard to overstate just like how uh I mean when you look at me today just like in in terms of that Pi like how big of a problem uh Arbitrage still is um and like how much that is costing I mean LPS and and and and by extension it's costing dexes in terms of liquidity and like users of those dexes in terms of spread and like overall it's making it's making decentralized exchanges less competitive and so uh it's really nice that we are seeing like some innovation

in this area I think um I mean the me texes mechanism we have as one example but then the app specific sequencing there's like more um there's like some things also happening on ethereum itself um that that are coming up and um yeah so I I think overall it's it's it's really nice that we're starting to tackle this and it has effects um also on the uh on the overall rollup ecosystem because like right now when you're in an application um that has meev and it's not able to internalize that to its users because the chain doesn't allow it um then you are I mean pract practically incentivized to like

go make your own chain if you can and um you know we're kind of stopping that leakage as well right and kind of stopping that that fragmentation from happening so you know actually you can be an app on a generalate purpose chain and still like you can have your composability like and you can have your me like you can have your cake and and eat it too and I think that's a trade-off that wasn't possible before yeah and by the way you say it's hard to overstate like we can when we last round the numbers the lower bound was like a billion dollars a year from Unis swop users of our product like you know with like the lower bound it's probably even higher so it's these are literally talked about billions of dollars worth of value that we're trying to return to users yeah and I wantan to if you're

listening to B curve you've probably heard me talk about this before but just a level set with folks who might not be quite as familiar with the lp problem specifically so LPS get arbitraged against the price of binance right so if you've ever heard people talk about this concept of lever loss versus rebalancing that's basically what this is is the fact that price Discovery happens offchain on venues like binance we have a 12C block time and so generally the prices are are stale on ethereum and so LPS basically pay Arbitrage yours to AR back to whatever the the current fair market Price is which is assumed to be binance and what you don't want is LPS paying tons and tons of money over a long period of time to those arbitragers right it's just you you you should pay Arbitrage some amount of money for that

service that they're essentially providing to the protocol but not too much and it's It's Complicated by the way because you know a lot of people also fr lping volatility harvesting and you actually also make money off the Arbitrage like if you're taking a directional bet as an as an LP let's say like I think eth usdc will be flat you can just LP and you know the Arbitrage benefits you if you're like betting directionally um it kind of is like a different form so anyway it's it gets very complicated by the way I'll also leak some some Alpha I think that there's a good hook design probably that uses priority is all you need from the LPS perspective where the lp can set their fees dynamically based on the priority fee if you have strong priority ordering potentially uh and basically like raise its LP fee the higher the

priority fee is uh such that you know it returns Arbitrage to LPS I don't think I've ever talked about this one explicitly with anyone but I I do think it's potentially a good hook design for for V4 and that's why like I said like all these pieces work together um also we'll say low block times extremely good for LPS because you know you're getting a lot more like LPS lose money on long-term price Divergence but they make money on chop along the way all these like tiny little trades like you know if it goes up and then back down they make money right if it goes up and it doesn't return sometimes they lose money that's the difference that gets it like you know you have fees are path dependent and Arbitrage and where what people call permanent loss which is this term I made

up and a lot of people talk about now but uh impermanent loss is you know per or Price Divergence loss is is like a onetime thing based on the difference and so the more small trades you have along the way the more money you make the more likely your feed your Perman loss and lowering the block time and people will be able to Arbitrage against centralized exchanges at this 250 millisecond block time uh that just creates far more lpvs um along the way which which should also be really good for liquidity providers yeah I mean it's it's easy to think about when when you consider Ur in an a stable coin pair I mean maybe you're trading maybe you're likeing tether versus usdc and like internally you just like book them like you just mark them uh one to

one right but then you have a huge buy and maybe you know it's pushing the pool out of balance obviously like you don't care if you're being Arbitrage back like 2 one to one that's in your interest like you're purely volatility harvesting and the lower the block time is the more sequences there will be where two you know two assets actually like mean revert and you're making the fees on that and so yeah the more you can have that the better for you and when speak to get to like compounding returns there it's also the lower the fees so we it's like a double double benefit here which is lower fees and lower block times equals far more Arbitrage there and now we're really going down rabbit holes here but there's there's a cool thing about by the way the rollup Centric road map which I

because the the trade-off right of reducing lever with shorter block times is you increase the risk of timing games like this is one of the things that uh like l1s with shorter block times have to mitigate against and so it is kind of nice to have to isolate those two different challenges to different layers in the stack which I find cool I find that to be cool hey everyone this episode is brought to you by Fuel Labs who's unveiling their new high performance ethereum L2 fuel ignition many of you know fuel as the launcher of the first optimistic stage two rollup on ethereum with fuel V1 and now they're launching ignition to address some of the creeping centralization concerns around some uh L2 architectures today and if you like nting out about the tech

there's a lot of reasons to get excited about ignition including how they solve State sustainability and performance execution they have this utxo asset Centric architecture that's very very cool but there's a lot of other reasons to love it super easy to use you can either use a fuette wallet or your existing ethereum or Salon based wallet like Phantom or metamask or whatever it is there are cool apps that you can experiment with and again like I said at the top of the program they're launching phase two of their points program so click the link at the bottom of the show notes head over start having fun on fuel ignition today just tell them that I sent you um all right but before we I I want to ask um you know some more questions about like uni chain Vision to you Hayden and actually an analog that I

think is sort of cool here as well is some of the uh yeah had a a talk actually with sunny agoal from osmosis um a couple of years ago actually and he was the one who first pointed out to me this kind of um Paradox in between being a permissionless app chain because and it was kind of that was the first step in my own journey of like blurring the lines in between general purpose and app chains and I've kind of oscillated like it was everything general purpose or is everything in app chain or is it kind of this purpose-built thing in the middle that's what my you know when I was listening when I read the white paper and the the announcement for uni chain it feels to me a little bit like a purpose-built chain for Defi and a liquidity layer which is somewhere in

between an app chain and a I've called this like sector specific block space I don't have a catch catchy name for it but it's somewhere in between app chain and and general purpose probably not that yeah probably not that yeah sorry dude not my forte to come up with these taglines uh no Dan Robinson there but that's what it seems like to me and could you just confirm or deny that Hayden and then I would love to get a sense of and then I want to get a sense of where this fits in with the other parts of Unis Swap and specifically maybe if we could concentrate on like V4 and hooks I'm very curious like yeah yeah I I think it's a really good question and it's it's it's a funny one because there is like this kind of paradox here where uh to be like I I think I think what you

say sounds about right it's like it's kind of like purpose built for for for defi um I think that we'll see over time like how that evolves it's possible that in you know 10 years from now we have actually single chain per pool on this off right we don't know what you know how or at least top pools might need their own chain because of you know the the horizontal scalability of like ethos DC just needs so many transactions per second right uh so we don't really know what that ultimately looks like but I think that uh in in the world of of today you know the you need like you you could do a version of of uni chain right where you have like all this custom logic such that the entire chain itself is you know is is the protocol right and

like that's actually not too far from um what dydx is doing for example uh right they have like their chain and their chain is also the order book for their purose thing and um that is like a thing you could do and there might be efficiency benefits to like making the entire chain you know at the technical level the protocol as well um but when it comes to chain and what Unis swop means right Unis swop is not just a single smart contract protocol it's like a world of things that interact with it and are built on top of it and you know when you think about even like the vision of unisoft V4 right and the vision of unisoft V4 is to say well there's probably actually not one best design for an amm uh there's actually probably many

different designs and we need to have like an open world of experimentation without everyone rebuilding an an am from scratch right and building you know something like unisoft V3 or V4 is like really hard right we had to you know we you know it's like the even just the process of making it safe and secure and whatever and so with unof before we we're saying let's build this like flexible implementation where people can customize their pool designs but still have certain like safety checks and and properties that are that are guaranteed by the broader system and but but hooks are just general purpose smart contracts that unisoft pools call into to to express their logic uh and so when it comes to like a uni chain you know how do you have support for hooks while not letting people have permissionless Smart

contract deployment and you know it's actually impossible right so you need to be a general purpose chain you can't just be you know all hardcoded uh but there are things that are are optimized for UNIS swap that might not be or for the Unis swap use case but Unis swap even as like a broader one thing is that like things that we do that are good for UNIS swop are probably good for other things as well and so we're not taking the opinion like there might be this other case we can make which is like this is only for UNIS Swap and things built on Unis swop but what does that really mean you know there's this there's like blurred line of like what should and shouldn't collocate and I think that like the right thing to do is kind of like let the market sort of

determine where things collocate um and build like I think that like things that need to be ultra Ultra Ultra Ultra cheap in like like in the long run you're going to need to have separate chains for like games or whatever right because a game isn't gonna you know want to like like it might need its own you know sequencing logic it might need its own you know thing to to you it might need to be like incredibly cheap and not want to collocate with anything else uh but you know uh really like I'd say you know it's what you said it's it's a it's a it's a you know kind of a a chain for a specific Community instead of use cases and you know I think that it will kind

of people will self- select into that and uh and we'll figure out what that means over the over the years so I've got a a follow-up question here and this is a little bit more directed to Hayden but host I want to get your opinion as well because this is something you've talked about when it comes to governance so something that I've noticed and you talked about this a little bit at the top of the episode already Hayden is that um there's this been historically I feel like there's kind of been the infra people in crypto and the app people in crypto and you see the line in increasingly blurring there and it always has blurred in ecosystems like Cosmos where you have to do both a little bit but starting to see this trend of actually people starting with building an application and then

reversing going down deeper into the infrastructure because this is one of the the chicken and egg problems when it comes to apps versus infrastructure is hey the infrastructure should support the application in the use case but we don't have any takeoff application and use cases so what goes first and I'm starting to see this trend of people that built an application like the infra for me doesn't exist yet and so they're building that and that's what it seems like Unis swap is doing and so and and what you kind of end up with is something that looks a little bit more word has always been dirty in crypto but a little bit more vertically integrated or full stack right where you control more of the entire user experience and I like fck more than vertically me too but do you agree with that is that roughly like a trend that

you see playing out yeah yeah actually so I would say that interestingly with un I say like for example coinbase is starting fully at the application layer and moving down the stack um other people are starting in the chain level and trying to move up a stack maybe like even like Salo with their phone or something which could be seen as doing that um I think unof is actually somehow like starting in the middle and moving both directions uh because we we have this like we we really start with this like lowlevel smart contract protocol thing and then we built an a web app on top of it and then we actually built a a wallet you know which is one St even closer to the user and then we started getting into chain development and and we have you know our EIP work and the

uni chain so we're actually kind of like pushing in both directions but I'm a pretty big believer that like some of the that there's a lot of value to building full stack because you get a bigger Pi a bigger set of the picture um and I found that like even like in the process of building a wallet in the process of building when we said to build a wallet for example we looked like the decision partially came from the fact that we were like looked at our some numbers right one of the numbers was like most users in ethereum were you making their first interaction on Unis Swap and even within our interface like and then we looked at our support tickets and most support tickets were complaining about a problem that was in a their wallet and outside of our control and we were like oh okay uh you

know we can't solve our we're like the on-ramp to crypto and people can't we can't solve our the our biggest problem which is like and so we we wanted to create a you know a experience where people can land on Unis swap use it for the first time and have that be like a high quality user experience without being pushed to to another you know app or you and when you think about like the mental model for like how people use the internet today people are not used to like like the the user experience literally used to be I went to unisoft for the first time okay now I have to go to coinbase create a coinbase account to buy crypto then I have to go to download metamask from the Chrome Store and send crypto from coinbase to met mask and then connect my metamask to Unis swap

like that's not a real user experience that's like a weird crypto thing and the support for third party wallets is actually valuable right I I believe that we should separate self uh custody from from you know exchange right like I think we should like separate out like I think that the the abstraction layers are important technologically but the user experiences still need to be cohesive and we need to offer paths for people that don't want uh you know this like extremely technical complex experience and it's not that different you know that that example I gave on the wall level is not that different from the chain level which is that like people are making technical decisions about their chains that are like massively affecting the protocols built on top of them without really understanding those decisions um and so what I'm what I'm

very excited about unop Labs is like contributions to the L2 scaling world is that we are really starting to span this full stack and we're able to think across it and you know it happens all the time now in our day-to-day where things that we know about our users inform protocol level decisions we do um we make and and things we do at the protocol level or protocol expertise informs what is and isn't possible and how we approach user facing problems at the wallet level and so I you know uh the the end result of all of this should be like uh you know much higher quality products built on top of blockchains in my opinion we're not saying that like sometimes when people frame these as like vertical integration they think of as like they're trying to build a monopoly and if we were trying to build

a monopoly we would do things very very differently right because monopolies you know have full control and discretion at each point in the stack right and so like you know for example like Twitter was able to shut off its API um Reddit was able to shut out F API and rug all the the you know applications building on top of the Twitter API because they fully controlled it right the cool thing is that we build on top of the unisoft protocol um right and and we don't have the ability to like shut it off right and that's kind of the so like the cool thing is that there's competition at each level of the stack right there's no Monopoly because the technology is decentral ized and permissionless uh but uh so you know building full stack isn't a way to like get power it's a way to

you know build better products and have a voice in the in the quality of the technology and the quality of the products and the abstraction layers that are that they're that they're building on top of them it doesn't really actually create more structural power the way that a mon like a vertical Monopoly would in trafi because there you own the API you own the endpoint and and and without and without the company the thing doesn't exist here you know if you swap Labs were to shut down the chain would continue to run the protocol continue to run uh and and everything would be just fine yeah I think sometimes you know it makes sense to unbundle and sometimes it makes sense to bundle and I mean flashboards has historically always like been a an unbundling company if you will right we've like unbundled block building from

ethereum validators and we've unbundled the meol in ethereum like in a sense we also unbundling sequencing in layer twos because you know uni chain is you know Outsourcing some of the sequencing or like the block building duties like now to a third party right so in a lot of sense like value can be created both ways right and um but in other cases then it makes sense to to bundle right and I think especially with regards to the user experience um it makes a lot of sense I like the lp problem for example that we were talking about all of that money lost to Arbitrage that that that just simply wasn't a like a problem that you can easily solve by waiting for someone and like you know CO's theory of

the firm you know um says you know it's easier to coordinate you know as part of the same company right or part of the same ecosystem let's say right but uh you know some some for some problems you just need you know to have a kind of smaller setting to solve them and then other parties and this is like the beauty of crypto as well right with open source and everything and then other parties can benefit from The Innovation that was pioneered in this more integrated model and so I think yeah it'll always be like a mix of both and both have their merits yeah and I think it probably does just depend a lot on the product like I think you and I have talked about this within the context of Lio and Ste which I know you're uh Le last time we caught

up about this would be very much in the camp of very thin layer um and very incredibly neutral so maybe there's no good one siiz fits all here just like many many it's really different depending on like what you're working on for sure and if you're if you're like if you're like talking about the trading experience of users then I mean that that that uh you know includes so many things right from like downloading wallet the the first time storing your private Keys uh account recovery buying your first crypto transacting on chain uh you know so many different things that if you're like if you're just qu Cote an application then maybe you control like 30% of that right and so really for the other 70% you're at the

mercy of other people in the stack yeah and yeah I I think that like um oftentimes people yeah I'll take the like strong stance that I think we need more companies building more full sack in crypto like I think that a lot of crypto problems today have do come and I agree with with hu brought take about like sometimes it makes more sense sometimes it's good to you know unbundle sometimes it's good to bundle but I say that like today in crypto I think too many there's not enough companies that are trying to span the stack and a lot of people are offloading their problems onto all their part the stack and or like even or it's not even intentional it's like sometimes it's actually they're trying to solve a problem and they create a worse one um you see this kind of all the time with

like approach you know even like um uh and I think that like you know it it comes up in in all sorts of like eips or com whatever like it's like well users want this and it's like well do you truly know that and you know it's you like uh it came up around like the C abstraction EIP where like you know I was kind of weighing into these debates I'm I'm pretty happy with the the end result of it I I think that there was but there was sort of like this conversation around like should you have revocable or irrev like should it be possible to have an irrevocable um implementation of account abstraction one of these account abstraction APS I argued that we should uh people on like like you know core developers through the like people were like The Stance from that is like well actually you know

users should always have the ability to revoke but by making it that the case you actually prevent people that Implement these EIP uh you know these these account of instuction smart contracts from offering guarantees about them right because if you can't if you can revoke then anyone in another wallet can revoke and then like let's say you want like a long-term account recovery mechanism like un wallet let's say un wallet we're deploying like an implementation of an account obstruction IP smart contract because they're all based on Smart contracts now that you know gives you key recovery uh well and then someone imports their key into a different wallet like metamask and then metamask prompts them to to revoke it to because you can only have one at a time the user thought when they created their account their you know their thing their

when they signed in unisoft wallet that it was like a permanent recovery method but now it got just deleted by another wallet and so like this was like a decision to say well users need more choice but actually it prevents the application from providing the guarantees to the users that they're looking to the applications to provide them and so it's it's like this very complicated trade-off space where you know and and that's why again we're we're excited to like try to bring a perspective that we think is sometimes missing uh to the table it's it's very interesting and maybe we can when we close at our more 10,000 foot vantage point we can talk about this a little bit but it's a very interesting question about who should be the voice of the user when it comes to ethereum you could make a strong argument that it's

actually the pure you know the core devs right who don't have these uh you know commercial interests but they're also much further away from the user and might have less of an idea of what they actually want so I don't know fascinating I think the right there's no right answer think it's like a conversation and we just and so it's I think part of it is just becoming part of that conversation and it's not really about like we actually you know I think that like there's a lot of value to the kind of like um fully technical kind of focused version of it or like trying to think entirely from first principles I think you just have to like marry that against like actual like you know we we took we've we've gotten a lot more pragmatic over the past like year or two like we we started like talking to our

users more we built out this you know customer experience team we started talking to our users we started like tracking data around what people want right like and it's like you learn a lot from just talking to your users and like like an example is that we used to have this kind of like uh very principled reason that we hated Fon transfer tokens which is like they're really annoying like they make every protocol more complicated there's a lot of weird implementations uh they kind of like break break like the clean functionality of like the transfer function transfer is what you expect um and you know so we like just basically didn't support them in our products for a really long time there was a point in time about like a year and a half ago where we like we're like looking at like failure rates and

numbers and we realized that 50% of all transaction failures in our interface was Fon transfer tokens and we're like okay our users clearly want these how can we make them the best version of it and so we like did a bunch of work to try to support as many as we could and you know like ultimately give the users what they want you know people want tokens that that take fees on on the transfer function for whatever reason and you know so I think that like there's just like value just like pragmatically learning about what your users want and talking to them and I think that like that is something that ethereum could get better at um but again like also there's a lot of value to these like the way like ethereum is where it is like ethereum is very successful for a reason and there is a

lot of really thoughtful technical decision- making and discussion that happens I think Cavs also get a lot of slack that is undeserved like I think that they for the most part like you know mean very well and build really high quality systems and so it's just like a conversation and then you know completely completely and Hayden I've got one more follow-up question for you on Unis Swap and then H I want to nerd out with you a little bit about me and rollup boost uh but there was there was a little bit of uh I don't know push back maybe I'd say on Twitter Hayden about uh you know this everyone was I think very excited about uni chain but there was uh one course of voices which was like hey was the DOW involved here and what does governance look like in between Labs versus the Dow Etc so just

any any kind of response to that about the ownership of unit chain and you know how that decision-making responsibility is split yeah maybe one place to start with that is just like like there are many different ways of achieving decentralization uh in the world I have like a tier list of decentralization in my brain and my tier list is like you know the best form of decentralization is like automation like if it just works right when you think about unisoft V1 on ethereum like it just worked and it didn't need like an active governance right and so like the best form of decentralization is like is just true automation it just automatically does the thing it's supposed to do um I'd say like the second best tier of decentral is like economic uh based basically like economic incentives like you know you

know that Unis swap pools will always be arbitraged because there's an economic incentive for it to always be arbitraged right so the sort of like supply and demand Marketplace that Unis swap creates is like it decentralized in part because of the economic incentives around it and and the kind of game theory of those incentives right and that's a form of decentralization and I say that like there's like and then it's like things that can't be and so like things that should be automated should be automated things that can't be aut automated if they can be incentiv like automatically incentivized should be automatically incentivized and then things that can't there that's where like governance uh you know are more like then then how and then the next year of decentralization is like how do you have like a more distributed form of of of human governance right like and

and I think that like you know the ways that you further decentralize human governance is you have broader distribution of power but I think that like a lot of the point of these blockchain systems is to to to reduce the requirement on you know human voting governance and so I'd say that like you know when it comes to UNI chain uh one thing I'll say is that like the intention of The Da was never to be like a technical Vision Setter or technical decision maker uh that's never really been its role it's really been about like a limited form of governance over a particular set of things right um it has like the you know the governance treasury it has you know uh voting over over things like the you know um the different onchain parameters it has like feers and whatever right and so you know

there's never really been a version of un swop that was like created out through like you know Dow delegates voting on the technical decisions um and I I don't think that's like this I you know so I don't know that's kind of and I don't think that a lot of I think a lot of there's a lot of people that seem to want that or have that sort of vision but I think there's a lot of people that also don't oh one other thing that I think is uh worth flagging on you that I think is pretty interesting is the uni chain validation Network um I was gonna say I forgot to bring that up yeah but that's and I think that like one thing that's interesting is is unof has always been this like kind of like governance token um you know for the community to

to participate in and uh you know in in uni chain for the first time there's going to be this uh more direct form of participation uh through you know validating the network and running nodes and I think that that you know it's almost like a transition from Pure you know governance to like actually it also does these other things like uh you know it has its direct utility it's almost like a network token within uni chain and I think that that's pretty cool as well um and kind of like a a cool opportunity and I think a lot of people same way that some people seem frustrated on Twitter I think a lot of people are also excited about what all this means um so yeah awesome I've got follow-up questions for you on that actually when we when we end but now hust I want I want to turn it over to

you and you know you and I did this whole season on me and we generally approach things through the lens of the me transaction supply chain on eth L1 and you know like I said I'll link it in the show notes here but about a year ago you put together this great talk on uh what PBS should look like on l2s and now the L2 space is much more developed um than I would guess at that time and you guys have been building a product to serve this like rollup boost in the background this whole time and so can you just give the audience a sense of you know I think most people their framework for this is thinking about how me works on eth L1 but talk to like what are the big differences in between solving or or working with Mev on l2s

versus L1 then we can get into some of the details around roller boost sure yeah so on ethereum layer one I would say say we have a very developed meev supply chain so we have the me supply chain has a lot of layers and you have competition on many layers of the stack um I would say uh The transacting Experience on ethereum has been getting a lot better over the last four to five years since we've started talking about me like applications have become much more me aware on eum layer one right like you see much more like me aware

designs like there's uh you know better liquidations like less Arbitrage fewer sages uh fewer like weird weird forms of meev you have um I would say like an unbundling of kind of the the public mol like you no longer see a lot of transactions going through the public manol at all both because like apps users and wallets understand the that you know the uh the information contained in their transactions has Financial value and so they they started sending it through these private channels um you see much better transaction pricing mechanisms like you see me that's like being returned to users um and I would say like in spite

of you know the block Builder Market when you look there now I think this has always been like the Achilles he like that you Outsource a lot of tasks like of these important tasks to uh like a market of external parties but this role has a lot of network effect and so you'll end up uh you know with like a small number of block Builders and so that's what you're seeing right now you have two block Builders building about 90% of blocks um that said these block Builders are very very competitive at like a per transaction level even at like a per block level and so uh what you're seeing is um these block Builders are not making any like outsized uh profits that said it still is uh a

potential choke point for the network and so uh our goal uh has always been and is to decentralize the block building role itself and I think that's that's roughly where we are like um in ethereum like I can talk about some of the exciting um kind of advancements that that that we have made and that that others have made like towards the score but very roughly in eum Layer Two I think you are at a much more kind of early stage when it comes to me and so you don't have really much of this infrastructure built out um rollups are very still like bundled or vertically integrated in the sense that

like in ethereum you have this separation you have the you know you have the testers you have proposers and you have block Builders um they are all kind of different roles run by different parties and keep each other in check in eum ler 2 you have this was the actually the the topic of my post for people to like start thinking about the sequencer actually has this like dual role it's a proposer but it's also a block Builder and uh there is like no rollups as of today that that have started to to unbundle this there's no rollup that is outsourcing block building so uni chain will be the first uh to do this and um the benefit uh is that you can

get a lot of you know kind of the same effects the same user guarantees um on the layer two that we now have on the layer one and so um for example uh what unit what unit chain is getting through um their block Builder is uh they can run an explicit MV auction every 250 milliseconds they can commit to a specific block building algorithm which then allows you know stuff like me taxes to be developed on top um but you can also have uh rever protection uh for transactions something actually we hadn't mentioned so far in this episode but which is also a very big deal um because when when transaction when

transactions uh you know fail it's uh it's pretty much among the worst experience for users um that you can have and um it's actually you know in spite of fail transactions also paying transaction fees uh actually turns out that uh your Revenue will be higher when you filter out reverting transactions yeah I had a a question on the list of benefits there because the the revert protection thing is actually one ux one thing on the ux side that I think ethereum has a bit of a leg up on salana there's no partial reverts on something like salana which uh for a user it ends up you you end up paying for a lot of failed transactions which the transaction Fe are pretty low but it's it's a ux thing so you're like click click click click click which is a

little frustrating um but oh well here I I got to before we get there I I want to land the plane on like the other big part of this that we haven't talked about which is the tees so one of the cool thing and I'll actually I'll link this in because host we did a bunch of episodes together that season but you had a great explanation of the evolution of the sort of flash Bots product Suite where you were in me in the beginning of ethereum when there were just a couple of big minor it was actually sort of much easier in some senses to design a solution because you could rely on the trust of that that small number of actors it was a huge challenge when it became how do we serve all of these you know you know tens of thousands of decentralized validators

and now it feels like we're actually moving back to a space where there are you know I'll put but like more a smaller number of actors right so you kind of open up the design space in terms of some of these solutions that you have and then you also introduce tees so if you want to maybe comment on that that first part and then describe what tees are that would be excellent yeah so I mean if you kind of vitalic had this really interesting post already now a couple years ago I think it was on the definition of decentralization and he draws out this dichotomy between um like uh logical uh centralization and Architectural centralization and ethereum itself is a system that is architecturally decentralized but is logically centralized and this you know allows it

it gives it all of its like unique benefits for example the fact that it has like a single virtual machine that it has a single piece of state that is maintained across all of these different nodes that is what that's what it's meant by like they all do the same thing right every block they they process they they use the same state transition functions to apply like new transactions to the state to arrive at the same new state that's the whole purpose of a blockchain and so it turns out that um when you have a architecturally when you when you have a system that is logically centralized you can get a lot of benefits uh without the drawbacks um of architecture centralization as well and block building is is one such example like

block building is an is an example where uh when you have one block builder then you can actually have a lot of benefits from this like this block Builder um Can much more effectively like price transactions they know of all all of the transactions that exist and they can for example charge the users exactly the lowest price they need to get into the block and like don't charge them them anything more um the block Builder can also enforce uh for example additional rules imposed by applications you know they can filter for Hax and like uh anything that like breaks specific invariant that like you know applications impose uh or you can have the block Builder like

perform like computation maybe just in time for the block to be built like for users to make their transactions really programmable like there's a lot of cool things that you can do when you have like an logically centralized block Builder but you need to design it in a way that is architecturally decentralized and you know that doesn't doesn't just mean um you know it's like run by uh a party you trust or that's run in a te it it also includes think like geographical decentralization because that's you know what everything eventually like collapses to like the lowest layer kind of accountability in the world is like you know where is your Hardware located like who can come you

know knock down your door and like hold you accountable to like local laws and regulations and so um this was a kind of a long windup but um the beautiful thing behind this you know what we call tees trusted execution environments um is that it's functionally it's a way to kind of separate um uh kind of the ability to run or like like what you gain is the ability to uh have trust like software that you can trust run by uh untrusted parties um and so this is a you know a very big deal because you no longer have

to trust you know the the person running the machine um and the way we do this is I mean we didn't design tees tees are designed by Intel AMD like many others um so is through these secure areas uh in a in a processing unit that are designed to protect the Integrity of code and data and so you can you know imagine a te is a special kind of server that also commits to the code that's running on the server makes it verifiable uh by anyone and it keeps the computation that happening in the server private so I mean Hayden I think you're in like previous podcast you use this example of you know like a a blockchain is this

kind of magic server with magical properties um like I sometimes say like a like a blockchain is a server in a straight jacket and so uh because it like it can't do any harm right it can't do anything that you don't want it to do and like a te is kind of a very similar thing but it you know it also has blindfolds on um so it can't even see what it's doing like it knows what it's doing but you know it can't Peak and it can't do anything with this data so it's in a lot of ways it's a it's a complement uh I think to uh the public blockchains that we have today and also to like cryptography techniques like zero knowledge um like very specifically with the idea of separating the trust

guarantees from the hardware provider and uh from the data that they operate on Yeah by way I'd love to quickly hop I know we're probably running low on time but I'd love to quickly hop in on tees for a second just because I'm sniped um I think that like I just tees get like a lot of undo like there there's I I've learned te Twitter as a whole contentious place but um I think that you know the way I think about it is that it gets right back to that like those like abstraction layers and full stack thinking um where a lot of people still like sometimes people view tees as like a lazy solution to like solving a thing that decentralized consensus should solve or that's like the criticism it gets um but I think what people don't understand is that like there's a lot of like open

like tees aren't purely repl in like they're not replacing decentralized consensus or nodes right like in uni chain you still have like the validation Network you also still have the um the uh you know L1 consensus right but what they are replacing is things that today often happen in a more centralized way right like blockchains are not just the chain they're like everything built on top of it and around it and today a lot of that happens in servers and by the way a lot of ethereum nodes are also run in things like Amazon AWS and right and so uh tees represent like you know a way to me they represent a way of like adding on better trust and better decentralization and better you know security Properties or privacy properties all of these things to things

that currently lack them not to things that currently have them right ethereum L1 isn't going to be like that much you know if we're not going to run all of ethereum L1 in inside a te right we're going to continue to have this decentralized consensus system but for this for for what we're doing at the Builder level right this is like replacing things that are happening in in various like you know block Builders I'd say that like you know the the properties you can get are are really desirable anyway all have to say like decentralization is like a spectrum and it happens at many parts of the stack and tees represent like the ability to further decentralize things that are on the fringes between what is decentralized and centralized today and uh I'll I'll also just add that like uh ever since I started adding them to my

mental models and design space I realized how many cool things you can do with them and that's why you get things like uh you know the flash Bots like teleport thing which is like you can actually because they're really just like you know uh you know you could start to like connect block like you can start adding decentralization properties to things that aren't even close to having them today yeah um and that you could do a better version of that right um and a very good example of this by the way is like you know the deploy process for like a an application today like those you know you almost like if you have something in the Chrome web store you almost inherently have to have like a chrome admin account to deploy your app or you have something in the Apple store right but if you could you

start creating these credentials inside te you could have a guarantee that the code is Open Source before it's published to the App Store so things like this are like ways of enhancing the decentralization of things that sit around decentralized things anyway so I I like tees they're pretty yeah so I think uh I mean kind of uh the the the the nice thing that we're getting here from tees is to remove a lot of Trust From the Block building process and so when you specifically I mean whether you look in into our layer one te block Builder so we have a block Builder that's competitive with other block builders on eum layer one that's also running in a te and then on uni chain um you know the

block Builder will also be running in a te so I mean what you get is um very specifically those block Builders they can attest to the code that they're running so you can you know how your transaction will be processed it will be processed exactly to the rules that the block builders that the block Builder says it will M and uh it's it's keeping nobody no Searcher no nobody can Peak not even the operator of the hardw can peek at your transaction yeah so it's it's uh you know it will at some point stay fully encrypted um within the te uh and then only the final block uh will ever make it out of the machine and so there you know in options in general

like there's the famous problem you know of residual bits you know for example like when a Searcher is bidding uh for an opportunity but they didn't get the item right so this is also like an extremely valuable uh piece of information that the block Builders have actually from The Searchers who submit bundles to them and so we like making the whole um you know just by keeping all of this data in a te and like making sure that it can never leave uh you're just making the whole stack so much more resilient okay guys this is this is great and um you know I got a million more questions here but uh yeah I'm going to link by the way both the the unit chain white paper and the rollup um boost white paper as well so if you have

more questions on some of the technical spec specificities here i' highly recommend that you read it and maybe ping Hayden and hu and maybe they'll get back to I'm not sure not volunteer your guys time uh but I want I want to end where we started and zoom out and ask you guys some more questions about just this like moment in time that we're in around ethereum and how some of the stuff that we've talked about in our episode today slots into that and you know one question that I have for you guys we started by reaffirming both of your faith in the rollup Centric road map let me poke at that just for a second here and I One almost philosophical level question that I find interesting to ask is what should be the Providence of like what should be within the realm of the free market to decide

as a solution versus whatever our solution is for like government here like public uh you know public versus private Market Solutions and you know I there's a really influential piece of writing that barnabe did uh from the ethereum foundation called seeing like a protocol which was very influential to me in terms of how I thought about this and you know basically I've got this dividing line in my head now like this is what the L1 wants to directly concern itself with solving and this is what it doesn't want to concern itself with solving and the I guess I'd just be curious to get your guys perspective on because on the one hand I like the idea of pushing more complexity to the free market I think that it that has historically shown that that's a better solution on the other hand the criticism of that would be well now you have all

these different people with different types of incentives so it's harder to move or maybe people are acting in the best commercial uh interest of their business as opposed to the broader ethereum L1 landscape so how do you guys respond to some of that push back what are some of the potential cons I suppose yeah uh potential cons of rollup Centric road map I happy to like you let me know if I'm like off get get off track or if I'm not answering because it's a complex question but what I'll say is that like one I think that one issue ethereum has is like people think pretty negative sometimes sometimes and I think that like fundamentally I believe blockchains are like positive some things where we can create like net new wealth and and and value and access in the world and I think that like a lot of the criticism I

see of the worldcentric thread map comes from zero some thinking where people think like well if defi is not on L1 then all that value go to some other token and E each should get all that value and I think that's like a really zero some way to think of it um I think that like you know uh the scale of what blockchains can be and what they can do and what ethereum can be be and what it can do is like so much bigger than you know the execution of defi transactions um right it's it's like you know and there's just like no like what what would be a huge failure of ethereum in my mind is if it just pursued like maximally like maximum node size at all costs to have everything try to happen

on the L1 because when you think about where we started and like horizont like horizontal scaling does massively outpace you know vertical scaling we need to have you know many chains to scale blockchains and ethereum needs to think about like what is its role within that world right and I think that it's very clear today what it's it's proposing it be which is like well maybe it's not there's there are some slightly like and then I think then then it gets at like business models and like economic streams which we can talk about in a sec but like the you know like ethereum today is basically providing data availability to rollups um and I I think that like you know that's that's what these like data shards are where

these blobs are right and I think that you know if it instead shifted its Pro Focus to like scaling up defi transactions on L1 I think that like that would come at the expense of all the chains that are building uh you know rollups on top of it I think you know from my perspective the you know I think ethereum should essentially try to be like money some I actually like I used to criticize eth's money I'm actually kind of on board with it I like the idea of like having an alternative stor value that is kind of like built for this economy um I but I don't think it should be like the I think the core focus of ethereum should be like should it should have like a a vision and mission for what trying to do and that should be like this like highly decentralized but

also highly scalable um uh you know blockchain ecosystem and then when a lot of people like criticize it they're like basically afraid that money will not all go to eth and I think that that's like a really bad mindset to have right because I think that like you know having other things that are valuable built on top of ethereum will drive value to ethereum right um and and the thing that will actually the failure mode is holding it back the failure mode is like preventing progress people kind of like are afraid of change very often um versus like you know like like ethereum's yeah we we all used to we loved when ethereum was sub one cent transactions and everything ran on L1 and we didn't have to worry about liquidity fragmentation because there

was only one chain but that's not the future that's not what it will look like in five years right and it's kind of like people kind of like want to go back to that and I think that what we have to do is actually figure out how do we make it just work in the future uh in a more scaleable model and that's what rups are trying to do and then like and then yeah how you know what makes giv e value well ideally it will be providing a huge amount of of data uh in a very very decentralized way uh one thing that I think that is like an underpinning of some of this I do think that ethereum should have an enshrined proving system in it you know I'd say that rollups rollups are doing a multiple things right they're doing proving systems but

they're also doing you know um sequencing there's a bunch of components to it I do think that in the long run like L1 social consensus should and will protect a decentralized pro like a proving system for l2s um you know we when you think about like um today uh you know if a rollup failed with ethereum fork probably not but if you had an enshrined proving system you know an in shrin snar Stark or snark system or whatever it was uh I think that then you could have like you could build things on top of that that you knew would Fork if if uh or that you knew ethereum's social consensus would ultimately protect so I do think that like ethereum should in some ways ethereum should play a more active role

and in some ways it should but it should also be doing it with like a positive some n mindset that like seeks to like you know spread and and have more and more chains not less because I think eventually we're gonna have literally you know millions and millions of chains and uh you know the the mindset of there should only be one chain is it really very good one Al what do you think yeah I don't know that I have that much to add I think um I still remember when EIP 1559 happened a few years ago I was one of the the bigger supporters I mean I thought it's like upgrading uh the transaction like the transaction mechanism like the pricing mechanism a lot of like very nice upgrade for that

um but in hindsight I think you know the idea of having the burn I mean it's like it was kind of you know two inone package like you you couldn't really have it without the burn uh but the fact that we got it I think uh in a lot of ways is like backfiring very bearish like it was very bearish for etherum because now all of a sudden like people are looking at this number and like you know oh is ethereum making money oh is it losing money like who gives a like e like if ethereum was a company in this like uh it's not but like at this stage of its life like it should be definitely not concerned you know with how much money it's making and uh or especially

BuyBacks yeah yeah yeah exactly like um that's a better framing yeah like it should definitely like not uh you know people shouldn't like look at oh how much money is it making because like you go on Twitter now you see um oh yeah we should actually like raise blob prices you know so right because right now the like the the the demand to cons blobs is like under the supply that we have like only barely but it is and so like the blobs are nearly free so oh yeah we should like lower you know the supply just to you know make people pay more and I think some of this stuff is like very very like self-destructive right like ethereum should be very focused on this you know long-term goal that we talked about in the beginning like I

mean i' say it again it needs to be very scalable with like a good ux and it needs to be Global and decentralized and um like every like bit of progress that we make towards this is way more important than you know any you know 10 so even like $ 10000 million dollar of buyback that we can make at this stage for E folders so uh I think this is what people should be focused on and this is also how they should evaluate um the ethereum road map to me this doesn't mean for example that we shouldn't like improve the ethereum layer one like ethereum layer one should improve like as much as kind of technical progress allows because rollups also benefit from the etherum layer one getting better right like Blue Chip defi like The Bluest of Blue Chip defi will probably

you know there will always some of it will stay on etherum ler one a lot of the liquidity will stay on eum layer one because that's the place where you where you bridge out from into these different uh ecosystems right and ultimately where I think a lot of things will get settled at the end of the day and so sorry a second yeah I was done anyway so I mean like that's so yeah mean eum should push the lay the eum layer one it's like not but it should also support the role and I think both of those are very important well it's just hilarious to me that like when you see like like we literally call the road map for ethereum the RO of centric road map and then the idea that like road maps are very like I'm like still like very bullish and optimistic about I think

that like high quality new rups are like like should be seen as a huge win and progress towards you know ethereum's vision and towards what ethereum is trying to achieve I think that like you know that kind of gets it like the the zero of some mindset like people you know I agree with hu completely like it's a complete failure mode for everyone to just focus on you know short-term profitability for ethereum um I think that like that would be Mega bearish relative to like the long-term opportunity of ethereum which is insanely huge and so much more expansive right becoming like this like settlement layer for the world is like so much cooler and more interesting and um kind of you know even like like the EIP 15559 debate is is such an interesting one um I kind of also agree it was like

potentially a mistake I again I don't think it was like such a bad mistake necessarily that it's like a big deal that we have it like I think it like works for in certain ways um but you know I think ethereum like when I if I were to like give advice to etherum road map it's like do things that make l2s better don't do things that make l2s worse and the more you put like constraints on like how transactions are constructed for example like I don't think we even realized like you know 1559 came out before we had all this intense architecture it also came out before we had like high quality like private rpcs and like a more sophisticated me market and so like you know a lot of like the ux issues around uh block pricing probably wouldn't exist today it would have been just solved

naturally in the same way that like we still had to people still had to build these like mware nodes that like and like we might have fixed that like that like temporary like you know like whatever like the main wallets weren't very good at pricing the gas and it was a little complicated because it was a little more volatile um but we could have probably fix that problem without introducing a constraint that you know the message do the transaction. origin of every transaction needs to have an eth balance which actually makes it way harder to onboard users to ethereum and like like a lot of issues that we deal with like at a UI level actually are like a direct result of that decision and we probably would have moved like away from like we would have probably moved and it also is like influenced

like the account abstraction conversation as well which is now like building around this constraint of you needing eth in your wallet yeah transaction you can no longer easily like sponsor any transactions for somebody else right it's yeah yeah and to do an approve you have to like to like to do like sponsor approve you have to transfer eth into their wallet but then you have to know that they're going to then do the approved because not just keep the eth it's like it's actually not again there's ways around it but we're like working around these new constraints so my hope for the ethereum road map is that it really focuses uh on its scaling road map the L2 Centric scaling road map and tries to like not introduce complexities that uh you know in in in an attempt to capture Mev and an attempt to capture you know an

attempt to like monetize eth in a more direct way and by the way side you know even if you pay a validator on the side to be a validator you still have to stake eth so like that value was already going to eth it's just like gone from implicit to explicit um like my hope is that we kind of really focus on what will make ethereum better for L2 and that's why like and and when it comes to the blob decision it's so freaking obvious you know cheaper blob data is better for L2 it's better for scaling ethereum it's not even close you know we don't want to artificially jack up the price of blobs we actually should be probably focusing on oh it's almost hitting the target we should how can we add even more blob data not it's almost hitting the target let's lower let's

let's artificially constrain it so that it's making us more money which just you guys are unpacking like my absolute favorite subject to talk about right at the end of an hour and a half long podcast so unfortunately we don't have time to unpack all of this maybe we can do a part two at some point but the fee discussion is actually I think one of the most interesting ones and the the other thing that trafi has up on crypto that we we just haven't had time to unpack some of the Nuance of these ecosystems and I think in a lot of ways l1's are pretty new things in general you can B there's nothing new Under the Sun and there's been a lot of first principles things that sort of looked like this in the past but it's okay to just update your mental model l1's are

not companies and it's okay to think about them in slightly different ways but you we just haven't had the time to unpack and have all these debates and discussion but all right guys unfortunately I think that's all the time that we have I want to be respectful we've already gone way over here so um look this was this was a great discussion from my perspective I had a whole list of questions I didn't get to with either of you but if folks want to find out more about boost or unit chain and get involved in some way shape or form maybe develop a hook or I don't know hu they're an L2 if you're an L2 Builder and you're like hey I want to W you know solve some MV stuff like what's the best way to follow both you guys or learn more get in touch um yeah

let's see uh to learn about uni chain you've got the new unich chain. org can always follow Unis swap on Twitter to see everything we're doing follow me on Twitter um I think that you know uh something that we were really Pioneer like all of these code bases are now open source and and or like you know or you saw before for example we we built a lot of in public and like people are already have been involved in the conversation for a very long time but definitely hook Builder should reach out to us or should reach out to the unisa foundation if they need you know funding and support um uh to get you know get involved in un chain like I think that like uni chain we're gonna be um you know this I think that like one cool thing that I'm very excited about with

uni chain is actually that one it's uh entirely MIT code based uh which is which is pretty sweet um and uh you know the the network is MIT the Builder MIT um we're really trying to like build like we're not trying to like keep this to ourselves we're trying to like build something we're trying to like prove out e a lot of ethereum's map we're trying to bring people along with us right if we make advancements we want other people to benefit from them um so I'd say that you know uh definitely uh we're hoping that people see what we're doing and build on top of it and improve it and contribute to it um so yeah and and that's one of the Reason by the way I'm excited to be collaborating with you know with flash Bots with optimism on this project uh you know a lot of people

it's sort of this like World of like like everyone is trying to build the same thing that's it's a thing that a lot of people say in crypto is every like everything is trying to every every company in the crypto is in some way trying to build the same things right because there's these sort of inevitable outcomes to to how some of this will be architected um and uh I'd say that like you know one reason I'm excited for it to be a collaboration is that like there's a lot of like I guess it's meant to just be an outro I now I'm now I'm monologuing but um it's just like like you know like having a bunch of high quality teams all contribute to the same thing is really useful for helping set the standards for those things and like you know we don't just need good

blockchain implementations we also need a convergence towards the standards for how they interact with each other and so part of part of like doing this not just like unisoft Labs by itself on a side mission but unisoft Labs with a bunch of other partners in the industry and bringing other people along and making it it it's all about like trying to bring the industry along with us and open up a conversation and and so yeah we want a lot of people involved in this in the long run aw yeah so I I I can second that I think uh like we've we are designing rollup boost to be this Innovation platform uh for rollups and um you know so we we want to develop a lot of different modules we want other people to develop a lot of different modules so we can supercharge Innovation

um on Layer Two uh and we have like so many cool things that we didn't even get to to touch on like Beyond this the stuff that's shipping with uni chain um we have uh you know te validity proof so uh you can get these uh multi-prover systems then you know upgrade roller Bridges to type two um uh you can have TECO processing Hayden you were touching on it with teleport and like I don't know if anybody was watching this you know T he bot that you know uh the other week um uh like a like a bot that was only its own Hardware in like living in a te and was tweeting all day was like some some

you're seeing some some very cool stuff I mean like a lot of this uh looks like a toy right now but uh you know a lot of great technology started out uh looking like a toy and I think every one of these demos is proving something very new and very fundamental and I think if you're if you're on te Twitter if you're kind of following these things then um it's a very very exciting time I think to be in crypto right now and I think um especially like uh crypto becoming more open to Hardware Innovation as well is like unlocking a lot of things that seemed very far away and like that seemed uh you know very much impossible I think for the longest time uh you know

like interoperability between like web 3 web 2 and web3 services for example it's it's like a major thing also that the tees can unlock and so I mean personally I'm I'm very very excited to be in crypto right now I think you know a lot of cool things will also happen around Defcon so if you're uh attending that or watching any of the streams like um look out for our events there and yeah if you're a rollup operator if you're like uh a developer if you're interested in contributing to this or running vola boost also um get in touch because yeah this is a public good that's uh you know built by and for the community excellent guys you've both been super generous with your time this was a ton of fun for me I'm sure L will find it useful as

well and yeah we'll uh we'll have to do it again sometime soon [Music]