BONUS EP: Stephane Gosselin on MEV, OFAs, and Frontier Research

Transcript

all right everyone welcome back to another episode of bell curve before we jump in quick disclaimer the views expressed by my co-hosts today are their personal views and they do not represent the views of any organization with which the co-hosts are associated with uh nothing in the episode is construed or relied upon as Financial technical tax legal or other advice you know the deal now let's jump into the episode [Music] hey everyone before we get into it today just want to give a quick shout out to this season's sponsor Rook close to a billion dollars worth of Mev has been taken out of users pockets and that's just on ethereum and that number is only getting larger unfortunately Rook thinks that it's time for a change and they've

built a solution which is going to automatically redirect that Mev back to where it belongs into your the user's pocket so you're going to be hearing all about them later in the show I'm a huge fan of this team and what they're building so stay tuned to find out all right everyone welcome back to another episode of bell curve this is a little bit of bonus content from the Mev season today I am joined by Stefan Gosselin today he is the founder of Frontier research which is putting out some of the best research in Mev in a in a past life he was a a Founder at flashbots and the Chief Architect over there you might know him as the ghost step on Twitter like that might be the best intro I've ever done so Stefan welcome welcome to the show man that was amazing wow can

you do my intro all the time yeah absolutely just uh just buzz me in I'm happy to uh don't save the silent clip and just like send it to people like that's a good idea um this will be a little side hustle for myself going forward but we've uh we've got a lot of ground to cover today I wanna I Wanna Give You a little shout out here you know Frontier puts out Far and Away the best uh the best content in terms of Mev whenever a new piece drops all of our analysts are always talking about it sets off some Lively debate for at least a week um and I kind of want to use this as an as a chance to kind of go through some of the topics that you've outlined so maybe we can kind of start at a high level and talk about

the transaction supply chain of Mev and just how it stands today how it's changed over the past uh you know years since you were at flashbots and you know wherever you want to take it all right yeah let's do it well first of all thank you so much for for having me on um I haven't done really media since I left flashbots and so I'm excited to sort of get back into the open and start talking about these ideas that uh that I enjoyed talking about a lot um so the the Meb Supply term the transaction supply chain I published initially this um this post it was in April of last year around the time of uh the dev connect conference and um in Amsterdam

we um we hosted I should say flashbots hosted a conference there called um math day and I sort of opened the conference with a presentation that was titled um you know Mev Utopia or dystopia what is the future of this industry um and in this presentation I outlined sort of these new sets of roles that I think are becoming very common in you know the vocabulary and the language that people use in blockchain protocols now and especially on ethereum by the time weren't you know very popular at all so um the movie supply chain I think a lot of people recognize now is there's users they have intents they communicate those

intents to wallets or applications those help them put them into transactions those transactions are put into some kind of transaction system whether it's an order flow auction a transaction pool Searchers see these uh these transactions and they augment them with their own transactions those make their way to uh to block Builders or sequencers those put them into into blocks and then propose them to the validators into the consensus of the of the system on which they a they operate so the last year since I sort of published this article I mean we've seen a lot of excitement and uh and Innovation happening at every layer of

this of the supply chain which was really really the goal of of framing um sort of blockchains in um in this way um I think for me it's also been sort of a personal Journey like going from trying to sort of Define um the way that these systems work in a fairly simple and and linear way um and one of the things that I had in there was the question of how will value flow is the value going to flow sort of towards the inside so towards the Searchers and the block Builders and they'll sort of aggregate a lot of of the value or is the value going to sort of flow towards the outsides towards users and uh and validators um I think for me what I've sort of come

to realize is that that was actually the wrong framing of the question like this question of of the Mev supply chain how is value going to be redistributed amongst the the supply chain is it going to tend towards Utopia or dystopia was really sort of the the wrong Framing and the correct one is actually like what can we create using sort of this model of the industry was the additional value that this allows us to uh to generate um and so it's sort of reframing it from a wealth redistribution which is sort of more zero-sum I would associate more with a lot of the questions that we have around Mev um to one of wealth creation uh which is

actually like how do we deliver you know systems that are more that are better for users that allow us to generate sort of you know genuine value for for the technology that is that is cryptocurrencies um and and I would sort of frame that one more as a as transaction Supply chains so how are we sort of um and enabling these systems to to settle the the kinds of of user activity that that is most beneficial um so that's sort of my introduction of like you know my journey and exploration of uh of the supply chain over over the last year um I would love to sort of dive into any of the components of it and and see sort of where has the energy been and and

where they're going yeah let's let's dive into that concept a little bit um you know when the first thought that I had when I was hearing you describe that is sort of this trend in Meb of returning Mev to users right from the intents which we'll get into that they sort of create and preferences that they want to express onto a chain right now the uh the vast majority of that value flows all the way down the the valley the transaction supply chain to the validators right now there's this big trend of returning that to users is that kind of what you're talking about or expand on that concept A little bit of moving away from sort of this allocation mechanism to a wealth creation mechanism yeah um exactly so the there's sort of some very simple

mechanisms that we can think about that like manipulate this this value distribution right like oh we're going to add in an order flow auction which allows us to sort of claw back a lot of the fees are being paid through transactions to validators and instead like send them back towards the user so you know there's this whole area of research and I published an article about it you know the design of order flow auctions how do you design them in a way that's incentive compatible and um and sort of helps helps return value to the users um I think the other place where a lot of this discussion is happening is um for like uniswap LPS right um or you know you could say Dex design

as a as a whole how do we design better decentralized exchanges that don't leak as much value either from users swapping and being targets of um of information leakage through uh sort of Sandwich attack or LPS being vulnerable to sort of slow price updates and and uninformed liquidity provision which leads to lvr and sort of large Arbitrage opportunities between uh between the centralized exchanges and centralized exchanges and a lot of the the conversation there again is how do we redistribute the value away from these Arbitrage players were bidding a lot on gas to be able to pay for the

block space and more towards these uh liquidity providers um maybe like a final one is the recent conversations around um Mev Byrne right how do we sort of redistribute value away from these like large spikes that are sort of random to to individual validators and more towards the network and and sort of spread out the the amount of of that value um I think a lot of these conversations completely miss the point that like we're trying to solve with all these these blockchains like we're trying to create systems not just that like move money around between different actors but rather are genuinely more useful and are going to be able to attract sort of

the next billion users um and developing mechanisms that allow for those better user experiences isn't just about redistribution it's much more about expressibility and like what can you actually do with this infrastructure um and and I really think that's where the the limitations of our spaces more so than any sort of you know protection or extraction or any other framing of it so can we I'd love to get your opinion you know on each of those things that you just mentioned sort of in reverse order I'd love to just get your thoughts on Meb burn and you know that sort of uh dichotomy between different stakeholders the way that I think of of burn and and sort of value that it brings well I

think it fits within like a fairly complicated question which is like what is the future of uh proposal Builder Separation on ethereum what is the future of roll up roll up adoption and sort of Mev activity on layer one versus Layer Two um Mev Byrne is kind of I think a desirable property when you have a system like we have today where all the activity is on layer one all the economic activity is on layer one there's a significant amount of Arbitrage um and um like sandwiching activity that is highly contested so you have high contention rate uh rates or you could call it

contention rates between um at every block for having sort of a specific uh State update um and what that causes is basically all of those opportunities are being paid out to the uh to the validator through um through block Builders and uh and math boost um that's somewhat destabilizing to the consensus of the chain and also centralizing to the sort of distribution of the nodes um the reasons are that you have a strong incentive to pull um to pull your uh your stake to be able to have the yield that comes from all this Mev activity that is highly variable right so my chance of getting

100 each block as a validator is very small because those don't come very often but I can have the same expected value if I join uh you know a validator pool that's able to spread it out exactly the same economics that we've seen you know on on proof of work ethereum and and uh and are sort of playing out here as well um burning in this case sort of creates less of an incentive to to join into into larger pools because um there isn't sort of this High variance uh uh value that you're missing out on as a as a solo sticker um it also means that you know now this Mev value is getting burned which is you

know great for uh for for e-stakers there's some or I should say East holders um there's some economics here that's sort of above my head and like understanding so I'm not gonna I'm not gonna LARP as like knowing what the economics of ethereum are um but anyways that's that's the the argument as um as I understand it there's a great if you if you want to really nerd out on this and get super In The Weeds on it Justin Drake gave a good talk on this on meconomics um which you can just look up on YouTube and sort of get his get his take on it um I also want to get your before we move sort of up the stack you know you mentioned uh sort of the future of Roll-Ups on ethereum and how mev's gonna look up there I want to get into that

with you but before we do I want to talk about because yeah you've written about sort of Dex microstructure you've talked about order flow options and that's just pretty cool interesting content so maybe can you talk a little bit about sort of your thoughts on the evolution of Dex architecture and maybe touch on you know take something like uni V3 or something like that what are the sort of leakage points of Mev and how how could we design dexes in a way that will be better for users so this is an active area of research and exploration for us um you know I should say I should give a shout out to my research collaborator here um Ankit who's been doing a lot of the research on on this uh this particular vertical as well he sort of

helped classify uh framing of Mev which is called EV ordering and an EV signal and it tries to sort of provide the basis for analyzing how a lot of these any of the opportunities are being targeted in various different mechanisms and so it's quite a it's quite a useful framework for um for doing this this kind of analysis now in the case of um of uniswap the way that we would think about this um and and the value leakage is let's say you have a user that sends a transaction through the transaction pool through the mempool and they have sort of a slippage tolerance that allows them

to get sandwiched well EV ordering says you know any operator any Searcher can basically take the transaction sandwich it and extract some value internal only to the system that is uniswap another example is of this would be sort of doing Arbitrage between uniswap and some other and some other exchange on chain um so that's a sort of one just certainly one piece of of the puzzle and where sort of the value comes from um there's a second piece of the puzzle which is um defined as EV signal um which is actually the value that comes from this activity happening um on within the environment of ethereum

but also taking into consideration the sort of external state to the ethereum and its mempool this looks at other transaction it looks at the states of other chains it looks at the state of centralized exchanges like like binance um and so here you have a much more Rich sort of field where you have Arbitrage between these chains you have Arbitrage between the centralized exchanges and and the decentralized exchange and you can also measure as part of this sort of Arbitrage between the LPS of of uniswap and um and uh and binance which is the second biggest source of leakage for uh for the protocol it's what's sort of described when

um when people talk about uh loss versus rebalancing um it's the same sort of value that's that's being explored here um so to your question about how do we do better than um than you know Swag V3 well to do better we need to solve those two things those two leakage the ordering that's based off the slippage and the um the leaked value through uh through signal um the design space for here is still being explored um I think one of the insights that's quite clear is that it's really difficult to um have price discovery on a system that is not very fast and has high fees um and so we're sort of in this

situation now where the price Discovery for a lot of these assets is on binance and this leaves all the LPS to be sort of targeted by by toxic flow basically our Arbitrage traders that are more informed and are able to get that um get those price updates uh reflected on chain um it's like you could think of it as like a very inefficient decentralized Oracle where you have a lot of LPS that just put money and like consistently constantly get bad bad execution and offer this service of having a price on chain uh to anyone else who wants to uh to trade against it um there's been sort of a lot of uh Innovation a layer up the stack on like

RFQ systems right so looking at okay well clearly the on-chain price isn't the best one right it's a delayed price that can be up to 12 seconds still depending on the situation um is there an opportunity to actually use this price as you know the best bid an offer have it as sort of a backup price but provide the opportunity for market makers to improve upon this price and that's essentially how a lot of these these aggregators work today right so they'll say okay well we see that the current price of this swap on uniswap is X let's put it out there and if anyone can provide a quote that sort of Beats this uh this execution then we will settle with them and provide some level

of of price improvements to um to the users and so moving up outside of sort of the chain execution into this environment has some benefits it does seem to provide better execution for um for uh for users but it also means that all this now it's sort of private market makers that have all of the liquidity and that's being kept off chain as opposed to being placed on chain I think there's still a lot of innovation happening the the biggest question is is it going to be sort of a a decentralized exchange on layer one that's going to be going to be the place where price Discovery happens or is there some designs for exchanges on

layer twos that um that will allow to actually have price Discovery happen within the decentralized context um and bring that liquidity back onto uh onto the chain and not necessarily have it out of the chain yeah I think it's uh all you know I think the the point that you make about basically LP's overpaying the arbitragers is going to be a pretty interesting one in the coming years in some sense that's sort of a political argument as well the other um you know the other thing our Point listeners to as well is sort of uh FBA sort of designs for exchanges like sort of the Cal swaps of the world um where very similar to kind of an RFQ based system there's sort of a a

baseline right of execution that Traders are willing to to pay and then I forget there's a I think solvers is the word for this this particular actor but they sort of go out into the world and figure out a way if they can execute at a cheaper price and that Surplus is actually measured for our Traders on the exchange um and that was we'll get into this later but I think that's kind of one of the first experiments in terms of intense um you know in exchange utilizing intents yeah I mean there's there's so many interesting components that go into this one of them is you don't necessarily need to only source liquidity from a single chain right once you have this the intent is especially liquidity from everywhere um I think one parallel that's not

explored enough is actually how frequent batch auctions add the transaction layer are fairly equivalent to payment forward or flow systems um they're doing very very similar activity in terms of auctioning off the sort of ability to solve it um or to you know to settle a set of transactions um but it's looking at it from a different time perspective um and there's some sort of auction theory around this that means that you might actually have sort of more collusion in a system that does uh frequent batch auctions Than A system that that does payment forward this is not by means by any means sort of a question that's that's fully settled

that's one of the questions that uh that we look at when we analyze these systems why is that Stefan why does that maybe later collusion so it all has to do with the type of uh of mechanism that's that's used so it this is the same case for um for uh for order flow auctions that sort of auction off individual opportunities um you have in the case of you know EV ordering or EV signal a fixed amount of opportunity but with some probabilistic payoff so you're not actually sure if you're going to get the execution that you want um and so when you're providing a bid you know that okay the max that you can make is is X but you have some risk factor on your execution and because

it's a single winner auction there's only the person that bids the most that actually gets settled and they always end up like winning because they overpay for the the opportunity that that they have um the end results of this is that you have sort of an incentive for all the participants in this system to collude and under bid and say like we will all bid with like 20 discount down to what like our real preferences as a way to make sure that we don't you know lose money um I you know one can argue it's not like a stable equilibrium and you'll have like bidders that um that over bid but the thing is that those bidders will eventually run out of money um and so if you have infrastructure

that you can run and that your um your cost of capital is sort of within the range of your your opportunity um the amount of opportunity that you capture you can you can run sort of a discounted uh bidding um profitably for for a very long time well have to it will be interesting to explore honestly very interested in the FBA sort of design space and and Cal swap you know which by the way stands for Coincidence of wants which I didn't understand didn't know until recently it's a funny little thing um we also talked to uh Henry David Lentz this season and he you know is building penumbra on Cosmos similar FBA sort of design with privacy in mind first which is super cool I want to uh transition here to your thoughts on

order flow auctions because you put together a great piece sort of exploring the design space there and you you know you just started to say one of the highlighted two of the differences between payment Florida flow and Order flow auctions which is sort of the time the time component and then basically the individual sort of transactions that that's possible to do in an order flow auction but maybe I'll just turn it over to cart launcher and can you sort of explain the what the design space looks like if you had to categorize the different types of water flow options and maybe highlight how it differs from payment for order flow sure um I think you know the the Top Line takeaway for that that piece is um a lot of these systems try to solve two

things at once they try to solve for Price Discovery and for settlement uh sort of together um and it sort of confuses one thing for the other um that ends up being expressed in like a lot of challenges and how these auctions are constructed around trading off value like basically best price you know return for the user best price best execution for user and reliable inclusion um and um and so based off of this this Insight I sort of proposed that there's actually four different types of um of OFA designs um and they can be sort of put on a four

by four grid between if their application specific so only for sort of a swap use case or generic any for for any transaction um and if they settle based off of uh individual transactions or based off of a group of transactions um and um and we can put basically all the different designs of of order flow auctions whether it's an RFQ sort of a cow swap a pfof model um a uh sort of a Mev blocker uh style uh style system or more of a MAV share type system inside of this uh this you know two by two grid each of them have their their trade-offs each of them

optimized for something slightly different um and I think crucially I don't necessarily think that any of them solve the problem that like users actually want to have solved which is like um you know being able to trade and event you that that is they're gonna have like high confidence is going to give them a good price and timely execution it's definitely an interesting debate I I would actually be that that two by two Matrix is I think a really good way of thinking about it I'd be curious just because um if you have thoughts between sort of generalized order flow auctions versus maybe more specific like maybe in the future like dap run order flow options if you imagine something like a uniswap trying to run their own order flow

auction uh you can certainly see the incentives to for them to do something like that and I would just be sort of curious how you would assess sort of the pros and cons of those two different approaches um I I am a strong believer in um application specific order flow auctions I think that's likely the way that you optimize for user experience um I think there's like some arguments you can make that you can have like a specialized order flow auction built on top of a generalized one but I don't see a way to really do this that doesn't introduce a lot of trade-offs um and so my um my intuition says that uh certainly the ultimate best user experience for

users who want to do the use case of swapping or providing liquidity trading in some kind of way which is you know one of the few proven use cases I guess of of uh of ethereum today um those will end up in a in a specialized order flow auction uh model um I guess a separate question is like is there still demand and and um you know value-added to um to long tail order flow auctions or generalized order flow auctions um maybe but I I see them more as um as block space aggregation platforms and sort of discovery of of contention on uh on state access then I see them

about uh returning value to the users or providing better price um so I wouldn't see uh in those system uh it to be very successful to build an order flow auction that Returns value to users I do think that it's likely that those systems are more used to like minimize the like block space fees paid by um by uh individuals trying to interact with the change so speaking of fees I would love to kind of get your thoughts on sort of moving up the stack um you know to the roll-up layer and how Mev works there today you know I'd be I'd be here right now obviously not a secret right that the all the Roll-Ups use one Central live sequencer which has

big implications for Mev and a lot of these Roll-Ups today sort of view Mev as a as a source of Revenue right for their for their roll up obviously this isn't sustainable um you know hasu's argued for something like PBS that exists on uh ethereum main chain today and sort of the form of MV boost eventually it's going to be enshrined but something like that needs to exist on on sort of Layer Two I'd be curious just to get your sense of how you're thinking about Roll-Ups these days and I would love to sort of get your kind of high level thoughts on how Mev is going to work up there um you said every roll up except for one uses centralized sequencing uh no no all of them I think all of them use multiple color sequencers yeah yeah um I'm still not convinced that that's a

bad thing I still like I don't think it's a a done deal to say that actually first in first out sequencers on a layer two are a bad thing the argument that we often hear when when this is mentioned is well then it creates a latency game and causes centralization around like a specific region and therefore the roll-up is susceptible to the censorship rules and Regulatory rules of that of the region in which the sequencer is deployed um and I think that's true but still the question is like is that actually bad so let's explore this together because I think it's like a bit of a controversial

or like uh it's a spicy take my friend I'm very interested to see where you go with this um this is something I've been trying to disprove so so you know if people want to add me on Twitter and argue with it I would love to um the the statement goes like this so if you have like a maximally decentralized layer one that probably means like a layer one on which there is very little economic activity like it only settles uh you know data and there isn't sort of contention rates right there isn't this sort of demand for settling um with a you know a on a specific position um you only have congestion rates that's sort of how we we

um we ethereum has been designed um when you start to think about economic activity then you have uh all the issues that comes with Mev and the latency games that are sort of ensued um if you have an architecture where the layer one is only settling um uh sort of data blobs and there isn't contention and you have all the activity inside of layer twos it reduces the um the centralization pressure on the layer 1 significantly um no okay how is it any better for the layer twos to sort of become centralized well if you have some way to have cross-chain messaging between these layer twos that is uh that is quote

unquote uh censorship resistant right you have some way to like push messages out from the layer two back into the layer one or maybe for it to be interpreted by some uh by some other um spin up of that layer 2 somewhere else you might be able to by when I say like interpreted by another layer Layer Two somewhere else I mean like a fork of that layer too um then um you might be able to you know be completely okay with having a centralized roll up that's targeted by regulations um because it remains um it remains censorship resistant it remains um you could say non-custodial um there's still a way for individual participants in that roll up to exit

their state and Bridge it over to some other roll-up in some other uh jurisdiction if you know there's enough demand to sort of boost up this other robot it's an interesting premise let me see if I can poke at a little bit and and see what you think there so I I hear what you're saying in that if there was maybe some sort of uh messaging system be that like an IBC or a shared sequencer or whatever and there are sort of guarantees right that the state or the TBL or whatever that you end up blocking on this roll up won't be necessarily Disturbed that said you know if the some sort of State Intervention basically shut down that that centralized sequencer it would certainly be a market disruption right and there would be like a period of time where

we'd have to migrate the assets and everything down into main chain and like back up onto another roll up I don't know how seamless you see that process being not tactical enough to have an opinion about it but I would imagine it's relatively disruptive I feel like if there's the threat of that hanging over these Roll-Ups then it will prevent you know tvl and activity from migrating up to the Roll-Ups as much as they otherwise would if there's always kind of that threat of reorganization hanging over them how would you respond to that yeah I think that's a valid um that's a valid argument I think the other argument is like well if you have some weight for like state to be able to exit back to the layer one if that does happen then you have a lot of contention on the layer one and so you have all the

same centralization pressure on on the layer one and and so I don't think that by any means it's it's perfectly it's perfectly solved um I do think like this does generate an argument for more fragmentation between layer twos which I would like consider to be a good thing I think like if there's a layer two that has all of the state on it and there's like a single winner in like the layer 2 world that seems really bad for Innovation it seems really bad for uh for like robustness of this system whereas if you have um multiple different layer twos with like liquidity fragmented across all of them um yes you know it does create some challenges on like cross-chain messaging

and like um uh like creates you know maybe perhaps worst worst price execution and and some framing of it through like there being more Arbitrage opportunities um but I would treat that as a as a separate concern than one of okay how do we have like a robust uh and scalable uh ecosystem around um around selling transactions okay uh it's actually a very interesting point it makes me think of you know we're recording this on uh May 26th so vitalik just came out with his piece don't overload ethereum's consensus and he actually listed you know one of the explicit risks of relying on ethereum social consensus on let's say uh one of the layer twos becomes really large

don't count on the fact that if there's not some sort of catastrophic hack that we're going to roll back the chain because then you get the sort of favoritism that you see in Trad five right between say like a large you know bulge bracket Bank versus a small Regional Bank and that doesn't seem necessarily fair so and actually I don't know if I have a good response to you but I have to think about the fragmentation of the L2 ecosystem because you know if I I had been kind of thinking about it within the context of the the optimism super chain right and there's you know we'll figure communication out if that's IBC or shared sequencers but I hadn't really thought about fragmentation actually being good for the robustness of the ecosystem that's an interesting take I'll uh you know I'll copy all this by

saying I'm by no means a layered to experts um I think I've known one thing or two about how ethereum works by now but Layer Two I haven't built one I haven't like worked with closely with the team story building one so I'm sure that there's a bunch of content there exposed to um but it seems to me like and this goes maybe into a bigger perspective on decentralization that I've I've sort of grown over time which is um you want the decentralized system to be as simple and and and dumb and uncontroversial as possible um and for those rules um to be sort of um not changing very much um if you want the system to to remain um to remain centralized uh uh

decentralized I mean um the the concern is any system that is complex and requires sort of global consensus is likely always the lead um to centralization um and capture and this is not even just like an economic or like Mev argument it's actually an argument on how humans coordinate and like politics where um anything that requires vast groups of people to agree to the same thing will devolve into uh into Political communication games and that will inherently cause the like system to um to concentrate and centralize whereas if you have a system that doesn't require

people to all agree on the same thing all the time um then you have a lot more fragmentation and diversity and and by by proxy decentralization yeah last last question that I have for you on the on the centralized Secret Service so what would you say I mean the if you look at how it's sort of played out with arbitrim so far which has taken a very different approach from optimism in terms of their Meb strategy although optimism has changed uh as well but if you look at arbitrum which was a sort of fair they were the fair ordering um sort of group they've already changed the way they're doing it a little bit because they have the same problem that Solana has with just Spam so I'd be curious to you know get your takes on

what you think about in when you introduce something like fifo or Fair ordering or whatever it is you end up getting a whole bunch of spam be curious what you think about that um so so the spam is a proxy for latency um so the um the question is not you know fifo is the wrong approach the question is more like how do you handle the fact that there's latency advantages like the dreadfy solution to this is like you sell co-location spots around you know your your matching engine um and you have you know you know you can boot people off of their co-location spot and so therefore you don't have the same uh sort of incentive to spam or for them to like try to DDOS you in some in some way or another

um so I think that there's there's other approaches if you're if you're sort of committed to this idea that there's going to be a single um a single roll up in a single sequencer and that's going to be first and first out um certainly there's a lot of uh established ways that you can make sure that this you know this server stays online um and you can deal with all of the sort of economic side of it in a much more centralized way um so it's you just have to sort of commit to go down the centralization path um if if that's the if that's the design that is being the alternative is to try to go down this sort of encryption path right okay

I will continue to use a first in first out system but it's going to be encrypted and distributed across a bunch of different sequencers are sort of all around the world um you still have a latency game they just become sort of at the edges of the encryption right so before the packet gets encrypted who has access to that information how is this sort of generating value and who can sort of capture that value and then afterwards where does the information becomes available first and that's where you have again sort of the latency game yeah interesting um before we move move away from layer two so I would love to just sort of get your thoughts on you know a big topic du jour sort of this idea of shared sequencers you know for a whole bunch of

different reasons part of it is the communication uh issue that you were alluding to before but the other sort of Honeypot that this opens up is cross domain Mev which has been a very sexy topic for a long time although it's sort of yet to be realized aside from statarp so be curious sort of what your thoughts are on is eventually do we end up getting you know this very rich ecosystem of cross-domain Mev on the roll up space or is that a little bit of a pipe dream for now so I still don't understand what is shared sequencing do you have like a a quick explanation for me sure I can explain it to you like your five because I'm also secretly five but yeah basically the idea is like I think about like three different models right for for how Roll-Ups can sequence

their transactions there's one centralized sequencer which is super easy to understand then there would be a decentralized sequencer set so multiple different sets of sequencers that only settle transactions for or order transactions for one roll up and then you can imagine a world where there's a set of sequencers that sequence transactions for multiple different Roll-Ups and in theory there you could guarantee something like Atomic inclusion okay I think I see this very similarly maybe maybe this is like the other the other take that I'm gonna have on this which is I think everyone is building the same thing um and I don't know if like that's what it is building the same thing um I agree with that

so so whether we call it shared sequencing decentralized block building oracles um all of these to me all have sort of the same flavor they're all sort of trying to solve this problem that you have a lot of information you have a lot of messages that are like floating around into the world um and then you have some consensus system that asks you hey can you find some like way to put all these messages together into like my next update um and everyone is kind of trying to figure out okay how do we represent all this information that's out into the world into some reductionist view of what the state of this roll up this chain uh is and we're creating all kinds of games around this right

um I I sort of put this cryptic tweet out the other day that said it's all messages and solutions bro um just saying okay all of these things just end up being exactly the the same thing um so sharing sequencing yes um I think that there's like value to be able to do things on multiple chains at once there's um there's sort of this this um concept that I've been trying to push since uh since 2020 of chain abstraction like if we want um the crypto ecosystem to become useful if we want it to be used like uh you know AWS is used to be able to like deploy application

um not only users should not care about which chain that they're interacting with but application developers should not even care about which chain that they're interacting with this is just what's required if we want to be able to enable the types of applications that are like actually useful um we need to create an ecosystem where it really doesn't matter um and at the end of the day like yes you're gonna have trade-offs in these different execution environments but it shouldn't be up to the the developer to sort of know exactly how does an ethereum transaction format work does like the layer two have shared sequencer or not shared sequencer they just want to be presented with uh sort of a set of features or an interface that they're able to connect to and automatically deploy um to these systems

um I do think that these shared sequencers or you know decentralized block Builders cross-chain Bridges um are all in the same game of trying to build and provide those uh those Services they're trying to say okay we have a good view of what each of these different execution environments are good for you know storing data maybe it's more decentralized maybe it's good for storing value maybe it's like lower latency um and they're trying to provide these services to applications that are trying to build on top of these different systems and expose them in a way that's the most intuitive um and you know provides the best guarantees um so I think whoever is going to be able to provide these best sort of user guarantees is really you know what the

shared sequencing should be all about um and the the mechanisms behind it for actually settling it um I mean there's there's so many different ways to build these things and that's that's not clear to me where it's going to go I completely agree with you on the point about abstracting away a lot of the complexity for developers and maybe that's a decent segue into you know something that is a a very sexy topic within Mev which is account abstraction and how 4337 is going to impact Mev in general in sort of the creation of this new interop and pool and it's um in that kind of you know that intersects with intents as well there's a lot of questions about what is an intent what is not an intent so can

you kind of just delve uh head first here into this whole thorny topic and sort of explain what 4337 is who are the new actors that it sort of introduces and how is this going to impact Meb the intense um language I think is a really interesting one um I think it goes to the core of this this question which is like we're not Building Systems that allow people to like update the state on a global computer like we're not trying to provide these Services we're trying to fulfill user needs and that's really how I started to use the the language of intent in the concept of like trying to provide better services to to users um I think it's being now used as a way

to sort of escape from the limitations of the ethereum transaction format um saying okay like you're it's very rigid to be able to say that this intent has to be programmed as I'm interfacing with this uh function on this contract and it's like yielding X result when you have sort of a design space of interacting with multiple different uh different chains um unfortunately uh I don't think that account abstraction today is really nudging us that much in that direction of uh of being able to achieve intents um and so I sort of have this this quote which is 4337 is is the future

um but not really um so 4337 is trying to develop this new mechanism that allows for um for account abstraction that allows for um signing uh basically call data and have that call data be routed through some distributed peer-to-peer mempool and be eventually included on chain um it introduces um the concept of a user op which is sort of a new message format again it's all messages bro and uh it introduces a new uh entity um which is a bundler um and again it's all solutions bro

um and so it tries to say okay with this new message format and this new solver format we're going to be able to enable sort of uh um you know account abstraction and unlock uh some some latent value and like build new kinds of applications I think it's true it's correct in the way that um escaping from the current format of how ethereum fees work and sort of the the requirement to have eth in your account when you're submitting a transaction to the chain is going to unlock a lot of um a lot of value I'm not convinced though that the um the standard itself the 4337 standard is going to be robust enough to be able to express um all the intents have to offer and

I'll give a very simple example of this so one of the one of the limitations of 4337 is that you have to be constrained in the state that you're simulating against um so there's sort of this attempt in this assumption that you want to create these messages in a way that is Dos resistant and the only way to do this is to fix the state that you're going to be simulating the transactions against so you're saying here's a subset of the state on chain this transaction is only valid if it only interfaces with this subset of the state and that's incredibly limiting from a perspective of expressing intents because my intent maybe isn't just about this limited

State maybe it's about some other contract that's like outside of the state that might change or maybe it's based off of the state of some Layer Two or some you know the state of uh of uh binance as a chain and none of those preferences can be uh expressed within a sort of 4337 um user operation um so I think it's a step in the right direction I think it's sort of a it's going to be adopted and I do expect uh sort of it to be the new way that um that users interface with uh with ethereum but it's not going to be uh the future um and it's going to be superseded by some some Superior transaction formats

the way the simplest way that I've heard how to describe intense is your hard coding the output instead of hard coding the input so you know what you just mentioned is like we're not 4337 doesn't perfectly allow that right but it definitely does solve some some big problems I think that have been uh holding adoption back from ethereum so one there is another entity that I think 4337 introduces which is a paymaster right and the one of the results of this this new format of each wallet being a smart contract uh well instead of an unknown account is that you can actually have another entity pay on your behalf right so you can imagine a paymaster that adapt acting as a paymaster to

subsidize gas fees for users right so you don't have that weird problem when users first open their their metamask right that they can't do anything because they don't have any Ethan there which is a very difficult onboarding problem for for new users um the other the other uh you know question that I that I have for you on that is what are the implications of for Mev when you think about this so intense it does something there but not quite everything that we would want from intense but it does create kind of this what's being dubbed is an alternative mempool and I would love to get your thoughts on who the bundlers of these user op transactions are ultimately going to be um does this sort of I guess is that the

IB Builders um or Builder Searchers and does this sort of raise the you know now you've got the ethereum public mempool you've got the user op mempool you maybe in the future have application specific basically application specific mempools if they Implement some sort of RFQ type system just seems like now to be a successful Builder Searcher you have to draw on all of these different sources of order flow right to build the most efficient blocks so I guess my question to you here is like what are the implications for Mev and does this sort of raise the barrier to entry for Builder Searchers does it raise the barriers to entry I think like one of the places where people go to when they hear this is

exclusive order flow is sort of a danger um and I used to believe so and then sort of stopped believing that that's the case um and the reason is that I don't genuinely think that like block building is a profitable activity um and the the argument is this way so to be a block Builder you have to um have some you have you have to have two things like one block space access and then two enough trust to have order flow being be sent to you um and the only way that you can sort of bootstrap this um is if you

make no profit and spend all of your profit subsidizing either the block space or the order flow that that you receive and so block building I actually see as sort of a race to the bottom um to maintain a certain position on the leaderboard to justify hey you can send your uh your transaction flow to me and I'm like providing all these different features I have to um uh sort of be willing to pay for that uh that transaction flow that's uh that's being sent um and so as soon as there is sort of a mechanism for Block Builders to be able to pay for uh various different types of transaction flows um they are going to be in a race to

zero to be able to sort of outbid other block builders for those those um and so under this framing sort of block building actually becomes much more of a sort of like a public good you know not not this similarly to how it relays on like the MAV boost context uh are being operated today um there just isn't like a natural way for uh for these systems I think to uh to capture value and they will be sort of competing in this uh in this closed game uh for um for for uh for being able to access um barriers to entry I'm not sure um I think it may be that it creates uh some barriers to entry there's like some

integration costs to be able to support all the different transaction types um this is like maybe the barriers to entry here is actually that um like there isn't like a good way to make a case that you can bootstrap like a new one of these things because um it requires you know subsidizing uh and not everyone has infinite access to Capital to be able to do this um so I'm not sure where where that piece of it is going to go yeah you might be absolutely right I guess you sort of answered my next question but maybe just uh just to poke at the Builder question you know one more time is there definitely have been you know builders that have made a lot of money right uh without naming any names but sort of these uh you know shadowy uh you

know not super public uh Builder Searcher firms that have done quite well in recent years so is your contention there that basically those profits are going to get armed away as more Builders come into the space and yeah it's sort of a race to zero in the same way that transaction fees are raised to zero and The Brokerage world is that kind of the idea um I don't think that trading activity is going to go away and like Arbitrage opportunity I think I think that's going to remain I think the Paradigm of like you have to be a builder to be able to like land your Arbitrage transactions isn't a correct one um I think that uh actually neutral block Builders are going to out compete um Searcher Builders

um because it's um it's a position that actually requires more neutrality um and I think that uh the neutral block Builders are going to compete basically on merging systems and who can subsidize the most the order flow that they receive from Searchers um yeah and and Searchers will continue to sort of send to every block Builder that they trust so long as they can sort of believe that the block Builder isn't also sort of using their flow and trading trading super interesting that just kind of yeah I do think that neutrality that just that point just clicked um that's super interesting I would love to maybe we could um before I I wanna you know give you the chance to talk to

the audience about you know sort of leaving uh flash Bots and you know what you're doing at Frontier research but before we go maybe we could close on you know this idea of latency as you know just really getting into the weeds of Meb for these past couple months it seems like that's sort of the topic that folks are super interested in and we kind of talk I kind of think about latency now in terms of two buckets there's sort of latency as a as a geographical centralization kind of risk for the health of the network and we talked to that a little bit about a bit about that on Roll-Ups um and you kind of have that contrasting against like a strong user preference for fast confirmations so that is kind of that bucket when you talk about latency and

then there's this idea of latency you know expanding time around uh you know the around the block so you have more time as a right now I guess as a builder researcher to get more transaction order flow and therefore you can build more profitable blocks and shout out to your team you guys wrote a great post on optimistic relays which I think uh you know you really got deep into the weeds of how this works today so I would just love to get your thoughts on latency you know whether or not you view it as this sort of inevitable Force that's going to shape the network in good or bad ways if you want to get into the technicalities of how optimistic relays work um and unconditional payments from

Builders I think the that's a good nerd snipe I think we could the people would like to hear your thoughts on that however you want to take it maybe we can just close on the subject of latency I started to mention like um my my thinking around the Mev supply chain and like recalling it to transaction supply chain evolved I am I thinking about latency also evolved um and uh the benefits and innovate inevitability well that's a hard word to say of uh of latency in the systems that we design is is sort of something that blockchain systems are going to to struggle um it's

I can make a I can you know take this from so many different directions we can talk about sort of Mev boost and how it came to be um and how optimistic relays are sort of inevitable as the next steps in the progression towards um enshrined um PBS we can also talk about how sort of latency means that for as long as there's economic activity on the layer one even in an enshrined PBS context there's going to be centralization pressures due to uh due to the the latency um and we can also talk about how like in distributed systems that have Global consensus because of general relativity it's inevitable that we have these

latency games that sort of emerge when you have information propagation across the network all right let's start with uh optimistic relays let's go because you wrote a great piece on this so let's get sort of your your thoughts on there and then maybe if we could segue into med boost why that was originally created and then talk about the roadmap for enshrined PBS your thoughts there optimistic relays um is sort of this this new relay type that um that allows for the faster propagation of blocks from block Builders to um to validators in um in the meth Booth boost context um so the state of of the art right now and and when math boost launched was a block Builder sends a block to a relay

the relay simulates it validates that it's a correct blocked and then forwards it over to um to the validators for um for inclusion in uh in the chain for the proposal and inclusion um optimistic relays basically just say well instead of simulating the block we can have a trusted relationship with the block Builder where they're bonded for some amount of capital and we can optimistically just pass the block over to the um to the validators for uh for execution um and we save you know call it 100 200 milliseconds on the simulation of the block inside of the relay this means that you can incorporate 100 milliseconds to

200 milliseconds of additional value into the block and by value here I sort of mean it equivalent to information right because time is money as we as we we are all taught um and so in this additional 100 to 200 milliseconds the block Builder is able to sort of incorporate a lot of new information that's being propagated in the world about maybe the price of of different assets on different chains um and update the the Arbitrage trades that they're sort of including in their um in their block which means that the blocks that come from these optimistic relays land on chain um and win sort of the block auction at uh at the validator level

um so this is like a very clear example of okay well every millisecond you can incorporate additional value into sort of the The Proposal that is being made to the validators and so there's going to be sort of a race for uh for more efficient um uh relay infrastructure to be able to to carry these blocks over interesting so let's let's segue um the the discussion of optimistic realize because relays are you know not a permanent State of Affairs right there's eventually there's sort of a a stop on the road into enshrined PBS and I would love to I feel like that's a pretty good place to lows this week Mike neuter actually came out with sort of his updated thoughts on what the road

map was going to look like in some and and how we should all be thinking about that I would love to get your thoughts on in Shrine PBS um you know what you think about the decision your ethereum is very selective about what it actually decides to take into the protocol uh PBS was kind of one of those uh very select decisions so you know maybe if you could just weigh in on that and then how you see ethereum consensus sort of changing um you know with intron PBS okay this is exciting so um I started looking at PBS in the summer of 2021 so I'm like two years into into PBS research I guess you could say I was just looking at it and I was like yeah

it's 2023 that could be five years ago I have no idea um and it was it was all in the context of okay the merge is coming right now flashbots is sort of working on proof of work ethereum um what is the solution that we're going to bring to uh proof of stake ethereum or on Mev is it going to be the same thing or is it going to be different um had some discussions with the EF team with vitalik around like what does you know a better system look like um and that's where sort of initial designs and ideas around PBS came into play separating out the the Builder role from the uh from The Proposal role

um and the question at the time was okay do we wait until PBS and Shrine PBS is ready before doing the merge or do we proceed with some other solution um at the merge that uh that doesn't you know require consensus changes and therefore isn't going to delay the timeline of when the merge happens um and the decision at the end of the day was okay well we don't think that we can convince um protocol developers and development teams to delay the merge and to like do the research that's required for for PBS to be a thing we think that there's just too much research there um and so we need to find a solution for having out of protocol PBS and that's

where I started designing Mev boost and over the course of sort of the following year sort of uh brought it to uh to the market um now there are some like very conscious decision in the in the design of of Med boost one of the main goals with it from my perspective was to be able to put a clear price on the value of a block um which was not a thing that existed in the past this like concept of there being a public value of a block and say this block is worth X did not exist because all you had before was a bunch of transactions being routed to a minor a minor constructing that block and it was sort of secure how much value they were they were capturing out of it and you had this issue where the mining

pools would be able to capture some of the value and not distribute it back to their uh to their miners so they would capture the Mev value and not distribute the rest of the value back to their to their miners um I'm maybe boost sort of solved this by saying okay we have this block auction the value of the block is public and this allows us to have sort of transparent price discovery on on block space value um it also sets us up nicely if sort of this continues and gets put into the protocol for being able to burn this uh this value so if you can have some consensus on what is the like highest value of a block um because you have this open price Discovery then you can like sort of do

whatever you want with with the value of of that block um that's sort of where the Inception of you know why why MVD boost is good why it's sort of a good precursor for and trying PBS and for eventually uh math burn uh sort of came from um I think there's still a few upgrades to sort of MAV boost that are possible before we uh we get and try and PBS one of them you mentioned a few times which is unconditional bids um so this just means that now block Builders have to sort of commit to a payment regardless of the validity of um of the block um there's some technical difficulty to implement this outside of a protocol you need basically some Oracle that's able

to say after the fact if a block was was valid or not um but um certainly when it comes time to and trying PBS that's sort of one of the main things you don't longer need to have an oracle or some external source of trust trust to validate a block you just have sort of within the protocol the attestation committees be able to determine if a block was valid or not kind of land the plane here around like when we eventually like so move past all of this and this is all put in protocol how is it going to look how is it going to look different from today obviously we're not going to have relays and then walk me through like what are the big implications of this is this going to be like a huge Boon for the network and you know really lead to a more decentralized

sort of validator set or is there kind of another side or take to it one of the things I found most beautiful about the merge is that you completely changed everything from like the RPC down um and the application developers and to users they like didn't notice a thing right they just went from one transaction transacting with like Miners and like mining pools in China to like the next day transacting with like stakers and staking pools are mostly concentrated in the US and like to their to their perspective they didn't you know have any ux uh ux difference and that was that was just like crazy to even think that you could do that I think in trying PBS is going to be

pretty similar right so it maintains the role of block Builders and all the features that sort of block Builders can can develop it changes um it changes the role of the relays and sort of gets rid of the relays and allows block Builders to to interface directly with uh with proposers um the question is then okay well what's the next step after that is math burden going to come around I think that might have a bit more impact so if Mev burn does get shipped then there should be less incentive for concentration of stake inside of um inside of staking pools and so maybe you'll have sort of more of a noticeable

difference there but in trying PBS on its own I don't think is going to have much noticeable difference to uh to the way that the system works all right Stefan we could I'm sure go on for for hours more here but I want to give you a chance to talk a little bit about Frontier research and you know what your plans are and what you're you're kind of doing over there yeah so I I sort of opened up the the conversation here by talking about um you know going from thinking about Mev supply chain to transaction supply chain um and Frontier research is really sort of based off of this idea an idea that thinking of sort of Mev as a you know zero-sum wealth redistribution problem

is not the right framing um it causes um causes thieves to think a lot much more in a closed mindset like not wanting to share information um and uh a lot less experimentation um and and Innovation across sort of across the board um and instead reframing things as uh sort of a pauses some wealth uh wealth creation uh game and allows us to think more about okay how can this infrastructure how can this supply chain be used to generate um uh real value for for the users and uh and and really prove that crypto and uh and these systems that a lot of us have

spent so much time building on um have some some True Value to to create for the world um and so it was it was based around this this idea and wanted to explore these these Concepts uh a lot more um and so we have a few different initiatives um we have a publication that uh that we we you kindly mentioned and said nice things about it um and this publication is not just the content that um that we at Frontier research create but we invite any actor that's building on the transaction supply chain to submit articles for uh for review um we basically uh provide editorial Services we review it and then we curate you know the content uh to make sure

that the publication remains sort of a place with a super useful mental models and and high quality signals um and so yeah anyone who's who's interested in in submitting article for for publication there we'd love to hear from you we're also offering sort of uh advisory services so part of the vision is for this transaction supply chain to um to deliver value to the world we need a lot of different teams building different things and like really going out there and trying new things that you wouldn't have if you just have sort of a single approach or a single way to view this uh the supply chain um and so we we work with with parties sort of across the entire the entire

industry to help them bring bring some of their their products and ideas ideas to Market um and then the final initiative that we're sort of getting kick-started is uh is what we call Frontier Adventures um and uh maybe you could call it a bit of a play on Frontier Ventures but uh but it's Adventures because um what we're we're looking to do here is really sort of Target some of the the components that are very core to to this transaction supply chain um and and incubate and and sort of build uh build some experiments um in-house um as well um and so there's uh We've sort of started collaborating on on a builder on a block Builder there

um that's called the the faith Builder um and uh and we're helping the team there sort of bring this uh bring this to Market and provide sort of some of the more interesting uh transaction types that I think are going to become um really useful for uh for providing these better user experiences um and then we're also looking at okay what does it look like um to um to provide services at the um at the messages layer um at the at the networking layer we we are considering sort of launching some additional uh efforts uh on on that side super interesting man um well honestly Stefan this has been so great I'm so glad we were able to do

this this was like uh just honestly just you know filled with gems and we on we got to basically everything that that I wanted to ask you about so I don't know um I guys listeners to the podcast I mean you just heard from Stefan uh for the last you know hour and change but I highly recommend that you go check out the site um I can link it in the show notes but yeah it's Frontier research and I think the site is frontier.tech so definitely go check that out and uh yeah man I'll be super excited to see particularly on the adventures the adventures are adventurous fun I'm excited to see what you got cooking over there so maybe we can do this again in uh you know six months or something like that and uh just see how things have progressed amazing um thank you so much Mike it was a

pleasure to speak with you um and uh yeah I hope it was some interesting content for everyone absolutely all right be well thank you [Music]