Season 4 Finale: An MEV Masterclass

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 alright everyone welcome back to another episode of bell curve today I'm joined by haasu and the final the finale of season four on Mev it's gonna be a fun one but a little Bittersweet this has been a really fun season for me yeah the grand finale I mean it's so much fun thank you for doing this with me thank you for pulling me out of my podcast retirement um yeah it's been a lot of fun to just talk about Mev every week uh yeah I agree um it's been a blast and I think this

episode will serve kind of as a nice book end and we can revisit some of the topics that we addressed in earlier episodes and maybe the the most fitting one to start with is just latency I would love to get your sort of closing thoughts on this we talked about this within a couple of different contexts both in terms of sort of geographic like a sort of a risk Vector for uh platforms like ethereum we also talked about it within the context of Roll-Ups and Solana so wherever you want to start first just on the general concept of latency yeah I went into the season um with latency as a big topic on my mind and I kind of it was on my agenda to bring it up to a bunch of different

people and just hear their opinions right how they think about the topic and how they are dealing with it and um so just to summarize I guess for our for our listeners so um systems that have very low latency um in in in in like in some cases they can provide um you know a better experience for users for example users get faster confirmations but also market makers can update their bits faster so that they can kind of reflect the real cost of liquidity and the outside world but latency low latency has this like really particulars like downside which is that it um it basically leads to a lot of geographical centralization

because in the latency system you're always incentivize to co-locate um or up to the point of like direct vertical integration with whoever has the ordering Monopoly on that chain and so it leads to kind of different validators co-locating with each other but also search as some big block Builders co-locating with validators and in a system that is supposed to be decentralized and kind of robust to any particular regime of Regulation um that's that sort of uh almost the death blow I would say so I I I'm we talked to with this uh about this with a

few different folks throughout the episode but I think my uh my View kind of uh has not changed uh from from kind of this new information that I think you cannot build the system that is both kind of globally decentralized and a very low latency so I I sort of think about maybe we could divide this this discussion into sort of two the way I think about the the latency price resistance and sort of two buckets one mental framework that I think you gave me at the start of this season was you can think of uh you know the pressure to basically elongate time around blocks right the more time you have to view um uh different opportunities the more transactions you can see and the more

profit that you can generate so you're starting to see that play out in a super interesting way or especially around kind of optimistic relays uh which is a pretty cool example of that and then the other type of latency is more driven by user preferences for fast confirmations so maybe we could start with the first the first part of that because then we'll get to sort of the roll-up angle on that second part but we sort of talked about optimistic relays he didn't use that word with Justin Drake in the episode with him and tarun and then we kind of continued a little bit with your discussion with Matt Cutler about like infrastructure and how we should incentivize that so what do you kind of think about that pressure for latency like how harmful is that is it totally fine and then maybe if you could like

re-sort of dredge up the the debate that you had with Matt and any thoughts there yeah so um something that we see in the math boost Market is that latency is becoming kind of a very important factor in who wins blocks and and who doesn't and uh so we see that uh both with the build us so for example it turns out that um almost all of the winning bits are basically coming in right before you know the the end of the bidding period why because that's the point when uh Builders have kind of the most amount of information uh about the state of the world they have also the most amount of transactions accumulated and they have also the least amount of risk

um kind of in terms of the price changing against them and um and so it matters a lot that you're very well connected to the network and very well connected to um to the relays um just a quick kind of shout out to also our episode on Searcher so I think what they showed us is like latency is important in terms of like getting your blogs and your trades to the network but it's also equally important on the other end which is initially like ingesting information from the word into your search algorithm and um and yeah I think they were alluding to the fact that even more edges basically in having like really fast and comprehensive kind of data

ingestion really good pricing models and like basically having like a perfect view of the world around you and that's that that that's even more important so um one Innovation that you mentioned Mike um is the optimistic relay so it turns out that one of the rules of relays is basically be this counterparty or this like this like intermediate intermediate between um the validator and the Builder and kind of the the goal here is that the block Builder can just send you know an invalid block or a block that lies about the amount of its payment to the value data and kind of steals money from the

well data that way um and so originally the idea was that uh that the relay uh just executes the block and just checks if these things are true if it's valid and if it if it pays the correct amount but um there are also other ways that you can do this that don't require YouTube to actually execute the block which takes around like between 100 and 200 milliseconds um and uh yeah so we call that optimistic relaying where the uh the block Builder actually makes a deposit uh so they deposit some ether with the with the relay and then um they can basically pass through the relay without being executed

um and if it turns out that they lied about that block the block was invalid for example then the Relay can just take their security deposit and give it to the very data who lost money so I would say this is overall a kind of a welcome uh proposal I think always like when latency becomes more important than the just like the decentralized market kind of needs to catch up and then we need to make sure that kind of the decentralized uh thing that's available has to be like not that much worse than kind of VR what what a relay and the Builder could do if they're vertically integrated for example with each other and so I I don't I don't think that's really like a way around optimistic

Wheeling but um it does have some um implications basically on barriers to entry to uh to the relay Market but especially to the Builder Market because relays now are in charge of taking deposits and like custodying these deposits and Builders Also may need to put up deposits with multiple relays um plus there's kind of the thing where um if if like the value of the block gets too high then basically um it can still be profitable to steer um for uh for the block that I might be we've seen blocks that are 10 20 50 even 100 Eve these are these happen sometimes and uh if you want to make a deposit that kind of exceeds that then it

becomes very Capital inefficient and if you don't then basically the in many cases kind of the like size of the bond is not enough to actually secure that and so you may need a kind of a dual approach right so you basically get past two for smaller blocks that are below the deposit limit and you have to uh simulate um for blocks that are kind of above the deposit limit so you know what you're describing I'm sort of um reminded of the concept of Med burn there with just in terms of incentives uh we again talked about that in the the Justin Drake episode but I would be curious what you sort of thought about that maybe if you could rehash some of the benefits the thing that reminded me that you just mentioned was basically you

know that 32 eth is not super effective collateral if the value that you could extract from a block is much higher than that right um yes and just yeah so I I would love to get your your thoughts on Mev Byrne because that's sort of it was talked about I feel like it's gained steam even since we started doing this season so I'd be curious to get your thoughts there yeah so um we brought this up in the context of this equivocation attack on meth boost that happened um I think one or two months ago where they let uh basically equivocated and stole it um they basically like stole Mev from uh from a block data and um so uh why is this important so yeah math boost basically had this like

um assumption um that um the the way that I cannot just make like a block of their own and kind of race with the the block that the reader publishes but yeah it turned out it was like not correctly uh implemented and so um yeah it's basically uh it was two at the time that um the like the proposal is basically only can only be like slash for relatively small amount and so if they can like steal more money then they definitely have an incentive to do that um because they may like steal more money than they get slashed for and ethereum's uh staking system is very uh yeah forgiving basically of uh of a convocation

um so how much you get slashed voice is a function of how many proposers um equivocate I believe at the same time um and so I think this is like so and this is like in order to kind of protect solo stickers from like the hardware having afford Etc and uh um yeah so basically if you just like if you're the only person equivocating then you get slash for almost nothing I think the I think the um the party that ended up stealing like 20 million dollars from from different Savage spots they got state for one Eve so um and it turns out that there is like a fix that is um you know very very robust to kind of

anything that a proposal can do so basically like it's completely secures kind of the like math boost implementation and that requires kind of moving from like a one a one step to a two-step for Choice rule so you would first have a phase where let's say the proposal committee uh kind of uh kind of a science science basically a block header and then like the proposer assigns it um or something uh I forgot like the exact thing but the idea is that basically the the um the proposal um The Proposal commitment becomes more credible so they it's not even possible for them kind of to to equivocate um in that sense and so um and ironically this is like the same

thing that you need for um for Mev burning because for any v-burn you basically need an Mev Oracle so you need to know how much Mev is actually in the block because if for example the block Builder or the proposal just was able to say oh yeah this is how much Mev is in the block then they would both be incentivized to just lie and understate the amount so they can kind of keep the rest for themselves and so you need what we call like a disinterested party like a party who's not the active proposal or the active block beta and that's what you would that's when you would usually rely on kind of this proposal uh committee and um so uh yeah I think kind of the the the solution that uh of of

burning mvv actually has a lot of parallels to like how we would Harden uh kind of PBS and um nonetheless I think it's uh like I used to be kind of a like a little bit of a Critic maybe of of Mev burning but I think if it's like implemented well then I think it's uh it's a really good idea because it just um it kind of removes this uh this instability in the block reward that can lead to um kind of incentives for uh equification of a kind of reorganization attacks uh stealing Mev from another validator um so optimally like in a perfect world we would just have kind of our rewards for well it has be very consistent and I

think this is pretty much uh what myth burn affords us and um also um in terms of the economics of ethereum we don't want to overpay for security right so if we think that um there is a minimum amount of security that we need to pay for the system to be secure then uh any money that we overpay uh over that amount is basically uh yeah it has a opportunity cost right it's it's like money that's lost from the system and so um yeah I'm with Justin there I think uh is great we just gotta be kind of careful about how we implement it so it doesn't create any new like uh trust assumptions in the system yeah I like

that that was a great explanation and I think it's a good reminder that you know oftentimes you see anything related to the burn and ethereum is kind of that ultrasound money sort of graphic of the supply going down but really there's so much more that goes into these decisions and often it's about eliminating threat vectors for the security of ethereum the network so it's just a good reminder that that's really why a lot of these decisions end up getting made and less about sort of the deflationary meme that you might see on on Twitter um I'd love to sort of just return to that latency here and just talk about the second bucket and maybe this will sort of bring us into the Mev in the modular stack episode that we had with John and Robert which I thought was one of my favorite episodes of the Season it

was just a great great discussion but one of the you know especially within the context of something like a centralized sequencer which is how all these Roll-Ups are running today the incentive there first of all just decentralizing the sequencer said is a tricky problem which we can get into as well and I would love to get your thoughts but also because users want fast commitments right that's a user preference and that does incentivize sort of a a lower latency sort of environment so you know you kind of started this episode by saying hey there's there's a lot of risks associated with low latency blockchains but it seems like based on user preference that that's the way we're trending with Roll-Ups so I'd love to get your thoughts there and how you see that playing out yeah so I I um the the main risk I see is that

um all of the Roll-Ups have have like one or two second block times and um so we kind of it's it's very clear that like as a user you want like faster and faster block times and um but if we go down this route then the fastest block time is basically like using a centralized database is like interacting with server we can just uh you know erect instantly and there's no need to like run the consensus on on anything and um yeah if we like to only follow the user preference then we basically just we just end up back in kind of this like very centralized web toward and the yeah the problem here is that um Roll-Ups have kind of gotten the users used to

just like very good user experience but now that it turns out that oh we actually need to decentralize our sequences um it's not that easy to do once you have uh you know once you have kind of logged into this low block time um because uh it's it's relatively tricky to run uh I've kind of proposed a better separation um with one second block times because this is very like a very short amount of time in order to kind of build a block and run uh like a block Builder option um especially if you wanted to have them in a kind of like geographically like diverse uh regime and so Bitcoin blocks originally like

set 10 minutes apart from each other so there's enough time to propagate large blocks between like China and Russia and the US and so on and um now we are down to like one second block times and it's it's getting it's becoming more and more important um for these parties to be like very closely connected with each other and um I think it's very likely especially as Mev plays a big role for the returns of Searchers and Builders and and the the way data Stone uh yeah you basically end up in that word that we don't want which is kind of geographical centralization I mean this has been you know discussed since at least since I've been in crypto which is basically there are different

trade-offs right and eventually you're gonna have to find some line that we're all comfortable operating in and uh I'm not sure what solution there is other than to just see how it all plays out but it's definitely a concern worth highlighting I'd love to get your thoughts again we talked about this a little in this episode on we sort of highlighted the difference in between decentralizing sequencers but each of those sequencers sort of rolls up within the same blockchain or roll up and then but then there's also this idea of shared sequencer sets right where you sort of have this set of sequencers that sequences transactions for multiple different Roll-Ups and I guess there's some interest there because in theory

this could lead to more cross-domain sort of cross chain Mev which has been a meme for a while it exists today largely in sex to Dex Arbitrage but in this instance you could imagine you know like a uniswap deployment on arbitrum and a curved deployment on you know optimism being arbed away and it just widens up the design space and is good for interoperability but there's a lot of problems with that as well so maybe if you could just comment a little bit on the talk about shared sequencers and how you think about the challenge there and just where you see that whole Space going yeah so I uh my view on this generally is that the name chat sequencer is a little bit maybe of a misnomer because it implies that

um the the sequencer actually has kind of an opinion um about the ordering the audos transactions and it kind of optimal way which is not the case so when we talk about these chat sequences then it basically means that that two Roll-Ups or more Outsource that the proposing of of the next block to the data availability layer to the same party and that party can be kind of the validator of another blockchain um and so this actually includes kind of both base roll ups but also other forms of Roll-Ups so ethereum for example could be the chat sequencer for other rollups and then would kind of be the next ethereum proposal who also

proposes the blocks for for rollups but it could also be a party like um espresso or Austria or or radius or Suave right so if you if you kind of zoom out far enough um and um so but what they all say is basically that um they don't uh execute the book they just uh you know take the transactions and include them and if you do that then of course uh there's like no such notion of for example like cross domain atomicity which they advertise so um James prestwich had I thought a really good article on this I I think it

it was something in the title around like shared sequences or sequences and but the main idea was that um a shared sequencer basically can't guarantee uh you know Atomic execution they can at most guarantee you um Atomic inclusion and um and the reason for that is that uh you basically need to execute the transactions in order to see kind of how they interact with each other because transactions are in many ways conflicting with each other contingent on each other and so if another transaction is sequenced before you then you know your transaction um that might be like buying something on unisop and selling it on arbitrum is no longer guaranteed even to succeed

because you can never know what the state um you can never know what the ACT what the state of the the blockchain is going to be at the point when you actually put the transaction in the block unless you put it at the very top of the block because there's always like transactions coming before you that actually change the state and then the transaction that you thought will succeed will actually fail or not sex not succeed in kind of the way that you think it would um so it's a little bit of an Arcane point but it's very important and I would encourage um people to check out the article um so I think the way that it's most likely going to play out is that these chat sequences will all adopt some form of proposal Builder separation because

although they all say like we don't execute the block the service is completely useless if nobody executes the block and so if they don't want to execute the block themselves then they need to separate this out into a Beta Market and then you just have big block betas to execute everything and kind of come up with an optimal ordering and then they the builders can basically give like soft they can give you almost like soft guarantees about um kind of atomic execution but you should kind of be aware that this is like this is a trusted guarantee so they can put your transaction so why do I say this is like a trusted guarantee because the block Builder can basically like you can't guarantee like you you cannot like cryptographically enforce that the block Builder cannot

put the transaction somewhere where uh the atomicity doesn't break that's because they can order the transactions in any way they want and you have no way of restraining that and uh you could for example mandate that like clock Builders take a bond for example um and and then maybe the blood Builder can get slashed on Suave uh if they kind of don't respect your atomicity but then all of a sudden like you get into the territory where things become very very cleverly and efficient um and so it's all about trade-offs basically and so I would say when it comes to shed sequences I think it's it's like a nice service but it's been it's been overhyped because it doesn't really give you kind of the guarantees

that I think people think that these shared sequences give I agree with that I think there's a lot of misunderstanding about this and shout out to Jill Gunter who's uh at espresso and she mentioned even the abbreviation SS you know that's probably not that is sort of reminiscent of the sort of secret police arm of the Nazi party which probably so for many maybe multiple reasons I agree with the SP sure uh shared proposers would be quite better or better yeah but I agree I think once you start to dig into the technical complexity of it it's difficult and and also there's the I mean one another thing that often gets cited here as well is sort of the economic incentive of the Roll-Ups

themselves so it's very technically difficult thing to implement um but also in a way you could measure the Mev being generated by the centralized sequencer today is revenue for the Roll-Ups and so there's this economic question on top of everything else that we just discussed which is why would the protocols want to do this so I think that's another question that these and I think the general consensus is there will be some sort of Revenue share but but you know I think that's still an open question for me as well totally you're totally right I think the argument that for example as espresso would make is that by opting into a chat sequencer you can make more revenue for your chain

than if you just like made your blocks in a kind of more siled way um but it's very hard uh so it's basically very hard to say uh what like a fair profit share should be between these two chains and uh sequencing is pretty much um the only way that like blockchains make money right and so it's uh it's like quite tricky to Outsource that and say oh yeah I'm getting like exactly what I think I should be getting because it's very hard to calculate uh how much each individual chain like contributed to maybe the final ordering um or the final revenue of a block so it's a very tricky problem

um I think you can like like solve it in a little bit of like hand wavy way uh just to maybe like governance agreements that you like adapt every now and then um but uh yeah I think we yeah it would be interesting to see kind of how these economics play out then in the real world yeah I'm in total agreement with you there maybe we can sort of segue this into the the cosmos episode that we talked about and I'd like to get to privacy sort of a primitive for Designing blockchains which I thought Harry did a great job or uh Henry did a great job of describing um but one of the analogies I'd love to get your sort of thoughts on because I

missed you during this episode but you know we sort of compared the the sort of Internet of blockchains and Cosmos right it's a bunch of different sort of full stack uh blockchains that interact with each other but the that environment of Cosmos and the Roll-Ups like so the sort of layer twos on ethereum I think there are a lot of parallels to um in that first of all you don't really have you know Atomic composability like you have on Main chains like uh eath main chain or Solana and you're also started so you sort of start to worry about um you know standardization across different chains and how do these chains talk to each other and maybe this idea of cross domain Mev which we were just describing before how did you how do you

think of that analogy between Cosmos and sort of the roll-up environment on ethereum yeah it's a great question I think um that's obviously I think a lot of overlap between Cosmos and the roll-up environment on ethereum um I think that the line will blur more and more I think there will be many Roll-Ups in the future and uh you can almost like mix and match between different uh components that you can plug into the you know different layers of the blockchain right different settlement layers different data availability layers different maybe sequences sequencing layers and different you know execution layers different virtual machine machines and like who else what we'll end up like stripping out like uh shared mempool

like Suave I think is a chat like block space auction I think is one of the things that um would make a lot of sense for many reasons um and when you kind of have the ability to unbund the a chain like that then you gain much more control over Mev than can be afforded by you know a chain like ethereum and that's for the simple reason that ethereum is like a very big has a very big state it has like a lot of applications a lot of users that already are built on top of it and changing anything in a way that doesn't break the system or makes it kind of breaks a social contract at least like towards at least some of your users is very difficult

um ethereum is also a system that's you know designed to be maximally credibly neutral and so we don't want to change it too much and that mandates that a lot of improvements to the system have to happen in a way that are kind of more Backward Compatible or also more maybe don't even require changing like core protocol that can be built on kind of peripheral layers so math boost is a good example right we could have done uh could have done and shine PBS from the start but then we would have to touch the protocol every time we need to update it and since it's a completely new protocol that's operating in a very adversarial environment where there's

too many parties incentivized to break it um we have filtered need to update it many times and uh so I think the decision to keep it outside the protocol until we think we have found uh an iteration that that's very robust and doesn't need to change also with the like changes in the market uh anymore um I think that decision is like was totally justified in in Cosmos or on kind of the more centralized roll-up chains you don't have these constraints and so you can move much faster you can you can kind of change much more things about how your blockchain works and um you don't even have this constraint of

um that kind of your like you need to um kind of assume that like your validators are um you know very like adversarial or something like in in ethereum basically we have to assume everybody can be available and If For example you can steal the money from like a search I bought then we have no way of preventing that whereas in in Cosmos they can say oh yeah validate us like you're not allowed to uh send it to user why not otherwise we just like uh make a governance proposal and have people vote with their coins and then you know slash your coins um and so this is like very far outside what's possible in ethereum and I think at these different constraints

basically make very different approaches to to Med possible and um you know given these constraints I think both uh both both like ethereum and Cosmos are kind of doing what's right uh you know given their unique situations I couldn't agree more I I really enjoyed both our Solana and Cosmos episodes the it sort of informs my general uh you know perspective that I I I like that there are different sort of L1 ecosystems because I think it does encourage some amount of just diversity of of thought uh maybe a slightly unpopular opinion would be I actually don't think in all instances same think is a bad a bad trait for

instance in a startup you sort of need to have same thing right you need everyone you need everyone rowing to the beat of the exact same drum that will literally be life or death but I think once you get large enough or once the problem in design space becomes big enough you want to start encouraging diversity of thought and opinion so you know in general people in Cosmos sort of think with a relatively similar attitude obviously not entirely in painting with a broad brush here same thing with Solana and same thing with ethereum so I thought it was super cool to like Cosmos is just a great example of something that's so opposite to ethereum because they don't have this you know unique problem and opportunity of securing billions and billions of dollars in tvl they often have more app-specific chains

uh that's you know the app chain thesis of Cosmos so they can do these decisions and the biggest decision always for me like that the difference that they make is instead of like trying to make validators sort of commoditized and and dumb they want their validators to do more because they understand that consensus layer but also the semantics of the application and I just think there's so much cool experimentation that comes from that so it's just yeah interesting I think um ethereum salana Cosmos I think these are the most like the three most kind of interesting blockchains or like blockchain universes I think because they make very like distinct trade-offs from each other like

they in in kind of a competitive market we would say they are meaningfully differentiated from each other and I think if you if you want crypto to win if you if you're not tribal if you just want you know decentralized applications to to you know become meaningful in the world and you should be very happy that there are different parties to explore different trade-offs and have very kind of serious communities behind them and like attracting Builders and so on on their own like in their own right and so yeah I definitely applaud like ethereum salana and Cosmos or all three of them um for just doing what they think is right and like exploring the design trade-off and yeah like a lot of things uh sometimes they don't work and you

have to like walk it back for example Solana um going without a fee market for example like uh some some stuff may be like more obvious than others but nonetheless I I think like what they do reveals a lot of interesting stuff um about Building blockchain Systems that yeah everybody can can learn from Quick break in the show here I want you to imagine something for me imagine swapping two stable coins on chain paying zero dollars in gas and instead getting a rebate of two thousand dollars this is something that's actually happened on chain to understand how I want to introduce and thank the season sponsor Rook zooming out for a second the current state of affairs at Med is billions of dollars so far have been

extracted from users Pockets using Mev Rook is coming in and saying enough is enough blockchain should drive value to their users and the applications they use it is time to leave the hobbyist era behind us if we want to move forward we want to get this right that's why Rook has built a custom blockchain settlement Network and it's one that gives you full control over the entire transaction life cycle today you can connect to an open source rep node The Rook protocol will automatically match bundle and auction your orders and transactions in seconds with zero gas overhead also any Mev that's discoverable along the way will be returned to you the user created as a collaboration between the industry's top mechanism designers and Mev Engineers Rook was built from the ground up to be scalable safe and programmable you can

get your own mempool choose searches and Builders and link your mempool with others to discover even more Mev you can Define how the Meb is shared and delivered as well and what could basically process anything from transactions to meta transactions and more this is the way that blockchains basically should have been from day one so if you're a user listening to this here's what I want you to do I want you to go to your wallets go to your favorite app your node provider and say Hey I want you to be working with these guys rock I want the Mev that I create to be redistributed back to me if you're a developer and you want to stay ahead of the game the best way to do that is to follow them on Twitter they are at Rook or even better yet slide into their DMS they are

lighting responsive they'll get you set up today and if you do slide into those DMS as always please tell them that I sent you now maybe we can sort of segue into privacy here and I would love to get your thoughts on you know I thought Henry specifically had a very cool take on privacy and you know Henry sort of found his way to Mev in kind of a backwards way he started actually as a researcher I believe in sort of the zero knowledge uh space and was actually basically trying to to find a way to give users more privacy is sort of a necessary primitive and then that wound up having implications for Mev so maybe um if you could kind of give your take

on privacy as a necessary primitive for markets functioning as opposed to kind of this ideology I would just love to get your thoughts on that yeah this is such an interesting topic because I would say there's three different camps in crypto when it comes to privacy that like had very different motivations so you have to I like the very ideological people who are motivated by you know the idea of like human rights for example everybody should have the right to to privacy to transact privately in fact have private thoughts and so on and um and then you have kind of the like very academic camp like everybody like all of this researchers studying kind of zero knowledge and cryptography um and trusted execution environments

and so on and um so they are in it almost for the intellectual challenge of it like in the Curiosity and then the third campus is basically like like mechanism designers and like Market structure designers and um that includes us I would say and then it actually turns out that you kind of you need privacy in order to build um you know credible mechanisms that work for example auctions just tend to work you know infinitely better when you can have privacy of bits for example right when like not every bit is is kind of public and people can react to each other's bits and so on it makes the auction extremely strategic and um and very very centralizing as well and also much

easier to break for the auctioneer um and this is one example but um privacy is kind of extremely important when you want to build you know a good Market structure and especially in the nav supply chain um uh privacy is very important because there's a lot of informational value in the bits so just seeing the intent of a person what they want to do um basically already gives you like a financial Edge because you can front run them and allow you to do harm harm to them basically um and it's also very important for collaboration so one of the I one of the goals that we have for the Meb supply chain is is basically for it to be very flat like we want to because we want

validation and like block production to be decentralized we don't want any actors who kind of are able to monopolize uh you know the med supply chain and the crew like an extraordinary amount of power and so we we want to you know make it just as possible kind of to be a small Searcher or a small block Builder um and not as it is for example to to be a bigger party and basically what that requires is that that small parties can also collaborate with each other as well as bigger parties in order to participate kind of in the same extraction process and block building process and this collaboration doesn't work without strong privacy because you

always have to be like very aware that others can steal your bundles um and and kind of steal money from you and so it turns out that privacy is is just of fundamental importance um for the Meb supply chain and yeah we couldn't we couldn't kind of achieve our goals uh if we don't solve the the Privacy puzzle yeah and Mev share you know which went live relatively recently I think this is super cool example of that as well um it's something that wouldn't be possible without the power of privacy yeah it's uh so any of you shares a new order flow auction that that differs kind of from all existing products in the sense that um uh searches can can basically run on

encrypted uh user orders so search or submit um their backgrounds um in without actually seeing uh all of the information about the user order so right now we revealed some amount of information not enough to front-end but just enough kind of to constrain the you know the search space um so that searches are not completely blind and then there's actually uh the block Builders are in charge of like running this simulation and matching the orders um together which is very cool and for search as it's a completely new paradigm um kind of searching searching on on private data is is not what they are used to um but we think it yields fundamentally better outcomes for users and if you

kind of continue this this line of thought then the next step here is kind of just searching on search orders not just on user orders so for example one Searcher could you know do like uh an Arbitrage and another search a good kind of um you know do another part of the Arbitrage that maybe the the first search I was not willing to take because of risk limits or uh because they had like a different like ability to hatch or like such as um kind of back running each other or like matching like complementary traits together and um yeah I think uh I think we're looking at ways that that kind of um order flow can be aggregated um in a trust this way that that doesn't

like require trust from anyone uh and also kind of removes trust in the auctioneer itself and uh yeah I think that's that's kind of the next big item on our agenda yes yeah I want to get into order flow auctions because that was a great episode that we did with uh quintus and Barnaby yeah but before we do I have one one question for you because this was I actually thought one of my it was a telling moment in in the series I think when you it was the Searcher episode and you asked uh both Dean and Anish are you using private order flow and there was probably a five second pause and a niche you know very diplomatically said I think you'd be hard-pressed to find any and uh one sort of you know Trend that it is privacy

sort of is definitely this good Market primitive I'm also to be honest to just be transparent I'm ideologically in the camp of privacy I sort of find myself rooting for more privacy I think it is a good human right that we could all probably get behind but the sort of flip side of that is you know people talk a lot about kind of private order flow and there's so much incentive for especially Builders right to to get this private order flow and on top of that it does seem like the there's kind of this maybe called like fragmentation of the mempool so to speak and used to be there was one sort of mempool it's not exactly it doesn't exactly work like that but it was one public mempool now you actually have protocol level changes like kind of abstraction 4337 where there's this user

op sort of mempool and uh there's a whole bunch of different whole bunch of different things out there so I would love to just get your thoughts on kind of private order flow and is that the destiny and sort of the way that um Mev and transactions on ethereum are moving or what are your thoughts there yeah that's just that is a great question I think right now the meta game is definitely that it's all about exclusive water flow so in order to so all of the the top block Builders are also Searchers and um the way that they can they can basically like be block Builders is from their own exclusive order flow from their search operations that they may not necessarily share kind of with other block betas and because

they don't submit to other block Builders but rather do the block building themselves on many blogs it's actually more profitable for them to to do that because if you share your uh if you kind of share your order flow with all of the other block datas then kind of in a very naive example you just increase kind of the bit of oil block Builders equally whereas that end up that they end up paying to the Valderrama whereas if you keep it private then uh your own bit stays the same but the bit of all other block Builders goes down and so you actually have to bid less you only have to bid the second bit right the the bit of the next best block Builder and so you just end up leaking less money to the validator and and then end up retaining

more for yourself and um so I I think um this is a strong strong uh incentive to attain um exclusive water flow whether it's from users for example by buying order flow from like a wallet or an application um in the form of like vertical integration or striking any kind of private agreements um or the other alternative is of course like running your own trading operation because then you almost have like homemade exclusive order flow that is in many cases very very valuable because of you know the high amounts of fees that it's able to pay because of the profitability um I think in order to kind of break up

the current meta game of the Searcher data um so if this was allowed to continue then you're basically going straight for that future where um only the biggest trading firms are able to build blocks on ethereum and today that's maybe like a winter mute maybe an SCP and these are like firms where you kind of know that they are ideologically aligned with crypto to a good degree but um like when it comes to I don't know like a citadel or like a tower or a virtue whatever like once it becomes profitable enough then you can be sure that like the existing like thread fire hft firms will also crowd into the space and they have more you know they have even

like better people and they have even like lower latency and proprietary algorithms and bigger balance sheets and then all of a sudden you know then you find yourself with like these new overlords that um you didn't ask for and you don't know how to get rid of them anymore and so that's why I think it's very important to break up the current meta game and make it so that searches can be searched us again um and don't have to kind of opine on you know the composition of the entire block and um and that basically requires you know a meta game where order flow can be shared more freely both because it's kind of uncheckered from the bounds of of trust um so you don't have to trust and that anyone would steal or mess with your

autoflow anymore um but also kind of make it economically feasible um for other flow to be shared right so removing this incentive that uh you have to pay the second bit so I want to get into order flow auctions you know I actually it's funny this is one of those Seasons where we could have done I think twice the number of episodes and it wouldn't have felt packed at all like there was just so much there was so much we could have talked about and I do wish we could have gotten into this subject a little bit more and I actually changed my I had my thinking changed about order flow auctions um this season and I'm sort of thinking about it differently than I was in the beginning and I would love to change it so I went into this season thinking and

actually an early iteration of our final episode was going to be focused on wallets and the impact that wallets were going to have and I've sort of changed my thought process there because and actually some of this was I went back and re-listened to the tail lipowitz uh episode that you mentioned uh is I pledge allegiance shout out to reverie they've got a great podcast and I hadn't really thought about that uh we got into it with quintus a little bit but sort of general order flow auctions versus a specific order flow auction run by a DEX or something like that and at least for me I like to think about things in terms of like incentives almost like if you start from this hypothetical like how will they play out

and I probably changed my thinking in that I think the dexes are the ones the applications are the ones that have the user connection and The Leverage much more so than the wallets which I think uniswap will be super interesting to see how that plays out because they sort of have both now um it's definitely one same with coinbase by the way uh definitely ones to watch and and then it will be kind of this interesting so I think wallets are going to have much less of an influence than I originally thought and then I think it will be this sort of interesting trade-off between generalized order flow auctions where maybe you have more information and therefore more like profits to be extracted versus a DEX

exerting its own influence um and even though you have less uh you know transactions and therefore less value that you can extract they do have more leverage because they're the ones creating that maybe so I that's kind of how I've changed my thinking about it I'd love to like I know you probably have a perspective about this you're not flashbots but what do you think yeah it's a it's a it's a great question I think um or there are several parties in the supply chain we have a lot of power I think I would not count out the wallets I think guys there are a lot of users who um basically would not change their wallet um and then I think the application is I would say it's like in about like a 50

50 kind of split um when it comes to like managing the user to to do different things um because a lot of users also Trend that turns act like through their wallet right so for I mean some people go to like the unisub front end um and trade there and in that case I would agree like you know swap can kind of direct the transaction wherever they want and this is actually a thing that if they were to adopt 457 then it becomes even easier to do so it gives application even more control over how exactly the transaction is executed um but for many users also the wallet is simply kind of they are portal to crypto and they rely on whatever the wallet is telling them like where to trade for

example or even executing that their trade outright as we see with metamask and so I think both have a lot of power s does coinbase so coinbase for many is even like the starting point um for for many in kind of their onboarding journey to crypto because you you couldn't even use like a metamask um without first like buying some if like you need to deposit Euro need to buy some even you need to go out and so I think they actually are all uniquely powerful in their right in their way and um uh yeah I think that it just differs by different users I would say it would be very interesting um to see how that how that plays out so right now it just really depends on how this all ends up

playing out right so when I think about okay what what impact or sort of Leverage can wallets exert across the value chain well a whole bunch of transactions flow the intent starts at something like a metamask right and right now if you have a solution like uh flashbots uh or like Mev share or something like that or the solution that you know Rook has as the the sponsor this podcast shout out Rook they you have to go who do you go Lobby right you could go to RPC endpoint you could go to the wallet and uh you're you're trying to it's sort of like the marketing you know strategy I would almost liken it to drug companies sort of marketing directly through like TV to the users even though ultimately their distribution mechanism is the hospitals

so that exists today yeah in the in the in the future I think it's telling actually if you look at the uniswap wallet uh the mobile wallet that recently came out that you actually don't have the option as a user to switch your RPC endpoint and in the future as we go past our little world of crypto nerds today how many people are actually gonna care about that you know so in in some sense the wallets will will have that but I also I can't kind of help it also like when I close my eyes and try to interface the exam or imagine the user interface I think for I used to think about there's going to be one wallet that wins and sort of organizes your apps and you start with your wallet and go there but now I I kind of think

that already exists and it's called your phone and you're gonna have your phone is really gonna be your your wallet at the end of the day and then you're just gonna go click through apps and then like coinbase's wallet as a service you know I think basically the apps that gain product Market fit are going to integrate their own sort of wallets that'll be white labeled and sort of built into the product experience and so I just think that yeah to me it kind of comes down to the user relationship and I think people feel more of an affinity with their apps than an interface but I think if you could predict who's going to own the user then and now you can make a lot of money being right I think in practice it's it's a it's very like hard to know B I think there will be

some distribution between them where if you look in traditional products then um it's not always kind of user-facing product that ends up making the most money in in kind of the supply chain right um for example I mean there's this famous example of like Intel and windows making kind of 95 of the profit in the PC market even though neither of them is kind of the like makes the actual PC or like contributes um you know any of the most of the hardware components nor are they the party actually selling the PC Etc and um I think account industry is going to evolve and and who ends up having powers is very very hard to know in advance and it also differs very strongly by by the

user so if you buy if you buy like an ETF like do you know like what's the The Profit split there between like uh I don't know like the like the index provider like how much money does like msci make versus like how much money does Vanguard make uh or like interactive brokers and how much does like Robin Hood make I mean these are like different kind of uh you know different parts of the supply chain speaking to different people as well and yeah it's uh I think it's like a similar situation here it's a very good point just a good reminder that's very early and it's super difficult to predict any of this stuff also we'd love to get your thoughts yeah basically on sort of an

app specific order flow auction maybe hosted by something like a uniswap versus a more general order flow auction where uh you know you have access to a little bit more information but you know you might be end up sort of competing with the unit swaps of the world in a way yeah I think um I think it's a good question I think um Tio Leibovitz in this interview with uh iPatch Allegiance he he gave some arguments for why um it's better to have and have specific order blockchain and I think the two arguments that he gave they come down to uh one economic alignment so he thinks that apps are not interested in Sharing any Revenue with a third-party provider and the second is

specialization uh so you can he would argue that you can design order for auction that's much more efficient for an individual use case than a generalized other floors it could be I think in the other direction uh you can argue also a few things so you can argue um complementarity or network effect of having more orders in the same auction I think that's a big case for the generalized options uh where for example trades that happen on unisop they might have um some kind of network effect with liquidations on a landing Market or trades that happen on other exchanges and so on I think you can kind of argue that many user intents can ultimately

matched against each other or bundled together in some way um I think other network effects apply as well so I think if you have one uh one other Fortune then there's only one uh one auction kind of uh for Searchers to to adopt um because kind of the the um the demand side from users is important but equally important is the supply side um I would say of like Services provided by search us and block Builders um and so um that's that's very key I think and then in the episode with quintus and in Barnaby we also went into uh I think some more uh arguments why it

can be more efficient to have a single block space option and I think this is the way how we are thinking about it with SWAT where um there's just this one giant uh block space auction where like it collects all the preferences from users and then uh run some kind of uh you know combinatorics on it um and and kind of produces um the optimal ordering across several different chains and um so we think if it's if it's one giant auction um there's actually uh more uh benefits in terms of user Surplus than if you isolated this into into two steps where first user transaction is kind of setted

or optioned off and then there's certain auction where the Searchers have to actually bid uh for Block space or the block beta or validator um and finally um I think also decentralization is a is a key part so it's it's much more difficult I think to uh to have these like individual admin pools so if we imagine at the extreme that every app has its own mempool then I think um this puts us much closer to uh a place of exclusive order flow or exclusive order flow like dominates the Mev supply chain um then if you had some kind of holistic solution uh where Searchers and block

Builders can access user order flow but only in a way that that is um in line uh with the user what the user specifies their their transactions are allowed to to do um so yeah for these reasons I think that um I'm betting personally that a generalized solution will be good enough that kind of these Network effects will kick in um and in terms of economics um that should then resolve itself as a result because if more value can be created from a generalized solution then it would still be more profitable for an application to direct the order flow there than finding a kind of vertically integrated solution themselves I I hear

you on those arguments and maybe you know it's just funny the the Mev space moves so quickly I heard you use the word intense there that's the the latest uh sort of buzzword of the day and um I I thought it might actually be valuable if you could sort of describe uh you know what an intent is and kind of how how that represents sort of a a change in the way that people think about order flow today and why are so many people focused on that particular word at the current moment yeah so um intent that word has come up in the context of um basically moving to uh from from executable transactions to more generalized um intents or kind of descriptions what a transaction is supposed to do so in

example of an intent would be or rather an example of of kind of an actual executable transaction would be you know um call this particular unicep will um you know with a with this particular swap right so you define exactly the the execution path your transaction is set out to take um whereas in the intent you would only specify something very general like um I want my transaction to pay at least at most let's say like a thousand usdc and I want at least 1.1 if so it's almost like a predicate on on top of your transaction what your

transaction is kind of allowed to become and then someone else can take this um this sort of intent which is not an executable transaction it's just like almost like like a small part of a transaction um and they can turn that into then an execute a transaction by substituting the actual execution path so they can find the best way to execute your trade uh on ethereum or on some other domain right and so this gives basically more control to third parties um to turn your transaction into something that can actually satisfy uh your intent and if you have that then so calcium was

the first um project that pioneered this right yeah um you basically introduce uh you unbunded basically like the construction of the execution path from the user's intent um and that allows you to create competition about constructing the best execution path so you can have multiple solvers or executors or search us um if you were um oil compete on finding the best path and then turn that transaction turn that intent into a transaction and only the best one may get executed so I think um this kind of division of labor in the Mev supply chain enables much greater flexibility for the

user transactions and better execution because um it's basically uh you Outsource the execution path the finding of it to a competitive market so that's why I think it's very common missing but intents also carry um certain risk that I feel like is worth calling out right now because it's very under discussed today so the risk that I see is basically disrupting uh the Mev supply chain especially um by making like fragmenting order flow a lot so um I think right now there's a vision or like a group of new parties that are positioning themselves to

become bandlas because they think that they can take this user intents and um you know generate any view from it um or basically become new uh you know block builders on the basis of exclusive having exclusive order flow from these user intents and I think that like while chasing these two x benefits for users we still have to keep in mind um that we need to build a decentralized Med Supply Chain and then we need to Min and that's only possible if we actually minimize uh the role of exclusive order flow and uh seek to basically Empower competition with all layers of this deck

and that involves uh developing solutions that make it so that user order flow can only be executed in a certain way but also keeping it private at the same time um and uh yeah I think um this is an interesting strategic question right now not just for flashbots but for for anyone I think in in Mev like what is exactly like the Mev x uh con abstraction uh going to look like and how can these new players be reintegrated uh into the existing supply chain without kind of degrading it I think it's a an excellent question maybe sort of a good place to start wrapping

up because exactly a sentiment that you just shared is maybe like a risk or a potential warning of sort of fragmenting uh you know order flow I had kind of had the same thought about you know 4337 now we will be used to you know returning to a simpler time right 2019 there's kind of one public mempool and that was that was a sort of a simplistic way of viewing things but now it's certainly more complex than that right so you layer in something like 4337 where you have a separate user op mempool that's sort of at an earlier stage of the chain where we're dealing with intense and then you also might have something like uh like a cow swap where you can plug in to be a solver and suddenly you have all of these different right sort of pools that you need to plug into and the one

of my big takeaways from the season is especially when it comes to builders the real differentiator is just access to information be that extending the block time through latency be that access to exclusive bits of order flow the more information and the more time you can give yourself the more successful you're going to be so my question to you is is that inevitable because those are two trends that we kind of see accelerating right like the latency sort of race and order you know user information court is kind of fragmenting do you view that as inevitable are we relying on larger Builders here and how much of a threat is sort of bigger Builder to ethereum's decentralization and incredible neutrality uh I think latency definitely plays a big role

um I think the other role is just the economics of exclusive order flow so if you are a search or a block builder then by having exclusive order flow you can just carve out the bigger pie a bigger piece of the the Mev Pi um right now uh so the the incentives are basically basically in some cases like stacked against sharing order flow and I think um what is necessary is that we find ways to create the economic incentives for order flow to be like for for Searchers and block Builders and wallets and so on it like that they find it more profitable to share the order flow in an open way rather than sell it or keep it

for themselves and um so it's I think this is one of the things that we're looking into um and this is one of the challenges that any kind of decentralized block data also will need to solve so I think both aligning the economic incentives but also basically creating Network effects uh from aggregating like a lot of other flow in the same place um and these problems are very closely related so I think um yes like size is a problem economic like exclusive order flow is a problem and latency is a problem and uh it will be yeah it would be difficult but not impossible um to I think build a system build an

auto fluctuation build a kind of block space auction that that can address all of these yeah and uh you know Suave I think plays into that and one of the the cool things for me about this season has been uh learning a little bit more about it and I now you know sort of describe it internally as kind of the most audacious project in crypto out of all the audacious projects that exist because the scope of it is just so massive and it's just so I'm just I'm gonna be very interested to see how it plays out in the coming years but what are like maybe if you could break down within swap what are some of the challenges that you're most sort of concerned about when it comes to Suave and then what are some of the problems that you think are most it's most immediately set up to solve

yeah so maybe just to give listen us a very quick recap so yeah Suave is basically uh we wanted to be the center and unifying block space auction in crypto so on the one end it takes in all of these user preferences and then it it Aggregates and matches them in a way and runs the auction in a way that produces some kind of ordering on the target chains so it basically distributes the supply of blocks space and in some way that is uh welfare maximizing for the users as well as Revenue maximizing for this for the sellers so it kind of satisfied satisfies this dual constraint and it achieves that um basically by by Searchers and other

parties kind of being able to upload um you know programs if you will that are you can think of them as like a loud user like allowed operations on user order flow so the system itself is kind of this like encrypted pool of transactions and then other parties can upload kind of actions that are allowed to run on user order flow so Mev share for example is one example where the user the user opts into using Mev share then what happens is that transactions get shown to search Us in in a privacy preserving way and it allows them to recapture their own Mev so I would say that Mev Discovery for example is one necessary function of this option right um another is maybe that transactions

get can get merged together or that the user can control exactly like the execution like the execution of their transaction According to some predicates or who's allowed to see what transaction and that is all kind of mandated through the use of these uh of these programs so in terms of uh what it what it most like immediately can solve I think as just preserving um you know the decentralization of the Mev supply chain in in the light of you know changing uh kind of user preferences um but also changing kind of search are Builder strategies so I think there's two big trends that we are seeing right now like on the one

hand like we see there's increasing user demand for higher efficiency in the global blogspace option and what this means basically users demand like lower fees they demand discovery of the Mev and they demand privacy from front running all right and so if the blogspace auction doesn't provide these things in a decentralized way then the users will increasingly go to centralized entities that can provide them that um and this like if if kind of a market is contingent on on trust in centralized actors then that's very problematic because it basically rewards the biggest and most like the biggest and oldest and kind of most entrenched parties

so flashbots itself for example can or like a gnosis for example you know they can provide uh like the service because they are very very ordinary trusted kind of institutions in crypto but it's impossible for like a new party to come in and say you know I want to receive autofill from users because they are less trusted and it's very hard for them to bootstrap this amount of trust um and in terms of the the like searches and build us I think um the system needs to find a way to break basically um the the incentive to kind of uh have exclusive order flow um but also benefit from latency advantages so it's all about kind of

uh like arbitraging away what what what leads to the current dominant Paradigm in order to make block building uh more distributed so I think if I had to guess then this is this is for me kind of the the main benefit basically it increases user efficiency and user guarantees in a way that also flattens the Meb supply chain and keeps it more robust uh and this is yeah this to me leads to much better outcomes for users but also like in a way that preserves ethereum's long-term decentralization of course there are many risks involved with that I think um getting to a point where it's genuinely

decentralized uh it's very difficult um so right now flashbots for example is the uh an important kind of centralized actor in the supply chain of Mev you know that many things like many things rely on uh same goes for relays for example we talked about that in our episode with Matt Cutler um in terms of kind of the efficiency side of the market I think we're very optimistic that we can you know by having like an open search on Market we can provide a better uh execution for users then you can have in pretty much any other system um on the on the Searcher side it's definitely harder right so if you think

about what makes someone an effective search a builder for example in today's meta game then you have to like think about stuff like you know far like low latency access to the large exchanges you know large balance sheets like um very strong pricing models Etc and then how do you Arbitrage that like how how do you write a protocol that basically makes it so that a small Searcher can do part of what currently like a big trading firm does and that like small Searchers can work together um that basically the lacrosse for example the sex Stacks other charge Market becomes accessible to to smaller players and that you don't have to be a block builder in order to be effective at sex text Arbitrage

because I think if there's like this economic like link between block building and sex text Arbitrage that's obviously a big problem because then over time whatever party is the best that sex other touch will dominate block betting and so um I think these are the important questions to ask um and they speak to like a lot of questions that need to be solved along the way mainly like latency is a big one I think uh finding a way to create um you know decentralized privacy is another big one um but also giving orders uh like arbitraging like yeah these benefits that that kind of the the big centralized trading for himself I think um I would say these are definitely the

challenges that we're gonna face over the next year so maybe as a way to just bookend you know this conversation sort of the entire season let's say it's two to three years in the future you know when you and I are getting together and saying hey it's been a little while since we did that Mev season a little while ago maybe we should do a rehash of that what are we going to be talking about is it relatively the same subjects are we going to be talking about privacy and latency two to three years in the future is there something new what do you think the kind of future any Trends you're particularly paying attention to something that's under discussed now what are we going to talk about in that season two years from now oh interesting so we I maybe we assume that you know swap

already exists and it has like a vibrant ecosystem kind of of mechanism spilled on top of it that like lead to Mev Discovery and like different ways to execute usage transactions and it has a lot of chains connected to it as well I think I hope by this point we'll be talking about how to scale the system really far like in terms of connecting more and more domains to it but also finding more and more uh Network effect between different order types in order to just drive more and more value to the user basically at a lower and lower cost I think that would be my dream um I think it will also be a big question how the system should be governed um so my personal

yeah mental model for this is Visa as I think I said uh in the past in the season um basically an organization that that is like that that creates uh you know cooperation from like actors that otherwise only engage in in competition and and kind of um creates a positive externality from that um and getting the governance rights of that system right will be very important as well as kind of decentralizing the stewardship um because whenever you build I think the Mev supply chain is naturally is incredibly centralizing across chains because of the network effects involved and because of the skills that you need to win on one Chain's Mev Supply Channel almost identical uh to whatever skills

you need to talk to like win on another chains mov supply chain right and in fact being in like two Supply chains makes you better in both of them right so that's kind of the particular effect of like scale economies as well as cross domain order flow and so on and um and that's why I think at that point we'll be looking at this like Behemoth basically like there will be no more talking about like this is like this one Cosmos chains and maybe supply chain this is like ethereum's mov supply chain and this is like arbitrum 7B supply chain I think we'll be thinking about them as one Mev supply chain and we'll be thinking about how can we Harden that as much as possible how can we make it better and better for this how can we govern it and how can we ensure that

kind of its stewardship is you know as decentralized and robust to um you know for example the uh the utility function of any individual regulator as possible I think it'll be very much like talking about ethereum at that point uh basically an open source project that a lot of other applications rely on in a lot of other blockchains yeah that would be that would be my guess I've sort of a boring answer as well and then maybe one that's a little bit more fun which is I think the the one that's the boring answer but I do think would be a part of the conversation two years from now would be regulation I think especially as it gets larger and larger and more traditional firms get involved and you know sort of it leaks out how much money is being

extracted and the absolute value of Mev you're going to get Regulators involved there and that's going to be more part of the conversation of you know how to uh you know what does best execution look like in a Trad Phi sort of World versus what we deliver in crypto and I think that's gonna be I think flashbots is already keyed into this with the emphasis on execution for the the user but I think that's going to be a bigger part of the conversation the other part of the conversation which is again this is so it's so difficult to tease out Mev versus protocol design philosophy but I don't know if you caught vitalik's paper that he dropped over the weekend and I'm blanking on the the title here but I can get it but it's very much what Barnaby was discussing which is where is the the

pomerium right the the line of uh ethereum versus you know other things that are built around and outside ethereum and I think there will be event I do think eventually people will start to look at like daps and Roll-Ups built on ethereum through sort of a political angle and um because that that article made me think about what if you know we're moving execution and users up to a big Layer Two what if there was a catastrophic you know day one bug in arbitrum or something like that and all of that pvl just gets wiped out that would be almost an existential moment for ethereum because you know to vitalik's point people would demand people would demand very loudly for there to be a fork of ethereum's uh code

similarly to how there was in the Dao and I think a lot of some of what we'll discuss in Mev is maybe like a softer you know one catastrophe move in forward of that is something like uniswap this was sort of Dan ellser's point about uni chain being inevitable where maybe some of the larger apps with pmf that are throwing off a lot of Mev well politically sort of try to you know push that clout right into uh you know favorable sort of regulatory decisions from from an ethereum standpoint and I think that would be part of our discussion two years from now as well so we'll see if if that ends up playing that makes total sense I think and regulation I think I like the like double term of the meaning I think regulation from uh from other

blockchains but also regulation from Regulators that exists outside of blockchains I think both will be in a very important topic um the first is already a very important topic today the second we haven't seen much regulatory uh touch points basically with uh for for the Meb supply chain I think fortunately because it gives us more time to preempt the need for regulation frankly by Building Systems that you know are better even than what can be built in thread fine I throw in one curve boy one thing that I I hope we will be talking about in three years which is we'll be talking about Mev Beyond crypto in the sense that you and I we touched on this like the

philosophical side of Mev right that Mev really exists everywhere it's like in any system where there's uh like a privileged party um that has you know access to information sooner that that can act on information sooner or any any like um situation where there's you know money value loss to this coordination between you know parties I think those all qualify you know smev according to you know the three EV model um that uh Sun put out and um you know Mev is really everywhere and I think it would be amazing if if this is

not if Mev doesn't stay a crypto phenomenon basically like I want Mev to become you know a global field of research that basically completely changes mechanism design and um and back the way that we design oil systems to make them more accessible and more fair and more egalitarian and just like work better for users than they do today I think that's a beautiful sentiment and a very a great place to close uh this episode and the season honestly this has been a a super fun one doing this with you uh I was you know saying to you offline but this is gonna bum me out I've really you know all the seasons have been fun of this show but I will be pretty sad to leave Mev behind this has been a really

fun one doing this with with you as well so yeah this has been great can only resonate uh resonate that um it's been a lot of fun I already feel that uh you know the withdrawal clicking questions [Music]