Panel I MEV: Fractalization of Block Building | Pragma Istanbul 2023

[Music] welcome to the final panel on the fractalization of block building um it is a meme as a tribute to tun who is um a master of abstraction at memes that said I don't know about that we'll see um why don't we get started so this is a a discussion primarily about block building but block building means a lot of things uh in ethereum probably more than you think so hopefully um we can uh share what we see and get on the same page so um why don't we do a quick round of introduction I'm going to ask everyone to um introduce yourself and your project and the role your project plays in the transaction ction supply chain if you play multiple roles please name them and also um please share one single uh surprise you have from 2023 um in the me space you didn't expect last year so Teran hey tan uh founder of Gauntlet um I uh we are our supply chain usage is we do risk management for a lot of defi protocols so we monitor a lot of mols and try to understand sort of when particular Protocols are more at risk depending on their liquidation mechanisms or other features um one surprising thing uh I guess I'm surprised that we haven't had more disclosures of L2 me if there has been some if there have been sort of malicious sequencers I I kind of was hoping we would have like had a crap rollup like someone just Forks a roll up and then it exists only so that the sequencer extracts and I'm surprised we haven't seen that performance art yet hey um I'm Anna from cwab and if you look at the M Supply Chain we at the very beginning at the application layer and our goal is it essentially to reduce the me surface as much as possible um because we really believe um or our mission is really to protect users and ensure that by placing an intent on C swap um and then also via Bing of different um transactions together that we can remove as much M as possible before it even hits any further step in the transaction management ah and a surprise um thinking about this I would say I I mean the hype that intense become maybe this year um I would say I mean we've been working on intense for a few years at this point and this year it really like sparked up and I didn't expect it to that extent hi my name is weha I am uh from to colaps right uh our position in the entire supply chain is that we are private market makers we are Searchers and we are also solvers right and I think the surprising thing this year is same as what Anna mentioned I think intense there start to be an explosion of Interest right so as private market makers we actually have been doing intense right for many years right it's just so happened that this year everybody started realizing that hey this is the next big thing so this is a surprise to me hi everyone I'm Kuby from Titan and we are one of the larger block builders on ethereum uh so we are in the middle of the supply chain and we also recently um released a relay so that's the second role that we are playing surprising thing um telegram BS don't what's surprising about it that they exist not not their volume yeah yeah so much volume going through them amazing well I guess I'm Tina um there is no alternative or Tina is not aligned depends on the time of the day um from flash Bots and um we are currently operating a builder and um a relay and we are also um operating in um orderflow auction protocol Asen experiment that said um I guess most importantly we're working on I guess trying to iterate the Meb game to the next level so um main job anyways so uh one surprise man this is a hard question I thought I had the crystal ball um one surprise that I think this year had been the like shedding over over intense because I think that it's um shockingly polarizing in terms of people's um view on it when it is in fact a um a technical um field that people have been working on for a while so kind of similar anyways so why don't we get um into some of the definitions first so we don't bike shed I actually just want to ask the audience first um how many of you are familiar with PBS please raise your hand great we can go really fast um so I'm going to ask call on each of you Anna can you define solvers in your own words and how they're different from Searchers yeah for sure um maybe to explain it first in simple words it's like um on cowall but also on Wen fusion and Unis swap X um user intents are collected and uh the Assumption essentially is that a lot of users uh are obviously not necessarily professionalized Traders or at least are also not super familiarized with transaction management or at least no professional parties in such and um we want to enable users to a simple ux where they can essentially just express what they would like to achieve and then have professional parties that we refer to as the servers take care of the transaction management for them because they're much better equipped in doing so because they know are the ins and outs around me how to protect transactions against the and so on and so servers look at incoming order flow and then they compete on this order flow and I think the disorder flow auction is depending on the application is slightly different on cup essentially what happens is that solvers submit a score uh you can Envision this as the score is higher the better the execution price for the user and here the interesting point is that a solver is winning this auction this order flow auction if they maximize the value for the user and if they do then they're selected and they can continue and do the transaction Management on trade and the big difference to Searchers that I would highlight here is that the orderflow auction within the application of cop essentially is maximizing for the value that's attributed to the user versus if you look at Searcher competition it's usually maximizing for the value that can be extracted because they compete on being um on their bundles being included in the block space that is an interesting angle um I was more thinking about from um I see solvers as protocol employees um uh just like our robotic friends um so like essentially they get kind of they operate under the constraints or confines of the um application do you agree with this view or um I I do agree with your take though in terms of like the differentiation between solvers and uh Searchers but I see it more as a uh the constraints that they're operating under yeah you mean basically the the rules of the competitions in a sense like what is the framework what are the rules that they need to follow in order yes to execute what quick question let's so would you say a Searcher that participates in an orderflow auction platform is a solver because no I mean it depends on the because you're maximizing the value that goes to the user or you're competing on how much you pay back to the user no I mean it's on a different level right like I'm just using this as a like the uh this servers are within basically the application framework so they are partially also not operating on chain they're basically collecting um intense and then converting them into unchain transactions thank you well um koopy so your turn there is a tweet yesterday um oh a builder is just a Searcher that grew up do you agree can you define block building and how that's different from searching um initially I would say I would I disagree um but then again we used to be a Searcher ourselves so a bit hypocritic I um I would Define a block Builder as a service provider um in a two-sided Market where on one side you are servicing the proposer so basically performing the difficult task of constructing blocks that are valuable and maximizing the um the block reward and on the other side um maximizing inclusion or the block space consumers service provider yes where a Searchers are where Searchers are maximizing value capture for themselves that's the sole goal fair fair um weo um your turn so do you see Trad five parallel to each of the three roles um Searcher solver Builders um and do you believe that all good market makers will end up running its own block Builders okay I don't think there's a Perfect Analogy to this right because of the uniqueness in dii which is why a lot of the transactions lot the value itself is variable right because of amm and other conditional transactions right but if I have to give a parallel I would say that you know uh solvers are like your OTC desk right you know they are very close to the users right they have a lot of flexibility in deciding what they want to do they can be principls they can be agents so the OTC desk right now Searchers are like your traditional market makers right they see the orders coming in right into a pool dark pool Bright Pool doesn't matter right it comes in then they decide whether to put in their own transactions you know to maximize their own you know um value their own profits right and then I would like to think that Builders as you said are service providers they are just the settlement layer you know finally when all the transactions needs to be settled right it goes over to them but because of the uniqueness of defi the powers given to each and every of these players is actually uh very different from traditional Finance because they can each affect the value by doing different things right so that would be my take right now if I were to look at it right would all market makers eventually become Builders was it the question um will they all run their own Builders I guess the incentive is there it's very strong right because uh having that vertical integration will give them a lot of Advantage right uh but I would think that I part of me would like to think that the competition itself would be very keen because people would see that there's a lot of value in doing this they'll attract a lot more players in you'll make it not economical for them to do that because if they have their own Builders and they only were to consume their own transactions they'll find that they are actually not maximizing their value by being at different parts of the supply chain very cool well this seg with right into um tun's question so do you uh do you see how these three roles May evolve like do you see them converging or become further Specialized or something else yeah so I mean maybe if we take a historical look um if we look at you know ethereum pre flash bots so 2018 2019 you just had tons of pgaas gas auctions lots of spamming of the network um and you had this almost completely disaggregated um system for ordering then you move to flash Bots bundles and now you have this sort of kind of a little bit annoying to to write down comori auction where I had you know different Searchers give you different bundles of different sizes so someone who has a sandwich attack gives me a bundle of three transactions but someone who's trying to win fomo 3D will give me a bundle that's the size of a block right and so there's sort of this this thing of I have to consider how to do this kind of Bin packing problem over all the different possible uh package sizes which are bundles and then you know we you know that auction was sort of centralized uh because it's kind of hard to to to run that type of auction in a decentralized manner so then you know we kind of moved into how do we have a single item auction and the single item auction is the block auction right of you know the proposer Auctioneer the auctioneer auctions to builders um and we had this in you if you look at the kind of flow over time we went from very very unaggregated to slightly more aggregated to completely aggregated right um um something my mom told me when I was five I remember vividly is the the following statement which is capitalism is just bundling on bundling repeated many many many times uh and so we've been in this bundle phase of the supply chain for the lifetime of ethereum of aggregating and I think most of what we're seeing right now especially in terms of people defining intents people um trying to Define how to split up block building so having different Lanes within a block so maybe the top of block would be auctioned separately from the rest of the block maybe there's a portion of the block that's guaranteed for private transactions only but they have to pay a higher fee higher base fee um that is the this is the disaggregation step so PBS was sort of the Pinnacle of over agregation and I kind of think we're seeing the the roles split up uh and we're entering the disaggregation phase of the supply chain so I think these roles will actually change dramatically when that happens nice and very wise mom um okay so with that I think uh that's uh concludes the definition that was quick good so next uh let's talk a little bit about our favorite topic intents so um while we're talking about this um I think is uh the first question why are we even talking about it um we uh there seems to be a return of the rfqs um but you know they've been around for such a long time and so why are we seeing this shift what's different now you want anyone's I can start you guys can add on to it um so basically why are we seeing this increasing attention attention to intense again I think it's was a little overhyped at the moment but I think it was also long overdue that more intent infrastructure would be being used because it's just essentially an significant ux Improvement that is needed um transaction management is difficult ux has significantly improved over the last how many years has it been now probably is 20 16 17 right yeah yeah yeah so been many years in the in the making and um at the beginning it was quite horrendous now I'm I'm not even sure how to judge it because the longer you've been in the space the more you get used to it so it feels like okay it's very easy at the moment but then when you do when you get new people into the space And You observe them while they're trying to navigate Simple transactions you do realize again oh it's actually we're still very far from where we need to be another to to get more people um engaged and adopt this new technology and I think intends a significant step to towards this future where we make it easier for new joiners to basically obstructor the complexities and um navigate through Defi and other parts of this technology and I think it's yeah it's uh essentially significantly improving it it's that giving users the possibility simply to say this is what I want to achieve and another parties take care of this I was going to ask way first sure so well I guess to me I think a large part of it is due to the you know good work that you know C protocol has done right you know you have brought it actually back into you know uh attention right but the the way I see it it's actually a reaction you know to um the novelty of uh def right you know from defi summer wearing off right so that people started looking at okay there were lots of inefficiencies earlier but because it was novel right they were all very happy obviously they all making a lot of money right so they actually didn't really care but over time as that you know the profits dropped you know as nov wears off right people's asking a question why am I accepting all these you know inefficiencies right all these things that you know we wouldn't have accept in a web two world when people started asking right why do I have to deal with you know N Management gas management what do with all of these different you know nuances over here that makes life so difficult and then you know people start realizing that actually it's possible right you know um it's just like you know in web tool space right where you pay with your Visa with pay your MasterCard you actually don't know what actually happens behind the scenes it gets routed to different payment processes goes to different banks you don't have to deal with all these complexities right then people start to realizing it's actually possible Right intense is a way of doing it that's why I think it's come back into Focus right now I was very I'm very surprised by uh uh that perspective um I was because you were one of the earliest RFQ market makers right um from way back so like are there actually more uh service provider like more market makers uh on rfqs these days are there more competition um are there more like what changed from the uh from the other side well if you ask me this question at the beginning of the year I would say that the RFQ market makers are at risk of dying out right A lot of them dropped off right but uh over time in the second half of the year I think we see a uptick there are more people coming back to the space right people are getting more excited they seeing that the opportunity is there so yeah this is what's happening fascinating sorry tun I um yeah I I I actually am willing to take a very different stance than everyone else here which is uh fundamentally the aggregation effect took place from ethereum's birth because in order to actually have a sustainable proof of stake Network you actually really have to have fees that go back to the proposed and in the earliest days of ethereum when it's proof of work proof of work is sort of a a nice way to not have to rely on St stable fee streams proof of stake however needs the fee streams to be dominating alternative yield in the same asset otherwise why would you stake why wouldn't you just go get the alternative yield and the kind of natural tendency to kind of aggregate yield to give to the proposers because the proposers have this sort of temporary Monopoly is a way to make proof of stake sustainable virtually none of the other per of stake chains are even close to sustainable if you compare fees and subsidies versus transaction volume and so the aggregation effect in some ways is almost a necessary condition to bootstrap the the shift from P to posos and I think now everyone realizes well we just made the proposers really rich and the users kind of didn't have a say in it and we're sort of having the disaggregation revolution for that reason uh but proof of stake I I don't think that the merge would have been a successful transition had there not been um all of these extra real yield for proposers and I think now it's like you know it's like collective bargaining like the users are becoming a union and you know fighting back for some of it interesting so your view is actually the what's driving this um uh uh you know intense uh wave or RFQ offchain execution has more to do with the uh turn of returning um me to the users when users now unionize and they choose these offchain venues yeah basically interesting Koy yeah nothing to add I I would say ux nothing to ux telegram Bots I remember your first answer um well actually I realized we kind of Dive Right into intense discussion um without actually answering a very important Twitter question um are intents just limit orders this is a question directed to tun so I think the definition of an intent that's sufficiently General to not have the mob come after you on the internet is any verifiable by all parties and multi-party verifiable conditional computation so what that means is I have conditional comput ations like if Pepe goes greater than $10 on eth Main net sell Pepe on arbitrum those that involves verification on the main chain eth that involves verification on arbitrum by the sequencer that involves verification by the Fulfillment agent so the person who's holding the risk of fulfilling that order on both chains and that also finally requires verification from the user so the limit order is the case where that's obvious I said a certain price in quantity did I get it or not it's an indicator function it's a very clean clear uh notion of verifiability uh I think you know obviously people have jumped to the farthest end of abstraction on verifiability you know I think anoma is sort of the farthest end of like every possible thing wanting every possible thing to be verifiable um unfortunately I don't think you will actually be able to get that uh but fundamentally multi-party verifiable conditional computation is really what an intent should be defined as um the simplest verifiable things are limit orders but there could be more I'm just not sure we know of anything that yeah people want I'm going to ask every single one of you to name a non-f financialized uh intent um KU youo first non non financialized hey on ethereum okay I can relax that constraint it's okay to be on salana can be off the blockchain right um sure that can be a stretch but buy me Taylor Swift tickets are you kid is this like a common comparison because I was literally on a panel like 30 minutes ago and that was exactly in the use case they came up first is this like a popular take but isn't that Financial still it is like I don't [Laughter] I think everything I can come up with is going to be class financialized so so so I I I think one that is financialized but neutral versus financialized but taking a position so like a swap is like I'm going long an asset short other asset right I'm taking a a position uh is these types of crazy ZK onboarding things like if you were at PR crypto and you saw ZK P2P which basically lets you have attestations on things like venmo that get fulfilled by someone on chain who then gives you stable coins or eth that onboarding fulfillment type of thing is like not quite it's still Financial like it involves dollars or E on both sides but it's neutral it's like it's not trying to like belong one asset and for another one example I can think of right probably would be you know conditional voting conditional voting right so for example saying that okay uh you know cast my vote for this proposal on this particular down only if let's say another proposal passes on another down right that could be a kind of intent good one all right well then okay so now we have uh seems to be this uh um Rising Trend uh whether it's Market chosen or it's some other forces um intense um becoming popular so how would that um will do do we think that it will have an olly like block building um and if so will it be worse I think yeah it's probably going to be like block building um because I think in the long term the most sophisticated solvers at least for the use cases related to trading are going to be market makers and so it's going to follow a similar trajectory I think yes I think likely going to happen right uh unless obviously uh there are more players entering the space and further increasing the competition otherwise we are going to end up with all the god pleas situation yeah I think there will be specializations um you know for like specific order types that don't happen so frequently so it's maybe not necessarily worth well for our solice to like um be able to support these use cases so they will be in that sense like not an oligarchy of like maybe just five players but I think it will be a little bit more decentralized but yeah for sure like usually to remain competitive in the game you need this more and more higher degree of sophistication and potentially private liquidity that there's definitely going to be some sort of um reduced market share there are thoughts about potentially um creating also Baseline solers by basically providing simply a continuity reward that incentivizes the community to also participate in the competition just to ensure that if for whatever reason there is some sort of um Monopoly of s try to um to how do you say like to to to merge their approaches to have like still like some sort of security backup to this but of course there won't be nearly as profitable as the more sophisticated ones yeah I think there is allop although I think permissionless oligopoly can sometimes look quite different and um again I I think the the PBS is more strongly oligopolistic you're really only B bidding on one thing repeatedly versus having to do many fulfillments and the there's kind of this trade-off of like how specialized can you be at One Thing versus like having to do many things and if if there's enough disaggregation at the intent side in terms of like different types of markets then there's some point at which it should be less at least on a relative basis less oligopolistic than PBS but it's still obviously will concentrate relative to Market size so this now brings a interesting point remember what we were asking earlier that uh will market makers uh inevitably all run their own block Builders and it seems to be generally yes uh it makes sense so we have uh you know uh an oligopolistic Market um with the solvers and they also run Builders so what would what would that entailed for the block building market like who is you know are the lines just being blurred or our solvers is eating Builder's cake or even more uh you know uh a hot take will um Builders just end up doing the routine sexx AR and some transfer um and call it a day whereas most of the work is being done by those um solvers so um I think both um larger other block Builders so our main competitors are large market makers some more than others um and both of them are quite competitive already when it comes to solving um and more of the solutions are going to be provided from um internal liquidity versus uh onchain routing is mostly going to be the long CH uh longtail assets and so there's already like a very Sy synergetic setup uh as is so and in general the more the Builder Market matures the higher the barrier to entry and so you have these entities that are competitive at building blocks that are already market makers and already solvers so I think that is more the problem and then the last thing is um even when it comes to oning routing the the block Builder is basically the only actor that really has access to State at the time of execution so they are theoretically only the the only party that can actually provide a optimal solution so then you have another Edge layer on top of that so yeah it it doesn't look great that being said um this is kind of where we come in and we are trying to be the the the credible neutral builder in the space and try and uh Comm commoditize some of the edge that the vertically integrated parties have and so we have we are doing that for currently for the CX Dex trade and we have some ideas how we could potentially do that for solving as well but it is it is harder to do because we need to interface with an external party but on the other hand we have um we have the edge of have being able to focus on the infrastructure and not having to do too many things at once and be more specialized so I I think that's the dynamic yeah well um I mean the one reason why right you know private liquidity is going to be more efficient than you know amm is simply because uh amm have to do a calculation plus a trans while private liquidity just do a transfer right so it's going to be more gas efficient anyway right so I think there's a natural advantage over there right but uh well private liquidity can do everything and amm does because they don't have inventory over all the gazillion assets that exist right so I think there's a space for amm right it's just that yes uh private market makers will have a significant Advantage right and yeah I mean they're just making good use of this Advantage when they actually go on to building and then when they do solving yep they can tap on this Advantage so it's definitely something to happen I mean that this is the case you can already see on uh uniso X today right like if you try like basically the UI switches between the uniso X and normal amm mode depending on what orders you type in and then on like the top 10 whatever trading pairs it would show you it would usually route it by Unis swop X because there's Market maker Integrations and then for all the long T tokens it would still take you to the amm so that's 100% the case um maybe to the other point I think there can also be really user benefits for if there's more collaboration between solvers and blog Builders because essentially um You by the more access you have or the more control you have over the overall supply chain um you can also take that at the advantage of like for your own within the sare ecosystem within that auction right because if you for examp example um execute a transaction and you know okay you can you have now access to all transactions that are in the public M Pool you could place transactions that are improving the prices for your users before um the transactions that as part of the solware competition you need to execute um you would already know well potentially actually that's the point that we talked about earlier you wouldn't know what the sex Dex price point would be at this stage you would only know at the very end of the block but still at that point you would know and um you would know what types of like other transaction you you could also use to background transactions and you could use all this to improve your price point that you're providing in the solware competition so it could be used at advantage to the users of that application um I think the maybe maybe the layer one portion is kind of moot that we've kind of already reached that's sort of where where the aggregation will be in terms of users but I think Layer Two is going to be kind of interesting if there are really decentralized sequencers or shared sequencers shared sequencers sort of offer you kind of crappy guarantees in some ways right they can only really offer Atomic inclusion uh I think like with some extra effort you can get you know partial sequences I'm looking at Josh right as I'm saying this because I want to see his face in horror as I say that but I think you can only really guarantee Atomic inclusion um the real question I think becomes uh you know will solvers who have to do you know if it's if a world exists where there's more multi-domain uh value to to capture versus the centralized decentralized Arbitrage now now I'm not saying we're anywhere near that world but if that were true I actually could see there being um less integration over time um due to having to interface with all these different components that uh you know you you have a carrying cost too and you'll specialize into some subsets perfect segue to the question for uh on specialization so it seems that um well the Builder or the market maker can do it all but clearly I guess statistically um I think uh Market makers have been very selective uh in terms of what venues they trade on and also like what kind of risk they take um and so there is um so what is uh what is there left for specialization or under what condition would um essentially specialized solvers or Searchers um Thrive well for example today right when we look at solvers right they actually mainly swap solvers right because they're actually looking at just swaps right so that is already a specialization right so potentially if let's say we get to more generalized kinds of intents right there could be for example specialist that comes for let's say lending protocols right you know the swap solvers probably won't be able to do that right but even within the swap solving space right you know different solvers would have for example different kinds of specialization because some would for example be very good at let's say stable coins some would be very good at let's say Majors some would for example choose to you know do certain you know longtail assets because they for example have let's say access to an inventory right or they could you know price the risk better so I I think there is room for specialization I think if we have any of these rollups that do non-financial is yeah I added the the weasel weasel post fix there uh you know like gaming kind of things like if there really are gaming rups they get usage uh I think you will see a lot of weird me that you don't expect just like wow gold Hunters you're just going to find the equivalent of WOW gold on chain and I that will be something that I bet you the swappers will not be able to do so um both of those conditions entails essentially more applications um um actually like and more activities um that is not what we see today so um maybe one question here is that if we assume the current like uh you know uh the market condition doesn't not change but rather is there a world where is setad a a builder or a market maker does it all um there can be a world where specialized actors um can bend together can coordinate um and out compete the the um integrated um Builder solver Builder Searchers I think it's going to it's going to resemble more um the current landscape of um if you think about the different M strategies and like we talk about these other apps and ecosystems but the majority of let's say the volumes or the value is going to be in a specific market and So currently that's the Unis swap binance trade which is 60% of all M or CX Dex are and so even though there might be more longtail opportunities in a diverse set of solers that are very specialized you still have the big boys sort of you know capturing the majority of the pie yeah yeah I agree I think it's actually more uh strategy driven right so it says how often do these opportunities arises right so if you're just looking at let's say sexx Arbitrage right it happens well fairly often right versus someone who actually specializes in something let's say liquidation versus someone that let's say specialize you know in in in something that's more you know long till some stress you know kind of let's say Bridge hacks and such right you know they will have Advantage because the you know the big boys also as speak right they are not focusing on this right they have no specialization in this they have no idea what to do right but the opportunties just don't arise so that's why you see more and more of oh yeah it's just the same bunch of people people that's why they are seeing more opportunities because it just happening often I think we're almost at time well we have two minutes what I'm going to ask is I'm going to ask uh um the panelist to ask pick someone on this panel and ask them a spicy question that's a copout exactly making us work taking a break do you think um solvers that are currently not major market makers will will have a chance at all to exist yes I think they will have an opportunity right because I think currently if we actually dive deep into what the solvers are doing right it's actually very simplistic there's a lot of different uh things that can actually be done right you know they actually not just put it they are just stretching the surface there's a lot more uh you know liquid resources a lot more you know uh ways of you know generating extra you know uh value out of the system that I think hasn't been touched so I I think we are still uh having quite a lot of room for them to survive my question for Anna right okay what do you think about you know the competition from 1 in and Unis swap that's good uh it's interesting um I would say to some extent it makes Now unisa volume more comparable um we usually compare car volume with Dex aggregated volume so we compared it with one 1 in Paras match because uh the majority of volume on an amm is usually Arbitrage volume um and it's like estimated at 80% so that puts Unis swap obviously right on top of all existing Dax volume and now if you only compare in 10 based trading volume the their market share shrink significantly in comparison so in a way it's like um improving the competition game and uh in general I think it's good also to see that they adopt intent based system I see it also as a um is showcasing our signal that we are moving in the right direction and again like I think it's a it's a needed ux change in order to bring all of us to the next level and ultimately we shouldn't be competing on the little market share that we have today but we should think about the market share where we want to be in 10 years from now Anna your turn thinking about the spicy question does anyone from the audience have one wow we're really just Outsourcing labor no no this is this is me if Unis swap um or metamask or anyone else who's like moving into an intent based system um was considering to strike a private deal with a block build of like hey I will only collaborate together with you um not titon speci any block Builder um in order to uh basically if you have high market share as if if you would consider all Unis SW transactions from today would be given get going into uh would be only forwarded to a single block Builder uh would obviously significantly boost their market share and their probability to be the one uh being responsible for the next block um um do you think block Builders would be like trying to strike such a deal or you think block Builders are like um in itself ethical enough to believe that the competition is needed in order to not try to pursue such a deal yeah I I I think uh there probably not a block Builder who would not try to make the deal like if Unis swap came knocking and said hey I want to give you all our order flow I probably won't say no uh I think it's more a question of um the ux basically so if you even had 70% of block space there's still like 30% of the blocks that you want to get into so you kind of don't want to send it exclusively there are some things you can do around exclusivity for specific slots given certain confidence levels and things but yeah I yeah I would like I wouldn't say no and the question is like ethical from what perspective well if I would say I mean at least you're honest U ethic probably from the perspective of um Des the system isn't when is the system failing at one at what degree of centralization that leads us to tun who will have their margins compressed to zero first Builders or intent solvers anyone can I would say intent solvers actually um I think um even currently like looking at most of the dashboards uh in 10 Sol this are net negative um I don't know about you guys PCI specifically but there a lot of subsidies and stuff going on already um I think from a perspective of just generating revenue from a share of the block reward that is certainly going to Trend to zero but there's a lot more a builder can uh generate revenue from yeah I have to agree right I think it's more in terms of uh the M that they build up right so it's easier to become a solver than to become a block builder at least a competitive one in this space right so so I do know right I do hear about a lot of the solvers right they are taking a loss right uh well luckily for us we have always been profitable so I think we are fine amazing well I get to ask one last question but this is also to actually anyone Builders um at this stage still competes um pretty much solely on um order flow so do you see that at some point down the road the tech matters like merging algorithm Etc or never I think it already does um and it does in the cases where you have autoflow parity uh so their latency and merging matters a lot and the give you have like three large or four larger block Builders and that is just based on water flow but amongst them it's um in a lot of cases um actually related to latency and merging um but order flow is still King uh so you know it is no black or whites that makes sense well one year down the road MH rank top three things that matters to win for black rer order [Laughter] flow um latency merging stay the same what about five years down the road probably the same I don't I don't think there's much going to change structurally in terms of autof flow and if you can't guarantee priority then that's always going to M matter well we can work on that great thank you so much for a great panel and everyone for staying over a little bit over time and will be I think happy hour or something I'll close don't worry thank you [Applause] perect