MEV: The Economic Driver of Blockchains

Transcript

before we jump into this episode quick reminder that everything set on bell curve is a meme and nothing set on bell curve is financial advice enjoy the app [Music] all right all right I'm gonna be talking about Mev Mev miles want to give us a little synopsis of uh today's episode yeah absolutely so very hot topic um along with the general app chain thesis is is Mev um but I think the two the two concepts are very interconnected because something we've teased a little bit um on other episodes but you know the ability to capture more value from Mev or the ability for applications to

recapture some of the Mev that they generate is one of the big Tailwinds of this app chain thesis um and so yeah today we'll be speaking to to some folks across many different parts of the Mev value chain who operate in general purpose blockchains as well as app chains um to get their perspective on on where things are heading and now this how Mev plays into the app chain thesis yeah absolutely um all right well without any further Ado let's just let's just hop right into it all right everyone welcome back to another episode of bell curve uh I'm joined as always by my co-host miles and we've got natash Magnuson Xavier on to join us to talk about Mev welcome guys

great to be here excited um so we've got a round robin uh folks on the call today um and we're going to be going really deep in the subject of Mev in the cosmos ecosystem um maybe just uh just to kick it off uh could each of you do a super quick round of we can just do a super quick round of intros if you could just give 30 seconds on who you are and kind of where you are in the Mev space and maybe I'll just call it Natasha on you since you're first on my screen then we can go clockwise perfect yeah so um I'm the founder of D flow it is a protocol for order flow markets so basically we're building infrastructure that allows order flow from various sources to trade at sort of a supply demand market price um and then also build some

infrastructure around things like best execution and uh reputation tracking of market makers excellent Magnus you want to go next sure uh Magnus I'm the co-founder of uh skip protocol uh skip. money uh we build Mev solutions for Cosmos app chains um our thesis is around Sovereign movs so basically providing app chain developers and validators the tools to construct uh their own Mev markets and customize them to sort of best serve their application so I'm the chief investment officer at chorus one so I've been now A Chorus for two years and of course one is one of the largest global note operators in the

world uh I.E a validator so yeah we do a lot of work looking into Mev how it impacts the node operating space and chorus one is quite often at the Forefront of some of these topics in staking so we were quite early with liquid staking a few years ago and uh yeah we actually released a sort of white paper about Salama Mev last week so we're always sort of in the weeds and what impacts validators so yeah glad to represent the node operating side and the podcast today awesome awesome so um just to kind of kick it off here I think to frame this this episode for listeners I think we really want to be getting into sort of the Nitty Gritty differences of the approaches to Mev in

the app space sort of environment as opposed to the generalized layer one sort of environment but before we do that with the subject as complicated as Mev and a little esoteric if you're not kind of directly involved with yourself I think it's good uh to just sort of set the scene for viewers like what is sort of like the the Mev value chain so to speak or who are all the major players or actors that are involved in in Mev and Xavier I know you've done a lot of work on this maybe I could kind of kick it over to you can you just kind of explain who are sort of the key stakeholders right when we talk about Med yeah I'll do my best Meg and Natasha and chip in here if I've got any parts wrong because it's not as religious for me as it once was but so in terms of the

chain itself so about that like corresponding would be at the sort of very end so a validator is the one proposing the block at the end and everything that happens sort of before the block and in terms of like ordering of transactions and everything that goes in the block itself uh in the in the recent past has been sort of modularized in a sense so in ethereum right now they tried sort of separating the difference between the proposer like asset chorus one versus like the Builder the one who's concerning the transactions and doing the block to be proposed to course or to a validator um but it goes you know even before that you sort of have a Searcher sorry I'm working backwards because I'm a validator so it might be a bit confusing for people but yeah so before the Builder you have searches so quite often

these are sort of people looking for Arbitrage opportunities and then before that you might have something like a broker for example which in this case in sort of the decentralized finance world might be like a wallet and then before the wallet you have a sort of user creating the transactions and what's sort of interesting which we might touch upon in this podcast is you know Mev is a topic and the value chain of Mev is quite well studied in ethereum the last two or three years and it's the value at stake in ethereum and the value that sort of impacts users is quite great of course one sort of released some research that as much as like 60 of Mev and ethereum right now is from front running whereas in Cosmos for example the users aren't so impacted at the

moment because volumes are a bit smaller but uh yeah I'm sure mag and Natasha can touch a bit more on sort of their specific role and the different types of actors in in that Mev value chain and the cosmos [Music] yeah I can I can jump out so I thought that was pretty spot on um maybe The Only Thing Worth adding there is just that a lot of this does depend on like the ecosystem that you're a part of so you know the value chain is slightly different on ethereum versus Cosmos versus Solana um and you know new ecosystems are popping up too so uh but high level I think that's you know spot on [Music] yeah one thing I would add to

um though I thought that was that was pretty accurate is that like value chains are seem to be very correlated to power dynamics so like ultimately for example on ethereum um proposers or validators have sort of like the ultimate power over the ordering of blocks and so necessarily they're paid all the money right um whereas like in other ecosystems there does seem to be a difference in power dynamics right so when you when you come to an app train world I think because like the app chain is really built for a specific application the dev team right takes a much more active role and the community takes a much more active role sort of in the direction of the chain and therefore that's like sort of disrupted the supply chain in some interesting ways that we've been

experiencing and being you know has affected our product roadmap significantly um so like for example in Cosmos you've seen for example delegators um choosing or validators choosing to give more of their money back to delegators versus in a hearing more it's a no-brainer you just you just take it all right so there's a big difference there maybe there's one thing dad from anyone sort of listening from this from an ethereum perspective so one major difference in Cosmos especially in app chains and Cosmos versus ethereum is that you know in Cosmos it's delegated proof of stake so at the end of the day a lot of these delegators are users themselves and they'll sort of vote with their stake in a sense so it's a much more complex Topic in the cosmos

um compared to ethereum where it's there's not so much delegation so even if you think of Lido you put one ethereum in and you get State East but this is going distributed evenly to all the power that is behind that and you don't really get to choose who your sort of bowel better is so it's less important and the cosmos as a validated like chorus one mev's a huge topic for us because it can actually be a way to differentiate and try to capture more stake in a network and earn you know voting power from delegators so you know what mag just mentioned before what's really interesting and in fact now chorus Works quite often with skip through some of these problems is how different is this in the delegated for mistake system versus like a pure proof mistake system like ethereum where there's a cab and there's not so much

delegation and the state doesn't move as much it's not as free so yeah that's a quite a big difference in that ecosystem as well that's great and I think you know we're talking about Mev here because it can be an application's best friend or worst enemy depending on how it's treated right um think about the issues that the economics of Mev can cause us around you know potential risks of network centralization with a few validators really owning all the Mev and getting all of the networks stake um and then there's the user side in terms of you know users that are getting front run by Searchers or sandwiched right so we have you know this public mempool which allows us to happen

um and you know the two of those things can can really Sky me the entire space I think so that's why all these these solutions that you know folks like yourselves are building have have popped up and and are really interesting um and so maybe just to talk about what like the perfect world solution would be um I think about it covering three main objectives that it's solving for um it is protecting you know mitigating the risk of net Network centralization number one um it would protect users from you know harmful Mev like front running or sandwiching and third it would actually allow the applications that generate the Mev to capture that value

um and determine you know internally how that value should be distributed and so maybe to get into some of the solutions that are being built right now and thinking about you know which of those three uh or all three they solve for um no I kind of bucket it into off-chain Solutions uh on-chain Solutions and and Order flow Solutions um so maybe I'll kick it to you Mac could you speak to you know kind of some of the off chain and on-chain solutions that you're working on at skip yeah sure uh so first of all I sort of agree with with those three framing right so like reducing centralization obviously important in decentralized networks

um and then also sort of giving chains the ability to express and then I would say like as part of that giving change the ability to differentiate by saying okay this is a domain where you don't have front running right so like osmosis founded or was founded sort of as the Mev resistant decks so to give them a solution that gets that makes all the users front run is probably not going to be in the best interest of them or their users and would be fought against [Music] um so like we we support two solutions right so one is um an off-chain Builder we call this skip select uh it's basically an implementations of a flashbots like Builder that has a lot more preference expression so meaning

the chain can choose sort of you know minimums and maximums in terms of payment the chain can choose to disable front running and sandwiching by sort of we do that on the Ingress Side by looking at uh transaction signing uh and then chains can also sort of you know basically choose you know how much of the validator set can adopts it and you know validators can sort of make their own decisions about these things the other side of what we've been building um which we so far have only launched on osmosis is a fully on-chain Builder um so this actually relates to a lot of the research that's been done by course one in terms of having sort of like side-by-side builders that sit next to the validator that basically build its own block I just lost uh a light okay

back um so basically what that means is uh you know that the protocol itself will try attempt to capture Mev this is basically like Impossible on ethereum right because you have like generalized transactions it's like there's so many forms of Mev whereas on a place like osmosis like what do you have swaps right like that's the only thing you really have and swaps are very back runnable right like they're very it's very easy to understand how to background them and so when we observe that all the Mev on osmosis was essentially back running we're like why can't we just build this like into the chain um I think an interesting design space for us at Skip and where we're moving next is like how does that generalize like how do you have the properties of

an in protocol Builder or an unchain Builder that generalizes to something like auctions um so you can like allow for a search or ecosystem to also participate that's that's great and that would actually you know the benefits of that is that it leaves the control with the application itself right um and the application can be ultra opinionated on what type of Mev it allows and and how that you know value is distributed um maybe I'll kick it to natash you you're taking a slightly different angle at this with building I would say more of an order flow like solution right um and your your primary customers are not chains like like mag uh but actually wallets and and other sort of you know

applications that basically any app that is not its own execution environment like a like a text um so I'd love to just kind of hear your angle at solving for Mev and and which of the you know dream solution you know one through three it is solving for yeah um I think you know it's it's difficult to look at this problem and and try to come up with Solutions without uh I guess trying to identify um who the users are that are generating this Mev at the end of the day um generally This falls into two buckets retail and and non-retail so institutional and that could sort of include you know prop trading firms or just anyone who's

um sort of you know a sophisticated model building type uh you know swapper or uh transaction sender um and so that that's sort of where we've started that's that's one thing we've we've looked at uh when we've you know began designing how we approach this problem is identifying who the users are and then um with that backdrop we have uh really positioned ourselves um sort of at the application layer rather than the validator layer so I think a lot of Mev type applications uh operate really close to the validator layer and we we like to insert ourselves as close to the um entities that are actually generating

these transactions and so that ends up being wallets um with that backdrop basically of one being able to identify who is sending the transaction and two uh it like operating close to the application layer uh we we get some you know really I think attractive benefits in terms of the design space one of those is that we're able to now provide really strong guarantees about Mev to the users who are generating that those transactions that cause mov and in particular with retail Traders which is sort of where we specialize we want to build this infrastructure to be best for retail um we're able to do stuff like give them a hundred percent of atomic Mev uh for

for swapping immediately without it ever without the transaction ever touching a validator um and this is sort of a really powerful Paradigm that I think flips flips a lot of this you know uh it flips a lot of the question upside down um and then sort of that second segment which is sort of institutional prop you know style uh Mev that gets generated which it does get generated if you're if you're you know a quad trading firm and you're trading on a decentralized exchange you're still generating muv but we don't care about that because and when I say we I mean the D flow protocol because um you're sophisticated you you know how to set up validator nodes yourself you know how to send transactions in a way which protects your interest it's just

that for that other group retail uh We've basically designed our systems to protect them as much as possible oh that's that's really interesting and you know I think you are you're definitely coming in it from the the the wallet level which makes the solution a little bit more generalizable you know to any sort of blockchain architecture um but we've seen you know when you're building up the validator level um or other parts of the chain you know the architecture of of the chain really does make a difference in terms of what you can do um and so maybe I'll kick it to to save you know on on Solana that this is a general purpose chain um and typically the only solutions we

see on the general purpose chains have been uh you know off chain uh PBS Solutions right um but I thought it was a really interesting piece that the course one published I think it was last week about potentially moving some of this you know to the actual validators themselves on Solano um so I'd love if you could kind of give a perspective there about you know maybe this doesn't maybe these general purpose teams are not limited just just you know off-chain Solutions or something like a wallet-based solution like like d flow and yeah I would love to love to hear a little bit more about that thanks for bringing it up so yeah at chorus we were sort of researching this for the last 12 months and also maybe

for the listeners who might not be familiar so Solana is a chain where of course there's very very low fees um and then as a result uh especially arbitrages really spam the network when there might be an opportunity available for them and there's at the time you know flashbots would Judo created a sort of version of flashbots in Solana but the way Mev Works in general is so in ethereum it's quite even you have sort of validators now and searches or Builders and it's quiet 50 50. so maybe the value that it takes half of the revenue that the Searcher would take if there's an opportunity there in Solana it's really the other way around where it's more that a Searcher will take almost 100 of an Arbitrage opportunity

because there's very little competition there's something like less than a thousand volts or something like that and also a very low fee so a validator and Solana will capture very minimal and at the time Solana had an ecosystem of quite a lot of volume which is also something that we might be able to talk about today like the volume and Cosmos so the volume in Solana was there but validators were taking nothing thing and I guess at the time you know we were quite big in Solana I think we would we were the number one node maybe or something like that and we sort of assumed that Mev would become quite a Hot Topic in Solana and uh yeah we decided to research this and sort of build uh it's sort of like what skip has where it's like in sort of network Mev for osmosis but we just decided to do

this actually in the validator itself so essentially when it's our time to propose a block we have these algorithms already sort of pre-created and generated and it'll search for opportunities just really simple Arbitrage opportunities on you know radium or some of their Solana decentralized exchanges I think a really important Point here though is that so we actually I can Source this and we publish the white paper and you can feel free to check it out on the correspond website it's called breaking Bots um but an interesting point here is that you know in terms of an app Chain versus like Solana which is general purpose so one thing we talk a lot about at chorus is sort of searching and and building sort of a bit as well especially when it comes to app specific chains so in our

specific change you have a lot less opportunities as in everyone sort of knows it's just Arbitrage or something that exists on osmosis for example so in a sense the actual searching is a bit of a commodity because even like you know someone at chorus is a research you could do it themselves so if you're looking at Solana or ethereum as general purpose which means there's a lot more long tail mov and by this I mean you know it's not just Arbitrage transactions and you're trying to capture extra value by background or front running there might be some nft men you know going live on someday which is huge and then you sort of want to get your transactions in first or mint all of them before everyone else and then resell on the secondary Market or something like that so these are quite clever long tail strategies that exist

on ethereum and Solana whereas versus an app chain like osmosis which skip is on right now uh you know it's it's quite well known the types of strategies that you can use to capture the Mev so I think it's a really interesting topic I guess when it comes to sort of Mev and where the value lies so we heard from a test that you know maybe is at the very start between like users and Brokers for example which I think is actually super interesting in general and then you have someone like skip which is a slightly uh more well-known model which is sort of but also new in a sense because they're sort of capturing it for osmosis itself so osmosis the you know Cosmos decentralized app chain specific came out and said that they don't really want bad types of Mev and

they don't really want searches they didn't want malicious Mev in the network which is quite a bold statement for a network to take and so they partnered up with Skip and Skip some an amazing job sort of neutralizing the Mev almost where it just captures the most obvious Mev and distributes this back to stakers and then validators and Skip have a sort of cool option where they can choose the commission on Mev being taken by Skip sort of and this is a way someone like chorus can differentiate but you know sometimes we might run into sort of questions open questions that we have as a nerd operator because if you think it's a validator we're at the very very end of the sort of value chain mov and so for us you could argue that validators like searching in some networks as well could also be a

commodity so a question that we always have any validator will have this question is how do we do differentiate it and clearly I want to differentiate can be through Mev why because if you distribute back more mub to your delegators you're giving them more yield you're earning more delegations you're getting more stake there can be Arguments for like is is this a centralizing factor so I am personally on the fence of like I think it's really hard to neutralize the sort of centralizing Force at any stage of the supply chain so if you take it away from validators it's going to move one down if you take away from Builders it's going to move on down if you take it away from Brokers it's going to move on down so also recently I've seen quite a lot of solutions that are focused on actually returning Mev back to the users

not just sort of so they pay no fees like maybe a d flow or some sort of payment for order Floyd protocol might have but actually probably like they will earn revenues from like sort of creating Mev so I guess what I'm saying here is that I personally think this whole Space is really early you know funnily enough you might argue that like you know someone like d flows someone like skip someone like chorus is sort of you know push and pull with who's capturing the Mev and where it's going and who can do what and yeah it's it's super early but I think it's one thing about Cosmos as well like just in general and you know you have a cosmos Maxine the podcast so I hope non-mind's a virtual Cosmos a bit but you know one nice thing about Cosmos is that

infrastructure it's really easy to build infrastructure from the cosmos which like fundamentally Mev can be a security issue on any network so I think you're seeing a lot more experimentation and Innovation at this sort of infrastructure layer um in Mev and Cosmos so what you're not seeing is volume volume is really poor compared to ethereum and now we're also talking at chorus about sort of you know there's a lot of building going on in Mev and Cosmos but now it's about sort of how can we get the volume there because you sort of in a sense you have to attract some of these searches to your network and right now a lot of them is in ethereum so a question I'm sure Skip and also you know D flow and maybe some others who are sort of asking like where does this value come from and how

can we sort of bring them in so you know I might just quickly mention as well there's another Meb sort of protocol when the cosmos not represented in this college in this podcast called uh Fair Vlog which is sort of focused on you know mitigating front running in the cosmos so then the question is like I'm a big personal fan of this solution for this problem but then the question is is this the actual problem in the in the cosmos right now because there's a lot more flexibility because skip might already be removing a lot of the front running so but then you have something like dydx which you also haven't talked about yet which maybe we might talk about in a second so dydx is a network of course a lot of volume on ethereum this could completely change the game and the cosmos then the question there

is who's going to capture that Mev and you know the order books that are in the validator so is it the validator that has the control here or is you know does someone like skip come in and uh you know maybe they offer to capture it sort of in protocol and distributed back to stakes in dydx so it's really unique how every network will take a different approach and super interesting as well because you're sort of always on your toes about not knowing what to do next and that's why we're frequently in conversations with each other about this topic that's great um yeah I could not agree more at those points I mean I think just how it app chains are interesting because the design space is for Mev is somewhat limited right to the application itself um we don't have thousands of apps

creating like this huge long tail of opportunities and so you know you can internalize some of the Mev that that is generated and maybe mag quickly just to would love for you to share a little bit about what the the proto-rev module is on osmosis um that will be going live soon and and to me I think the big difference is that you know you are not at the mercy of the validators trying to differentiate um you know through through kicking this back to their to their delegators this revenue is actually going to be controlled by the Dow who can you know let it accumulate in their treasury they can distribute some to stakers um and to me I feel like this is a this is a big differentiation because it makes this more of a you know a viable

business this model for apps and so maybe before going any further would you just give a quick overview of what the proto-ref does um and and how it will benefit osmosis sure but also I am in favor of validator differentiation um and so uh Proto wasn't designed to remove that by any means so yeah so so basically what protarev is as an in protocol back Runner right so similar to what so he was talking about like it it because all you have on osmosis are swaps it has the ability in protocol to uh detect sort of like what the price changes by any individual swap would be and then to cyclically back run it right

afterwards so we launched this thing called satellite.skip.money where you can sort of like see over time also shout out course one has a great Mev bot on osmosis you should follow but it basically records the amount of cyclical Arbitrage both do um and we realized it was quite substantial like seven million dollars you know since osmosis started um and so basically what it does is it you know monitors every single transaction and then at the execution stage right when it's going through the actual processing of the state transitions it looks and says oh wow like this would move these Pool prices this way let's Arbitrage it back right and it does that per every single transaction so um it doesn't insert it as new transactions it actually happens in the

processing of a single transaction so every single Swap you ever do in osmosis going forward will have like perfectly balanced Pools by the end of it so there won't be Arbitrage opportunity afterwards or maybe there's a little bit because we'll miss it but for the most part there won't be um I think it's it's worth noting though that uh this solution doesn't um this solution is is somewhat dependent upon already existing inefficiencies right so like when you think about the routing of Osmosis right when you think about sort of how like why there even is a back running opportunity you could argue and this is sort of in favor of a d flow-like solution that there should be matching or there should be like built-in back

running at the OR at the you know front end level so that it wouldn't make it to like the actual um so like the execution wouldn't create it sort of when it got to the execution stage I think that's eventually sort of like where we want to start collaborating and thinking about going forward because so far what we've really focused on is okay Searchers were getting all this Revenue it wasn't even a question of validators getting it or you know um users getting it Searchers are getting all of it how can we distribute it back to the network since like the network really can do it themselves and then how do we make that like a governance driven process where of course validators have it say so I expect that this will receive some compensation for it but then also you know users and individual delegators

um but I think like a big a big Focus going forward uh is really starting to think about the order Flow side of Mev in Cosmos and what that looks like I want to maybe just kind of bookend this part of the conversation before we before we move on but uh and give listeners like one sort of takeaway right for a difference in between like an approach to Mev in like kind of from the eth framework or mindset and then from kind of like the full stack app chain mindset and Magnus I kind of heard you say something that stuck out to me at the beginning of this episode which is you know wherever value sort of occurs and that Mev value chain on ethereum land it's kind of based on Market forces right um and then there will be some kind of Leverage and

push-pull between that different value chain but the one of the differences in like a full stack uh app chain sort of environment is you it's a little bit to borrow a phrase actually for miles is you can have more of an opinion on it right and there's kind of a more holistic approach for how do we kind of give some of this Mev back do we have an opinion on what sorts of Mev are good Mev and what sorts of Mev are bad is that like a relatively okay generalization to make I think so yeah I think it's like a big part of it is also what Dave said right where it's like validators are first of all public and a lot more beholden to their delegators and Cosmos right there's only 150 slots so if you up right or you do something that like is

evil then you're gonna lose your delegation and it's gonna go fast so I think there's generally a sense of like okay um you know the chain Founders are saying this is good for the protocol like why would I go against that right um and and sort of like users are saying they want this like why would I go against that um I think there's like there's definitely more voices at the table versus how I think it is in ethereum where there's I think it's just a different set of power dynamics so then I I guess I'd like to get maybe just open question to to everyone on the call here but in in sort of the the perfect solution that Miles laid out right where you mitigate risk of centralization uh protect users against

harmful types of Mev and then kind of the third one being allowing the application to capture some of the value that's generated from Mev how would you prioritize so let's just say you can't have all three of those things how do you think those are going to get prioritized thank you so if you look like got an answer I'm not calling you man I uh I'm not sure I've got an answer because I think there's there's no answer yet and it might sound cliche but it's quite interesting because first of all is a very tough question which one out of the three you want to prioritize the most but if you look at ethereum right now you can already argue that one of them are de-prioritizing and that's front running like I mentioned earlier in this episode you know some ridiculous

amount of capital was being taken quite unquote from sort of users with no retribution for sort of what they've done or anything like that so I think networks like Cosmos for example clearly this is a major prioritization especially for osmosis which I'm pretty sure right now is the biggest X you know they've come out and said that we don't want bad Mev and so they want to avoid front running you know and they also want to sort of reduce centralization you can sort of reduce um centralization in a sense because if validators struggle to differentiate let's say it's a bit harder to say like steak with chorus okay you can differentiate with governance and stuff like that as well um but in a sense potentially it makes a

validator game more even because you know say chorus one is a more advanced validator we might be able to capture more stake and you might be able to argue this is bad for centralization so osmosis here is sort of prioritizing you know decentralization and also bad Mev and also what it sounds is capturing itself so I would say they're doing all three when it comes to like uni swap or something like that on ethereum they're not prioritizing you know removing front running because that's just not a thing for them right now you know in terms of the actual sort of centralization or decentralization of it you can't really even compare to osmosis here for example because they're not a chain yet which I'm pretty sure they will be at some stage but that's another question topic but um yeah I think it's it's hard and

it's like I think another way of thinking about it is it's not just those three things it's also about how can you keep every actor in The Chain happy almost because someone's gonna like be kicking and screaming if they're not getting enough I think that's the beauty of Mev right now because there's always going to be someone kicking and screaming somewhere and it's interesting because I could sort of Imagine potentially in the future like maybe even one network goes we want to use D Floor and Skip like that's so cool and also pretty crazy and you'd never be able to do that ethereum for example you sort of restricted to something like just flashbots you know maybe it's like a fair block and a d flaw and a Skip and there's like five different Meb Solutions and you're like plugging and playing and then it goes

hell modular and then you know then it's like you know then it's a party so yeah I think right now it's sort of about preferences of the network especially in the cosmos and Skip like I think here as well does a really good job at it's sort of like a product in a way where they're going out to these networks and asking them you know what are your needs and what do you prioritize instead of skip saying like we prioritize this or even deeply for example saying we prioritize this you know it's going out to dydx or osmosis and saying like what's important for you and how can I build around that there's a question maybe there of does that scale this is a great question and problem to have and if you can sort of solve this I think this is a great right

sort of possibility and potential opportunity for who can do that in the cosmos one quick thing just to add to that is like I really so just like to dive into one of the points that you said today which is like I do think there's going to be two more hits right there's going to be a market for Block space and then there's going to be a market for order flow right and I think like those markets are not like in my mind they're not necessarily Superior right it's like there's going to be markets for you know tenants and there's going to be a market for apartments for them to live in right and there's going to have people representing both saying well I need my tenants need this this and this and then people say well my apartments are going to require this this and this and I

think there's going to be auctions and ways of sort of obtaining both so if you're a Searcher you're going to need the flow right to actually back run and then you're also going to need to land it on the Chain um so like for on both ways I think you're gonna have to participate in both auctions and so I'm very excited about order flow Solutions coming to Cosmos I think I think they're a big missing piece um to add one thing I think Mag is pretty spot on like the right way to think about this is to separate these two um you know high level markets out into their own distinct things as as you try to reason about them um and one of the reasons I think that's like so important is because something that pretty much no one really talks too

much about in crypto when it comes to this question of like what are you know what are the priorities among those three things is um what actually leads to True institutional crypto adoption and I think right now the number one thing that leads to that is uh Fair infrastructure for liquidity providers um I think you know liquidity providers market makers are providing a service to the community and that service is providing liquidity you know 24 7 obviously they're collecting the spread and if they're you know good Market maker they collect more of a spread but they are providing a service and in a situation where they have to compete with other um takers such as Searchers that have like handshake under the deal under the

table deals with validators it is impossible for a market maker to profitably make markets and provide that service to the community so until we get to a point where um you know market makers are able to make markets fairly without getting you know uh run over because they don't have some sort of under the table deal we're not going to see true institutional crypto adoption and so I think uh back to mag's point which is once we're able to separate these two things out order flow markets and block space markets we're able to create I think more efficient infrastructure for both parties uh you know both Market participants in these two types of markets so that that's what I'm kind of most stoked about in 2023

oh that's um really really interesting especially hitting on the the separation there because I think you know no one's really attempted this so far um and so it's hard to kind of visualize what it would what it would feel like for a user at the app level um but maybe just to transition a little bit here you know the whole season we've been talking about a spectrum of app chains or app spaces um going from Full stack app chains on one end of the spectrum you know ethereum l3s uh that you know settle to a common L2 on the other end of the Spectrum in something like Sovereign Roll-Ups built on Celestia falling somewhere in the middle and there's a lot of different flavors there and so

you know I think we've talked about today we've just general purpose l1's and full stack app chains um but in you know this feature that we see is uh ethereum and Cosmos on somewhat of a collision course in that ethereum apps on ethereum will be wanting their own you know app specific or sector specific Roll-Ups um and so I would love to get into kind of what what the Mev space looks like uh for the Roll-Ups um you know right now we there's pretty much no Mev because every all these Roll-Ups are run by a single sequencer right um and so I would love to hear maybe I'll kick it to mag because I know

you've been working with the Celestia team a little bit um would love to hear what sort of solutions you were thinking about for Roll-Ups in particular um whether these are you know coming from shared sequencer networks um and just you know how trying to start to tease out just how opinionated say a roll-up can be about how it treats um uh Mev versus a full stack app Chain versus you know an app built on a general purpose L1 um so yeah we'll kick it to you Mac sure I I can take a shot but I mean full disclosure um this field I would say is very early um so I mean if you think about Roll-Ups on a shared da right which I guess you

can also put all the l2s and ethereum in the same categories and say you know use ethereum for their da uh there's a question which is sort of like when do you post the data right like like what hap like can you coordinate somehow between the different Roll-Ups so that you could say uh like have cross chain atomicity rules so that like when you actually have the posting stage could you sort of combine that together and give Traders down at the roll-up level um some kind of increased preference expression over doing that um one design that invite anyone watching this to to take a look at is Evan forbes's shared sequencer model so in this model basically you have a set of nodes that does sequencing that sit

on top of a bunch of Roll-Ups and then they all post their transactions and the shared sequencer is in charge of basically ordering each one while also sort of extracting Mev and you know even doing like cross domain extraction between them I think when it when it comes to like the sovereignty question uh we come to this like idea of like Fork Choice rules right so this is the concept that basically like a roll-up can decide to selectively choose Forks based off of some set of preferences right so like let's say let's say you had 20 sequencers for a specific roll-up and then they all posted um a state route and then they returned back and they also and they each

individually said okay this is the state that we individually think should be next the roll-up can say okay well that one's based off of Front running that one's based off of sandwiching and that one I know is good because like they're from let's say a trusted sequencer that you know follows the rules that I want to so I'm going to choose that one and all of you guys just paid for data but you're we're not going to include you on chain so I think they're you in Sovereign world you still have the the opportunity sort of to have Roll-Ups make decisions for your fork Choice rules about where they where they want to you know how they want to sequence their transactions I think this space though is very early right right and so it does feel like the value capture

there is still at you know away from the application uh and or app specific roll-up and more you know happening at the actual uh shared sequencer uh Network level um and I think it just will depend on you know how competitive these networks of shared sequencers are because then you know they will actually have to offer you know maybe more of that value flowing back to the roll-up side um yeah so maybe I'll kick it to save have you could you speak to a little bit about you know what the role of the sequencer is here um and and how it differs from you know the validator side and if that's something that you guys are exploring at chorus one or not

yeah it's a good question so I have a sort of opinion here on what it could look like in the future which I think is potentially not thought of so much right now and it's somewhat related to the work that chorus has already done on this sort of invalidator Mev so I've spoken with a few execution layers in the past who have been sort of interested in potentially utilizing let's say like a validator that takes all the Arbitrage itself which in this case a validator is just the sequencer so you have like the layer two you know here and then you know you have a sequencer here and then the segments it posts everything back to the to the layer one and quite often in ethereum right now there's just one sequencer and theoretically it's actually okay to just

have one sequencer you know it just censorship resistance is a problem here because you know you sort of it's somewhat reliant on them actually doing the right thing even though if they do the wrong thing you can always prove it but it's better to like right now it's a topic in general of dude even decentralized these sequences so even for us at chorus like we're not even running any layer twos right now so it's so early that we're not even writing the layer twos let alone think of it thinking of like Mev here but I think what you could start seeing is this sort of idea of let's say there's one sequencer and the sequence let's say it's an app specific Layer Two and it's just an app like osmosis has a layer 200 Theory I'm just theoretically so we all know the opportunity skip knows all the

opportunities so in the sequencer itself almost you could just have like working with skip sequencer is there they take all the opportunities and then X percent goes back to token holders you know and this is actually what token holders want and then maybe skip takes a cut and maybe the sequencer takes a card so it's like 10 10 80 split so realistically what a lot of these applications want is a split of they just want to return some value back to token holders and I know on ethereum right now this is a massive question in fact I feel like every week I'm getting pinged a bit more about like oh this is happening in Cosmos or Solana like how can we implement this uh in ethereum so the becoming a bit more experimental um but I I don't think in this really

Layer Two Meb space it's going to be the same as like a pure Builder or something like that I think if you're just going to have one sequencer right now then you know I think it makes sense especially if it's out specific the problem space is less as I mentioned before so you know the opportunities put it in the sequencer then there's a question of like okay how do you create this deal and is this deal off chain of like the X percent split going to token holders versus validators versus someone like skip who's doing the sort of building itself this is a another question perhaps it's like osmosis where it's like a dow boat or something like that but I do think a major difference here and mag sort of mentioned it before and we haven't mentioned that much in this podcast is like the community of the

cosmos is different to the ethereum community for example so if you want to implement a strategy in Cosmos of like 10 10 80 you take that to the community and you're going to have a political fight for like a week or two about if you should be 10 or 5 or 1 or whatever you're just going to be on the forums like all day or night in ethereum I feel like a lot of it is sort of not under the table but it's not as obvious so if there's a protocol up specific layer 2 you know they want to introduce X percent fees no one's going to really discuss that that much or protest against it let's say so I what I guess I'm saying here is that in ethereum you have a bit more leeway I think to implement new Solutions without pushback from the community because they just

want value to their token a lot of ethereum token holders and letters in Cosmos it's like it's this ethos of like we really want you know many many chains and you know the community has a loud voice and then validators have to listen to their delegators as we mentioned at the start as well because delegated private state so it's it's harder to sort of manage all the parties there I guess is what I'm saying so it's super early it's sort of confusing response but I do think one takeaway here is that it would look a lot different in ethereum versus Cosmos and it's like when Celestia is out and later two is a sort of arriving Cosmos this year oh I think that that totally makes sense and it's it's pretty easy to you know get excited about some of these Solutions when you can you know part of

the the benefits is that all of this value that was being captured by only one party it's now being you know recaptured by the application right and and then sent to users um maybe for Natasha because you are taking a different angle here um could you speak to I guess some of the benefits that users will see even if they have no idea that they are sending order flow you know through through D flow um and and what the value prop is there for for wallets um on behalf of their users yeah I think the biggest value prop is that these wallets who want to provide a good user experience for their users and also uh monetize without necessarily you know slapping fees onto the order flow

uh the benefit there is that they have a really simple mental model for how they how they should think about order flow monetization with d flow because we uh abstract away the actual infrastructure where transactions go and settle um to us any blockchain is identical because we have isolated common properties of each one and just interface with that interface that we defined so uh for for wallets that's probably the best benefit obviously this system is you know it has a trade-off which is that we can only um reason about a certain transaction type it just so happens to be that the transaction type that we reason about

and that we specialize is Market orders from retail Traders which is also like you know 99 of all transactions that have Meb so I think we we solved the problem we we don't aim to be a general purpose solution but we are specializing on a class which is pretty much the entire category um and so for wallets it's you know it becomes really simple with a solution like d flow that's that's great um and I think you know we've touched on it a couple times here um but if we are moving to this app specific World um you know app specific chain World either app specific Roll-Ups or you know these Sovereign Roll-Ups or full stack

app chains um there's going to be a need for for these cross chain uh Mev Solutions um and so we've you know we've learned a little bit about uh what flashbots is building with Suave um and some similar ideas have been introduced on the cosmo side first in the atom 2.0 Paper with the interchange scheduler um but yeah I would love to kick it to mag again here to talk about you know what is the purpose of something like a cross chain you know priority block space Marketplace uh what are some of the use cases that can be solved for for the applications that you know cannot be solved with these intra chain uh Solutions and and kind of teasing out I

guess where the side of the space is going yeah definitely so so uh Crossing Mev is very complex I think it's like something that you know is is easy to understand you know conceptually but when it comes down to actual implementation uh quickly becomes almost seemingly impossible at least on Cosmos where you have like a bunch of separate l1s that don't have any um communication uh like Baseline except for asynchronicity so basically there's no way to coordinate like okay you do that and then I'll do this like you can't say that right atomically whereas in some place like you know interchain

or let's say on ethereum you have that kind of composability you can say do this then this then this otherwise don't do any of it at all right um and that's what we call like atomicity and it's as simple as like a transaction in Cosmos you can have multiple messages right one might be a swap one might be a flash loan um and so you have that natively which enables a lot of this Mev that looks like Arbitrage right so if you don't have that right uh like how do you how do you even start to reason about like Mev so like the simple answer is okay I can keep a lot of money here and a lot of money here and then sorry about that um I can basically Arbitrage between them by taking swaps on either side right uh another way I

think you can think about it is you can sort of like um go more of like this Suave and then don't want to butcher deflow but I think they're thinking about this way which is like some sort of General cross-chain market making right where you can like aggregate the order flow and then give it to market makers and then they can do the cross train Mev in a way that's actually good for the ecosystem like Arbitrage in a good way um I'd say so like in terms of solutions that I've seen for cross Gene Mev that like potentially really would work the scheduler is one but it's quite a bulky solution because you have to like you have to buy the right to be the proposer in multiple different chains over the synchronous period of time so it's like what if I just want one transaction like

why do I have to buy the right to build everything um and then also sort of like uh these preference matching uh Solutions so like Suave um I think the shared sequencer and then you know anything that sort of deals with like cross chain order flow um I would say like one other thing to note is that I think generally most cross domain Mev is not really between chains it's mostly between decentralized exchanges and finance or centralized exchanges and I think like one thing that we've discovered over time coming in very excited about Crossing Mev is like wow this is not really happening because the Christ the price Discovery

is all happening on the deepest liquidity pool of them all which is the centralized exchange and so I think in the short run the the way these play out in my mind is there's going to be a lot of sort of like sex to decks uh Arbitrage tools versus sort of like decks to decks given sort of the complexity of that as well as the fact that it's just not where prices are being discovered right um yeah I'll say that much I I don't know if anyone has any uh response to that but I I want to um kind of conclude actually with with sort of a discussion of um even within sort of the cause so I want to talk a little bit I mean there's a lot of discussion in Mev

around kind of uh like like they've got a lot of really great points from kind of the validator perspective um on this call but I'd love to talk a little bit about uh kind of the role of of middleware and maybe specifically wallets um so like one you know one sort of metaphor or idea that we've sort of drawn on kind of time and time again throughout this series this idea of like the proximity to the user is ultimately uh extremely extremely important in terms of Leverage right over an entire value chain and if the ultimate like kind of capture or lock-in for the user ends up being at the wallet the wallet layer you could see them having an enormous amount of uh control when it comes to Mev and maybe even dictating right what makes good Mev versus bad Mev which is kind of an interesting

philosophical discussion in and of itself so maybe Natasha could kind of call on you like I mean is there kind of some future state right where there's like a wallet super app right which ends up having like an enormous amount of sort of control uh I just love to get your your sort of thoughts on that yeah um there is definitely a future state where you know maybe a single wallet or a handful of wallets have so much uh user activity that they sort of dictate ecosystems and it's it's a really dangerous path for crypto um because of that power you know yielded to some centralized group of people that operate these wallets

um and I think really the only solution to this is creating uh economic incentives that basically prioritize the decentralization of order flow over the centralized distribution of order flow um which is I mean coincidence in in the name of D flow that's kind of what we try to do is decentral create economic structures that decentralize order flow um but but really this is this is quite a big problem and um you know creating these economic incentives is definitely I think you know the number one priority of infrastructure projects in crypto or at least should be happy to talk about how I see that playing out but um I wanted to pass it on to anyone else who has

thoughts as well yeah I would I would love to get uh everyone's thoughts because even to maybe just compound a test on what you're saying for like how this might be able to play out it's like one of the Dynamics that's made this so difficult in in web 2 to kind of Point At You know kind of our web 2 Giants and say hey what you're doing is anti-competitive or not good is because they keep doing things that are ostensibly good for the customer right like if you take a look at like one of the super apps being like Amazon right they do things that are very good for that consumer right like you can return anything on Amazon with you know for a while you can do this right no pictures no anything but who's getting squeezed on the other side of that this is kind of their network of

vendors right that have these really strenuous sorts of terms and they're exerting their their enormous leverage kind of down the chain and you can see a similar Dynamic there happening with with Meb so I don't know mag or zave if either of you guys have thoughts on this um or if you just want to close with like a philosophical discussion of like what is good Med versus bad mme I'm sure you guys both have thought about that quite a bit maybe before we move on to that topic I just have a quick point on on the wallet side and it's sort of when you think about the chain here again so you know at the start it was minus and they were capturing all the Mev then you know then the builders and so just came in they were capturing the majority of the movie

and then now it's like pushing down again and maybe it's like the wallets that end up being the ones that are centralizing force and they're capturing their movie and it's like at what stage of the chain or what you would like at somewhere it's going to get clogged up I guess so I do really like deflated solution and actually I was going to mention as well back to mag's point before on this like cross-domain mov so here potentially is like where even a d floor or something like that could integrate like cross chains maybe and like maybe this is sort of the way like and that's how it works in Cosmos and maybe skip specializes in just like a chain by Chain sort of basis and they sort of work together so yeah again here I think just super early and really interesting and then maybe they close in

the good versus bad muv I think you know the fact is Cosmos is about choice and you know all networks have the choice and so sovereignty to choose if they want bad anyway or good Meb or if it's a topic at all and this is the beauty of the the cosmos in general um yeah I'm sorry go ahead sorry I was just gonna make a quick point on what Michael and save just said which is I I think it's really important to note that um in this future state where these wallets have been yielded so much power it's important to understand with you know the Amazon comment that what they're doing is not an effort of trying to be nefarious it's ineffort of trying to give their users the best experience uh

it just so happens that right now the way things exist today the best experience for users is to under the table sell their order flow to some centralized thing which will guarantee like great execution but is also a centralized in force um so I I think yeah that that's maybe one of one of the big things that we're trying to work on is just how do we get you know something that's better for users ux than uh like a really you know tight-knit but ultimately you're centralizing under the table deal with yeah Quant Market maker or something like that sorry go ahead go ahead Mike that specific example is a really interesting one by the way because even if let's

let's just give Amazon the benefit to the doubt and say the reason they do that is to protect the interests of their their users their customers it ends up kind of very indirectly not being good for customers though because what ends up happening is in this like terms of service right if you're a merchant that signs up with Amazon the price that you sell on Amazon cannot be different from the price that you quote anywhere else and the what the retailers say is well I don't want to give up 30 of my margin you know sometimes that's more than their entire margin so what they do is they just Mark everything up so that they can sell on Amazon and they have the same margin as before but now they're selling on Amazon so basically what you've just done is you've marked up the entire internet and then even

though these users think that they're getting a deal actually nobody's getting a deal and it's just more expensive everywhere so just like it's a funny example of how even if even if they are doing this uh supposedly for the right reasons you kind of still end up screwing people too so it's just an interesting yeah in the context sorry I'll go I'll let mag you know make his point because I interrupted him but the last thing I'll say there is uh in this point about like you know um I guess to tie it back to Mev and crypto uh and not you know sway too much to the Amazon part of it um really the important part here is to allow market makers to provide liquidity efficiently and and that's that's ultimately the driver that will make sure that none of these centralizing

forces are able to accumulate and snowball into something that's pretty terrible foreign My Philosophy on these things is sort of like three three big things that sort of go through my my mind these days one is we don't really know what the right solution is like we're all trying to figure it out and presenting sort of different ways of structuring the market and ultimately I don't think I think the worst thing we can do is be like this is the way right like our PBS is the way or you know Suave is the way or or you know Skip is the way whatever it is like I I think making sure that there's chance for exploration about how these markets work um is is really important at this stage

and sort of not stifling that you know Innovation um I think the second thing is there's like a there's also like a legal conversation around a lot of this stuff right that I don't think we got into here but like you know we recently talked to a lawyer and they're like you might become a money transmitter right or flashbots might be a money transmitter like validators might be I don't know say what are they saying you might be something scary unfortunately not we try to avoid those type of conversations okay cool uh well anyway there's like there there is some you know there is some conversation about how these things work and you know a a far version of this might be that users are required to be given best execution right which is

how it works in additional financial markets and everybody sort of works around um you know Works around that sort of Baseline and then I think the third thing is like I think mev's history is not yet written in terms of how it relates to user experience I actually think the most interesting things I've seen Mev is how it affects like the basic user in terms of using their chain there's like a good article written about this that invite everyone to take a look at by press which um about like Mev wallet or sort of like using Nev to like put yourself up to the top but I think what we're going to see is like a an explosion of um sort of like understanding like the the connection between sort of the user experience whether that's refunds or

free transactions or whatever else it is or you know atomicity your prioritization and then also Mev sort of how those things fit um which I'm very excited to sort of like see a lot of solutions for well I want to make a quick Point here before we get too entrenched in this though which is on sort of the regulatory thing that mag was talking about in terms of best execution it is a very crypto thing for us to basically discuss like you know uh redistribution of Mev in a sense where a percentage of the atomic Mev goes to someone who is not the retail Trader like the analogy in traditional equities markets would be uh Citadel keeps you know gives you

terrible execution and keeps some of that spread against the best market price and it is in my opinion obviously not a lawyer but incredibly unlikely that when the SEC comes to look at Mev and how these swaps are getting executed that they'll be like okay cool you know 80 goes back to the user no they're gonna you know hand on table slam slam their fist down and say uh 100 needs to go back to retail um so yeah that's it's gonna it's super weird crypto is crazy yeah maybe the uh maybe the conclusion is Gary is coming for us all so we best all be careful um guys

yeah yeah um guys this has been a super fun hour you've given miles and I uh a lot to digest um from this conversation if folks want to find out more about each of you follow you on Twitter follow your projects uh and keep up with the good work that you're doing uh what's the best way to do that for all of you I'm not going to call whoever's feeling bravest here you go call it out okay I'll I'll kick off uh D flow.net is our website and at D flow protocol is our Twitter uh you can also follow me at uh this hash math is my name and uh usually post MVP type stuff there too yeah and I'm uh I'm zero xave on Twitter and chorus1 chorus.1 is the website and

the Twitter is just chorus one with uh no spaces or anything in between no dots foreign I'm at Ox Magmar um used to be a Pokemon streamer so that's that that's from leftover days uh and then the companies at skip protocol on our site is skip. money wow that's awesome mag we need we definitely needed a couple of minutes to uh to dig into that but uh we'll save that for the next time we do a pod together um guys it's been a super fun one uh you've been super generous with your time so thanks a lot this was uh hi partner great podcast what do you think yeah it was a lot of fun

um I thought you know it was great to have this conversation with folks across you know different parts of the Mev value chain um we have dentist who's really building kind of for wallets um and really focusing on on order flow and then you have skip that is building all sorts of block building software and solutions uh both on-chain Solutions and off-chain solutions for app chains um and you of course one who is at the validator level and and participates you know actively across general purpose l1's app chains and so they can see the full spectrum of approaches you know to Mev that are that are being pursued here

and they can have a you know a great perspective on kind of the trade-offs that we're trying to tease out yeah yeah absolutely I think one of the challenges of this discussion you kind of kept hearing it from Magnus is like hey we are venturing into new ground uh a new territory hey we're breaking new ground like a lot of the I mean part of the reason it's exciting to be talking about this now is it's happening but also necessarily there are a lot of the sort of Big Ideas that I think you and I want answered sort of still sit in the realm of conjecture so just just kind of bears mentioning right that the stuff it's very similar to like a lot of our other conversations we're talking about you know what sort of stacks work best right we've only seen the the cosmos

full stack you know in in action we've only seen ethereum and you know Salon in action same with Mev we've only seen really off-chain PBS Solutions in action at scale um and that happens on ethereum and and Salon it to some extent um and we're only now beginning to see really what app chains full stack app chains can do with with say you know osmosis is kind of leading the way in partnership with skip um and there's still so much that you know has been we spoke about today but has not even you know development is in the first inning right we're at much less you know there's no usage of it yet even um that's not it's not not developed so

I thought it was I thought it was pretty interesting to get zave's perspective as a validator right talking about where value Crews and he kind of did it uh chronologically which I've never heard it broken out but like first there were only miners right so all the values sort of recruit to them then there were the The Searchers and then the uh the builders kind of the relay and uh and so as and as each new kind of um component cropped up obviously there was less competition in the beginning so an enormous amount of rewards sort of uh sort of flowed to them um you know it's just interesting to hear is almost like a whack-a-mole type description uh and then obviously wallets haven't really had enormous amount of success yet but there was a big question mark there and I think everyone uh you know one once we posed

that question or done a lot of thinking about that it also is interesting to hear him say that all the validators you know something they always ask is how can we differentiate and I have to I actually have to uh not not that this is necessarily the case but it it sort of rung a little PTSD Bell for me coming from the world of Consulting and you know a bunch of our friends are in like Investment Banking I'll tell you something that investment bankers I'll ask like how do we differentiate what's our value add and usually when you have to start asking like what the value ad is it's not quite as obvious you know not to say that uh you know course one is is phenomenal and like on The Cutting Edge of all this stuff but it was interesting certainly to hear that uh that Dynamic

voice right right I mean it everybody all all validators are providing you know somewhat of a commoditized service in terms of securing the network and validating blocks um and so it's it's more about what else can you do for the network that that makes you more valuable for your delegates will attract more delegates right um and I think part of what he said that was also interesting too is is really just the difference in communities and and what communities care about um because I think you know for app trains you're just dealing with a a smaller Community right that is only cared about you know really prioritizes their own application uh and the user experience for that

specific application um and so they can be you know much more opinionated uh naturally more opinionated than than say a general purpose chain that is trying to appease a community of thousands of apps and millions of users that all have different you know priorities and interests and so it's there's the technical components of what is possible and not possible of you know with an app chain and a general purpose um a general purpose blockchain but then there's also kind of the social components of of you know just how far an app chain is willing to go in order to exert you know control and influence over these questions versus how far the community that is trying to appease as

many Builders as possible will go um right because as soon as they start prioritizing one application or one you know sector of applications and they've got all the other parties and stakeholders you know start to get upset and that that you know potentially leads them to move to their own environments which decreases the value of all the Mev being generated on that on that general purpose L1 um so just Mev is so interesting because it does really intersect you know between Technical and social um and and there's you know a lot more to dig into here but I thought that was interesting yeah me too and you know I I think we're a little biased here I think because we had the perspective of three

people who spend a lot of their time in in the cosmos ecosystem obviously but it did sound like just the nature of the you know many different layer ones right uh kind of makes it a little bit more complicated especially when it comes to the the sort of cross-chain Mev domain right there's definitely like I think Magnus said it best like in theory it sounds great right like when you're reading the atom 2.0 white paper you're like this oh yes huge but I totally see it this makes all the sense in the world but then when you get down in the weeds like this is actually a very difficult you know boarding on Impossible problem yeah right right right but you know they're building at the Forefront right this is this is where we know what we're trying to solve for right in those

objectives um and we're trying to find solutions that solve for as many of those as possible um yeah and and unlock new use cases potentially as well something that I wish we we like got into it right at the end but I would love to have gotten uh more of their thoughts on on you know this is less of a technical but more of a philosophical discussion about like what constitutes good versus bad Mev and usually when that subject gets brought up people are like oh yeah well sandwich attacks and front running are bad and back running is good but the reality is there's an enormous long tail right of other forms of of Mev that exist out there that much more Fallen kind of like a nuanced gray area so I I think I would have loved to get a little bit deeper

into the Weeds about like what constitutes good and bad and maybe yeah yeah I think that's fair and and but it's also worth mentioning that that that long tail um is really only of material value on general purpose chains um because you know part of why you can have a solution that is so opinionated and customized for an app chain is because you know there is a pretty narrow design space for Meb that's possible um you know we use osmosis as an example but say you know 90 95 of of Mev it generated on osmosis is you know from potential back running or front running of swaps right it's a DEX

um versus you know Solana which has an enormous long tail and so you know I think the difference there is like from a practical standpoint is that osmosis can build things like the protarap module that is basically a protocol owned our bot right that is that is run on every validator and is capturing that you know Lion's Share of all Mev generated where you know if you were to try to build something like that on Solana there would be days and and or blocks right where it's looking for what it thinks is you know the most common forms of Mev but it's leaving enormous amounts of value on the table because there's no way it can possibly You Know

cover all the different sorts of long tail opportunities um and so I think that's that's what I thought was interesting about course is one course one's paper last week is that it it's implying that there should be kind of you know a Marketplace of of Searcher you know um strategies uh that that validators can pick up um and then potentially kick back a revenue share to those Searchers um and that would allow the validators to run some of these systems you know in protocol but but also have be able to capture the longer tail of Mev and distribute it you know to the to the Searchers that to generate it basically what did you think about Natasha's point at the end where he's like it's all well

and good uh you know so that we can say hey like we'll have horrible execution but we're going to kick back like 80 percent of this to you know to users and that'll be fine and the SEC coming in saying uh hey guys we actually have uh rules out there about best execution you know giving best execution to retail and you can't just have bad execution and kick back 80 and call it a day you know what was your I know that's kind of a far off question but like what was your sort of thought on that yeah no I thought that was interesting along with his his point around really you know how we're going to uh there will be at some point a separation of Mev you know between uh execution uh auctions and block space auctions because you know those two things can can potentially

fill very different needs um and I think you know it's unclear how this is all going to play out in terms of whether you know kicking back say you know 80 you're only letting the user capture 80 versus kicking back you know some sort of Revenue share with with other um providers how whether or not that will fly or not um but I do think that the the point is that's really interesting is that you know You by separating execution from from block space auctions um you know you could offer some some pretty significant benefits to the users in terms of you know getting your uh trades done basically immediately and guaranteed at the very best price

um and and to me you know it raises some questions around okay if that's all being controlled basically at the wallet level you know is there a way for apps to also you know see some of that see some of that value as well maybe it's an app specific wallet that you know becomes available long after account abstraction is implemented um so folks you know users are not creating new seed phrase for every app specific wallet they use but um it would you know potentially allow something like that you know would present a different future than say five wallet providers or even small fewer wallet providers really dominating the whole Space mm-hmm

yeah it's tough I mean it's once you kind of have a uh it's that there's a specific phrase for this bias but like once you've seen a problem solved in a specific way it's very hard to unbox yourself from that problem and you know I have to I have to admit I've been sort of red pilled on this uh you know the the analogy that Sunny gave us in the beginning of this series which was um you know how this sort of played out in web 2 and the the power that apps at the at the very uh that have close proximity to the consumer of the user the amount like the natural tendency The Market Force tendency for centralization and you know when I think about it my heart of hearts I'm like okay you know they're like what do I think is most likely to

play out here right like there might be these other Solutions where apps want to control the the wallet right and there's a beautiful full stack sort of synchronous there but in reality like someone right is gonna start a wallet that's like pretty good they're gonna be phenomenal at distribution and marketing they're gonna grow that baby and that like as soon as you have that like critical mass of users which you can argue metabask is like close to having right now but in the grand scheme of apps there it's still very early days very few users and I just see that that natural convenience being a huge Factor but you know a little voice on the back of my head is saying that's a very kind of web 2 Paradigm and who's to say it will play out like that again yeah yeah which is true and I think we we talked about this

a little bit but you know as the user relationship with the wallet or was it with the is it with the application um and you know I think there's probably strong arguments of both but I I by my hunch is that it's more with the application uh that that really has the brand and and it's been you know the reason you're up you know a user of a blockchain application in the first place is to because they're providing some service that you want um and the wallet's just the middleman in between there um but yeah I think it does it does definitely you know pose some big questions um because even even something like d flow like a you know basically an execution environment

um is competitive with with a lot of these applications that we're talking about like dexes um and so maybe you know Dex has become a thing of the past and and all of this has happened all of this is handled at the wallet level with these with these different you know type order flow routing Solutions um I think it's probably that's probably a little extreme um but yeah it's definitely definitely really interesting point to raise that I hadn't hadn't thought about um too much before before this conversation and by the way um you know who it's kind of played out differently in uh there are two different models right like there were there's a great uh paper from Z Prime Capital that kind of used this the

fappening example and the the I the what they're using as an analogy or inspiration is like kind of the super apps that exist in China right uh but that that doesn't really that doesn't really exist in the United States there are there are in like kind of uh there are lots of different small apps right there are some big ones like Amazon is huge but there's also you know you don't there's not one interface where you call Uber and Airbnb and Etc you have a relationship with each one of those apps where in China it'd be one super app that you do everything everything's from so I don't think it's uh Destiny right that it plays out in one specific way it's I I don't know yeah exactly and maybe just for for the audience the fappening

refers to the flipping of uh value from the you know L1 base layers to more value being agreed at the application layer yeah it probably is good to provide that disclaimer yeah so um yeah exactly it and what that gets into is these apps that you know start with one service and capture users and then eventually expand horizontally into into lots of different services so that the user basically has one interface for anything they could want um or maybe anything they could want within a certain sector um

and then a question is you know do they expand vertically up into the wallets do they expand down into their own you know app specific roll-up or chain um you know and what is the sequence of that because it does you know again my hunch is that they will you know once you hit critical mass you will want to expand you know one way or the other um in order to capture more value and offer an even more you know differentiated product experience for your users yeah you bet all right buddy I think that's all the time we have here but this is another one we'll see you same time next week partner sounds good foreign