Ethereum Sequencing and Preconfirmations Call #14

Transcript

okay um welcome everyone this is uh sequencing and pre-confirmation call uh number 14 today we have uh Kevin lepo from Eve gas who will be talking about uh pricing pre-confirmation tips so the the same subject as the the last call and I guess this is the the second call in the series uh I think we're going to be going down the rapid hole and have many different speakers talk about this this very interesting and somewhat complicated uh topic but before Kevin speaks I want to share um my screen real quick so um I want to uh highlight that we're going to have a sequencing day um

on November 15th uh at at at Defcon is the very last day of defon so do do feel free to apply it's it's completely free uh we have an amazing venue in at the haatt uh which can accommodate hundreds of people and uh we're hoping that this is going to be uh the the canonical gathering for many of the rollups and the the companies building around shed sequences and and Bas sequencing um I guess the other thing I wanted to very very briefly mention is that there's a a few more logos on this slide um so one of them is a new base rollup called facet that that is uh that

is live on Main on Main net I guess that is the the second live or maybe the first I don't know uh base base roll up uh with with Tao and there's a few more additional names here as well in in the precom projects so I've I've added e gas uh in interstate and uh possibly also HDA which actually have been have been around for quite a long time um and would overdue to to be part of of of the slide so that's it from me um Kevin feel free to go ahead and and take it from here sounds good thank you uh good evening from Hong Kong I just uh realized today is Friday the 13th um but I hope nothing that we talk about today

is scary so uh we'll see how it goes let me just pull up my slides here give me one moment okay all right so on my side I have my deck and I have uh the chat on the side if anyone has any side questions as I'm kind of going through this um Al righty let's get get right to it so uh I dropped some notes in the in the chat earlier about uh you know I guess two things one is um you know using using the V markets to price gas and separately a bit of background on what we're doing at eth

gas and so um there's a there's a lot to unpack but I hope to give a bit more clarity today so we'll start first with this concept of v and pricing ethereum blocks this is more of a um uh again just like a a thought exploration or exercise um and what what I want to highlight is that U it's a it's very different in terms of how one would like conventionally price ethereum blocks um and so we'll get into the detail shortly but uh quick disclaimer it's a pricing discussion not a ideological one um you know I I shared uh the the the notes on this earlier and it uh we quickly went down the rabbit hole which was fun and

interesting but um maybe if we can save the IDE ideological part till a little bit later but U if you have questions on pricing feel free to jump in and just ask uh so for background there's two parts to this um for more information there's an article on sorry about that I'll make sure to mute people that's that's um okay so there's the article which is on the eth research site and then there's also a pricing model um attached to the article itself so if you're wanting to kind of look at that while I'm chatting through it feel free to kind of bounce through that um the way it starts off is basically well looking at ethereum or the network itself as a financial instrument and the

the context here is that there is a um a a latency kind of Arbitrage or opportunity in between the prices of uh Financial products or between the prices of tokens on centralized destinations uh versus the prices on ethereum and so conceptually if you were to say um if somehow you were to buy 30 or 60 blocks of ethereum for example then you are afforded um you basically kind of lock the the last state of of the chain uh you could lock you don't need to but you could uh in which case you have this window over you know 6 12 minutes where effectively you could watch prices in

the centralized world and you could just see how those prices evolve versus the price the last state on chain and so at some point if those prices deviate by some threshold maybe in this case say 4% then you know a opportunistic person would basically um sell some assets in a centralized world and then they would buy those at the last stated price um on ethereum and so you know that for that likelihood of a 4% gain or an X percent gain uh multiplied by the potential volume that I could put through this U that helps to kind of form a basis for um quantifying the the actual price of blocks as in like what is the likelihood

that I could make a certain amount of money over a certain period of time um and and and so that that's kind of the discussion again so likelihood of um a certain profit over a certain period of time now in uh on a real-time basis this is a latency or Stat R game so you know there's a lot of players kind of doing this but what we're looking at uh again is pricing and pricing is a is a forward looking indicator so in this context we need to find forward-looking um assets or instruments and in this case it's volatility um so basically we can look at the V of assets as it relates to what our expectations around token volatility in the next you know one day two day ideally in the next 12 six 12 minutes you know what are the instruments that

we have to uh what are the forward-looking instruments we have in order to price forward-looking blocks so that's why we look at the V Market uh so a bit more specifically what we're doing here is the the trade and again there's more details on the the research article but the trade is you would short a straddle or strangle um we're using a strangle just because there's transaction fees that are baked into this you effectively short a strangle on say derit now basically if um the price of eth for example in this case if the price of eth moves by some percent um then you're going to lose money now as you short of the strangle you get some

premium up front so if the price doesn't move you get paid let's just say $10,000 so I short the strangle I get paid $10,000 that's my base case I take that $10,000 and I now buy blocks or pre-ms on ethereum and so that then basically closes this kind of Arbitrage so I might lose money on the top side I might gain money on the bottom but that basically kind of closes this um this trade so it's kind of a close form trade and so in this context if I were to um you know make uh 10,000 from selling that option I would be willing to spend up to 10,000 to buy these blocks on um on the ethereum side and so that's the trade so

therefore we can basically now use V uh in those markets to price blocks over a some period of time on ethereum now so we went ahead and did that and um there's there's a lot to kind of di dive into but at a high level we assumed that you could sell 10 million um or the market would bear $10 million of tokens on the option side and liquidity on the the on on the Dex side on the ethereum side you know we look at derit and we see oh the V is 75% and interest rates are whatever um you know we assume some fees and now we can basically plug into your standard black shs um you know what is the price of

this strangle in between 1 to 32 blocks and then that's basically it so what is the price of a strangle between 1 and 32 blocks and that will then back out what is the price of the pre-ms or what is the price of the blocks for that same period now one thing to highlight is that in in this exercise there's no concept or um awareness really for Mev for other types of Mev for standard deal flow you know in this case I'm not talking about other conventional Mev front running back running or whatever else it is um I'm not looking at the meol I don't need to and so here on the on the bottom side where I am

long ethereum again these options offset one another but on the long side there's actually a positive basis meaning I'm not ascribing any value to the ethereum side therefore this is a floor price or a minimum price for these uh for these blocks okay so again anyway so so this is the curve that we have uh basically we took into account what are the uh so every like one in 10 blocks you kind of get like these solo uh self-built blocks vanilla blocks and for the rest of the time everyone's using Mev boost so if you assume that the Mev boost guys are economically rational as in they want to get paid more then you

know we assume that what if every time it was possible to sell a block or multiple blocks validators would do so um and so in that case we looked at the frequency of consecutive blocks we then look at the expected pricing as implied by the V markets uh um and it comes out to around one and a quarter to 3% in yields for validators which is that high is it low you know uh we can kind of debate that but the point is this isn't even factoring in um any of the like the orderflow mechanics right any of the priorities fees from orderflow any other other types of Mev this is purely just interplaying the V markets and the um uh

and kind of the block space market now um that's the theory part there's a bunch of considerations kind of under the hood and um you know we can maybe unpack these here or uh somewhere else but um some of the core Dynamics are basically that um fees or transaction fees make this difficult and so if I'm going to uh you know sell on derit and I'm eventually going to buy on the dexes well what are the transaction fees like maybe Unis SW swap is like 25 basis points that means I need to factor in the prices on derit moving beyond 25 basis points so I need to factor in fees into the strike prices of these options and so fees make this difficult however

time makes it easier if I can buy more than one block or multiple blocks then um you know it's more likely that I'm going to have that payoff you know I guess as a expected higher Vol would mean higher prices on the right side here basically execution is possible you know one could do this today if one was able to buy blocks um but it requires basically an active options book you know some expertise around how do you manage how do you manage short dated options it's not as straightforward as kind of longer dated options uh it's less of a theoretical exercise much more of a practical one um you also need to have a sufficient amount of inventory because you're selling in one place buying in another I use 75% V just as like a

random number on the previous slide but historical V for eth is actually more than 100% for very short date ad options in some cases it's 150 or 200% doeses that mean it's higher maybe but you offset that with basically liquidity and you know when the vault is 150 200% you're not getting huge liquidity so what we did is we ultimately said even if the V markets are 200% we're only going to be able to execute at 75 so we assumed a whole amount a huge amount of slippage in that process again this example it only assumes eth you know one could um this is actually a portfolio really that we should be looking at I would be selling

you know eth Bitcoin uh uh xrp again salana as well basically basically all the the centralized assets versus the their equivalents on the decentralized side bunch of other considerations but um but this is it so uh I wanted to walk through this because it's a it's a thought exercise and I think it's relevant and important for the market that we talk about you know we're talking about uh pre-confirmation uh and proposer commitments and when you extend from a basic pre-confirmation to selling a whole block of confirmation pre- confirmations to multiple proposers selling pre-confirmation it invariably

kind of moves in this direction and so you know I want to present this to just have enable us to like think about how the future may evolve so I'll just pause here for a quick moment um I'll get into the eth gas product shortly but I'll take a pause to see if there's any questions thanks Kevin so I guess this was a e research post that you you posted recently maybe uh today or or yesterday I guess part two is more of a an actual auction mechanism that you're trying to build and it's is maybe like uh more on the mechanism design side of things as opposed to the to the because I guess for pricing this there's two

things we need to do one is that we need to design good auctions and then two is that we need Searchers to be able to to place bids and I guess part one is more on the Searcher side of things uh and it could get very complicated with mathematical models but maybe part two could be like a very simple auction mechanism yeah so that uh well I guess it depends how we look at it so uh uh so there's a question here zerox princess not sure if I'm mistaking anything but you're basically trying to do sex deex ARB shops R&D and set it as a baseline for pre-com pricing yes to zerox princess but um there's two elements of um sexex are basically there's the real-time kind of um the

trading the delta or trading the gamma of it um and then there's basically what's implied through the options market so it's almost like a forward look um Market versus a uh like a like a real-time active Market itself so um they're they're kind of two similar um two parts of the same Pi yeah okay so with that in mind I'll jump into our product and U again if anyone has any questions just U put them in the chat so eth gas uh what are we we're basically a uh actually I I'll I'll change the word addressment you're using it's less about auctions and more about a Marketplace and um the way we see it is uh you know we basically have a

Marketplace for real-time commitments and at at a high level what excites us is basically that you know we you know the moment that we can the moment that anyone can put through any trade on ethereum and get a confirmation within three milliseconds that basically means that you know uh ethereum can compete on a latency perspective um on a settlement perspective with the centralized um counterparts but but but of course you know it is decentralized and arguably better in that process and so for me that that's kind of the singular moment for ethereum that's what excites us and the team uh you know if we only applied real-time commitments to uh ethereum itself we'd be looking at maybe 130 TPS on ethereum this is assuming that we use

the most efficient most efficient kind of transfer mechanisms for example uh intent protocols so when you combine intent and real-time settlements that's when we're looking at 130 TPS in the future when we look at base rollups then you know you're not limited by that L1 gas you have a more capacity and so we'll see where that goes um but um what we're doing can apply equally to you know L1 L2 and uh based rollups as well as other l2s itself um we are uh centralized so we will have a server somewhere uh sitting on a rack uh in discussions with the various Market participants we haven't had issues so far in terms of this centralization but uh if it's an issue we can open source um look at some ZK

compute or um or other approaches if it's required through governance on the topic of governance we are soon to be governed by a dow or a token so you know to the extent that you wanted to change the incentives the products the features with what we're doing one is you know this call itself and you know feedback from the community but another is basically through our token and the the Dow design where we could dial up or down you know how we kind of skew the product that we have um important for us as well is kind of giving people a choice and so you know we're using a commit boost which we can talk about later that gives um proposers choice but um in in what we're doing we

want to give the proposers the choice as well to to to to choose default settings perhaps or to structure how they want to sell block space you know I think this whole initiative around pre-confirmation is in kind of maybe shifting some of that power to um to the proposer or validators and kind of letting them decide whatever it is and how they want to sell their own block space um okay and then so as relates specifically to us in our kind of more micro uh thought process the as I said earlier it was basically empowering validators or empowering proposers U but specifically maximizing their fees and in maximizing their fees what we wanted to do was look at the the PBS flow um but

basically just reduce fragmentation across a big chunk of that um close information leakages we wanted to increase access and I'll talk about what this means later on um but we basically achieved this through some type of product standardization um after sharing the article last night uh I've learned that um we need to engage more with the community in terms of what is perhaps good meev versus bad meev uh maybe we can discuss this later but I think it raises questions over like you know who should be policing whom perhaps in this case you know we are we are of course out of protocol but you know who who should be policing proposers is it us as a um you know as a as a Gateway that you'll see or is it the is it the is it

is it the network itself is it ethereum is it the community you know to to be decided okay uh so the PBS flow as I said is uh from our perspective it's quite fragmented U there's a number of aspects to this but basically the one of the main parts is that with the Builders you have uh they're basically selling kind of like like conditional commitments or conditional guarantees um you know the the fact that there's no guarantee for them to actually get into the block you know the best they can do is um the best they can do is try uh but when we shift to uh this other box here is called pgs I call it proposer Gateway separation I

don't know if that's the right term but it's my next slide uh but basically when you delegate the commitments from a uh from a proposer to a Gateway uh where the commitment is specifically specifically from the proposer then these are absolute guarantees and I'd argue that absolute guarantees are worth more than a collection of conditional guarantees I'd also pait that with an absolute guarantee from the proposer there are new types of commitments that we would see that currently don't exist or currently not feasible um and potentially we can talk about these later as well now structurally uh I think Justin you gave me earlier I think the the more

appropriate word um a unified Gateway if that's what it is um we call it internally like a bit of a proposer Gateway separation but what we've done is um basically we take uh well where do I start here so a proposer basically delegates this block optimization process this um control of commitments and how they're going to work they delegate that to a Gateway in this case let's say it's us um and they delegate it to a Gateway per for a certain slot so in a previous slot it could be someone else this slot it's us next slot who knows it's up to that proposer now um within this Gateway is some type of place where we would sell and other

people would buy commitments um as well as building blocks itself now whether we eth gas build blocks or someone else we're kind of indifferent um but what's important here is that the Builder has to be neutral um and so I'll also get into this as well before I get into that maybe just some definitions so we have three products one is uh inclusion pre comps and this is just to kind of um agree for the purpose of this discussion on on on how these work so an inclusion precom is uh I think as most of us would generally agree um you it's a commitment from the propos proposer to guarantee just to include your transaction somewhere in the block that's it so the way that works is proposer number five in this

case sells a certain amount of block space they they sell 10 million of inclusion pre-ms um into the market now let's say I've bought that 10 million block space um yeah I bought that 10 million pre-m now maybe two slots later uh maybe I don't need 10 million anymore right um so you know when it comes when that block comes to now maybe I don't need anything in which case I kind of just don't do anything with that 10 million and it goes empty now what I'd like to do ideally is if I'm not going to use it I'd want to sell it because then the absolute cost of these pre-ms is not the for way that I paid up front for the pre

the pre-com it's really just the the carrying cost so if I could buy that precom at you know foru but then a moment later sell it for three and a half then that gives more value to the pre-m itself uh because there is a secondary Market to it so what's important here is that it's not just the primary the ability to sell a commitment but it's the ability to trade it and basically um off lay or or or uh get rid of it to the extent that you don't need it um what this does is um I think am I missing [Music]

uh we can't hear you anymore I think we can see you but we can't hear you hello hello that's better yes okay sorry I think I hit my mic there um so I I think where I may have left off because I hit my mic was at um by selling by selling the precom and having it trade in a secondary Market it basically provides price Discovery and when pre-ms trade in a secondary Market you then basically get a curve of pre-m we have trading at this block that block this slot that slot many slots and that basically answers the question over what is the fair price for a precom and the fair price for the precom is whatever it is trading in a secondary Market um so so that's one of the key design

decisions for for us so that's the inclusion precom on an execution precom this is only for the the next slot so basically like slot one uh and in this case here what happens is that I am um I'm a Searcher I want to put in this bundle and I'm going to bid uh 50 gay for this bundle someone else is now going to put in a bundle which conflicts with my bundle so bundle one and bundle two conflict now I put in a bit of 50 that other guy put in a bit of like a 100 what we do through this Progressive auction is we provide a feedback loop to the person who is uh whose bid is

insufficient so through websocket we basically tell me that hey your bid of 50 is insufficient would you like to update it so at that point in time I can then update my bid to a value higher um in which case we would then notify the other person that there is is insufficient and it basically goes Higher and Higher and so this recursive or Progressive auction it's uh it's intended to capture as much of the expected Mev of that bundle itself uh U but there are restrictions around how fast can you simulate these and uh you know kind of how many iterations can you do as well um there was a question earlier about um why don't we use like a second price auction or some other approach um and the reason

is because when you have execution pre- comps the ability to support a single execution precom it often depends on what else like the other components in the block itself and so it's never just one execution pre-com by itself it might be like um like one execution precom might conflict with two others and these two others might not conflict with one another so we basically need to create different groups or batches of pre-ms and uh sorry batches of execution pre comps and optimize those independently so it's it's a bit of a optimization exercise we're not that sophisticated yet but um just highlight that that there's this um feedback loop uh in in the execution pre-

comps finally there's basically like the entire block itself um so extending from an inclusion pre-com if a proposer says you know what I just want to sell 30 million then that's basically the entire block and um so in this case that's what uh proposer number eight is doing they would just sell the entire block and of course they would sell it in the primary market like the original sale and once it's sold then those people can then um you know onell it within a secondary Market itself so in uh inclusion and whole blocks primary and secondary execution is through an auction okay so a little bit more detail on this uh proposer Gateway separation if that

is even the right term U so the key points here are that the participants in this market uh it's it's basically everyone everyone is in this single unified market now the par participants they trade commitment products and during this time no transaction information is revealed to anyone so I can't see exactly what you're doing but I know that you know there's x amount of inclusion preon there's a there's a an execution pre-m which may or may not conflict with what I'm doing I can see the public meol um you know but but I don't actually see any trades and therefore I'm not able to kind of specifically FR front run anything um what's important for this is um basically product standardization and so in the way we've designed it again

you have the whole block you have inclusion you have execution um you kind of don't even need to know anything about PBS or Mev in this context like you could be just like a Standard Market maker you could buy a whole block and you could then choose to um decompose uh it into its um individual Parts as you like um and so what this invariably does in the process is you know whereas today we have kind of four block Builders you know arguably it basically opens up this entire process to you know dozens if not you know like 40 or 50 like sophisticated people who just want to Market make on blocks and so um I have a

slide in terms of some of those Dynamics shortly so the the idea here is that people are trading these commitment products at and then finally at 11.8 seconds or whatever that threshold is then everyone will have submitted the actual transactions um to that block to to us and then there's some type of building that's going on okay now in this process you know how is this built do we then basically um distribute this information externally to other builders again neutral Builders do we distribute it externally to neutral Builders who might combine it with their own order flow maybe but the difficulty is the execution pre-ms if we have execution

pre comps in that block which are then negated or challenged or conflicted by someone else's private order flow then it makes it difficult for us to kind of to optimize and honor that execution precom in the first place so um this is a part that we're struggling to figure out right now the optimal approach is to keep it internally but would love to work with some some of the block Builders to see how could we share this with them as well but uh again there is some difficulty around the the execution pre-com part one more slide um entire blocks so I I know there's some discussion around you know is it uh good or bad um but um the way we see it

happening in practice and you know this is for discussion is again you know right now we have this process where four people are basically controlling what goes into the blocks we see this as shifting from four to maybe like 40 people a bunch of which are just like very financially driven and just market makers they'll just buy the blocks and then they'll then strip them apart and in this process if I am let's just say GSR for example like I'm a I'm a market maker GSR I can buy a block at some point in time and I can hold on to that block until you know that you know 10 11 second point where maybe I can use it entirely for myself and I can make

more money out of that block so how much the block is worth versus what I purchased maybe you know I I can I can do that alternatively maybe I'm actually not good at monetizing what's in this block maybe I I don't know the the various Mev opportunities out there my my my book is limited I don't need to do trades in which case I would decompose or I would strip that block into its individual components and I can do that because I can see people bidding for execution pre-ms I can see people bidding for inclusion pre-ms so at any point in time I know that there's these people who want parts of these blocks and I can basically look at those prices versus my own price itself um and so that basically gives me the option again

to to decompose it and strip it to its parts and of course if there's really no bids anywhere then you know might lose money in this process but um but at least I don't lose everything and that's where again the importance of that secondary Market comes into play and so there's some questions on you know Telegram and elsewhere but the the dynamic I think this is the last slide yeah I'll come back to that but the dynamic here is largely that we've intended to capture basically everything through this Gateway and within that process there's uh it it's a very efficient process you know we can basically optimize the entire thing however however fast like a modified R Builder

can work um but then you know maybe raises questions you know how independent are we versus the proposer so you know we would attest that we're neutral but you know maybe there's some way that we could um you know make that a bit more um build some more trust from that from that element but um but this was the way we basically put together how we can optimize block space and so one final slide here is basically starting with the the optimization exercise you know we feel like a a block that is part of a consecutive block that's worth the most uh following that you have like an isolated block if I cany an entire block that gives me

optionality to do whatever I want with that block again or strip it into its individual components which may be worth more or less um if uh that optionality if I don't have that optionality and I just have a mix of execution guarantees pre-ms and Mev boost then that's worth less and then you know as you kind of strip away the components it's worth less and less so I think this is kind of the the logic gates to determine you know what is the what maximizes the value of the block um but of course there are considerations are around like what does this mean for ethereum and so you know what is what is our role in building and supporting this you know do we do we enable people to buy multiple blocks um uh is that for us to decide is

it for our Dow to decide is it for is It ultimately for the validators to decide um but um you know ultimately we've built this technology we want it to work in the best interests of the community and we'd like to see kind of what that means so I'll stop there I rambled for 30 minutes I'm sure some people are pretty bored but I'll I'll stop there great thank you so much Kevin so I guess we have about 25 minutes for Q&A if we have any questions feel free to raise your hand or just unmute and and get get speaking I mean for me one of the the

one of the takeaways here is that uh with commitments there is an opportunity to make blocks more valuable and there there's several different reasons for that one is that we're dealing with unconditional commitments as opposed to to conditional ones uh we're we're unifying the market and so there's more price signals that kind of go in in in a single single auction and there's there's less leakage um there's also an opportunity for multi slot but that's you know potentially extremely controversial um and yeah there's maybe also you know the the the possibility to to get premium from from from the ux benefits of pre-confirmation

or or maybe from the hedging benefits of selling block space uh futures for for rollups that buy uh block space in BK so all in all like I guess my intuition would be that if we have a sufficiently um you know mature market then we we might see an adoption curve similar to MEF boost right so MEF boost went from 10% adoption that emerge to you know 90% today over a period of you know a couple years and uh and hopefully or yeah maybe we see a similar thing with uh with commitments have your hand up sorry go

ahead uh yeah so just really quickly earlier you mentioned that this enables different kind of commitments that you can't get today with the builder Market could you explain some of what those are because I couldn't come up with with any I'd like to plead the fifth on that one at the moment okay possible um there's uh so we uh we we have a pretty good Network amongst uh a mix of uh market makers and Traders um they basically came up with a whole Suite of these other commitments which I I was like well anything is kind of theoretically possible but um you know we also want to be good custodians of

like what we're doing here uh and the broader ecosystem and so it it's uh there's some in some in those territories uh in that territory but um I I'll kind of leave it there for now yeah all right thanks um okay I have another question if uh if no one has one um so you talked about how you know this uh Marketplace could potentially be used for not only

ethereum but also based rollups and and actual L like non- based rollups in the future um I guess I'm trying to think how so like for example a lot of rollups today don't use the typical PBS pip line and they're very controlled in how they sequence or who who they let sequence um so I don't know maybe you could just give some more detail on how you envision it working with like non- based l2s yeah so the I guess the question is who can be a sequencer or not it's basically where we can where we can basically do exactly what we're doing on those chains or l2s that we can kind of unify the market between here and there um and so that's that's kind of the the

broader goal um I think Bas is straightforward uh the L2 side we're we're looking into it and on the surface we don't think there's anything that really prevents us from merging somehow uh your B your L1 ethereum rollup with the with the non- based um L2 as well um you know there might be different timing associated with with the two but um you know at the end of the day once you have a commitment you it's kind of like an asset uh you can almost like tokenize it and trade it and it can kind of transfer from one place to another so I see this kind of future where commitments are a un Universal kind of language between all the l1s and l2s um and you can kind of like pull settlement and liquidity from all these at the same time um yeah

so there's this there's this intersection between kind of all this commitment market and all the intent stuff going on as well when you put the two together there's some good stuff there we need to unpack it a little bit more but um very excited to explore that we have a question from Francesco if I go right the base case scenario for E gas is that every proposer delegates to e gas so you can have consistent execution precom and also enable Market

maker to trade the commitments ultimately if gas would become the sole Builder on ethereum question mark isn't centralizing yes yeah that's centralizing um the coming back to the this part here U you know the goal is to maximize fees at least for the purpose of you know how we've designed it you know we could um we could we could uh remove our self hypothetically as a centralizing centralizing entity um by let's say open sourcing what we're doing so let's say now we have multiple people doing what we're doing with perhaps like almost as good infrastructure devops and software um so in that case you know now you have

multiple people doing it but then you basically lose um largely anything to do with multi-block um uh you know M multiblock mechanics um and then you also take uh risk as well so the issue with um one of the issues we had is that uh just around like fragmentation and adoption so a lot of our endpoints are kind of your your typical API endpoints as if you're going to trade um on like okx or binance or uh what have you so we just want to make it as easy for people to adopt but when you have Traders Mev and you know like a bunch of people who now need to integrate with multiple other exchanges that creates kind of hurdles and fragmentation for them and then that invariably then creates again more fragmentation just

means lower fees for validators so yes to your question but open to exploring decentralizing if you know validators the community our Dow says like hey you guys need to decentralize more and like reduce control or reduce like centralization yeah we're also exploring like a fallbacks as well so in the event from a um just like a devops p devops perspective like AWS goes down like what happens uh we haven't fully fleshed that out as well but um but realistically that's like you know 0.1% or you know whatever that uh that frequency is I have a question um is it correct to

say that running one of your gateways is essentially pure devops assuming that the code is open source there's no need for you know connection to dexes or fancy algorithms or you know Capital no no so so uh short answer is no the capital part becomes difficult so let's say that there are um 20 different gateways right so in this case there's 20 different gateways now how are the proposers posting collateral to back stop their commitments do they now need to post collateral to 20 different gateways because if I have if there's 20 different gateways and there's only a

12C block again it's the whole the the logistics around managing that collateral process and the slashing process it gets a little bit more complicated and potentially more Capital intensive for these guys I don't know how that would work within an igen layer context maybe there'd be like a unifying collateral contract which we all subscribe to and we kind of all agree to but um but uh yes it would largely just be a devops to your to your question and then in that case um you know coming coming to La's questions how that may evolve is that uh where is it here is that if there were multiple gateways hypothetically mult multiple

gateways might have different types of commitments right so you know we might have a certain type of execution guarantee with a certain number of conditions someone else might have some really nasty stuff in there or someone might have some better stuff right but you kind of get maybe you might have more diversity more choice but uh perhaps more risk as well yeah some like the the current path that we went down for the the collateral is that uh we have this notion of a um of a deposit contract where the proposers can deposit collateral on chain and that

would be collateral which would work for all the all the gateways uh equally and all the gateways would would know follow a similar standard to make those commitments so that uh you know you can have slashing conditions across all the gateways simultaneously and the nice thing is that you know the the collateral can be reused across gateways and it can also be reused across validators so if you control many different validators uh you can have you know for example one e of collateral across across all the all the validators MH I mean work go ahead go ahead no no go ahead I was going to say that would work for us conceptually so the way we have it is just it's a collateral smart contract um um but then

one one of the issues is that when you have multiple gateways then um if the gateways have different types of products uh have different types of commitments or products then basically what is the what is the right amount of collateral for those contracts and so right now we have one uh we assume uh for collateral you need to post one eth per one million block space so you need to post 30 million sorry 30 eth as collateral um per per slot um but then uh uh what we were looking into is that when you get into some of the uh special execution guarantees you probably need to post a lot more eth but um you know things we can discuss you can also not you can also

not post more eth and then the buyer can just factor in the the default likelihood and price it into the contract itself but uh TBD something else we were looking at was the ability to delegate or pull your collateral as well so to the extent there were other validators that think similar to you you could almost be able to post more collateral kind of on on demand um yeah and that was largely for the uh for the solo stakers because the solo stakers don't have 30 eth but maybe they uh maybe collectively we can kind of create pools of validators and uh we would Slash from that pool versus from any specific

individual so in terms of of sizing collateral like the the thinking that I have is that um you can post whatever amount of collateral you want it doesn't only have to be if it can be like any any asset um and then the the type of commitments that the Gateway is able to sell is is is a function of how much collateral you have um and uh you know my My Hope and you know this is kind of roughly um you know confirmed by the fact that you know most of the block are worth less than than than one e today from if you look at at at MEF boost that almost all the time let's say 99% of the time

for 99% of slots you only need you know roughly one e of collateral and so it's only like in these these tail cases where uh you need large amounts of collateral but if you don't have the large amounts of collateral then that's fine you just you know opt out of providing these commitments for very large purchases for example the the problem there is that those one 0.1% of times are the times also when people probably want to buy the commitments as well right so in periods of like extreme volatility that's like especially when you want that commitment so that's um it's just a Counterpoint to that yeah I mean I guess this is where like maybe we uh we disagree a little

bit so my view is that the main buyer of these these short dated uh commitments will be will be users and wallets for for pre- confirmations and that will happen you know roughly it will be roughly spread across all slots uh there won't be like some slots where you there's there's tons of of requests um and on on the Searcher side the my thinking here is that as a Searcher as an arbitrager I'm only willing to pay for commitment at fair price minus Epsilon where Epsilon is my is my Edge is my Arbitrage opportunity whereas the Gateway on behalf of the proposer is only ever willing to sell a

commitment at Fair Value Plus Epsilon where the Epsilon is is is the cut that you know the Gateway and the proposer combined uh make and so if if you have rational actors on both sides um then then then the the trade of the commitment would would never happen uh we we can't hear you Kevin I don't know if you acknowledge the oh we can't hear you no I'm still

here okay perfect yeah yeah uh someone just messaged me as well from who's in the group here as well but um we're on um we we've released all our docs on our website uh eth gas um in terms of timing we'll have uh basically our test net will be probably next week sometime so we can ping the group to actually uh do this on test net and again if there's any neutral

builders that want to work with us let me know we're we're not experts in that space so on this slide here um like this one or yeah the the thinking that that I have is that you know you have the Gateway in the center and whenever it's making a commitment it's basically broadcasting that commitment to all the builders whether or not they're they're neutral and then at the at the very end of the slot or towards the end of the slot the Gateway says here this was my very last commitment and so now all the the builders you know have end commitments that they have to satisfy and you know if we're working with with pre-confirmation then most of the time

these commitments will be top of block commitments and so really what we're asking the other builders to do is just build the the bottom of the block that that's still respects the the top of the block as as constrained by the the commitments from the from the Gateway uh so I I think that works in the near- term but with some of the discussions around those the types of commitments again I think that some other yeah I guess that the risk is just that there's some trades that go into the block that then violate or you know conflict with the with the with the guarantee with with the execution guarantee but um

well it's always the Builder's resp responsibility that's basically the non-gateway Builder to respect whatever was done so if there if they have you know private orderflow that conflicts with the with the commitments then tough luck they won't be able to to put it on chain yeah I guess there was a thought that we um uh at at the moment we would be broadcasting it we are then exposing um substantially you know the majority of all the what was then private transaction flow we would be then exposing it to all the builders and at that point then perhaps like integrated Builders would use that information to potentially Force certain execution

guarantees or users to uh force their transactions to fail so we want to kind of study that part further where again where we where typically the the transactions are all private but if in this case we're broadcasting it to non neut neutral Builders um then there might be bad doctors who would you know Force failure of certain contracts uh Drew in in your uh Drew to your question on what is a neutral Builder yes they don't they they're not part of a prop uh they're not part of like a hft firm they don't have their own trading business they're just a a builder yeah uh question from Harry execution uh

guarantees have to be top of the block doesn't it take away the sexx Arbitrage profits which indirectly reduce the overall me of a block um I don't think they have to be at the top of the block I mean they could be at the bottom of the block as well um there also may be cases where it could be in the middle of a block you know depending on what is at at the top at the top the block um doesn't take away I think execution guarante uh I think it depends what people are putting into an execution guarantee and that may take away some of

the Arbitrage profits but um um yeah I guess I'm not fully understanding that question Harry but you know feel free to message me and we can we can chat about about that I think the question is that the Gateway should make sure to only sell execution commitments if the price that they're selling at is greater than the opportunity cost of of selling it and so if the opportunity cost is like reducing a sexx Arbitrage you know by one e then the Gateway needs to make sure that it's it's it's selling the preon at least one e one e plus

Epsilon yeah um yeah I guess in that case we would just uh let the market decide I think what the price is there yeah sorry maybe it's terminology I'm not fully um not fully grasping that well I guess it's it's about setting the reserve price so if you just let the market do its thing then sometimes it will go under you know the fair reserve price and so really the Gateway needs to set the reserve price so that the the proposer always gets Fair Value Plus upset

on but but in my case the fair price is the market price um so if you take a case where there is like one bidder you know arguably that's um the person is put in like a low bid um but you know if there's a if there's an environment where you have like multiple people at the table with bids then that would be that would be kind of the fair price right so um irrespective of what they think of what a proposer thinks the fair price is you know the fair price is obviously what the market would bear and so if everyone's if everyone's only willing to bid like threeway for a certain type of commitment um you know the proposer is then expressing a View and they're saying that I know more than

the collective intelligence of the market to Value this block differently and I think that's that's not a a utility maximizing decision for a proposer to set kind of arbitrary prices I I think the proposer in that case should basically say I think the decision logic should be around we want to sell you know X Y and Z or Zed of products um when there's a sufficient amount of Market information um when there's a sufficient the definition of sufficient might be you know when there's at least three bids or you know something to that effect um but um yeah for them to say like no the price just isn't good enough I mean do they know more than the collective intelligence of all the players out there you know in most cases

I'd say no most the validators don't have that um that ability so I agree in the end game where you know all the Searchers all the builders all the alpha is kind of concentrated on this auction but there's going to be a transition period from MEF boost to to this and initially during the bootstrapping phase there might be like a lot less Alpha you know in this in this commitment Market than that currently is in MEF boost and so it might be a suboptimal move to to make the transition as a proposer if I don't have guarantees that it is net profitable for me to connect to this new auction Which is less mature uh true so I think that's not uh

distinct to us NE neily I think it's kind of everyone um who's looking to kind of enter that block space so um yeah that's kind of like a cold start issue I think all that we and other participants that are building kind of commitment platforms can do is to draw a sufficient number of parties to the table and get a sufficient number of validators or proposers to say you know we we support this initiative we're happy to take less than the optimal income over a week you know whatever that period is and over that week then basically uh you know Searchers Traders everyone else who wants to transact in a certain Block they'll be forced to go

through this Gateway all right so so what is that teething period it's like maybe a week or a month I don't know um but I I think that's practically how it might evolve for you know whether it's us or or anyone else I mean another issue is that there's a you know and grows super linearly over time and so if you're selling you know 12 1 second chunks you know that might be worth less than selling one big chunk of 12 seconds and so if if you're catering for pre- confirmations where the whole point is to just break down the slot into these these shreds right which might be only a few milliseconds

long then you know the the natural bidding mechanics won't be enough like you you need to have some sort of of guarantee to the proposer that that that selling the parts is is greater than than selling the whole perhaps Yeah in our case we we the only thing that we can do is provide the products that are inter interchangeable so whether it's a block or whether it's a constituents we think that the collective Market has more experience than any proposer and so the collective Market together should be able to buy these and bid for them and strip them down at whatever price so um in that process we think that the

proposer would never need to decide um what is a fair price for either a pre-com or a block or an execution guarantee uh they just know that it's a competitive process and a unified process across across all participants all the time but again the um I think I'm I'm kind of going at circles but you know it's it just comes back to like is that a is it a fair price is purely a function of is there sufficient information um in the Market at that point in time and and as you pointed in the early days there may not be that sufficient information okay uh we've reached the

the top of the hour thank you so much Kevin uh for your presentation and Q&A