The Multichain Future - TOKEN2049 Singapore 2022

hey everyone yeah so um nice to be here at token 2049 uh gonna be talking about the multi-chain future so um maybe just a brief introduction um like I'm doing research at the blog and like I focus on like layer twos crochet and stuff so uh really excited to meet you guys and um yeah maybe you can introduce yourself right hello Ryan I'll I'll jump over for a second you good yeah okay oh there you go back up Mike all right hello all right uh I'm Ryan Zurich co-founder and CTO of layer zero Labs layer zero is an omni chain interoperability protocol that lets uh applications on disparate blockchains communicate with each other uh things that could be built on top of layer zero are uh like bridges on the Chain nfts cross-train governance things like that we also built the first bridge on top of layer zero Stargate which is a native asset bridge on uh currently on seven EVMS hey guys my name is heart lamber I'm the founder of a project called Uma uh Universal Market access we are an optimistic Oracle uh what that means is we're an oracle that can trustlessly and permissionlessly uh get any verifiable data on chain uh and one use case the reason why I'm on this panel one use case for Uma is that we can talk between chains and we can verify data on other blockchains um trustlessly and permissionlessly and because of that we like uh Ryan at layer zero we actually built a cross Chain Bridge called across and we can talk about that too but across is uh is a cross Chain Bridge it's a native asset Bridge like Stargate uh that focuses on Capital efficiency uh and trying to be the most Capital efficient Bridge which we think will ultimately lead to the lowest fees and best usage my name is Richard Muir head height yeah everybody can hear me great uh they're killing it on the intros uh here it's got a lot of fantastic energy in the room um I'm uh managing partner of fabric Ventures so been focused on uh web 3 150 since 2017 um kicked off investing back in 2013 actually but you know um I'm extremely feel extremely privileged to be sitting up here with these guys who are very very deep in uh you know the um the details of what's going on this particular space um uh my history dates back to uh wrestling with the configuration of of Cisco's routers or routers as some people like to say um which of course you know made their name by being multi-protocol and so what's interesting to me is a an investor um is is there's a couple of things one is trying to look at you know this whole Evolution from you know Bitcoin is the only blockchain to oh ethereum's kind of got it right it's nicely kind of you know programmable and uh touring complete and we can do great things on it through oh actually there's a bunch of drawbacks who didn't have on anyone ever compete oh that's happening to oh my gosh we've got Bridges and you know and baskets of bridges and all sorts of things on top of that um and trying to navigate well what is this you know is there an analogy with the development of programming language or or uh of you know infrastructure as a service as we know it today or or routing and Communications that helps us work out where great startups are going to arise and and I I do think that we're a really exciting inflection point in this particular space where there's a super talented people in the space it's a big Market but yet it's still uncertain and I'm hoping to learn a bit more in this tap exactly where the use case is going to emerge yeah so I guess um yeah uh like to find out you know what are your guys like Vision about uh how the multi-chain future will look like you know like we've seen so many like layers layer ones and layer twos uh emerged just over the last year alone so like uh in future like towards the future what do you guys think you know like the blockchain space will look like uh you know where we have a lot of layer one still or well we all converge on like uh you know a few settlement layers what are your thoughts yeah I mean I think it's clear it's going to be a multi-chain future uh there's just tons of different trade-offs that uh you have to make between each chain and depending on your application there's different use cases right some gaming applications need speed uh like uh like Solana or you know the future Aptos and so easy some need a high level of security like ethereum uh likely layer twos uh that where you get that security and have Roll-Ups or potential in the future uh and they all offer different benefits and so uh building uh applications of the future will be multi-j and likely where people will build uh you know application that exists on multiple chains and takes advantage of the speed of say uh Solana and then the security of ethereum where you have keep your assets and say the storage of an R weave or ipfs I think like there's it's still so so early what's becoming clear to me is that there are like zones of sovereignty to steal vitalik's phrase where there's kind of like this evm ecosystems around ethereum and there's the salon ecosystem and then there's uh you know this kind of Cosmos ecosystem but what this multi-chain future looks like I think so depends on the usage patterns that we just don't know about um people talk about like the app chain thesis right now all these different applications are going to run on their own chain maybe it's a layer two that rolls up to ethereum maybe it's a cosmos uh app chain um but like if that happens we're going to have a very specific multi-chain future where I think people are bridging between these app chains frequently um I I think it's just so early and we don't have the data to understand all the all the plays here yet um but very clearly we are going to have different systems that people bridge between it's just for me is it going to be a super high frequency thing or is it going to be like hey I moved to this chain and I stay there a while um yeah yeah so so the perspective maybe I'd like to bring on that is that we still see a lot of um so you've got the beginning of applications particularly in the D5 space that have managed to you know span across multiple chains and um like Ave would be an example that's on a few different ones um but you know generally speaking if you're thinking about uh the next hot area that's really taking off it's in gaming and if you're a kind of a sort of AAA kind of Studio who's trying to work out who's trying to sort of got trade-offs in terms of time to Market and and how beautiful they make the game and say well um in general they're sort of saying well I want to keep it as simple as possible and then let me just stick to one one chain um and um I guess what that leads me to believe is that you're gonna have to get to the point when there's enough games that are out there that are competing in terms of the user experience and therefore going to need to kind of fine tune like you know whatever exactly that you're selling some transactions over here you're onboarding some some players over here you're maybe even running your physics engine on a blockchain over here if we've heard of some people doing that I'm trying to work out how that would work um and because that end result is a better game for the users and it doesn't feel like we're quite at that point yet um and you know I think there's some fundamental challenges we were talking backstage about the uh um atomicity um that you can achieve in one blockchain that is pretty hard to achieve when you're spanning multiple blockchains yeah I mean so just to rip on that too the other thing that's interesting in this bear Market um gas fees are down and I think I'm curious if Ryan agrees on this too but I think high gas fees is like a forcing function for the multi-chain future the higher gas fees go the more it's going to be like we got to find ways to scale it's just going to be like pushing really hard to have people expand horizontally into the chains so I I think that's like an interesting point um but I mean gas fees are down now and so if you're building there thinking oh this is how it's always going to be the next bull market you're in for a surprise uh and so then you're putting a spot do I where do am I going to build that so you're going to pick another layer one or a layer two and will that be here in the next couple years and then are you gonna put yourself in a spot where you pick one and that chain dies what happens to your users how are you migrating over uh thinking on the Chain first and thinking you know even having having it built in where you have messaging uh and you're able to say use like layer zero for messaging and you don't actually deploy another chains but you leave that out and that optionality so that uh as the these layer ones in Layer Two is mature uh you can decide if you're going to stay where you are if you're gonna expand and use the benefits of some of those other chains um yeah I could I can buy that back in um in web one I can remember the phrase I used to use was actually The Meta the better of course in this case it's now the meta the better as long as you're not actually matching um but but so so so exactly for developers if you can have that convenience and not having to you know make those trade-offs early and make those choices makes a lot of sense but there is another um if we're trying to work out how we get to that point um there's a sort of school of thought um that say look at at the moment and this is evidenced by what you've both sort of built is that it sort of value Bridges it's all about and that at the end of the day that liquidity you know is Will trump everything forget messaging and kind of like you know uh composing apps on a sort of cross-chain basis it's just liquidity and therefore the kind of you know the deepest liquidity that's Wormhole or whatever it is is gonna is gonna win out and that's gonna also happen in the short term do you agree with that I mean that that's assuming that like all crypto is just about you know defy and assets right and it's not about providing also building other applications on top like you know social media applications or uh you know gaming applications that are actually fun to play and for the game and like secondary part of that is oh cool I own my actual assets and I could sell them later if I want to but the fact that I have control over them and not the game that uh is issuing them to me means that they're mine and mine forever uh is like that's a different the next that's your onboarding the next you know 100 000 100 million users right is is making it usable not just making it about uh you know dgen crypto so so there is like an Arnold you can interrupt us with questions here too oh yeah um yeah because I think one of the things that like uh I was thinking about when you guys are talking about all these uh design and the trade-offs uh one question that comes to my mind is that you know uh you guys are they at zero have like um Stargate whereas Uma has built like across right so what's like the difference between the two and like you know what are your design philosophies and the trade-offs that you're going to make you know yeah I I I'll let I'll let him speak to you uh across so Stargate uh is a native asset bridge that is actually it's intended use even though consumers use it is actually to be infrastructure uh we built layer Zero's Imaging protocol and the first thought we had when we built layer zero is oh we need the ability to move assets right and we don't believe uh wrap asset rapt assets are dangerous right that's essentially it's not really rap it's an i we call it an IOU you owe someone owes you money on the other side if that you know that contract's hacked gets hacked which you know happens every month nowadays uh you know that IOU is worthless and so we wanted to have native asset pools on each chain and you were when you were crossing the bridge or any like like you mean you cross any other Bridge or uh you were basically that's the only time you're at risk and when you get off the bridge you hold the asset right there's no uh you know no IOU no potential chase case where you could get hacked um and so we built that and the concept with it is that you could send and call and so uh Sushi swap did this recently did it they launched Sushi X swap where you can go any asset to any asset on any chain and this is a perfect use case of what Stargate was built for in that uh what they're really doing is swapping on a sushi swap so to usdc or usdt bridging that usdt the other side and then swapping on Sushi swap on the other chain uh to get whatever asset they want and from the user's perspective it's one click one transaction and they get up on they get that asset on the other side whatever they want which is the best user experience you could provide so there's a lot here that um Ryan I really agree on and I actually I'll go back to just one way I explained it as I look at there as being two types of bridges in this world one I call the lock and mint bridge where I take an acid I lock it on its native chain and I mint an IOU like a representation of that asset on the other chain sorry yeah all right sorry I'm into representation of it on the other chain and that model I think is also dangerous um we're in agreement there there's some use cases where you need to do it there is no such thing as Solana on ethereum if you want to do that you need some sort of representation of it but I think I I prefer like Ryan our design philosophy is we prefer this native asset concept where instead the bridge is actually a convenience metric method to make it easy to move between canonical versions of the asset um on different chains um so I think that's an important thing for the audience to understand the risks of these different Bridges across then we designed to be the most Capital efficient Bridge we could possibly do while maintaining kind of Maximum uh trustlessness and decentralization and in our mind Capital efficiency here just means we can do more volume with less tvl in the bridge itself and the thought here is that further reduces our like attack area or Surface attack surface and um maybe I'll Loop this back though to Richard's Point around just like like what is the actual crunk concrete use case for all this cross-chain stuff today and you know it is asset bridging that's why layer zero build Stargate uh that's why we're focused on a cross not as an arbitrary message bridge but as uh an asset bridge and um I I think it is really interesting to Think Through uh and maybe there's a thesis here we could begin to kind of form of two different multi-chain Futures one where you aren't you're you're executing primarily on one layer one layer two one execution environment and bridging there is primarily assets and occasional messages and the other is this kind of like Omni future where I'm actually building on this infrastructure that can kind of under the surface it abstracts away and uses different chains at different times and I think that's a really interesting future or debate bait to have between those two Futures well I one thing I want to say like we uh we didn't build surrogate because we think asset bridges are like what people need right we built Stargate as this like underlying utility because what's the purpose of bridging you want to move some acid over there not so you could just hold asset over there right you don't want to hold the usdc oh I want to hold it on uh you know uh polygon for some reason right I want to do something with it right and so going over a bridge uh uh and just hold it that's just a multi-step that you're adding right at the at the go do something at the Swap to a bridgeable asset then I have to leave my that page uh Bridge there and then go to another you know am or Farm on another side but being able to put that together and roll that all up in one single transaction uh is the purpose of messaging right that's what we that's what layer zero is doing and did Stargate is just further enabling that where you can now become this defy Lego where every single D5 application that's cross-chain or wants to be cross-trained Builds on top of Stargate and is able to compose it and hopefully you can get educated from from the answer one thing that we wrestle with a lot is it's one thing uh should we say kind of D5 for D5 kind of moving existing defy assets around um and that's fine you may want to be able to borrow against them and and and hope you know go um leverage long eth um who makes gone most gone okay back hello you're just louder now okay good okay and and so that's fine but but what if we want to try and get some you know get heaven forbid some real world Assets in here um what what do you what does that mean for for your environment if you've got like uh maybe an invoice that's been tokenized for example or obviously some I mean you can have nftr but nft anything else and you want to take that asset and use it as collateral for for loans and and what if you go a step further and you've got automated market makers for those nfts and a protocol for whereby the nft might try and uh sell itself does that argue in favor of a um that level of interoperability argue in favor of a more multi-chain world or you know what have you contemplated that kind of use case I mean I think makers already doing like some of this right where they have like real or elastic that they were they're pulling in uh I don't know if that's necessarily a requirement for like to pull in real world assets if you need to use multi-jain it could be an interesting one um where uh you know you're you're if you're pulling in constant real world assets uh you could tokenize them in some type of nft or something and be able to move that between chains right um and go to marketplaces like because you know you're pulling it onto it say you're pulling onto a faster chain for some reason like you want because you're gonna have a lot of volume where you're pulling in a lot of them and they're transferring hands a lot but there might be a mark the use case of uh multi-chain there would be you want to be able to move those assets to another chain and be able to that has potential uh you know more liquidity right access to liquidity is what multi like this Omni chain multi-chain future uh enables where you can it may be today it's ethereum but ethereum has a expensive cost and it could be a layer two in the future and being able to move there freely uh and have access to that makes your assets more valuable it is maybe for me at least related questions like this privacy preservation have a role to play here uh like from your perspective tried to kind of mask be able to understand some parameter from some transaction on one blockchain but and know that that parameter is you know factually correct on another is that is that also part of your requirements I mean I think that's in I don't know if requirements the right word but privacy preservation is something people want right you're going to be able you just don't want everything public all the time um and I think that that's unless we completely change the world we the way we think about Society where we just like have our asset less super public for everybody to see all the time that's not going to be the type of thing that people are going to want um so and I think that is like something that will evolve or iterate into and I yeah um yeah I know the Privacy for for us it's just messaging so if someone wants to go from a privacy chain to a public chain and they and and they're for us we're just moving bytes payloads so like how they encrypt there how they do that is is their business uh but layer zero uh the actual message and being able to see how it moves between chains because layer zero exists and validates completely on chain will always be like that part is public yeah I think privacy is something that uh many users are concerned about and um I think I might be able to like say something about that here because uh I do a bit of research into zero knowledge proofs validity proof so like I think that there might be a space where you know you see a cross chain messaging that is being validated with validated validity proofs in future although I think that's you know much further down the line I think what I'm more interested in is you know the potential of like uh cross chain applications that you guys were talking about and something that like I've been thinking I've been meaning to ask uh you know what do you think some of like the biggest challenges would be like that you face you know if you're gonna build something that has uh different uh functions on different chains so let's just say for example you have a game that uh maybe stores the nft or itself on ethereum but like the game interactions are done on Solana like that like you know what are some of the challenges that you know uh a messaging layer itself of my face and it says a different like Pro overall design pattern so most you know developers in the space are used to atomicity right and so having to think about it a different way uh it's just like there's a learning curve there right um just like you know most when someone goes to GPU programming and learning how to write parallel code uh is it is there's some learning curve there and it's just a different there's different design patterns that get used there that you would never use uh on like a normal uh normal CPU all right yeah yeah and I think this is something we were talking about backstage that I think is really important for people to zoom out and look at um we use the word let's see if I can do it atomicity um but Atomic transactions are um what enable like if we go go for a very D5 specific concrete example um if you're going to liquidate somebody on defy in in current D5 protocols you can do it with zero Capital because you can take a flash loan and repay the flash loan at the end of the transaction and it's all executed atomically or it's not executed at all um and Flash loans like I come from a traditional Finance background flash loans are a crazy concept like I can borrow basically infinite Capital to do whatever I want to do and I don't have to post anything it's like wild right um so if we start moving across chain and you you break that you can't do atomic transactions in the same way we were debating backstage whether there is a way to do it and there may be right but regardless you are um breaking some of the programming patterns that people already have and so I think the challenge here and again this goes into like the thesis for what this multi-chain future looks like it's possible um developers don't want to break their current design patterns and then they are building primarily in one execution environment and the cross chain stuff is around bridging assets or occasionally bridging messages it's also possible that there's use cases where they really do want a program in a different way and I think gaming is a good example of something like that that like isn't really done high performance gaming isn't done on Current blockchain Designs maybe it would be um in that form of a multi-chain future um so I think it's really fascinating to think through that too yeah those are really interesting thanks thanks for thanks for sharing yeah so I have a question for Richard you know more on the Venture side right like uh how how do you you know navigate the uncertainties when it comes to like deciding you know which of these layer ones or layer twos are like uh investable what what I like some of like your biggest doubts when it comes to you know evaluating uh which chain might you know stay on which chain will not like what uh you know your thoughts on that so yeah good question so oops sorry but I mean uh obviously it's been um sort of investing in layer ones maybe several layer ones investing in layer twos maybe several layer twos has turned out to be a reasonably good bet especially if you you know it came in at the right time and remembered to get off the the train at the at the right time as well um uh so in that case you you could say um a lot of people look look smart um but weren't necessarily um I I think um and so for that game when we're trying to think about infrastructural components and work out which one uh we want to want to work with um I mean I think it's all of the the normal kind of criteria that you have as a kind of venture investor you're primarily spending time with people um and and understanding whether or not these are the the at that point technologists who are going to build uh the better mousetrap um and then your um your uh perhaps getting into the next level of understanding whether there's you know there's this uh you know hydrogenous sharded approach or is it a you know a homogeneous sharded approach and you know what are the trade-offs involved in that which obviously was one at at one point in time um and uh maybe you're hedging your bet I could say this though from our perspective that we tend to back uh founding teams and try and be as loyal as possible to them so it actually has presented a little bit of a interesting conversation at least for us because we kind of tend to go deep with a certain team and then want to um and see if we can support that ecosystem but in a multi-chain world that seems like that's a little hard because there's components everywhere so we had to kind of um uh adapt our stance a little bit um then if you switch to the kind of the the um the app side we've certainly seen um uh the kind of starkware ecosystem on ethereum do you know extremely um uh well and that has been primarily in our portfolio that's primarily been driven by choices from the the application teams themselves so you know we're led by them as to which uh ones they want to uh work on we wouldn't be coming in with a Viewpoint per se um but rather maybe to link it back to this point about kind of development patterns and and engineering trade-offs and and so forth um and even you know specifically engineering trade-offs about complexity and security risk um uh I think the biggest uh opportunity for the industry the ecosystem the whole remains getting developers just to come in and develop anything in the web 3 environment um and you know to just start um I think uh anyone who's had a security breach is trying to keep things as simple as possible uh although arguably if you're not doing well it doesn't matter you've had a security breach and if you're doing well then then then often you can you can navigate out of that that particular hole um but I I think um uh I was curious actually and I don't mean to kind of entirely bounce the question back um when you're thinking about your go to market and we wrestled with this a little bit backstage as well you know you strike me in both cases to be quite Broad and horizontal you know um or maybe you're just currently serving the D5 use cases we've been covering oh should we be doing more in the ecosystem to try and identify the uh say The Gaming use cases and work more tightly with them and then think more thoroughly about the go to market that always used to be a really important part of uh early stage Venture thinking that through and I actually doesn't come up as as much as I think in the more mature web 2 markets that I've ever invested in historically so we started layer zero we thought it was mostly going to be D5 that was building on us but actually 50 of our inbounds since we've come up well since we've launched has been gaming nfts uh and like like any nft marketplaces and stuff uh and a lot of them right are building games that uh need their own specific chain right because they're mostly afraid of even fast chains right that some nft drops gonna bring the chain to its knees during that time and uh affect everyone and how many of them are already using multiple chains those say again how many of them are using they've got their own chain but they how many of them so what they're what they're doing and and why like they want layer zero it's like great when you make a gaming chain you have nfts right or you're gonna have items in the game and now those users are going to play that game want to be able to sell it so they build a Marketplace and now you need to get capital on there in and out right and those aren't your core crop density right you're doing things that you don't you don't want to be doing you're Reinventing something that someone else already does a lot better than you're ever going to do and so layer 0 enables it so that I could move my assets off that chain and have them natively exist everywhere else so I could bring it to say ethereum and list them on openc and then have access to all that Capital liquidity and and the users who are buying them are then going to buy those nfts and bring them back to the game so now I have more users I'm able to touch and I can go to all of those chains at notches ethereum and go you know Solana and I can go uh to Magic Eden or something or openc there and then have access to those users and bring people into the game and then I give my users the ability uh on like so they give their users ability on the gaming chain to then uh you know go wherever they want and have the asset is more valuable to them because it's not just like this closed ecosystem uh I could bring my nft or whatever whatever asset I have off that chain to be clear right you guys aren't bridging nfts yourself right this is like you're just again doing the messaging of this and letting the application do the nft bridging so we came up with this a standard we call uh Omni chain fungible tokens which is ofts we call them and then Ami oh amichane nfts right or onfts and uh it's basically what what happens is these uh creators deploy the same contract on every chain and it's like a 721 for nfts or 1155 uh and then whatever that version is on Solana right the Solana has a built-in nft and they communicate with each other so uh whether and it's up to how they want to do it whether they lock and mint or they burn or or you know hold this destroy it I guess and put it on the other side right you're moving between chains uh and it's the key there is that since they're deploying it and there's their smart contracts it's not a wrapped asset it's it's it's you know maintains that provenance because when they deploy their uh nft in the first place they deploy it on all chains and they're connected via layer zero messaging yeah I bring it up because like I think it just also shows how early we are I don't think there's yet like the design pattern of how you're supposed to move an nft between chains all figured out oh yeah it so we were because the demand was so high for these gaming uh changes like well how do we move these nfts we realized we needed to make something yeah so we created that uh what we call the Own St standard uh and um basically they need that it was out of necessity and and they are using it uh now and and a lot of new gaming chains Avalanche subnets polygon supernets uh are coming out with this uh built in yeah it's still just such an interesting concept too because like if the idea of there not being one settlement place for this thing like a native asset does make most sense it does make conceptually the most sense for it to live in one spot and then if it's something that's non-fungible like an nft moving that around can kind of make your head hurt but these are all the use cases that I think are still getting like sorted out and figured out right now sure and I think the most important part about it's not how they move but how it's viewed right and so making sure that these exchanges can uh or or marketplaces can properly see like so like open Zoo they could see if they could they if there's a agree to prone way that we all read uh between all the chains where an nft unami chain nft is deployed right if they don't they can see okay there's this many and these are the ones on on ethereum and this is on arbitrum and this one's on optimism and they showed us one collection right like that that's like that's the problem that you're you're kind of getting at their heart is like if there's no unified standard like there is with erc20s and you know uh 721s 1155s uh then the the Mark it's hard for these uh marketplaces to kind of view all the different versions that are out there which is why we we went ahead and made the standard yeah like I can definitely see the utility that you know like a messaging protocol like that can provide um I guess one uh last question that I might have here is uh with regards to with such a you know messaging layer right like what exactly is the value of cruel mechanism when it comes to interacting you know across chains because like on ethereum you pay gas for transaction but like when you're using uh layer 0 messaging protocol um how exactly does value accrue on layer zero so layer zero uh messaging I guess is so the way it works is there's an Oracle and a real error and oral Oracle emulator is a kind of misnomers because they could be anything it's an open and permissionless system uh and an oracle's job is to move a block header uh from one gene to the other and then a relayer's job is to move a transaction proof and validate other side now those relay the Oracle and relay are abstracting gas in the destination chain and so gas is volatile especially uh can be a volatile special ethereum and so they have to buffer in some type of uh you know uh I guess amount for for volatility and so usually it's a couple percent maybe up to five percent to buffer that and so uh that is where their the profit for oracles and relayers lie is that that volatility and and buffering properly and having enough to because they have to deliver it no matter what and so if they don't properly price it or they underpriced uh they're eating that cost on the destination chain and and and not making uh and losing money so it's a balancing act and then since it's open and permissionless and we will have um I guess metrics for them to choose from when you're picking an oracle or a relayer uh you know how much more are how much are they uh how fast are they what are they waiting because they you know they under charged and now they're waiting for price of gas to go down so they're increasing adding latency right things like that uh and uh allows for this Marketplace to then get pricing in the right place uh for applications to choose from and that's where uh the layer zero Labs runs a relay and then like that's where the layer zero could make money right right so yeah well I was just going to say so like for an asset Bridge it's pretty obvious how you could make how the protocol could make money they like take a sliver of the asset getting bridged and that makes conceptual sense um it's actually interesting here Ryan described this so if I'm going to summarize it like you're basically have the protocol slightly overcharge gas fees um at some level but you're saying it's an open architecture so where does that go it doesn't necessarily go to the protocol but maybe layer zero Labs runs a relayer that does a good job and people then pay for it it's an interesting question though of like I don't think there is a model that's established for how to extract fees from a messaging protocol here because you don't know what messaging what's going across it right it could be uh you don't know the value the bytes payload so how do you properly price it you can't like uh and and so it matters basically comes down to pricing on gas like we're paying another side we're performing some service and there's some risk that's being taken it's a market maker you're in gas and so figuring out what that what that is uh and and buffering in the in some some amount of profit there right um yeah so I guess that's it refreshing to hear um questions about business models in this environment but yeah interesting bear marketing questions yeah so uh uh that this is for us yeah thank you thank you guys so much thank you everyone [Applause] [Music]