Crosschain DEXs: The Next Frontier | Season 6 Episode 6

Transcript

oh my God I agree so strong right like the lock and Brit Min bridge is super I think it's super dangerous I'm I'm not a fan and I don't like it and I think that there is such risk of vendor lockin um and centralization risks that are super scary and are the antithesis of like what the systems we're trying to build are all right everyone we will be back to the program in just a moment but before we do I want to share something that blockworks has been cooking up for these last couple of months March of this coming year in London blockworks is hosting D London the largest institutionally focused conference in all of crypto Goldman JP Morgan 72 all in one room so you can know what the big money is doing so click the link at the bottom of this episode it'll take you

right over to the homepage and use Bell 20 for 20% off I will see you in Sunny London Town in March this episode is brought to you by Maverick protocol a suite of liquidity tools built around an Innovative amm Maverick helps token projects Dow treasuries LPS or basically anyone in defi shape their liquidity with efficiency and flexibility how you might ask stick around and you're going to be hearing about them more later now on with the show all right everyone welcome back to another episode of bell curve before we jump in quick disclaimer the views expressed by my co-host today are their personal views and they do not represent the views of any organization with which the co-hosts are associated with uh nothing in the episode is construed or relied upon as Financial technical tax legal or other advice you

know the deal now let's jump into the episode hey everyone welcome back to another episode of bell curve today Dan and I interview Hart lamber of Uma and across this was a really fun one we explicitly covered interoperability in the role that dexes play there so we talked about some of the different flavors of bridges that exist today why some might be better than others and how interoperability all going play out but Hart is also just a very deep thinker about Market structure and the future of crypto in this whole design space so a lot of it we focused on the inter operability and deck side of things but we also went down some really fun rabbit holes as well so without any further Ado let's go right to the program all right everyone welcome back to another episode of bell curve today uh I'm joined by my

co-host Dan Robinson and we've got Hart lamur on with us today so Hart welcome to the program thanks guys thanks for having me thanks coming this is uh this is cool um we were just describing that Hart and Dan have never been on a podcast together so technically this is not only a podcast episode but this is a moment in uh podcasting history so um we'll I'll just try to soak in moment and enjoy it together our relationship deepens Dan hard and I hard and I have been have been friends in the industry for a long time and so uh and uh have had a lot of great conversations but none of them recorded to date so Dan has also never invested in anything I've built so there's also that sore Point we' never had to do that disclaimer before disclaimer disclaimer

uh Paradigm is not invested in Hart I'm sorry Hart you you maybe borrowed ideas but have haven't invested yes many times all right guys so um why don't we just start I want to want to uh kind of Broach this subject that Dan and I have covered um sort of in a secondary capacity so far this season but I want to directly address it with this episode which is this idea of inoperability and I want to spend a lot of time digging into the intersection between interop and dex's and the role that Dex is play in interoperability um but before we even get into that specific intersection just at a high level heart can you kind of walk us through the current state of interoperability in crypto why it's been such a thorny sort of issue that you know why why haven't we figured this out

by now yeah well I definitely can't walk through the whole state of interop because this is like a weird long multi- podcast thing to probably go through um and the semantics here make it all awkward um but what I would say is like why this is a problem um as we know uh achieving consensus isn't the easiest thing right like a there's we built blockchains and spent years doing it to figure this out and interop is trying to achieve consensus on one chain about what happened on the other chain that's like not an easy thing to do um and I think a lot of the approaches I think there's a lot of room in this design space for how we should go about doing

it many different approaches and um people are still exploring all the trade-offs um I might keep that as like a really high level issue of why interrupts interrupts hard um curious if I don't know if Dan has anything yeah well just to um I think Hart will agree with this but I do think you know one one type of interoperability certainly involves that but there are many ways to get what people would would like to get out of interoperability without having to having to achieve consensus and um getting value from one chain to another doesn't necessarily have to involve um any kind of like high stakes consensus on one chain about what happened on the other chain it could be done you know this and this goes back to htlcs in some of the original ways that that um people have thought about interoperability between blockchains

where the only parties that matter are actually like the two parties that are doing the Swap and they and they just have to come up with a protocol for actually getting value across chains so I do think there are you know if we brought in it a bit there's a lot of interesting um there's a huge interesting design space here around how how minimal can we make the requirements for actually being able to get what users want out of interoperability which is getting um uh value from one chain to another and just my FRA My framing for interoperability I think would be and you know I think of a lot of things in in Dex terms and of course since this is a DEX season we we'll I think we'll be covering some of the in uh intersection here but um I think of interoperability

uh from a user perspective as as very similar to what they want from a DEX where with a decentralized exchange often what you want is you have one asset um possibly on on one chain and you want to get it uh get value in another asset on that same chain and with interrup ability often you have one asset on a particular chain and what you want is uh is some value on another chain on some on some destination chain and so that you could think of that as um as a message being passed which is often how people think about interoperability but you could also think about it as a as a decentralized exchange transaction where I'm I'm starting with I have something that I have and I there's something that I want um and I think if you think about it in those ways you can often think about

find find different solutions that are maybe uh yeah are less technically imposing very much agree on that point and I think actually that might be a good way of also focusing like this conversation where I am much more interested in how you move value between blockchains not how you move messages um and not to say that messaging is not important it's got its use case but like 90% of what people are doing on blockchains today is like basically Dex transfers and I think like 90 plus perent of what people are going to be going to want to be doing in terms of inov is moving value it's moving tokens and the messaging thing is a I think if you focus on moving value your design space uh is different and we

can talk about I mean some of the trade-offs here and I think intense based architectures and kind of what we're doing um I think that there's much more clever Solutions if you focus on that sub that that subset of interoperability so I tend to agree before we even get into some of the Nuance there could you guys just give for folks in the audience who might not be as deep as you guys are into the interop sort of conversation you know when people even talk about something like messages even even for me actually I would love to just get a sense of like what is a message within the context of how maybe like a layer zero or sort of General messaging protocol might might talk about it and then even within like what is kind of the Orthodoxy right now within um sort of the bridging or you

know cross-chain messaging ecosystem like what are people generally trying to optimize for just kind of catch us up on on what people are talking about yeah I'll I'll jump in and Dan interrupt right so a message message is I want to send some message I want to send hello world from one chain to another got to figure out how to do that cool um I actually think the baser Primitive here is one blockchain wants to read state from another blockchain so you write hello State on blockchain a and you want to wait for blockchain B to read hello hello world on blockchain a um and so that read state is the basic primitive and you want to figure out how to get that back and forth um when it comes though to bridging specifically um

I'll give you like kind of my framework for how I look about this and then Dan can critique it and kind of add to it um but what bridging is usually around bridging assets or let's focus it around bridging assets and in that context the idea typically is I deposit asset on chain a and then typically some message happens that goes to chain B and I get asset on chain B um and I'm going to frame this General framework the term I kind of want to start using is actually a trafi term term called like delivery versus payment comes from like trading bonds so I'm going to deliver payment and uh I'm gonna make delivery of like

my cash and I'm gonna receive my bonds and think it really is like an exchange term so I'm gonna make a delivery on chaina receive payment on chain B um and again the the kind of standard way of thinking about this is that these this this deposit transaction happens and I send message it's kind of messy middle thing um um and this there's definitely two flavors to this one flavor is these like lock and mint Bridges so I lock asset in my deposit contract send a message and I mint like a wrapped or synthetic version of that token on the other chain um that's one flavor that I I quite dislike because of the security

implications there um of this like big lock box of assets On the Origin chain um but you can do a flavor of this that's also um like what's called a liquidity Network where I have pools of the Native assets on both chains I deposit in a pool on chain a send message and release asset on chain B um and this delivery versus payments sort of structure is the way I describe basically every bridge out there except for across um across does something differently uh does does something different with like an intense Arch Ure um maybe I'll pause there Dan what do you think you agree with that yes so I

think uh uh that's a good description I think of this of this lock and mint structure and I think that's you know I think it's a very important distinction between how bridges work and hard and I you know I'm preaching back to the choir here because I've learned a lot of you know how I think about this from heartt but um this you know this way of bridging assets where you where you lock an asset on one chain and and Ming on the other chain I think of as like it's like bridging uh via like a particle where you have you know you have some some um asset and now you've got a representation of that asset on the destination chain you've like moved the actual asset over in some sense like this the sort of like matter has been moved from one chain to another um but

there's another type of of uh interoperability um when someone just wants to move value across chains which I was alluding to before which I think of more like a wave so you know like the wave um particle duality in quantum physics um I think of this as like as moving value like a wave where I'm I'm doing a a trade on one chain and what matters is that I get some value some particular asset that I want on the other chain but at no point am I um am I locking up the value that I have on this chain and Mining some representation of it on the other chain I'm actually just um swapping in some sense with someone and we have we have some mechanism in there to guarantee that there's that there's atomicity here um so one one thing to note about this like wave uh uh

method the the latter method here is that it doesn't actually require um any message should be passed between chains it doesn't require an Oracle and we know this because like one of the first ways we've ever that's ever existed of having interoperability across blockchains um htlcs was invented by um in one form by Satoshi Nakamoto in a um in a post uh like over 10 years ago and this is uh an idea where where if I have a counterparty I can just lock up value on one um on one chain um he locks a value on the other chain we do this two-phase commit protocol basically where Alice locks on that one chain Bob lock on the other chain and then um Alice is able to unlock Bob's value on Bob on Bob's chain and that gives Bob

the information he needs to unlock Alice's value on um on the other chain and that's one that's a a way of actually doing a a fair exchange across These Chains without having any uh uh any kind of message pass between between those chains and so I think that I think the the wave theory and I'm curious for heart's take on where exactly AC cross fits between in this Duality here um but I think one interesting thing is this allows a lot of different kinds of interoperability that aren't possible um or makes it a lot easier for example with with the with the matter thing uh the the particle uh uh kind of interoperability you need two-way message passing between these two chains you need some Oracle for the state of the source chain on uh or the origin chain on the destination chain you need

some Oracle for the state of the destination chain on the source chain in order to be able to bridge the asset back um and you don't need any you you need none of it technically for um uh for having uh the sort of like wave uh Fair exchange with using hlc's but if you have one wave message passing you can then there's a lot that you could do that also looks more like a wave uh and that's that's how unof X works for example um which I can which I can get into at a moment but I'm curious to Hart does that does that Duality make sense to you and do you would you how would you characterize AC cross in that u in that pattern yeah I mean not i' go deep I'm not sure I love The Duality analogy but I do agree that there's two different

ways of doing this and like Mike I don't know if you know but like Dan can teach a master class on htlc so you can Google like all these old like talks he gave uh from way back um and the htlc thing the the downside to this right is just the user had to do a bunch of actions like you have to go back and forth with this two-phase commit and it kind of has this like user experience that never really worked um I I do think across is closer to this but the the trick which Unis swap X also uses is that we introduce a third party here um so unlike all these delivery versus payment Bridge designs we introduce a third party that Unis swap X calls a filler a cross calls a

relayer cow swap calls a solver the this third party though they have this unique prop property where they're actually willing to like lend assets on the destination chain um on the users behalf and um what happens here is we've like completely kind of broken the the way the the the the particle Bridge Works where the users um users deposit um it's not released to the relayer until we ver like the relayer goes and basically fills the user with their own Capital with their own money on the destination chain and the user assets On the Origin chain are not released until we verify that the

relayer actually did what they said they were going going to do and it's this really interesting tradeoff where by introducing this third-party actor that's willing to function in this like sort of Market maker role um by extending Capital we're able to break a lot of the usability issues that htlcs had um and actually offer I I think a strictly better per uh user experience uh in the design and we can get way deep into why that is and how that works what do you think Dan yeah so uh so absolutely I agree and you know my I had a talk on htlcs a few years ago that I if people are curious about hlcs would encourage you to check out called hdlc is considered harmful and it's it's it's

very anti anti- htlc as a as a general method for for doing cross chain um uh swaps but in that in that talk I actually do propose a way another way of doing cross-chain swaps that doesn't and and offchain payment networks that doesn't involve um any oracles between chains that's this packetized payment approach that we don't want to get into necessarily but um uh it's you know another protocol that doesn't yeah allows you to trade value from one chain to another without having to um uh yeah without necessarily having to do uh have have an oracle between them but yeah so I think what uh what's hard hard is pointing out here is that both both across and Unis swap X um which are two methods of doing this kind of value exchange across chains uh both of them

take advantage of the fact that you may have a an oracle trusted by at least the two parties um that are that are doing this swap um an oracle for WEA settlement has happened on the Destin ation chain um an oracle that is readable by the by the origin chain and so if you have a one-way a one-way message passing Bridge well will suffice for this uh if you are able if the origin chain is able to the chain where the where the swapper has their asset and wants to move somewhere else uh if you're if that chain is able to read state from the destination chain in one way or another in a way that is that is satisfactory at least to the participants in this transaction um then uh that it's possible for a for the swapper to um have a much more efficient method

than yeah than an hdlc um which which requires much more interaction between these parties and yeah here you can have sort of a minimal a minimal interaction in order to transfer value I do want to mention since I think neither of us would naturally uh be in favor of the or or be be uh shutting to the rooftops the benefits of a um of the sort of two-way lock and mint St Bridge one advantage that the lock and Minal Bridge does have is the ability to end up with a representation of the of the assets from the orig chain on the destination chain so if you want to for example be trading eth on Solana you need something like this or you could May or you could have a synthetic but that that you know you have you have Capital efficiency um uh uh downsides there um but if you want to

just have like a capital efficient synthetic you know Bitcoin on um on ethereum for example you know you it helps to be able to have this two-way message passing bridge and be able to like lock a lot a lot of Bitcoin on one on one chain and uh mint it on the other um so that you can actually have so you can go trade with that with that and do and do various things with the asset on the destination chain so that is that is a benefit that you don't get with these one-way Bridges so not that basically if you have no no so if we think about this just in terms of what are the requirements on the on the um on the platforms if you have no oracles at all you can do an hdlc you can do this kind of complicated two-phase commit uh

procedure between these two Traders in order to transfer value from one chain to another um but it's kind of inefficient um it has it has these downsides um uh it's it's takes takes multiple steps and like if one party aborts in the middle of it it can be just a disaster for the other party um locks of their capital for a while if you have a um a one-way message passing bridge then you can do these protocols like Unis X and across and if you have a two-way message passing Bridge uh then you can do these lock and mint protocols but yeah actually hard I think U helps uh radicalize me against the two-way message passing Bridges do you want to uh talk a little bit before we get into the details of the of the oneway of things like AC cross do you want to talk

about the lock and mint uh protocols a bit uh for a second yeah sure I think one thing that would be clarifying though for the audience and and all this is um when you're using the word Oracle here you're you're using it basically any message that's passed between chains is an oracle like an or so and I don't think that's actually the way um everybody thinks about it in the space although I agree with this um every message sent between chains is an oracle you're reading from another chain it's information that's not native to that blockchain um one Nuance I'll throw in here is um that most layer most layer twos and many blockchains do have like a canonical Oracle which a canonical

message passing bridge that is designed to have no or minimal additional trust assumptions over the the chain itself and so um one of the things like where a cross tries to go is we we basically only make use of this isn't entirely true but we're we're trying to go back into how we only use canonical Bridge oracles or the canonical message layer we're trying to use the most secure message layer at all times even if it's expensive or slow um we can talk about that more too so yeah Dan what do you want to talk about in terms of Lock and M well yeah actually I think that that's a good one so thinking about can Bridge passing like where how do we get Bridges um so I think there are uh there are a

few ways and sort of building up there's some it's interesting there's some oracles that are have the same uh security requirements as the as the origin chain of the message and there's some that have the same security requirements as the destination chain of the messages and and either of those sometimes are called canonical Bridges so if we take something like IBC the the you know inter blockchain communication protocol that uh um uh the COS the cosmos ecosystem uses that's one where the the bridge is canonical because it's as secure as the as the origin chain of the message so if you want to send a message from one Cosmos chain to another the consensus protocol for the origin chain um for the chain that's sending the message basically is signing that message and that's that's how IBC works is that it's so it's it's as secure as

just like the core proof of State consensus protocol for that for that chain um and then if you think about something like uh uh like a rollup any any2 essentially on um on ethereum there's a there's a bridge you know there's a complicated uh fault proof or validity proof based bridge to get from the rollup to the L1 to get from the L2 to the L1 and that's what's enforced by by all like the you know the security there but I think a really underrated interesting Bridge um with uh with l2s is the L1 to L2 Bridge like what what is that bridge like what does it how does it work like everybody every one of them has a built-in um bridge for it but like why how does that actually how is it secure why is it secure um and I think ultimately you know this this gets into

like some of the philosophical uh debates about what a rollup is but I think ultimately it's it's secure because the consensus the protocol that determines what's on the L2 everybody who's validating the L2 just has to validate the L1 and if the L1 were to roll back then the L2 State just would roll back um Canon and I think you know ultimately in some sense it's ultimately all enforced by the uh in some ways by the by the bridge back like the bridge back depends on um the the bridge there having having validated all the messages that's part of the of the validity for for that bridge but ultimately what it really means is just the L2 is built so that it is always reading what the L1 state is and so that's that's a case where the canonical you know it's canonical because it is actually just built into

whatever is running the L2 is that it has to respect the the L1 State and and messages from the L1 um um but then you know so I think those those are some kinds of canonical Bridges there um and then I think there's there's a lot of non-canonical bridges which are you could think of usually there there's some kind oh I'm sorry one more type is is if you can prove the consensus mechanism of the um of the L1 on uh of one chain on another chain and so I think one one great example of those is a bridge from Bitcoin to any other chain and you know you can build a a bridge from Bitcoin to any smart contract chain um by validating the the proof of work that was done on on the Bitcoin chain you actually just validate the block headers and so BTC relay one of the

first Bitcoin um to ethereum Bridges was was done this way validates actually the uh this the block headers to make sure that there's been a lot of of proof of work done on the on the Chain so yeah like so Mike I think like we should pull ourselves out as as you see we started this with like what is interop right like and I was gonna I was gonna wait for you respa there yeah I here's here's the way that I would sum the situation up so far right so there there sort of we can divide our bridges into two separate categories here maybe type one and type two bridges that each each have its set of sort of pros and cons um let let's broadly call type one sort of the lock and mint model um and the disadvantage of this and that's

basically mechanically what's going on under the hood I want to draw actually a through line here back to an episode that we did last season on liquid staking with Sam kazian who drew this comparison in between stable coins uh liquid staking protocols and bridges actually where you have one side um on sort of the demand side of those protocols people that deposit assets and what these protocols sort of spit out is a sort of a claimable uh a claimable liability and you're sort of managing um assets and liability similarly to like a bank or a money market fund or something like that um and that's what stablecoin issuers do that's what liquid sing protocols do and that's what these sort of lockin mint Bridges do and the disadvantage of that obviously is the vast majority of hacks whenever you hear

about these billions of dollars are getting lost in defi these mostly come from these these lock and mint Bridge hacks because there's this Honeypot of locked tokens on this one blockchain the advantage of that though is that if you don't want to just exchange value in between ethereum and salana sort of trust zones but you actually want to go into the salana ecosystem and use your ethereum then this is kind of the way that you have to do it right so those are kind of like pros and cons of of that that approach um and Dan was just getting into some of the different flavors of what even those constructions could look like but then there's this other maybe we call it like the the particle type Bridges where actually we say you know what uh instead of trying to it's not that important most users

don't actually want to take their ethereum and move it into salono land they just want to be able to swap between ethereum and Native ethereum and Native salana and for that for those sorts of instances we can use a completely different sort of minimize construction where even like the metrics for a bridge like that are are totally different right we're not trying to optimize for you know in the through the lens of someone like a nomad or I don't want to pick on anyone specific bridge but the idea would be like I want more assets to manage here so I want more messages being passed from one chain to the other but that's not the case for this for this new style of bridges so hey everyone wanted to take a quick second to shout out this season's partner Maverick protocol now many of you probably know Maverick as an

Innovative amm which they are but in reality they're a lot more than that as well Maverick is a suite of tools for defi users and builders that allows them to put liquidity where it will get the most work done since Maverick launched in March theyve been gobbling up market share and at the time of this recording which is the end of September on a trailing 7-Day volume basis Maverick is now a top three decks by volume and they support over 50% of the volume on the L2 ZK sync a chain Maverick enables LPS and token pairs to process higher volume with limited tvl which allows them to support some of the highest levels of capital efficiency for lsts like rap Ste another very cool feature is something called Maverick boosted positions so

that allows protoc calls looking to bootstrap their token liquidity to Target the shape of liquidity of any token pair with surgical Precision Maverick is backed by some of the leading institutions in crypto Founders fund panta coinbase Ventures Finance labs they are all backing Maverick in their Vision to revolutionize the next generation of defi daps and helping them build their liquidity in all market conditions click the link at the bottom of this episode let them know that I sent you thanks guys maybe we can get into and and that would be like kind of Unis swap x uh and then there elements it seems like across is almost like a little bit of a daywalker there are elements of both um from what it sounds like so maybe maybe uh Dan can I can I pick on you actually I think we've done

a couple episodes and I've never actually asked you to give me a formal definition of of Unis swap X but if if you kind of walk us through the the formal definition of Unis swap X how it fits in this second camp and then we can maybe segue into more intense based models and and where that all fits into this whole equation Dan I'll just add we should you should make clear that this is like the cross-chain unit swap X yeah single chain Unos X is is similar but different right yeah yeah yeah yeah yes we've talked a little about how um single chain Unis swop X Works um and you know it's an it's an offchain uh protocol where uh one with a swapper who wants to swap into another asset they sign a limit order or an intent um they sign something uh that uh allows allows

any filler to actually um fill their trade and and and deliver them some assets um in exchange for the they that they were offering um so that's that's how it works on one chain on um multi-chain Unis act which is described in the white people but isn't hasn't been implemented yet or or um it's not it's not live um cross un back work similarly except the when the order is filled it has to be delivered on on another chain um so someone signs an offer um a filler can can claim that order they and they have to claim it on the on the origin chain the chain where the swapper um wants to basically like send their value from this where the swap ER already has assets but then when they delivered that on that uh order they have to deliver it on some other

chain and the general idea is they they delivered on some other chain and then um they are able to claim on the on the origin chain based on um based on on some kind of proof that they've that they've made delivery and in I think the the general form the I think the most the most powerful and useful form of of unop X that's an optimistic proof is that is that um after a little time they can just claim on um uh On the Origin chain because nobody has has challenged that they that they've made that delivery um but the the general idea here it requires a a one-way message passing Bridge it require or or Oracle for the destination chain the chain where the you swapper wants to have assets it requires an oracle for that on the on the source chain um on the on the

origin chain uh that is readable by by the by the or chain where the where the user already has assets and so that's that's the general um uh outline of Unis a new XX is is agnostic to what that actual what that Oracle is or how it works um it could be a a uh canonical bridge between an L1 and L2 or an L2 and an L1 it could be some chain of canonical Bridges between chains or it could be some um some other kind of bridge you know a a a some kind of multi Bridge or or something it really just all that matters is that the filler and the and the swapper are are willing to agree on it and I like this design uh because as the unop x white paper references it's inspired by inspired by across certainly having thought about

this a lot um I think the overall goal here is actually to minimize the number of messages we're sending across these Bridges and um uh the way uh Unos swap X and across work is really quite different and the way I like to think about it is layers of this sort of uh intent-based Bridge settlement system which is quite different from this delivery versus payment architecture um and I look at it as like these these three layers of the stack we we start with um the the order generation where the user creates their limit order or their intent if we want to use the Buzzy word um then there is this network of fillers Rel layers or solvers whatever you want to call them that are actually

the ones that front Capital to fill the user intent and the bottom layer is this settlement layer um where the user assets are escrowed until we verify that the order has been filled um and what's interesting about the structure here that's completely different from these like delivery versus payment uh type networks is we can actually do much of this without very much messaging almost minimal messaging and when we do do messaging we have the option of only using canonical oracles or canonical messages um messaging layers which I think offers a lot of security advantages while still maintaining like the the speed and usability um that that users are demanding and so maybe we can

go a bit deeper into that yeah so so I want to uh um dig in on that because I you've talked before about wanting to minimize the number of messages um but I I might think of it a little differently which is so with any if you have any Bridge right you're supposed you have a canonical bridge but it's it's slow um possibly because say it's a bridge over from an L2 to an L1 that requires like a onewe you know fault prooof uh challenge window or something um or it's really expensive because it requires yeah consultation um of you know multi or something like that it requires a lot of onchain transaction to do um you can basically always bootstrap this into a into a reasonably fast and and very cheap bridge by just making it

optimistic and this is another idea that I think that heartt has been pushing for a long time and um is deeply sunk into me is that you can you can make a lot of things that are very slow or expensive um fastish and and cheap by making them optimistic and only and only having someone challenge them so is there you know is this just saying oh yeah we don't have to send messages on the bridge because in most in the happy case we just won't use it and we'll only use it when challenged there's actually two optimizations here um that fit with each other and um Dan I've pilled you hard on the optimization or so the optimistic approach um which is what I've done with like Uma protocol stuff for a long time um uh I've begun to pill you on this batched transaction approach too why we

batch things and I actually think the two of them fit together very elegantly and quite nicely um so like it's worth getting kind of technical here for a split second and it's actually worth going into some math here too so um first off um what is the like it's actually what's the cost of in this design we have this tradeoff so we're going to do this intent-based bridging design but the trade-off here is we have this relayer that's going to front money and so there's a loan a cost component here of fronting money um but to give you guys some numbers the cost cost of lending money at a 10% rate for one hour is a tenth of a basis point so that

means it costs like uh one1 100th of a penny to lend ,000 for an hour it's like a DI Minimus cost and I don't think people quite realize this and why this is so powerful so to the optimistic point to Dan's Point saying if we had a bridge that took an hour right but was really secure and let's we'll go into cost Savings in the second if we had a bridge that took an hour but is really secure I could fix that by building this intent-based framework where I have a third party lend money um almost instantly it comes across as almost Atomic for um a tenth of a basis point and then they collect their 10% yield

and we use the slow bridge to slowly verify things and we ret retain all the the the security advantages of this very slow bridge but give the user a fast experience for a very small amount of money so first of all Dan we'll go into like the optimistic stuff in a second but like agree kind of makes sense yeah yeah so I have a question about this I've actually had this question about all of these sort of like pre-confirmation type things what happens if there's ever a dispute there right in between the sort of fast confirmation that you'd get from sort of a market making type entity and the the confirmation process of a the settlement layer so okay very simply right um if the filler uh didn't send the user their funds uh they get their funds back after

an hour the fast the slow Bridge says okay they didn't get their money back and like there's different ways of kind of parameterizing this or constructing this and maybe you want to make it a little bit painful for a filler to not fill the user so maybe the filler actually has a bond so the user gets back like a 10% bonus or something like that if they don't fill the user and Unis swap X actually calls this the filler Bond and they've kind of have a design for how they this might work but you do something like that so it's painful for the the filler to not do what they said they're going to do um got it and conversely also if the filler just like um fills the wrong person or something like that the filler just doesn't get paid back like so if the

filler goes and like they don't they send the money to the wrong address for example then um the filler loses their funds and the user will get reimbursed with like this 10% bonus so you can design pretty pretty robust incentives to keep people honest here um cool okay so there's that so that's like a very cool shortcut how we've now taken a slow bridge and we've made it fast for what is like a very small amount of money um on the optimistic approach what Dan is saying here is like well and what what Uma protocol has done for a long time is like well okay let's let's say that we we don't even send a message we just have a challenge period

where we don't send a message between two chains we just say hey if the message didn't show up there's a challenge period And if somebody says I didn't get my funds and maybe anybody could say this depending on your design then and only then do we send a message proving whether funds were or were not sent and so if you design this correctly maybe everybody does what in the happy path everybody does what they said you never even send a message which is pretty cool um but then there's another aspect to this that is sort of this optimization AC cross has fallen into that I didn't really appreciate when we were building this but it's actually turned out to be really valuable um and this just goes

into like gas cost optimizations so right now in this design we have user funds are escrowed on the origin chain um filler fills the user quickly they're happy but then the user has to come back and ask for their funds out of this escrow account and there's like an onchain transaction associated with that like cost some money to do that and actually on like ethereum mainnet it costs like a non-trivial amount of money just to get your funds back out which increased user costs um so the observation here though is like well if I've already waited an hour if the relayer has already waited an hour to get their fill and it's cost them a tenth of a basis point what if they were to wait another hour um but then batch

together all of their repayments and instead of paying the gas costs say they did a thousand of these transactions instead of paying a thousand gas costs to reclaim their funds on the origin chain they do one and they get all their funds back in one optimization that that seems pretty cool and those gas costs are like non-trivial um they really add up and it's part of the reason why it crosses so fast and cheap and then across goes one step further where actually we aggregate we batch together the repayment instructions across all the chains so relays that fill these users they they get to batch the repayments on ethereum and arbitrum and optimism and polygon on

all these chains and they can get repaid in one step which has really substantial gas savings uh of how we do this whole thing um and so basically we've taken this intent based model where users are fronting money and then the relays are saving on all their gas costs by um by getting paid back in one lumpsum transaction which just amortizes all these costs over many many many fills kind of ranted there but does that make sense to you yeah so to sum up yeah I that's a good point so um if you have a cross Chain Bridge that that is slow and expensive you can make it a lot you can make it somewhat faster and a lot cheaper by making it both optimistic so you only actually have to uh go to the bridge if if something goes wrong and by

batching by having a lot of messages um sent uh in you know one in one uh message if needed I I think that I think that uh uh that makes sense I have yeah I think we hard and I can get into some quibbles on I think some of the um the micro structure of how these of how these things work but ultimately like I think all I do think all all bridges are going to end up um and all these Protocols are going to end up uh uh moving toward whatever the most the most efficient um uh minimal minimally trusted uh architecture is just because I think the competition on Cross chain um uh uh you know Bridges and and and costs will will end up being like it is with Dex so maybe maybe we should time to sort of take a take a step back and

talk about uh talk about some of that I think and and how we see the see the landscape playing out yeah and I guess on that I don't even know if we call this a bridge anymore like that's my kind of Point here like what is this architecture I go back to my three points it's like user like limit order cross-chain limit order or intent then Super competitive network of fillers that we want to have solve this and then this other layer that is your settlement layer and I think where Dan and I very much agree is um like if we want best execution for the user which is what we want uh we want to have the most efficient uh RFQ or orderflow auction system that gets

the user the best price for what they want to do um we want a hyperco competitive solver Market reler filler Market that's going to be really really competitive at filling that user and filling them quickly and by the way Dan we should talk about speed is the other thing we should do here too in a second um so so most efficient orderflow auction most efficient network of uh fillers or solvers that fill them that should be permissionless 14-year-old kid in their basement should be able to be one of these guys Ash should Jane Street like really interesting Market here and then the settlement layer at the bottom should be the least expensive but trusted settlement layer so whoever can

do the escrow and settlement stuff do optimizing gas costs do whatever they do at that layer um that should be the settlement layer that's used in this system to to just give the user the best execution and to me that's like where interop should go and that has a lot of really cool implications all right everyone we will be back to the program in just a moment but before we do I want to share something that blockworks has been cooking up for these last couple of months March of this coming year in London blockworks is hosting D London the largest institutionally focused conference in all of crypto we are gathering 12200 of the world's largest asset managers so thank trafi macro funds Crypton native funds big allocators and financial institutions so

Banks payment processors Etc all in one spot it's very rare to get the likes of Goldman JP Morgan 72 whatever all in one room so you can know what the big money is doing we're diving into the themes that they care about so we're talking about the intersection of macro and crypto where we are in the cycle real world assets so everything from stable coins to unchain treasur lues to tokenize assets it's going to be a blast but the other reason you really want to go it's London baby center of the world at one point you got Pub culture you got fish and chips great beer it's going to be a blast so because you're such great listeners to bell curve there's a Code Bell 20 that's going to get you 20% off so click the link at the bottom of this episode it'll take you right over to the

homepage you'll see all of our speakers and use Bell 20 for 20% off ticket price are going up soon make sure you go use that code I will see you in Sunny London Town in March let me ask you a question about that that middle layer uh which should be this really competitive uh market for sort of uh we call them Searcher Builders let's let's just call kind of hedge funds but it could be a Jane Street sort of hedge fund and then it could be some like 14-year-old kid in whatever country one of the one of the things though that um that sort of strikes me that the the Searchers or these fillers or solvers or whatever are going to have to do uh they're going to need inventory right to basically create those loans or sort of Warehouse that risk and and I guess my you know how do

you generally feel about I feel like if if you make that their role the extension of essentially credit um and that looks like a very Market makery type role of of warehousing and managing risk um how how could you as a as a 14-year-old with no inventory compete on something like that especially in sort of an offchain way where you might not have access to flash loans or something like that that was sort of a democratizing influence for Mev back in the early days of searching I mean I care a lot about that point Mike um and to be clear you can't do this without Capital like the reason why this structure works is that um those Sol the the I'm just use the word filler we'll use Dan's term you know swap X we we'll we'll call it filler

because in the yeah the filler here needs to have some money to front with their own capital and unless somebody's willing to trust them with like unsecured credit it's got to be their own money um that said what we've seen with a cross is super interesting so um and across has a a feature here that is different from the Unos swap X design where fillers when they do fill a user um they can get repaid on the same chain where they fill the user they don't have to just get repaid On the Origin chain they can actually get repaid anywhere in the network which helps them with inventory management so it's like I don't if I can't do that I need to have funds on every chain I'm trying to

provide liquidity for and I might end up then having CH funds in the wrong chain and I got to move them around and there's like a cost to that um but across by letting fillers re get repaid on the Chain they just sent funds we help like we don't solve it but we like mitigate it um and we really do have at the small fill size so like smaller orders it's hyperco competitive like hilariously competitive and it is really truly like some of these smaller guys are filling a lot of the small or ERS by doing really sophisticated things that we didn't even think of um so it's not like a perfect answer but given that many many many of the trades that are

going to happen here are $500 $1,000 you know 50 bucks when we start doing L2 to L2 stuff I actually am not sure it's as big as a of a problem as you may think I think it's possible that for relatively small amounts of capital or inventory you might be able to to do significant market share yeah and this is it's not a very satisfying answer but I do think at least in the short run um for in the cross chain environment I think a lot of you know traditional market makers are very wary about holding funds on a lot of different chains and so I think fillers might you know probably I imagine why you've seen some of this on across fillers might be more you know more risk uh you may have these more risk-loving sort of independent um fillers being more willing to go to

different chains in order to and holding you have to hold some inventory on other chains in order to to do these Bridges and people may just be more willing to do that in the long run once these things you know ultimately like hopefully we're trying to to make it not incredibly risky to be holding assets on these chains so that risk may not actually um last that long but then then I think we fall back to what I've been saying all season which is ultimately if the game is sufficiently competitive um and you and you don't have high barriers to entry I think it it might actually be the case that that sometimes uh more competitive parties just uh end up end up winning and that might be okay as long as again they can't actually extract that much value from that because it's a it's a hyper competitive

game and they have to they have to you know um uh they can't like sort of you know uh charge too much for for users because then someone else will come in and and out compete them I think ultimately as long as again there's is low barriers to entry that might still be an okay situation and here I think with cross chain liquidity provision it's like less maybe less uh danger to say the core consensus protocol of particular chain for example to have um to have this be something that's uh that's L Prov provided by professionalized actors although I don't know maybe there's some Downstream effects of it yeah and and I think that's where Dan and I agree and there will be people on crypto Twitter that disagree with this and I think rightfully so like but having no barriers to entry and perfect

competition if that's where you've got you're going to get great great execution in great fills and nobody's going to be able to extract excess rent and maybe at some point the 14-year-old kid of the basement just doesn't have the resources to do that but like that might be okay that might be an okay trade-off I I think it's some point these two two different conversations got intertwined and I'm not actually sure they deserve to be intertwined so you know on the one hand there's like how competitive do we want this process to be like are retail swappers getting their best execution that kind of got intertwined with this conversation and this idea that I think Unis swap especially V1 and V2 lit under everyone's you know under everyone's ass like hey a retail person can be a market

maker and honestly I feel like now one of the things that you and I are starting to ask Dan this season is like there's probably going to be a base of retail LPS hopefully or like sort of passive more retail LPS for a while but I don't know I feel like at some point um you know everyone in crypto loves this idea of democratizing Open permissionless Access but I do think it's probably a right and good question to ask like should just an average you know person be a market maker I think Market making is really hard um like if we want this this system to scale and be really large I'm not really sure that makes a whole lot of sense actually from just a first principal sort of thought so that that's kind of where I always end up shaking out on this like I I

understand that that people don't like this idea of kind of moving more things offchain but I I I sort of push back against the idea that you know retail people should somehow we should be designing systems so that retail people can be market makers I I'm not really I don't think that makes a lot of sense if you sat down and thought about it I also think especially in this context of cross chain bridging I think the the offchain the centralization risks of these sort of more offchain protocols um pale in comparison to the potential centralization risks of the sort of like onchain lock and mint style protocols where you have you know say one one canonical version of a of a bridged asset from uh from one chain that's being that's being sent to another I think if you have these onchain um Bridges those have much more significant

lock in effects they have much more significant um uh uh entry barriers to entry um and they potentially have have uh much greater risks if they're if they're compromised so think about what these with these offchain protocols um right like if one you know party happens to right now be be providing the most there's a a truly actually competitive system for for providing cross chain bridging where whatever filler is providing the best price always um ends up getting the fill um if that party disappears someone else maybe can come in to take their place someone you know if there's still a competitive um uh market for it if you instead have everybody uses one you know cross chain me message passing based bridge where like you know all their eth is is represented as a as some um locked you

know wrapped eth on the other chain if that party starts to charge a higher fee for um for bridging or if they if they become malicious or if they start um imposing particular you know like centralization constraints you know on who who's allowed to bridge or what you're allowed to do with the bridged asset um it's a lot harder for everybody to switch off of that um than one where you know you have this competitive system we not only is competitive But like everyone could be using a different one or you can you know like people may not even notice when this when this Feller actually disappears and people can start using a different one because you're relationship with a fellerer only lasts over you know like one hour it's the duration of transaction at most um and so I I think these you know these

kind of uh subtle centralization risks are important um and worth and worth noting but I think again we should we should also look at the very real centralization risks of of these kind of onchain um lock and minti bridges I'm sure I think heart I'm preaching to the choir certainly here with Hart oh my God I agree so strong right like the lock and Brint mint bridge is super I think it's super dangerous I'm I'm not a fan and I don't like it and I think that there is such risk of vendor lockin um and centralization risks that are super scary and are the antithesis of like what the systems we're trying to build are um we have shifted the conversation from your original question again Mike so it's like lock and mint bridges are dangerous and have centralization risks

on in terms of the 14-year-old kid in the basement being able to provide uh Market making Services look if they can compete with Jam Street and win and honestly ly one of them might be able to I don't know like go for it all I care about is perfect competition and no vendor lock in and we want systems that have perfect perfect competition and no vendor lock in yeah what I've got a question for you guys and I'm sort of thinking through this live which is always dangerous to do on a podcast but I sort of wonder what that ultimately I I heard um Preston Evans of sovereign Labs described this sort of framework for you know what he calls trust zones you know his his perspective was that like like look we're never going to fully solve

the bridging interupt problem in between these these different L1 ecosystems and instead we're going to have these kind of trust zones where you know like the sort of rollup ethereum main chain sort of ecosystem for something like eth and Salon will have its own one as well and on the one hand I I think I'm sort of torn in between this vision of you know at the end of the day it's you know sometime soon in the future like apps are just going to we're going to abstract away all this stuff underneath and you're not even going to notice any of this stuff but I guess what my question to you is is what is the time frame that you see that happening on because my user experience right as someone who like does stuff on chain you you do it by uh ecosystem you know you

start with like okay I've got some ethereum and I'm gonna go like play around what can I do oh I can go to a rollup and I can I can trade some stuff and I can you know what I mean and what I'm hearing from you is that there's a a fundamental problem with lock and Min right and what that restricts you on is you can't as an as a user of ethereum I can't take my ethereum and go use all the salana cool stuff right or all the cool stuff on on Cosmos so it does if we just accept that principle that lock and and mint Bridges and is not a really good way to solve interoperability then at least for the time being I kind of do see these at least from a user perspective right like I still do see my worlds as being segmented across these

different ecosystems and I'm curious if you guys agree with that or or push back on me and if so when do we get to this Vision that we've all been talking about about you decided to aack and you can trade across whatever or do whatever you want well you can trade across them but like I don't I don't get why I want my ethon salana I don't get that that doesn't really make sense to me what I do get but this is a different thing I do get that I might have like uh 10 grand of eth and I want a loan on salana fine okay that could be useful um I do get that but I don't get why I want I don't get the other version of it like trading eth like trying to do like

an eth uni trade like uni token trade on salana doesn't really make sense to me I don't know Dan what do you think yeah I mean I think bridges are sort of the least of the of the constraints right now like you know people right now you have to switch wallets you have to switch a lot of things uh to switch between one you know between in like the cosmos world or salana world or ethereum world so I think those those are a lot of the constraints right now um I think you know ultimately uh I I think you can go a long way with the with just sort of like getting paid in Soul or something um you know being able to swap your eth for soul and I think I think we will unlock quite a lot once you're once you're able to very easily move um value from one

chain to another um and like often yeah when you actually want to be doing stuff on on um salon and one of these other ecosystems you probably don't actually want to be doing it with e most likely you want to be doing it with soil or maybe with uscc or another or another thing um and we should actually we should maybe talk also about like the elephant in the room here which I think the US usdc bridging which is um a very very powerful uh thing but potentially you know uh uh kind of contributes to to some of the um the network effects that stable coin issuers have um but ultimately I think yeah I think the right now the idea that we want to actually be trading bridged assets on these other chains rather than going into native or going into other assets there I think is um I I sort of want to

interogate that a bit um yeah I do have I do have like like one thing I would want to do that I think is quite doable under this framework we're talking about is I sign an order on ethereum that says I want to go and uh do this action on the salon of blockchain that requires me to pay and S Soul like let's just say it's like the action is I wanted to go play Blackjack on this application on Salon of blackchain blockchain I have to pay and soul but I sign this intent on ethereum and it just gets handled for me like that makes sense to me and that's exactly what this framework we're talking about enables because you have the filler just go and do the thing on

the soul chain for you without you leaving your native wallet and if you extend this forward this goes a yeah sorry go ahead no no go go on go on that's interesting so every every transaction like I just do a lot of transactions on ethereum every time just basically paying out of my ethereum wallet yeah I mean Dan I don't know if you and I've I don't know if you and I have talked about this but something that I think is like when you take the intense stuff to its logical conclusion I just am signing messages that like let's just say I have a piggy bank like a venmo account um on a chain and I've actually thought about like you have venmo chain you have an account on venmo chain that has some funds in it and I just sign messages that say I want to

spend out of my venmo chain to do this action on um on optimism on Casino chain on salana whatever right and you have the network of solvers or fillers that are perfectly competitive they just go and do the thing for you and they get paid back out of your venmo chain after a delay right and the well sorry they get paid back after the delay your action happens almost instantaneously and if you think about that this is what I think is like a cool version for interop where I'm in one wallet signing one set of things and the solver network is the one that's just executing for me and then I think you're into a cool world of like chain abstraction that is fascinating I'm not

I'm not sure I believe that that that full extended version will be will be sufficiently efficient you still have to do some transaction on the source chain I would think every time although I guess you could have a payment Channel anyway this is interesting why well just just in order to like prevent me from double spending on the on the origin chain I'm gonna have to do some do something that's going to prevent me from from you know spending on on multiple on multiple things so you're gonna have to like somehow uh lock my eth on that chain um but it could be yeah sorry sorry your your your ven mode chain your ven mode chain would have to be like time locked where like I could still take all my money yeah so if I could like open a payment Channel with someone with someone who's doing that on

the other chain so maybe that's very interesting um but uh yeah I think ultimately look this yeah that which would be very cool and I I think this is true about the generalized intense thing it's like this would all be very cool but right now it's kind of scandalous that I can't even like make a payment to someone on another chain right now like there's no easy way for me to make it to pay someone so on a chain in fact the example that's really getting me agitated right now is usdt on Tron of all things um is that you know like apparently uh you know a lot a lot of places in the world you can pay someone in crypto you can pay someone in USD on Tron um at like a convenience store which is sort of wild that you can do this and you know but like I I don't

want to hold usdt on Tron personally but like maybe I would want to pay someone who wants to receive it and so like if I want to make payments if I want have my ethereum wallet like why can't I actually just scan one of those uh uh links and and make a payment to them um and so that's that's a very clear intent of just like I want to actually deliver money to somebody on this um on this chain that I would never want to touch with a 10- foot pull I don't want to have a Tron wallet even um uh let's say but like I do actually want to you know make a payment to someone who who does that should be completely abstract for me like I should just be able to pay anyone on any chain so Dan this example you bring up

is hilarious because I am here in assemble and um usdt on Tron is everywhere um like everywhere you could pay for it everywhere and you go to the Grand Bazar that's nuts Mike Mike like there's a whole other podcast like I went to the Grand Bazar and got a tour of these money businesses you can exchange $10 million of cash money like USD cash for Tron USD D on Tron in the grand Bazar and in symbol like they have C money carts they roll it's like absolutely wild um there's a whole other world there that's a different podcast but my point is it's a maor that podcast that's a nuts that's crazy next season next season should TR it's completely wild and like tun

like like Dan trun went and he actually did uh we should ask him about this he did buy something he like bought like a beer or something with like usdt on Tron um and I don't have a Tron wallet like I couldn't do this and they don't like they like it here a lot um but you're right we need a cross protocol to support Tron which is weird and I have to think about so that you could do have a relayer fill your beer purchase on the Tron blockchain near instantaneously um and okay it's like that'll go on our product road map um but we should be able to do that that's all right that I mean that's I mean that's a whole other that's a whole other

C together but um I I wanna I want to ask you heart because um you know maybe maybe just in closing here I would love to get a sense I always love talking to operators um and kind of the metrics and uh you know what you pay a lot of attention to so we've sort of talked about this idea of um reducing the amount of messages that get sent which kind of goes against the Orthodoxy of more traditional sort of Lock and mint Bridge designs you know you are really really big on this idea of capital efficiency so I would love to get get a sense from you about like what what do your customers find that are super important I would guess just the cost of bridging maybe time I know you tend to focus a lot on Capital efficiency and then if we could just get into a little

bit of the strategy of what it is to to run one of these Bridges and how do you compete and right and whether or not you see you know big Power laws across these different types of bridges as well um maybe we can just talk about that in closing yeah well okay what does the user care about they want to bridge assets uh canonical assets they want to bridge them as quickly as possible and as cheap as possible and those are the things I think the space needs to care about a lot more than anything else and capital efficiency here really matters um but It ultimately matters into like what are your costs here um versus speed and all that so in my example that an hour loan at a 10% rate costs a tenth of a basis point um in price terms um means

that okay there's that if you can use a loan to offer a user a faster experience and actually save money because your gas fees are cheaper when you're like um when you're using this intense based bridge design that's like a good tradeoff so our entire time in designing our protocol is how do we make the right tradeoff here to give the user like the long run sustainable cheapest bridge design um and like small shell but like Mike a cross bridging on a cross is cheaper than the canonical bridge on optimism or arbitrum or polygon like we the bridging costs are cheaper than the native canonical Bridge because of all of these gas efficiencies and it's way

way faster too so my metric is just like how do we make this thing cheap and how do we do smart things with money how do we actually use these loans effectively to give a great bridge experience what I don't think is useful is looking at the metric of like number of uh Bridge transactions like or number of messages passed number of messages passed like every message you pass is expensive and or in costly or secure it's like it's the wrong thing you want to do you want to stop sending messages because as Dan and I just walk through you can do a really elegant cheaper faster bridge design without sending any messages at all so you know I I kind of I kind of laugh a little bit like when Brian from

layer zero posts on Twitter the number of layer zero messages are like up and to the right I'm thinking you're just wasting gas Andor doing inefficient things and that's like not the right thing to to focus on so that that's my take all right La last two questions for you we were one I I cheated a little bit and asked you before before the podcast here but um you know my my mental model again for for for Bridges the traditional lock and Min Bridge the strategy uh very similar to like a liquid staking token provider where there's very large power laws and you're trying to you know compete on brand and liquidity something like a Lio and you want to be the canonical representation of stake deth something like that I would guess that's what you want to do

for lockin Min Bridges as well right if you're GNA have ethereum represented on salana even if we think that doesn't necessarily make sense you want to be the canonical version of that right so would assume sort of insane power laws and you also kind of want to have a relationship with the the consumer I guess these these other alternative Bridge designs that we're talking about Hart Who ultimately has the relationship with the the consumer and then in terms of bridges do you see the same kind of power laws that we might see in something like stable coins or lsts or does it end up being a little bit more uh fragmented or a different Market structure okay trying to do this quickly right um I think the majority like the biggest question for the last five minutes

there I think the majority of value that's going to be moved between chains is the is Major assets it's going to be usdc usdt and E and we still maybe need to talk about the circle bridge problem in a second but we'll come back to that I think major assets are going to be the major thing you're moving between chains I think you see that in Dex volumes today like 80 60 to 80% of Dex volumes is change is trading between major assets um so those assets are not going to ever be in um like they're not ever they don't want to submit to they don't want to be controlled by some other bridging infrastructure that's like scary for them um and I I I don't think you like ethereum is never going to be

really cool and like really on board with some other protocol controlling a huge amount of ethereum Supply in a lock and mint design on another chain that's just GNA be like really scary and I don't think that's what you're you're GNA want so um okay so I think that the lock and mint Bridges maybe make sense for longer tale assets um that have this idea of them wanting to move between chains I'm still not convinced that that's like a huge opportunity but I could see some developers thinking that's interesting so that's where maybe they're okay with like a vendor lockin concept and using something like layer Zero's of standard even though I I don't think that's like I think there's some

risks around vendor lock in there leaving that aside the other question is who owns the relationship with the user here and I go back to the relationship with the user is owned by whomever is generating the user intent or the the the limit order so if Unis swap X is the source where users are trying to bridge between chains they own the relationship with the user if it's a bridge aggregator like socket or Lei they own the relationship if in my example there's some of like my ven mode chain lets you do something on Tron they're owning the relationship with the user and uh like where I think a cross fits in is where this settlement layer that's behind the scenes that's like deep in the stack but is providing this like uh escrow and Settlement Services for this

intent based bridge design um so that's me speedr running the question awesome that was really helpful uh Hart this has been awesome my man uh really glad we could make history here with with you and Dan on on one podcast this is this is great um if folks want to find out more about you or follow the work that you're doing at across what is the what's the best way to do it um I'm on Twitter at how to 2001 h2001 um and a cross protocol on Twitter too um yeah and thanks for having me awesome thanks guys this was a fun one all right well Dan that was a that was a ton of fun with har um that was just a lot of energy on that episode very great

um yeah I love I love chatting with heart I guess uh you know the first question that feels appropriate to ask now that hard is in here is why aren't you invested in his startups man come on you guys smart guy right right right it's just the timing's never been right um uh for various reasons sadly but uh but I I love Harden have a lot of respect for him um and a yeah I mean he and I have talked about a lot of ideas and a lot of stuff including stuff like uh the yield protocol which paper I worked on a few years ago was very inspired by conversations with him yeah well came across to this conversation and I guess you know first takeaway you know kind of RP Mitten lock chains um you know I think you both you both you and Hart not not big

supporters of that of that design I guess you know it it felt a little almost not one-sided and if that's just not a great architecture it's not a great architecture but if we had to steal man on the other side of that is there is there any additional Nuance that we might have missed or or an additional point for why that design might actually make sense yeah I think there's there's one really clear one which is it it is often very useful to have a representation of a of a foreign asset on a chain and I think one example is is Bitcoin on ethereum right is it's very useful to have some representation ofit coin um uh on ethereum and inherently there sort of two ways to do that one of them is with a synthetic that is backed by on by collateral on ethereum and the other is to have some

kind of Lock and mint bridge where you can lock up um Bitcoin on uh uh on bitcoin and and just it's backing one to one some token on on ethereum um and the latter kind is you know it's how uh centralized um or somewhat centralized systems like wbtc work it's also how more decentralized ones like T DDC work but ultimately you still have this have this lock and mid pattern um and it sort of is the it sort of is the only way to do to do that Capital Vision the the alternative where you have collateral and a sort of a synthetic um works but again that that requires a lot of um potentially a lot of capital and someone to uh to provide a lot of uh of collateral on the on the um that chain so if you want that model and I think

some there are cases where you want it um it could be it could be very powerful but when someone just wants like a user just wants to actually move from one chain to another I think it's usually I would say probably not the most efficient way yeah yeah I tend to agree with you um so one thing that we didn't really get to talk about during the episode but I actually want to ask you about two sort of cross-chain interupt solutions that have popped up and are gaining a little bit of seam so we started to talk about the usdc Centric solution so it's cctp Circle Cross chain transfer protocol something like that and then also chain link has a CCI so um you know I guess my question to you is we didn't really talk that much about uh Circle solution Hart mentioned just at

the end of that interview that he thinks there's basically going to be power law of assets if most people don't really care about Bridging the long tail and most of the volume that people are going to want to transfer or swap into and out of are you know usdc eth you know kind of the big ones so you know I when when it then when you talk about Circle having their own native bridging sort of solution it's like well they get a good bit of the volume just there so what do you think about Circle cctp yeah so I think it's a it's CC so cctp it's basic on a very powerful realization which is if you have a custodial asset that's being issued by some party um on one chain the best way to bridge that asset um you know generally is to do what is to do it in a

way that's native uh to the asset and just has exactly the same trust requirements as the um uh as the asset itself or arguably that's the best way and so you might as well just have you know burn new usdc on one chain go give it you know have Circle effectively act as a bridge for you so that you can burn it on one chain and and mint it on another chain um and then once you have that it's actually it's very powerful because you could imagine even if the asset you want to bridge isn't Circle um you might be able to trade isn't usdc you might be able to trade your asset for for usdc on that chain and um uh and then trade it somewhere else or even if you know the asset you actually don't want to trade it for usdc funny enough like if you are

just willing to trust Circle they're kind of operating a bridge right now so one funny hack I've seen used to uh suggested as a way to um to take advantage of this bridge is just to you could imagine sending a message I'm sorry sending a message between chains by just sending usdc to a like one way of usdc to a particular address um and having uh and having that address actually be the hash of the message that you want to send and so just sort of hacking Circle that'll use them as a bridge um but because you know if they're already running this really high stakes Bridge there you know maybe maybe might as well use that so it's a it's a very powerful idea um I do think it poses some it poses some some risks for the same reason that having a a dominant

stable coin on any chain poses risks for the the centralization of that chain which is you ultimately putting a lot of power in the uh in in one party's hands functionally um or on one particular bridge and so I think I think it's a it's a risk to have that but that that said you know um we already have that risk on on any one chain with usdc right if someone were to compromise the usdc key minting keys um then the the C the consequences on on ethereum at least would be very severe um and so to some extent we already have that you already are trusting that um uh to to a great degree which you know doesn't give me a warm fuzzy feeling inside but does does make me think okay maybe maybe uh you might as well just do it for for a cross

Chain Bridge do you think that ultimately I mean is isn't this eventually some something that we're going to have to come to in face because isn't the whole point of the smart Contracting platforms to develop apps that have product Market fit what is circle if not an app that has product Market fit and then eventually in the you know hopefully this doesn't end up happening but it's kind of a feature of blockchain right is so you have these sort of hard Forks uh and eventually then it seems like C the circles of the world or the coin bases of the world or the US the tethers of the world right are going to have an enormous amount of say in terms of what is the canonical chain do you do you subscribe to that idea that you know sort of inevitably these apps are going to have a lot of

power in that regard I think for yeah well I think for custodial apps it is a big concern for obvious you know for for an app that isn't that isn't custodial and I think it might be a sort of a different a different situation um you know I I think it it asks maybe an even bigger question which is if um you know if we start trusting essentially usdc to run a bridge between these chains right then why don't we actually just having usdc run a ledger like why why aren't we just using usdc like why can't we just n transact natively in usdc with on their Ledger and have them keep track of everybody's balances um and sort of functionally just have them run a you know their own chain like a oneof one um chain and it's it is sort of a it's

maybe an important question um uh that's it like I think the the value of usdc especially is in is when you start using it with other things and so um to some extent you know the what actually gives it its value is you're going to be using it you know trading it for for Ethan on some other chain right and that's something that you wouldn't want to trust um uh Circle entirely on is like to to run The Ledger that has every all the other assets on it so I think in some sense like this is you know having this interoperability Zone where where usdc is responsible for usdc transfers across chains and you could maybe imagine a world where they where they were able to just operate uscc payments between people um might be fine if again like the point is then I is

interoperability between this Ledger that us that Circle effectively is running um with other ledgers that are running different or more decentralized applications man that's a good question I hadn't really taken it to that extreme but yeah that's probably when we get to us on Tron I mean like if that's like how everyone's transfering usct and it's because it's so cheap like you know I don't know why isn't tether just running a ledger right and I think there may actually be regulatory reasons why why there's a difference um but yeah it sort of does make you wonder it's like all right like actually why not why not why don't they just keep a keep a database um uh and I don't know some if if mostly what's being done is just simple payments yeah tether is a that one's an enigma for me that was a that was a

really interesting antidote that we got from heart yeah um well that one maybe will remain a mystery for a little while longer but um what do you think about ccip which is sort of the uh I mean it was interesting by the way I'm glad Hart sort of uh you know clarified I was a little bit confused with how you guys were referring to oracles but I mean you know chaining has its own sort of uh you know cross-chain interoperability sort of solution which is what they're dubbing ccip I don't know if you've poked around at that at all or have any thoughts I've only looked a little at it so I can't talk about the details but it it makes a lot of sense to me logically because I think of um from a technical perspective a bridge is a one-way message passing bridge is just an oracle

um it's just an oracle on one chain for the state of another chain um a two-way message passing Bridge which is all you need to to create like a you to wrap tokens in your lock and mint pattern um across These Chains is just it's just two oneway you know oracles between these particular uh chains right and so if you have an oracle protocol then it makes a lot of sense to use it as a uh to use it as a bridge and then to natively buil uh build that in um I'm I'm not yeah I'm not familiar with the details of sort of how of how all the rest of CCP CCP is built around that but like it just makes sense to me that if you have an oracle you might as well use it as a bridge um or that that that that

would be a compelling possible possible use case for it but like you know I would I would flip it around and say just like actually when we talk about bridges often Where We Are talking about his Oracle that they're all actually competing in this in this Oracle sphere um one thing I'll say though is with Bridges you actually have a lot more power um than you do with if you have an oracle for for something that's a completely offchain fact um right then like it's uh you sort of are going to need some kind of trust assumption or some you know multi-party something or other uh for it some subjectivity ultimately there I do think if you've got a message going across blockchains blockchains are to some extent legible um uh things they you should in theory at least one blockchain maybe should be

able to for example like parse another or interpret a zero knowledge proof about the state of another and if you have an A cross-chain messaging system that doesn't make use of any of that um uh or or at least or of all of it as much as you can that seems unfortunate in the long run it seems to me like ultimately the ground truth ideally of of cross blockchain communication sort should involve for example the consensus Protocols of those chains or the validation rules of those chains um and I don't know if that's I'm not sure whether maybe maybe in the long R chain link does does do that but I think just treating it as the same kind of problem as being oh you know learning about about like this you know who want a football game or something I think is um it's reducing it to to maybe a harder

problem yeah let me ask you you know one of the things that people have been talking about in crypto for as long as at least I've been in the space has been sort of this um you know uh idealized state where you and I ideally don't talk about any of this you know we're not on a podcast talking about interoperability because it just works and no one even understand understands that it's a problem and you just interface with your app and then it does all the stuff that you want it to do um I mean how far do you think we are out from from that standpoint uh just just because I remember being in like 20 2018 time frame where people thought that this was not going to be a problem you know within like a year or two or something like that and here we still are in there

that what was not gonna be a problem that uh interop like that we Bridges were going to come and it was going to solve the some some of the breakdowns in between these these different blockchains and the future was going to be mult M chain but that was okay because there were Bridges and you know here we still are all these hacks and we're still debating about different families of bridges and what the best what the sort of trade-offs are so I guess you know if you have to put a time frame on on really you know solving the 80% of these problems I mean how far away are we still yeah I think there's there's sort of two problems one is from a user experience perspective um can we actually have the experience of a user switching between chains being a lot better um I think some of that comes

from uh uh Tech like like Unis swap X or or I think across um just sort of abstracting away a lot of the of the root complexity for a user and making it easier some of it's going to depend on like on wallets you know and and just uh other parts of user experience from from switching chains um that I think are really challenging but I think all that we sort of have a have a line on um the other part of it is like is at at the the base level like what are these what are all these um protocols uh sort of built on top of and I I do worry that we'll end up up with a lot built on kind of RI more rickety foundations than we would have thought we would have I think even when you look at um L2 Bridges or cross

L2 Bridges uh you see a lot of kind of of uh bridges that don't ultimately resolve down to the canonical Bridges which seems like a shame um and like for for various reason uh reasons of efficiency or convenience I think often people uh protocols end up end up using different kind of trust assumptions for that um I think in the long run I would assume that we that everything ends up using using the can Bridge as a source of truth that's I'm not so sure because you know the um ultimately actually like that's a canonical Bridge even is still is still potentially a big uh uh uh Honeypot right and so actually having maybe having multiple sources of Truth or multiple Bridges is is actually not

um not as bad so I'm I'm I'm not so sure um ultimately I think I don't think we're going to have like one big solution for the for the for the uh that like second problem of the the ground truth I think it's going to be a lot of hacks built on top of hacks and just things that kind of like barely R rly hold together um for the user experience I've more confidence that like if this if all this stuff is working it's because we've really abstracted this away for users and it's actually pretty simple to move money from one chain to another yeah I would agree with that I think it's it's very it's very confusing from the perspective of a user because a like just imagine you don't know these words like optimism or arbitrum or ethereum or whatever and you get your

first mobile wallet and it's like I have these options to be on which chain you're like well I don't know what this means I mean is one of these chains better than the other chains like what am I deciding between it's if you if you really take yourself out of it for a second it's wildly confusing to users at your first experience you know yeah and I I have this I have this strong opinion now that I think especially something like payments which we talked about a bit on the on the podcast about like payments I think just should be I should be able to pay someone no matter what chain they're on from my chain without having to go bridge over to their chain and then make a payment to them or have assets on that chain and that's something you know yeah and then I think we get into hard had an

even crazier idea for for um building on this this was cool talking about cross chain intense um but I think that that's sort of a start where it just shouldn't actually be part of the user experience to have to like bridge between chains in order to use an application or make or certainly make a payment to someone who on the other chain yeah well I mean I think his and what I think he sort of called it venmo chain right where you know if you take intents to their logical conclusion you can just sit and have your balance of let's let's say ethereum and whenever I want to make a payment you kind of submit to this this Hive of you know Searcher filler solver whatever word you want to describe and they go off and kind of do that for you in a really optimized way and you know

your balance just gets debited or whatever um so that kind of does seem like the the ultimate future and I actually think there was a a podcast where you were describing even from the perspective of a mechanism designer you know when you would kind of designed these these these solutions that required um you know kind of left some money on the ground for people to pick it up there was a time in crypto's uh recent history where people wouldn't come and pick some of that money up but now uh now that's a little bit different and people will generally come pick up money if you leave it around and maybe that's the same analogy for these sort of Searchers and this this uh you know these third party networks of people that will go and do this stuff execute uh payments or transactions on users

behalf we just got to wait a little for it to be a thing yeah I mean I think it's gonna start with the very the very basic stuff if I have an asset on this chain I want an asset on this chain or I want to have an ass on this chain I want I want someone to pay someone else who's on this other chain I think a might be getting ahead of themselves um with the other stuff but like yeah I think in the long run may maybe it's entirely abstract and I can use an asset from any chain so some of the I think there are some challenges to that um and so one of them I was talking about with with uh uh on the podcast was just if I if I have to do a transaction on both chains in order to interact with an application

and I want to do a lot of like interactions with that application like that's that seems very inefficient um and if I have to if I have to make a trade of like you know one asset for another every time I think I'm going to going to pay more ultimately in like spreads uh if I'm if I'm trading you know usdc for ether something every time I do a transaction I'm going to be end up paying more than if I did one big transaction I would expect um Al that's not that's not necessarily always true from from an exchange structure point of view so so maybe not but um you'd think there'd be a lot of there'd be a lot of overhead in in a lot of these um and like there's reasons you oh yes and and if I want to do something atomically

right um uh it just helps to be collocated basically on on the same chain um in the same execution environment as as something but um yeah it's it is interesting think why why do I actually why do I need all that I don't know um and so uh yeah I think from a I I just hadn't really thought about it about this uh that much until I was thinking about it live today and I was like huh this is this is actually kind of interesting so I want to follow up with heart a bit about that yeah I think that'd be a good idea um all right Dan this was a a really fun episode um and I will see you uh next week next one see you next [Music] week a