How Intents Unbundle Assets from Protocols | Season 7 Episode 2 | Ft. Christopher Goes

Transcript

crypto will be successful if we can provide an alternative infrastructure that doesn't require communities to choose between autonomy and interoperability so at the moment you really have this dichotomy where communities even ones that want to be you know more economically self-sufficient must choose if they want interoperability to use like Global Fiat currencies they must choose to use the dollar or the Euro or the Yen you know one of these large currencies to interoperate economically with the rest of the world and using those currencies makes them really like dependent and subject to you know censorship or sanctions or just manipulation what's up everyone we are now almost one month out from D London the largest institutional conference in all of crypto that's

happening March 18th to the 20th obviously in London this one's going to be a blast we are almost 10 times over subscribed for tickets which is pretty nuts so again we've had to lower the discount to Bell 10 and better yet make sure you bring your friends we sell a four pack of tickets find people in your company bring your boss bring your family bring your girlfriend or boyfriend whatever it is just go you're going to get a discount if you use that team pack run up walk make sure you go get those tickets today and cheers and see you in Sunny London Town 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 has Financial technical tax legal or other advice you know the deal now let's jump into the episode hey everyone welcome to the first interview episode of the multi-chain endgame season today uh Hart and I are going to be speaking with Chris go one of the co-founders over at enoma now the reason why we wanted to start the season off with Chris is because he is a superpower where he can talk about Concepts was a very high level philosophical 10,000 foot Vantage Point kind of view but he can also get really in the weeds and the nitty-gritty details and that comes from his background as a protocol designer at tendermint where he worked on IBC and

currently what he's doing at anoma so we got Chris's thoughts on everything from the current issues with blockchains as they exist today um to where he sees this industry going in the future some of the design decisions around enoma this is where we got really into the Weeds on intense and interoperability in between different blockchain trust zones we talked about protocols starting uh to basically unbundle and competing on the basis of what the protocol actually does versus the currencies that they export finally we ended this episode by asking Chris what he would see as a satisfying endgame for crypto and he gave one of the best answers to this I have ever heard so really hope you enjoyed this interview with Chris we'll get right into it um yeah enjoy guys all right

everyone welcome back to another episode of bell curve today I'm joined by my co-host for this season Hart lamber and we are going to be talking to Chris ghost the founder of anoma Chris welcome to the program thank you for having me yeah we've been uh really really looking forward to this conversation and we want to use this uh as an as an opportunity to unpack um some of the big themes of this season and uh for listeners who haven't heard Chris before I think Chris you're very good at um kind of speaking at all levels of the stack so high level philosophy but also being able to go down into very very technical specifics and you know not to put you on the spot but I think we want to make full use of your range today and kind of get your thoughts on some of the the highle

architecture and almost philosophy of where we're going in crypto but also ask you some very specific technical questions and for listeners who are trying to get a chart of uh the agenda for today's talk we're almost going to do a technical sandwich we're going to start with some of the higher level stuff get into the meat of the technicals and then end with some of the the broader implications but Chris I thought uh just as a jumping off point you have mentioned a a talk uh given by Mox Marlin Spike the founder of signal before as kind of this um very influential uh discussion about uh and you something that you took away for a lot of stuff about blockchain architecture so you give us an overview of what this talk was and how it influenced or helped you see uh helped

you understand some of the stuff that's broken with uh how we think about blockchains today well I'm very happy to plug this talk again uh more people need to watch it because it will save you years of your life at least if you're working on blockchains so the talk is a talk by Moxy Marlin Spike the founder of signal although he no longer works there um and it was a talk given at 36 C3 I guess the 36th chaos computer Congress in Germany talk is called the ecosystem is moving you can find it online it's about 30 35 minutes it's definitely worth watching and I don't know all of the context but I think this talk was given by Moxy basically in response to a bunch of people complaining that signal wasn't fated so early on in the life life cycle

of signal the protocol and the application you know really before it was like anywhere near as popular as it currently is you know that emerged from a kind of you know hacker cryptography community and a lot of them were into building Federated chat protocols as in protocols like Matrix where there are a bunch of different servers operated by different people and there's a common protocol that connects all of them so it's like pretty decentralized different people can use different server providers and they can still talk to each other using the same protocol and Signal early on experimented with ious uh you know approaches to this um but decided not to be decentralized decided to operate a centralized server uh still build of course all the encryptions so that the server can't read your messages the server has very few permissions but

uh they decided to operate it in a centralized fashion um and this talk was basically a talk by Moxy explaining why he made that decision and what he explains in the talk is that he understands signals competitors as basically centralized chat services like WhatsApp and um I don't know Facebook Messenger stuff like this and all of those services are controlled by companies and one big advantage of centralized control uh in fact maybe the primary advantage is that it allows you to change the protocol quickly uh because you know Facebook is a company because um WhatsApp you know separate back then was a company because Google is a company they can upgrade the

protocol and they can upgrade the protocol in particular to give users more features like Emoji reactions and images and faster chats you know all the stuff that people want right and the critical disadvantage of decentralization in this sense the reason Moxy decided not to Federate signal according to this talk is that it would make it basically impossible for them to keep up with that speed of change in just what users want and what users expect out of a chat protocol and if signal wasn't able to keep up with the centralized Services then everyone would just use you know WhatsApp because WhatsApp would have Emoji reaction and images and automatic memes and all of the stuff that people find fun engaging in chat protocols and Signal wouldn't

because signal signal would be deployed on like 150 independent servers and those 150 independent servers would never agree with each other uh on protocol upgrade you know any more frequently than like once every few years um so the relevance of this talk to blockchains um I think blockchains have a very similar characteristic um as the thing which basically democracy was trying to avoid they're distributed systems and once you deploy them in practice you know they are trying to be decentralized right um is very very difficult to change them uh this is perhaps doubly true in the case of uh blockchains with tokens where cryptoeconomics is involved due to a selection effect fact so once you launch

something like ethereum or Bitcoin or Cosmos Hub or Salon of these different chains that chain starts out with an initial sort of cryptoeconomic model and people basically join the community in part because theyre with the model right it's very transparent what you know how at least for the most part not always to whom but typically how much of the asset there is how it's being issued what the rules around gas and fees and redistribution are stuff like this is quite transparent so people join the community um you know in part on the basis of those expectations and this me makes it very very difficult to change both the kind of actual computer protocols just because you need to get so many people to coordinate and the economic distribution functions of these systems that means that if you make a mistake or if something that you design

is not exactly what your users end up wanting uh it's very difficult to change it often what happens is the users just shift to a new and different system which can make choices which are better for them um so the the does that does that make sense is kind of a summary of the talk yeah so you have the Bitcoin core Dev problem nothing changes right I mean this can be a feature or a bug depending on how you look at it I'm not necessarily saying that it's bad but that it uh is something that's helpful to be conscious of because I think some people have this idea that they'll like deploy blockchain to like a sort of iterative MVP process that say a Silicon Valley startup might use and that works really well for Silicon Valley startups because they can pivot in like a board meeting on a

Friday evening and blockchains can't no it makes perfect sense and I totally get it there are Pros to it if you want to have digital goal that's super stable and doesn't change but if you're trying to invent the future kind of hard so so how did this lead to aoma like I I I kind of see where you're going here but like what's the how' you go from there to what you're doing now well uh yeah the path is is definitely not linear but there's there's some I can't tracing back all the causality is is a herculan effort but this talk was definitely a big inspiration I think in part of may be very cautious uh like I really didn't want to make the wrong decision and

deploy A system that would uh you know for example I had the experience of working on IBC right the interoperability protocol used by the cosmos ecosystem and starting to expand to ethereum rollups recently um and IBC makes a lot of decisions uh and some of them I think now with the benefit of hindsight were wrong um uh you know uh in terms of how it considers concurrent State machines and just how the protocol is architected and I think it's basically impossible to change this so it's a bit unfortunate you know I spent a lot of time working on that um and I would rather um uh you know work on things that will be around for a long time so um one way one lesson that anoma takes away from this at least is that we

actually try to make as few decisions as possible in the architecture of the protocol as in particular we try to not make decisions on like very specific Primitives that are probably going to change in the future a good example of this is something like a zero knowledge proof system so zero knowledge proof systems keep getting better which is fantastic except that it means that if you standard one it's going to be out of date in two years definitely there's going to be a faster one um so when we think about AA of course like the actual instance will run with a specific Zer knowledge proof system at a specific time but we try to incorporate that heterogeneity of components into our design so we expect that different people will be using aomo with different proof systems and the proof systems will change over time and we don't tie any of

our architectural decisions or design decisions to a specific Choice there I think that's our meain takeway yeah and that's where I think we start to to get into this idea of intense as well Chris and this where maybe it would make sense to why why don't we try to compare and contrast I think with so we we' sort of outlined one of the core problem statements and this has been a problem statement going all the way back to the days of Bitcoin where there's something that's a feature and a bug at the same time depending on how you look at it which is this immutability this idea of immutability on once you build it it can't change um you have to couple that with the idea that these are essentially and this is a

metaphor that I've an analogy that I've heard you use quite uh quite often is we're building operating system there's technical component we're trying to onboard users and obviously we need to iterate make decisions and make changes there so I think one of the we're at this sort of interesting point in crypto where there's a little bit of a a fish who doesn't know they're in water uh type thing you know like uh old fish goes up to uh old fish goes up to Young fish and says you know how's the water today and the young fish says what's water and so I I would love to understand from your perspective because I think whenever you get into something like intense people like why couldn't I just do this on ethereum or salana can you before we unpack some of the specific design principles of a Noma and

what that actually is can you help us understand what are some of the things that might be invisible today that you see as a massive problem or ma major design decisions that we might have made that are incorrect um you can you point some of those out because I think it'll help contrast when we sort of unpack your vision of the future so at a very high level I think a lot of the progress in blockchain design has come from a sort of unbundling of components um originally with Bitcoin Bitcoin started with everything integrated there was literally one piece of software written by you know we think one guy or girl or small coordinated group um and that piece of software did

everything it had an asset it did ordering it did storage it did execution it did verification um Allin one sort of glob without clear boundaries between any of these things now over time we've seen you know those aspects start to become unbundled with certainly with kind of the rollup Centric direction of ethereum with different aspects of modular design um but I still think there are some assumptions baked in that may not be suitable for the design context um one of those relates to the design of virtual machines so most virtual machines in blockchains at least that I know of are built on something called the vono and architectures this is true of the ethereum virtual machine it's

true of the salon virtual machine it's true of the message passing system between Cosmos modules which is a little bit like a virtual machine um vanman architectures are are uh pretty old they date from the 60s and they were designed I think primarily for mainframe computers um and they worked you know they were designed for people to execute programs do computations using computers so they already had some aspects which are relevant to blockchains they had maybe multiple parties using the computer um but they did not necessarily have they what they did not have was distribution in particular they had one you know one unique clock and one of the things one of the assumptions baked into this architecture is the idea that

there's like a single place where ordering happens and this is not true in the blockchain context there are a lot of places where ordering happens this is in fact where like me comes from is that there's not just a single place where ordering happens um and uh this uh kind of you know for example in some of our work with aoma we are working on what we call a resource machine which is kind of like a virtual machine that doesn't make this assumption um it assumes that there can be multiple stages of ordering where each stage of ordering is kind of like the partial application of a state transition and at the end you better get one that's fully applied so one I guess maybe blind spot that I would see is just this um virtual machine design that's based on a kind of logically

centralized execution context of a Mainframe computer and blockchains don't have that um another concept you know certainly the concept of privacy has become pretty Salient in recent years um personally I think the word privacy is kind of misleading as it pertains to blockchain systems for two reasons one privacy is not a system property it's a property of a specific interaction so if we think of a transaction I might make my transaction public to you Hart but not to you Mike so is that transaction private well it's it's sort of it's both private and not private right it's like private to one person public to the other person um so it's not really a property it's more it's a property of a

specific you know piece of data with respect to a specific Observer not of like a system right um also I think privacy has this connotation of hiding it's like it's it's it's comes from this world where light is bouncing around everywhere and if you don't want light to bounce on your thing you know if you don't want your neighbors to see what you're doing you have to pull down the window blinds so there's this like implication of an act of hiding and this is not not how digital systems work at least not how these kinds of distributed systems work you know by default your data lives on your computer and the only way it goes somewhere else is if you like send it somewhere else so the thing that we're really trying to program is disclosure not privacy um now a limitation of some existing

systems uh is that they couple these decisions so if you use a fully transparent chain like the ethereum main chain at the moment salana main chain at the moment um you can only choose to disclose your transaction to everybody or nobody right if you don't send it to the blockchain it's private but it's also not very useful and if you send it to the blockchain all of the data is visible to everybody so that is a kind of programmable disclosure it's just a very limited kind and uh what we're trying to work on at anoma with the resource machine is to allow for much more fine grain distinctions using among other things zero knowledge proofs where users can select specific parties who they want to see specific parts of their transaction data which is you know how most people are used to interacting in

the real world anyway right you have like a lot of chat groups and they have different partially overlapping intersections of people and different purposes and you share you know different information in them depending on contexts and I think that's also the direction we might expect uh blockchain systems to evolve yeah Chris there's so much to unpack there so one um one concept that I might Point listeners to as well as this idea of skew morphism which might be relevant here Chris Dixon has talked about this at length but maybe one difference um for listeners you know when when we've talked about building Opera operating systems in the past it's kind of like here's this software you you know log onto your computer and then it's just there whereas what we're

talking about here I think is building a distributed operating system here so there are some principles that apply and some that don't necessarily but where I really want to dig in is this concept of unbundling which is frankly a core uh Core theme of this season and I think people are only starting to understand and where I want to to maybe end up in this discussion is especi most people who are listening will be familiar with this idea of modularity as unbundling settlement from da from consensus and execution I think people are pretty familiar with that as a mental model model but I think the thing that folks are only starting to Grapple with is unbundling like a protocol from its asset from its security model and you have talked about competition you know

from the standpoint of a protocol competing with an asset and I would love to kind of unpack the the vectors of competition there but even before we get there you have this really great analogy of us Mal appropriating ideas and business models from Silicon Valley um and you kind of talked about us maybe having this mental model of platforms as the primary vehicle for Value capture and SAS so can can you even before we get into that unpacking can you talk to us a little bit about some of the preconceptions that we might have for how uh value is extracted on the internet and that are kind of taken for granted in Silicon Valley and how we have misappropriated that within a blockchain

context so I think a pretty common model of how value is ACR in um silicon sort of in web two companies mostly is this concept of a platform and a platform uh seeking and deriving value from usage and fees that are associated with usage so a few canonical examples of this are something like um Uber something like the App Store something like even you know um on like Facebook Marketplace stuff like this um in all of these cases there's a basic theory of proportionality in play the theory that like the value of the platform is proportional to the amount of usage and that's usually directly

true in these cases because all of the usage pays like fees to the platform right um and often this model you know it seems to me in the in the discourse is uh applied to um blockchains and their assets as in there's like a theory that the value of an asset is proportional to the value of usage fees paid on the platform and in particular the value of like applications directly running on or perhaps even deriving security from that platform um but a critical there are a few critical differences between what happens in web 2 and what's happening on blockchains one of them is that there's a distinct asset so you know uh the Facebook app store or uh Google um you know Microsoft sorry Google and Apple's uh stores on

their respective platforms um they don't have their own currency now they might want to issue a currency Facebook did try to do this uh but but they're too big and they're too anti trusted to um manage that and perhaps too badly coordinated internally whatever I don't know but they're still they're using a different currency so they're you know I think the platform like thesis of value for web 2 companies is like pretty much correct like the value uh you know whether or not it's good for society it's like correct for analyzing what's going on and that set of value flows but Bitcoin and ethereum Cosmos salana these ecosystems have native assets and those assets can be used and you know uh I mean basically the value of an asset is like uh not determined by anything other

than who chooses to buy and use that asset there's no I don't think you can reduce it to some like physical law or some usage of a platform or something like this assets can live in many different places and at least I would say you know to me the primary value of these assets let's say for example ethereum really comes from the community and people wanting to be part of the community people finding the services provided by that community valuable people choosing to invest in uh the asset to accept the asset you know really to kind of demonstrate um and increase their alignment with the community um so I think this model is kind of you know if maybe just accept that as a hypothesis then this platform

value capture model is not particularly synergistic if the actual value capture mechanism is the community because the platform value capture model tends to lead to this kind of attempt to create usage right um you see this more in new ecosystems uh than than directly in ethereum although you saw it more on ethereum initially where a bunch of the issuance of the asset is used to kind of encourage applications to just use the platform and in order to qualify for that issue and the applications have to like I don't know pay fees to the platform or inherit security from the platform or something like this um and I'm not sure that that's I think that's maybe not the most efficient use of capital if the primary value of the

asset is derived from the community not the platform if the primary value of the asset is derived from the community then you want to like make the community better make it a a community that provides more value you know provides more education make it a community that more people want to join and invest into the future of um stuff like this what's up everyone we are now almost one month out from Das London the largest institutional conference in all of crypto that's happening March 18th to the 20th obviously in London this one's going to be a blast we are almost 10 times over subscribed for tickets which is pretty nuts so again we've had to lower the discount to Bell 10 still hooking you guys up get you a 10% discount on Bell 10 and we've onboarded a whole bunch of new speakers so that's

Dan tapiero of 1rt Pascal gothier of Ledger Anthony scaramucci the mooch himself Michael Sunshine of grayscale Brad garlinghouse of Ripple s nazarov of chain link Matt mcder of Goldman Sachs their Global head of digital assets stany kulov Danny Masters the list goes on this one is going to be an absolute blast make sure you don't miss it and better yet make sure you bring your friends we sell a four pack of tickets you're going to get a discount on that so find people in your company bring your boss bring your family bring your girlfriend or boyfriend whatever it is just go you're going to get a discount if you use that team pack run don't walk make sure you go get those tickets today and cheers and see you in Sunny London Town so Chris like okay I want to push

on this and I want to draw it back a little bit to intense and how like intense uh how intense enables or improves interop here too but to regurgitate a little bit of what you're saying here let's just to make two examples we have like the ethereum platform and the salana platform right and um I'm gonna I'm gonna frame them as web two businesses which they're not but let just bear with me for a second so we have ethereum platform and Salon platform and this kind of platform value capture model we're saying hey um uh users go use the ethereum platform do things exactly this way here's how ethereum execution Works here's what an ethereum transaction looks like uh you use ethereum to pay for it and at the

end of the day um in this particular framing we're GNA say this looks like a web two value capture business we want more usage of our platform um it's good for the asset right um and on salana you have the same thing it's just different um uh infrastructure it's different op codes different uh execution environment here's your salon transaction here's how it works go let's use Sal C value that way um and as I've been reading and even prepping for this with you we're like okay these two things make sense um but you're making a couple interesting points here one point is the actual user and I'm going to use the word user and make it more singular you're kind of

talking about communities which are maybe more PL pluralistic but like the user from their perspective they don't really care if they're using the ethereum platform or the salana platform they just care if they can do the thing they want to do like whatever the action is they want to do and maybe I'll pause here and let you kind of go into intense and how you see intense as a method to unify uh these different platforms right I mean to your point um you know I think as protocol Architects or as a protocol architect my job is to be humble and accept that users want things

and I have no influence over it and users don't want to like use a particular I don't know blockchain it's like you know think think about computers do you want to use like the Intel you know i7 processor because you have a specific Affinity to that particular processor maybe if you were like the designer of that processor or you're like the dad of the person who designed the processor maybe it has a special you know significance in your life but for most people they want to use the processor because it's like faster or because it has more cash lines like they don't care about the specific thing they care about what it provides to them um and you know intents in our conception are a way for users to describe what they want the blockchains to provide to them and what they want

the other users to provide to them in a c in the course of a multi-party interaction so with well yeah let me go deep let me lead you or let's lead our audience into like a more technical understanding of why that's true right so let's let's keep going with this uh and we talked about this uh on Monday a little bit in in the preview or the prep for this but um let's let's go with the computer processor analogy because I think it's quite useful um and we're going to say uh the ethereum is a computer processor and maybe some um some listeners will know like the difference between like Risk and cisk architectures these are like different op codes for how computer processors work um and let's just say ethereum is like one set of it's like

the risk computer processor here are the op codes here's how this processor works and salana is the cisk processor here are the op codes here here's how this processor works and um in order to do a computation you can do computations on both systems they absolutely work but literally the op codes you're using to do the computation are different and at the end you can run your computation and get your output at the end and um my mental model for what you're doing with intense year is you're saying I don't care care how the C computation works I care about the output I care about the thing at the end and so I can define an in uh a constraint an instruction scent

you can probably I can mathematically Define the state transition that I want to have happen as an intent with like an output and then I can run and verify that that intent was fulfilled on any execution set on any processor and I don't care how that happens first of all do you agree with thatch technically what's wrong with that um and yeah yeah does that make sense yeah that makes sense I mean the user in writing an intent at least in the way we Define it defines basically what properties they want the computation to have so those could be a specific output those could be a relation between input and output they could be like it's the user is

purposefully less specific so with a transaction you say I want the CPU to execute instruction 10 then instruction 12 then instruction 34 then instruction 40 and then it's done and with an intent you say I want the relationship between the input and the output of the transaction to be that like my balance increases by at least six right and you don't care whether that happens with instructions I forgot now which ones I said but 10 20 34 and 40 or like 52 and 65 as long as the uh resulting State basically which the only thing you could verify anyways um satisfies the properties that you Des in your intent so just just to underscore that point like for me this makes this intent

architecture or this model uh a no-brainer for a lot of interoperability because like if I want to run on different processors if I want to run on different types of execution environments um I don't want to have to Define different instruction sets like you just said I don't want it to be like you know instruction 43440 on one system and 24 and I don't just completely different on this other system I just want to Define what I'm trying to do and then it can go and get executed wherever and so I do think when I've been reading or listening to your stuff there is this obviousness to me that this uh less prescriptive way of what I'm trying of defining what I'm trying to do enables this modular multi-chain

World um because I can Define one thing that can get executed anywhere um agreed yeah yeah I think so I mean there's still difference in the an to take the analogy of CPUs which I think is pretty good um there are still differences in CPUs like if um you know over time CPU architectures change and for specific programs different CPUs might be more or less efficient there is a relationship there you can't like completely abstract it um but still I think that in the context of um intense um and kind of intense flowing around all of these different different you know modularized abstracted blockchains um users are going to be delegating or kind of

expressing most of what they want most of their preferences for selection of particular CPUs in the intents just as properties so in my intent I might say that I want you know best price execution defined in such and such a way fairness defined in such and such a way and I want like these assets you know I'd better name the exact assets I want but maybe I'm fine with solata or ethereum or Cosmos like those all have enough security for me um and maybe I don't you know I don't even care like I don't care where the assets were initially I just care that they get to the place I want them by the time my intent is settled so there will be some like Market in some sense it's like if you had an operating system on your computer where your computer had a bunch

of CPUs that were competing with each other to run your computations and you as the person writing the program said okay I will give my computation to like the CPU which satisfies my properties and uses the least energy or something like this um and in you know your computer doesn't do that because it doesn't have an internal economic system at least not yet but blockchains do so they can yeah I guess to be really concrete right you you are right you're adding in the processor technology breaks down because there aren't really trust assumptions on your different processors they like are just they work right um whereas you're saying with a blockchain system there are trust assumptions and so I could Define my intent but I can also be granular and

specific about which processors AKA trust assumptions I'm okay with and like uh to pick on Tron I can say hey I'm okay executing this everywhere but Tron I just don't like it for whatever reason yeah yep that's right so okay so that to me um yeah that's that's super cool and we now have this maybe for a second we can talk about how um the the in this sort of solver Network concept so once we have intents um and we're saying users can specify the thing they want to do without specifying the processor AKA

blockchain system um that it's going to be done on although they have granular preferences to pick which ones they like um then the idea here is you have uh a network of permissionless decentralized actors that you call solvers that effectively compete to uh execute or fulfill their intent uh do you like expand on that for a second yeah that's the idea so the way um anoma segments the world of applications um we think that mostly users are using applications in the distributed systems blockchain context because they want to coordinate with other users and coordination with other users almost always has two phases which

we call counterparty Discovery and settlement and counterparty Discovery is the process of figuring out who is a suitable counterparty given what I want to do with my intent so if I want to trade uh eth for usdc I'm looking for a counterparty who wants to trade usdc for eth very simple the second phase is settlement settlement is posting that transaction once we've agreed on it once I found a counterparty and someone has um you know perhaps me or a solver has made the transaction a settlement is posting that transaction somewhere where can be ordered at least with respect to relevant state so that everyone can verify that that transaction and not some other transaction actually happened um and in our world solvers facilitate counterparty Discovery so you could you

don't need always solvers you could just do counterparty Discovery by sending each other messages on Signal or Twitter or something we've seen a little bit of this in the wild in particular with nft trading there was some counterparty Discovery happening on Signal uh on Twitter and some t groups um I saw some of that um but for the most part users would like like computational counterparty Discovery is a lot of computational work you know you have to like message all of these people on Twitter and see if they want to buy your nft or not and it would be great if someone could um kind of do that computational work for you and keep a track of a whole bunch of ATT tents and figure out when some of them could be matched to figure out what like the available liquidity is and in our

conception this role is called a solver so solvers are both they're handling a lot of intense intents will never become transactions because there is no suitable counterparty um but solware are handling a lot of intents figuring out when intents can be matched crafting transactions to do that and sending those transactions uh to chains or consensus providers for settlement uh Chris I've I've got a question for you about um you know something that I've heard you discuss quite a bit which is this idea of flow control or standards um and and I want to start to lead this into that like how is this better from intents as they're instantiated today in for instance something like Unis swap X or cow Swap and I I think the answer to that kind of

lies in this idea of Standards um and the amount that like a a user can kind of granularly dictate preferences across this network of Searchers so can you say a little bit about the idea of Standards within this world that you're describing right so I mean intents are are in many ways maybe just a word that has become more popular rather than like something new that has come into existence you know maybe word and concept um but they've they're around as you mentioned already they've been around for a long time um all the way I would say all the way back to like what was it ether Delta was one of the original ethereum dexes that used something which looked a little bit like intense of course a very specific kind

and now you have a lot of uh what I would call sort of distinct um sometimes they're called offchain or or kind of separate intent systems application specific intent systems for UNIS swap X for cow swap um for openc I would describe as this kind of thing um different nft trading protocols um different sort of orderbook based fungible asset exchange protocols um and those are um one those are those are fine and it's clear that users want intents but one disadvantage of having many different protocols is that uh they're not very interoperable and users typically have to pick one when they're doing so kind of application interaction so a user has to choose to send their order to Unis swap X or choose to send

their order to cow swap or choose to send their order to you know openc or whatever and uh typically that means that users will get a worse price just because like less liquidity is immediately available now to some extent solvers which understand like multiple protocols can maybe Bridge things but it also depends on how the intent system is built uh some of them you know may not allow settlement with like other things included in the transaction automically or something like this um so there's a lot of friction that I think is is unnecessary if you have intent level composability then users can just send like it's like there's just there's one giant pool of intent sort of conceptually the pool is it doesn't exist all in one central

location it's split up across solvers across this network but uh you can reroute intense over time to like put the intents which are most likely to related to be related to the other intents in similar places so you could do this kind of locality optimization and do that on the basis of what state intents actually want to touch right and that should in the end lead to better prices for users so like specifically just to make this super concrete right I have my swap from X to Y from usdc to eth um I could send that intent to Unis swap X or to cow swap or to onein Fusion I'll just use really concrete examples um but I'm basically constraining the

space and who can fill that intent to a specific application even though they're all doing conceptually the same thing and what would be really nice would be and that's worse for me I get worse price execution because there's less competition to fill my order um from this solver Network so what I want is a standard where I Define it once um I Define my intent of what I want to do and anyone that can any application that can fulfill that intent can go ahead and fill it yeah yeah that's right I also think that I mean this is maybe a sort of or marginal difference but it takes a lot of work to build an intense system like Unis swap X cow swap openc these are Big well-funded teams most independent D hackathon developers cannot build and

run like an intense system um and I think that just limits the supply of applications um and standardizing more of the protocols means that you don't have to write like all of this offchain sort of application specific solver infrastructure anymore you don't have to convince all of these like market makers to use your or bespoke protocol um if you're using a standard so I agree with that fully um and so you know I don't know how much we talked about this but I I have this uh intent-based cross Chain Bridge um so I'm like very Chris I don't even know I'm like an intense maximalist so this is like all uh all all kind of kind of fun for me but the thing that I think is worth pushing on pushing you on

is like why intense the examples we just had are all single chain mostly single chain examples where I'm like doing a swap on in our cow swap one inch um Unis swoft X it's on ethereum but um I think you have a lot of views on why intents help interop or moving between chains or Bridging the gaps between chains and I think it'd be really useful to get your perspective and your insights as to why intense uh work better or maybe even are required um in some of these uh multi-chain environments ments yeah so I think it's it's helpful to frame like what is going on if you

consider the whole ecosystem as one network so if you consider the whole ecosystem as one to be one network you know users are submitting intents all of the time they're already doing this transactions are intents they're just like more specifically limited because you force the execution path to be instruction 10 then 40 then 32 um so users are already submitting intents all of the time then these intents are getting kind of sorted and routed around um and matched and settled somewhere and to me this is really a scheduling problem as in if users now if users have specific trust Affinity you know if there's like one class of users who only wants to use ethereum and a completely disjoint class of users who only wants to use Cosmos

and they never want to interact with each other you know then you don't need a cross chain system but that is not the world we observe we observe a world where people want to interact with each other and they want to move move assets around and they want to like coordinate with all of these other ecosystems and I think that's great but it means that when we look at like let's imagine so that we're at time step you know one in the giant intense Network and we have a whole bunch of intents and those intents have different preferences as to where they can be settled and what they're looking for so one intent might say you know Cosmos or ethereum or salana another intent might say you know must be ethereum and another intent might say salana or Cosmos right and they are looking for different assets and so just

to like maximize throughput we want to kind of schedule these intents on different um you know different solvers and different execution backends so that we uh get everyone the assets they want and like don't it's like it's like a parallel scheduling problem like you don't want to send all of your transactions to a One Core database if they don't need to touch the same state um so you know to us one maybe this is a different a different and someone you could understand this is like a different different view on scaling than sort of modular or monolithic modular and monolithic are both focused on scaling like one logical domain in some sense you know maybe a security domain uh you know certainly in the case of a

monolithic system it's like an everything domain in the case of a moduler system it's you know at least one logical security domain um and uh you know I think both of those approaches are valuable uh but we need something else and to me that is like a a scale system that considers the scheduling problem of figuring out which intents need to be routed to which domains you know dynamically all of the time based on what users preferences actually are one thing as well Chris I think is is worth you know pointing out for the listener if we can maybe even just like telescope out for a second here and try to summarize what we're discussing what we're discussing is again if you had this Viewpoint of you know I live my life on ethereum versus salana versus Cosmos what we're talking

about is a future where you could consider those trust zones or you know different networks of compute or whatever it is and if you have to IM if you're a 5-year-old like me and you need to imagine things actually in space consider them like these little concentric circles right and they're all like kind of bumping up and maybe there's like a bigger vend diagram over some circles some circles are completely separate right and that's like what we've so we've got all these circles that like kind of touching and what we're talking about is moving up we're abstracting a layer to this intense sort of layer where there's counterparty Discovery and there's a a a set of shared standards that all of these different uh programs sort of plug into and users can dictate their preferences

to this network of little Hornets or worker bees or whatever that take these desires off to different networks and then they kind of get routed and executed and I think that's what you're talking about with this scheduler problem what you're describing is a system which is very accommodative of heterogeneity we are assuming that there are going to be multiple different sets of logic security models different types of assets different user preferences all that stuff and I'm wondering how much of what we're describing here is a technical problem versus a social problem and what I mean by that is like if you again now telescope back in as as where this is broken today it's probably on ethereum rollup land where we

fragmented uh liquidity and users across different chains that don't interoperate with one another and it's my strong suspicion that that's not a technical problem that that's a social problem and each one of these platforms don't have a strong incentive to actually build something like IBC in between because again they wouldn't say this but it is my strong suspicion they're like well we could accommodate for heterogeneity right we could be like yeah there's going to be arbitrum and optimism Etc but I bet what each of these things are thinking is I'm just going to scale so large that it's just going to be one hom block space and that it's going to work really well within my ecosystem and so my question to you is do you agree with

that broad characterization of heterogeneous versus homogenous and how you know how tactically do we get these applications to agree to this shared set of Standards because that seems to be the Big Challenge from from where I said right um yeah I I I I I that that makes a lot of sense to me I mean I think I would also tie this to the platform value capture thesis I think people think that there's not a lot of value in building interoperability protocols protocols which allow other people to like use different Assets in different systems and different domains because it's like it's not their platform uh and they think that they'll you know acrew more value by just getting more people to use specifically

their platform where their platform is understood as like their rollup their L1 chain like a specific trust domain right um you know there is a great I forget who set it maybe someone could find this in the comments but there's a great law law of software architecture and organizational design that the architecture of your software will always mirror the architecture of your organizations and you know in ethere the ethereum rollup planned is full of like I think largely well-meaning but definitely fragmented organizations and the ethereum rollup Protocols are perhaps largely well-meaning but definitely fragmented um so I think it's difficult to you know uh as as you put no one at least feels like they have an

individual incentive to do it and the organizations are fragmented so the protocols will be that yeah so I I want to One More Concept that I want to make sure that we cover because I do want to unpack I think this is this will be a concept that many listeners haven't you know explicitly thought about which is again this is kind of one of those assumptions that you didn't even realize that you were making that of course the value of eth the asset is going to be dependent on the function it and utility of ethereum the network and I actually still have some questions even hearing you describe you know separating these two as distinct Concepts and different arenas for competition I still even in this routing example like I still feel

like ethereum has successfully gated its trust Zone with e with ether the AET right to to gain access to you know the security set or whatever of ethereum I feel like you still need to use ether token so I still have some questions for you on that but before we get there I do want to cover the last you know bit of architecture here which feels important which is the existence of zero knowledge proofs and how that has like eliminated a massive technical barrier for getting these very different systems to speak to one another so C can you talk a little bit about that right so just talking here about kind of protocols like the evm svm even the Bitcoin Bitcoin does have a virtual machine it doesn't get enough credit um

limited but it it does you can do a lot of scripting back it's having a comeback is it is it okay great I'm not maybe not tuned in but um these Protocols are basically defined by their verification functions because this is what you need to know in order to verify the history of a blockchain in order to run a node in order to operate in this distributed context and those verification functions you know just to run computationally are very expensive if you want to verify the history of Bitcoin you have to replay all Bitcoin transactions from Genesis same preum um and that uh you know that V that cost can limit interoperability between different protocols because different protocols you know if you're running your transaction on the Bitcoin chain

that chain because the validators who running it have already done those computations that chain already knows like the current bitcoin State it's already implicitly verified that history same for ethereum but if you want to say use the ethereum state on bitcoin uh you know someone you you need to like it's I mean you could just try and convince all the Bitcoin full nodes to also run ethereum like that would work if you could do it um but it's difficult because running ethereum is very is expensive so all those nodes now have to replay all the ethereum transactions and verify them as they continue coming along so in some sense you're like your verification cost is the same cost as it was to run the transactions in the first place um with zero knowledge proofs and here it's really not the zero knowledge part that's doing the work but rather

the succinct part um once you do a computation you can someone can create a proof which is like a few a few orders of magnitude overhead still but it's getting lower I think it's like 10 to the 3 um at the moment um and then anyone can verify that computation in constant cost o of one even for a very long you know eventually for a very long history like the history of ethereum or Bitcoin and this I think really changes the calculus of protocol competition and interoperability because before these kinds of succinct proofs you know if you you could Cho if you wanted to run a transaction you really have to choose like either the Bitcoin history to run it on top of or the ethereum history you

can't do both cost is too high but with Zer knowledge proofs as soon as any changes are made on any one of these systems a proof can be sent anywhere right so think of like also think of like a zero knowledge rup which is using also the cinct part not the zero knowledge part U typically um these zero knowledge rollups sometimes we think of them as being related to a particular chain like we might think of them as being l2s of ethereum or l2s of Salon something like this um but really intrinsically they aren't once you do the computations and produce a proof that proof can be verified anywhere um and the ability to create these kinds of proofs which can be verified so cheaply I think it just makes it much cheaper for different protocols to interoperate with each other right as long as you

know you have to pay some computational overhead cost to create the proofs but once you have them you can verify them anywhere else with a few elliptic cve multiplications or some hash functions um so I think that will have some interesting implications yeah Chris I want to like uh Hammer Hammer in and underscore a lot of this because it was a big conceptual uh unlock for me and I think this is really freaking cool um so I want to relate back to just to help all the audience understand this like we go back to our processor analogy so you have like the uh ethereum processor and the salana processor right and um you can kind of think back about um you know some of us maybe a little bit older like emulated old computer games on

processors it was possible to construct the like Nintendo architecture on your x86 machine and you could emulate it but everybody knew it was kind of expensive right because you're sort of doing this like a lot of extra work to do this emulation thing and I think the analogy here if I'm understanding you correctly is like um you know salana uh the salana processor could completely trustless emulate ethereum if it replayed the entire ethereum uh blockchain history all the way up to the the tip of the chain um you could do that and Salon is a powerful machine it could probably do that but it would take up a lot of

resources and it's obviously a lot of work to do that um and what you're saying here is yeah these protocols can theoretically emulate each other or verify each other um but the succinct property of zero knowledge proofs let them do this much much much more efficiently um and then maybe it's worth underscoring what the succinct property of zero knowledge proofs really means here what it means is that uh we're not talking about privacy at all what we're talking about is compressing a whole bunch of computation into something that is easy to to prove so I could say I could

generate a zero knowledge proof for the current state of the ethereum blockchain and um it would be like o of n like one computation or linear time comput not even linear time one computation to verify this compared to actually replaying all of the ethereum history and so yeah for me I want to like I kind of jump up and down at this idea I think it's just super cool that the succinct property of zero knowledge proofs allows us to communicate between complicated protocols in uh like o of one complexity time very efficiently and that is conceptually a very cool way to think about interoperability first of all did I get that right correct me like where

I'm technically wrong or or or anything yeah that's right I mean I think maybe there are like two um distinctions from the case of emulation so I I used to have a oh my God it's like a Windows mobile 2002 pocket PC and I managed to run a Game Boy Color emulate because I got it for free because I was like 10 in dirt for and I managed to run a Game Boy Color emulator and played Pokemon Crystal on the emulator at about 15 frames per second now there are two differences between that situation and what what's going on here one is that in that emulation I actually wanted to do the computation like I needed to compute you know what the result of the Pokemon battle was um and in the case of blockchains we're not trying to do the

ethereum computations on Solona or do the Solona computations on etherium we're only trying to verify the computations that have already happened so the thing that we're emulating is the verification right not the original computation and then the second difference is that this these succinct proofs allow us to do that verification in O of one instead of O of n whereas previously before susin proofs verification was like the same thing as computation in order to verify you had to redo the computation now you don't and that's an order of you know just like a order of I don't know order of infinity level difference it's a many order of magnitude Improvement for how these otherwise complex protocols can talk to each other and that um yeah seems really

cool period yeah I I'm and and for those who uh are slightly less you know technical minded can you just give an overview Chris as well of like why you know I generally you know from what I know about um these sorts of proofs I mean they just represent such a massive Improvement um to how like interop might work today like what what is the reason these things haven't been fully implemented yet and like what are kind of some of the major issues like I guess I'm even just wondering in like a a sort of a cross chain transaction like how quickly does the proof need to get generated in order to do something that looks like Atomic composability across multiple Chains would be like immediately where my mind goes but walk

me through what some of the initial holdups are and then we can kind of end on this idea of you know the asset firste protocol we've been talking about right that's a good question I mean I agree that proof generation latency matters a lot for interoperability but um proof generation latency corresponds to the cost of compute and the built in latency bottleneck and blockchain systems corresponds to the speed of light and one of those numbers goes down over time and the other doesn't so I think we'll probably be okay at least give it enough time um why has it taken so long to deploy these system I I think it's actually going pretty quickly like blockchains are distributed systems they're hard to change plus they they you know maybe one reason it's slow is

that there tends to be a lot of like at least nominally US dollar value flowing around in these systems so people are hesitant to deploy untested cryptography as they should be um you know I also there's a lot of the zero knowledge field was really pulled especially by zcash from like Theory into practice in a very short time and there's still a lot of innovation going on and we don't have you know uh like we aren't trying to standardize a single zero knowledge proof um system right because there's still so much Innovation happening um we're trying to standardize things in anoma at a higher like comp compiler level so I think a lot of the stuff will get played over time it just takes time Mike can I get one more technical question in here before we go into the

um protocol asset stuff um and and Chris like keep this short I think but um I am still curious how aoma enables like uh cross-chain settlement like how does that conceptually work if I and just to give you an example my intent here is is sell asset on chain a make it ethereum um uh or I want to yeah I want to bridge asset from ethereum to salana like let's just say how how do the solvers solvers will compete to do that but how does aoma enable um this cross chain settlement right um so to be clear there's sort of no Magic Bullets and you still you know in a consensus system you

still have to run transaction any piece of of state has to be owned by like a specific consensus Quorum so we can't like magically do atomic settlement across completely different chains um but anoma can do a few things that should make users lives easier one is just the more declarative nature so users can specify what they want in their intent what ass they want on wit chain and sometimes there may be multiple steps involved in actually getting the asset there but the user specifies what they're willing to give and some solver can do the steps beforehand as soon as they see the intent right just based on that Declaration of like if you get this asset to this chain I will give you this other asset and the solver confront the capital there and move the asset so this is this is similar for example to I

think the Unos swap V4 optimistic bridging design something like this no it's not this is very similar to my day job so this is like which is unis Swap and Unis swap X's cross chain design is similar to this too so you have uh this third party front money um in across it's the relayer the front money um uh and then you verify so they make a loan and then you verify this is how you emulate composability um and then you verify that that loan was filled um this is frankly why I'm asking the question because I'm like did you figure out something that I didn't know about or some magic bullet here right but um yeah keep going sorry yeah yeah well sorry

for getting the protocols wrong um the that's one you know one thing that can happen Anda doesn't uh I suspect it's not you know not competitive with what you're working on it doesn't specify a particular way of like making these loans or whatever but just by the way intents are written you know solver some kind of already do that right they don't need like extra they don't need an extra protocol um yeah the other main difference is just that the scheduling is happening so users in their intents instead of saying you know uh I want like you just can say that they're fine with having an asset on ethereum or Solana or Cosmos a strong example of when you would probably probably do this is with something like usdc like what

matters for Redemption of usdc is what Circle considers to be reality it's not the etherum or salana or Cosmos blockchains it's like Circle so for the most part most users I think should be fine with receiving circle on any of the chains and if you have a bunch of intents with basically less specific constraints in the intent pool then um solvers can figure out like which batches of those intents actually need to be executed atomically and send those to their chains um so that is kind of like figur it's it's not solving cross chain bridging by like bridging between existing Chains It's actually figuring out what the chains are dynamically and then making less bridging necessary if

that makes sense yep I do love it and yeah I got a whole bunch more to talk to you about on that but it's very cool and I remain an intense maximalist so let's go let's go baby all right Chris I I want to end on on this idea of separating out uh competition vectors for protocols versus asset so I I first just want to understand you know how in an intent Centric view of the world or an intent Centric topology to use your phrasing uh unbundles these Concepts because even when I think about ethereum the ethereum trust Zone within the context of the system that we've outlined here I still my mental model for it is you know a Searcher if the intent is hey I want to deposit eth in a

which is smart contract on E you know my mental model is still like a Searcher well then go essentially route and do that on your behalf but you are still like the Searcher is still subject to the gating of e the asset to you know access those trust assumptions and probably a the app itself so can you just mechanically explain again for listeners like how these things are getting unbundled and how this might work differently in the future right so I I to be clear I agree that at the moment if you want to post post a transaction to the ethereum like main blockchain you have to pay ether for gas that's just how the ethereum protocol Works um now it could work differently it could say like um you know you could pay a different asset

that's not ether but somehow it goes to Ether holders ethereum doesn't do that but other chains like Cosmos do um Cosmos Hub um but in a world where many users are not they're not necessarily sending their transactions to ethereum they might want ethereum security eventually but the ordering and the kind of application Level you counterparty Discovery matching is probably not happening on the ethereum main chain because it's very expensive to do counterparty Discovery with eth gas prices um then these users are uh you know they're probably paying like some other asset or they're paying implicitly with slippage or something like this and if they want ethereum security someone is paying you know for an ethereum

transaction eventually now this this already kind of implicitly happens with sending transactions to rollups stuff like this um and the sense in which an intense Center topology unbundles it is that it allows the users to make their preferences in what security they want to pay for explicit and it gives ra it kind of inverts the supply demand so at the moment there's a fixed specific supply of Security Options there are rollups which have different security models there's the ethereum main chain which has its security model their other chains salana Cosmos which have their security models and users can like pick between those security model models by sending their transaction to different places or by as as you said by sending something to a solver that says you know

uh I want eth to be deposited on a on the ethereum main chain um but with a sort of more General intense language users can describe security models that don't actually exist yet like that that no one has produced a block for right users could say a signature by these validators is sufficient for my interaction because I'm trading like 50 bucks and I don't need to pay for ethereum main chain security another user who like wants I don't know is buying a um I don't know small country could say I want security of well don't you laugh but once the prince ofenstein tried to sell this country to Bill Gates or something like that um the you um anyway that that Prince ofenstein Lenin

could say that he wants ethereum and Cosmos and salana security for this transaction and like he's willing to pay a lot right um and then that would provide a basically a market signal to the validators of these systems that if they sign a block which includes this transaction like they'll all get paid right uh you can understand an ethereum transaction as it exists today as an intent on the peer-to-peer Network that just has this commitment that if you the ethereum validators include this transaction by the protocol rules ethereum you will get some ether um so that is already a signal it's just like a very Del limited one and a more General language allows for users to express pressing they want okay Chris help me understand I One Mini question for you on you know how do you

think about reaking or shared security actually now like there's sort of a more explicit attempt if whether that's a mesh security over in Cosmos land or interchange security and now ethereum reaking even Babylon and Bitcoin I mean h how should we understand is that almost like a protocol you know is that renting out uh the ability to you know access its security model would it be Gasper in the case of ethereum like how do you kind of think about um this idea of shared security within your sort of model and where that belongs in this kind of multi-chain future that we're discussing yeah I mean I think uh there is a wide variety for example of reaking systems and they have pretty different guarantees so I think it might be

helpful to pin down a little bit at least in this context how I would understand security um and I would understand security as um basically how difficult it is to revert a particular history so when users pay for ethereum security they're basically paying the ethereum validator set to agree that if the ethereum validator set refuses to consider this transaction in its history in the future the ethereum validator set will like lose some amount of money or some amount of ether in this case um and different reaking protocols have different like they have different bonding systems and different stake amounts and so they're providing very different levels of guarantee right and I would say that the guarantee actually

has less to do with the asset and more to do with the specifics of what is locked and what the slashing conditions are for example and whether there are other locks also made on the same Capital One you know worry I have about the way reaking is happening in practice is that it's kind of fractional Reserve like um if you stake the same Capital many different times or many different possible things you could lose it for you don't get additive security right like if several of these things happen at once you know uh they don't they all they don't have like any more security than just the assets that you actually put up at stake um so it's difficult even to reason about this independent of like what is the application usage yeah

the seven Sigma event becomes a real risk now right yeah well especially because it's like transparently exploitable by me Bots like instead of instead of this being you know some financial markets typically sort of try to regulate these things out of existence they fail anyways but they like maintain the Illusion for a few decades at least they often do um and I don't think we will have that luxury which is probably good in the long run but maybe chaotic this isn't aside but that is an interesting point where in traditional financial markets you're seven Sigma event which happened way more often than segment s like like five orders magnitude more often than seven sigmas was supposed to imply um you couldn't uh model it because you didn't

know the state of the world like it wasn't on a verifiable blockchain but you're quite right that if you can see the state of the world and verify it all maybe you could actually exploit or engineer a seven Sigma event which seems a little scary yeah I mean the good thing is that at least these commitments in order to work typically have to be quite public like if you were going to use a system you want to know what the validators have bonded what they have committed to when you use for example ethereum implicitly at least you know this um and that I think would allow for some more you know especially if the the protocols for these kinds of commitments can be standardized some more measurement of how much you know Edge risk there is in the system um and even you know for

example with intents you could say something where uh the validator set that you want your intent to be settled on isn't specifically identified it's actually identified by what they have staked so you could say I want you know four validators with a combined stake of 100 eth to sign this transaction something like this um or and you could say that a combined stake of 100 eth and like no commitments that conflict with blah blah blah you know you could get more complex to try and avoid some of these problems so that might be help that's pretty cool all right I've got two closing questions for you here Chris which is one L I want to get your thoughts quickly on the asset side and and you know what's made this more interesting

for me recently is I feel like their Bitcoin has always been explicitly it's kind of like a money app chain right like that is their export is Bitcoin the Tok and that's that's explicitly been the goal from day one but it feels like a bunch of other l1s recently are starting to compete on that same basis so you know if whether or not it's the eth is ultrasound money meme and then they've got modular money in the form of Tia and I would just be very curious you know I think the mental model for a lot of listeners maybe this is the way they'll frame it they talk about something like block space profitability and they're talking about very different things and you know if you're in the ethereum camp you'll say look at the fees look at the fees that we generate and look at the

Block space profitability that should translate into more adoption of easin if you're on salana you say hey those fees make it unusable and actually you can do so much stuff over here on our protocol therefore salana should go up they don't say that but that's the implication obviously right and what I'm hearing from you is that actually what these protocols do themselves is like much less relevant for the assets um and their adoption so can you talk a little bit about a like how would you view like this kind of this idea this is a long-standing debate internally for Block Works about how you like the p&l sort of versions of these l1s and then how are assets going to compete with one another as a currency or money-like

asset as they're increasingly divorced from the protocol yeah so first I think just to illustrate the um decoupling of assets and protocols a simple example will suffice imagine that there are a lot of people who want to use eth the asset but on salana as in they want to like buy ethereum they want to trade with ethereum they want to generate a whole bunch of demand for ethereum either asset um but they want to do that on salana because I don't know they like the Berkeley packet filter BM um if you know this doesn't benefit ethereum it doesn't like generate more transaction fees on the ethereum main chain it doesn't generate more usage of

the ethereum protocol it doesn't generate more solidity contracts or whatever but it clearly does generate a lot of demand for ether the asset so um I think that example is like enough to demonstrate this tension right if what you want is for ether the asset to be valuable then you should support uses of ether the asset and demand for ether the asset that is not dependent on ether of the blockchain um because it helps you and if you want to support ethereum of the blockchain then uh you know these applications on etherea might pay gas fees but they're not necessarily like you might be missing out on a whole bunch of other things that could generate demand for the asset um so to talk specifically about Assets in competition um at least my expectation

is that eventually this kind of these protocols as sort of infrastructure as rails will be quite standardized and there will be many assets which users can choose from so in other words users will be able to choose assets and protocols independently like you could use ether the asset but you could use it with anoma or you could use ether the asset with Salona um and I don't necessarily you know this may or may not be tied to the security domains but users will be able to make those choices pretty independently um and then I think the kind of differentiators of assets are really just the distribution function like how the asset is issued to whom it is distributed um and the community like who holds the asset um you know users ultimately you know you

choose to you know someone can give you a dollar they can give you an either but you like choose whether you do something you know you choose whether you give them a sandwich you choose whether you write them a report you choose whether you uh you know help them out um so you're the one making the choice and in making the choice you can choose which assets to accept and at least personally I want to accept assets that are kind of like compatible with my values and things which I want to support so you know by accepting ether or by buying ether I'm supporting the ethereum community and I might want to do that uh so I think that the kind of distribution and Community will be the primary asset differentiators nice uh Chris final final question for you to round down here this is uh something that Hart and

I have been wondering because I think you know if you if you almost look chronologically if you start with Bitcoin Bitcoin is probably the most principled crypto Community right they have some very very strong principles they didn't compromise with a lot of their design but some of the stuff that were discussing and this you know comes up around intents like there are a lot of offchain kind of market makers that are fulfilling these intents and you know as soon as there's this tendency in crypto I feel like as soon as you start to make something that feels like a compromise everyone just throws their hands up and says like well we might as well be trafi you know and and I think one of the things that we're trying to highlight is there's plenty of gray room there's plenty of like tons of room for

incremental improvement from what exists today but maybe just in closing you know Chris how would you zooming out into this like future end state of crypto assuming that we've made some amount of compromises what would have made the crypto experiment successful for you like what does success look like at the end of the adoption of crypto r large that's a good question um I think to me the answer would be something like crypto will be successful if we can provide an alternative infrastructure that doesn't require communities to choose between autonomy and interoperability an alternative

Financial infrastructure coordination infrastructure that doesn't require communities to choose between autonomy and interoperability so at the moment you really have this dichotomy where communities even ones that want to be you know more economically self-sufficient must choose if they want interoperability to use like Global Fiat currencies they must choose to use the dollar or the Euro or the Yen you know one of these large currencies to interoperate economically with the rest of the world and using those currencies makes them really like dependent and subject to you know censorship or sanctions or just manipulation by the very small number of centers of power that are issuing them um and of course they can choose not to use those currencies but then they compromise interoperability they can retain

autonomy you know not use any of these currencies and sacrifice interoperability or they can retain interoperability but sacrifice their autonomy um and to me the kind of Promise um not yet realized but still possible promise of crypto is to provide uh you know just a set of protocols that a community can run and communities can compose on each other and run with other communities that allow them to preserve both of these values autonomy and interoperability man that was a great answer yeah that was a real cool answer could talk about that for a while yeah um I actually get the feeling Chris we could have gone for another hour and a half here but uh you know I want to be respectful of your time this was super

valuable I think from I speak for hard here from both of our perspectives if folks want to find out more about you or the work that you're doing at anoma or NADA as well like what what's what's the best way for folks to give more information uh you can find out more about anoma at an.net or NADA at.net you can find me on Twitter sometimes at CW Go's um I I also want to plug we just recently released in in public this internal peerreview system we've been working on called anoma research topics which you can find at art. ana. net um and we're trying to create kind of also a peer review system that works you know for the needs of blockchain protocol

designers would welcome contributions participation there awesome Chris this has been super valuable thank you so much for coming on about to do it again soon thank you cheers guys all right har man that was a great interview kind of had High Hopes honestly that's why we put Chris as the kickoff of the season but I think he even exceeded my expectations or hopes for going into that what you think yeah I mean I'll be honest I was like a little bit nervous because it's like a big brain like a really really big brain and um a a big brain in multiple Dimensions right there's like all this kind of like economics and philosophy

like you said at the beginning um there's like the economics and the philosophy and there's the technical aspects and like deep on kind of the history of computer architectures and so there's a lot there and yeah I think he's also a very articulate guy um and I legitimately um both in prepping for this and then going deeper with Chris I really have some real insights that hopefully we can talk about now um that I think do do impact my view of the multi-chain end game I I agree let's let's just leave right off with that like you said you had two or three you know before we hopped on here so what was you know kind of take it take it wherever so okay let's talk intense like we kind of labeled Chris as the intense

guy um yeah and I on the on the show also said you know I'm an intense maximalist which uh which I am but it's interesting to think about and have um an architecture for why this intense concept works so well and you know maybe what we can do in this this kind of um summary bit is is we didn't talk about like the alternative to intense um and what we got into in the episode was this idea of like okay you have ethereum and salana and think of them as like different processors that work in different ways um and one way you could communicate between the two would be write at a program specific to ethereum um that specifically sends messages over

a different protocol to salana and then write a program in salana that can receive those messages and send them back to ethereum right um and this fits within the kind of like transaction like mindset so now if I want to communicate between these two protocols I'm going to execute a transaction on ethereum it's going to get to my my interop Pro program and it's going to send a message over another protocol to uh salana my salana program is going to receive that message executed send something back um and to me I think that this is kind of the the first way people were thinking about interoperability like this is how we're going to connect these ecosystems we're going to write programs native to

each of their architectures and introduce another protocol uh that communicates between those programs that are native to those architectures and um I I think like what Chris is saying with intense is he's basically going no no no no no wrong wrong way guys like instead let's define the protocol at the level above the abstraction layer above which is the intent what does the user want to actually do here and that intent is going to specify constraints and the state transition and all that and then I don't actually care if ethereum exe how ethereum executes it or how salana executes it I care about where I get to at the end and to me that seems like

such a such a powerful way of architecting or designing the system where you know now if I have a third blockchain or processor I want to slot in like Cosmos or I want to slot in um Aptos or or Tron or whatever else like they all work on top of my intent layer whereas in the previous version I'm talking about I'd have to build new connections between them and update my protocol to like send messages to them and all this other kind of stuff so I think it's just a really um powerful mental model for how to abstract interoperability and like abstract away the chains that users are are

touching I would completely I would completely agree with you on that and it feels like it first of all it makes for a much better more unified user experience but it also my philosophy on crypto for a long time has been run every experiment just try everything and and I think this allows you know within that context of flexibility it allows different protocols to express opinions different design decisions and sort of compete it out and like ultimately that's where you know a a Searcher will will route something if it if that protocol ends up producing something that's super valuable so like I I completely agree with you on I just think it's a super

powerful framework and I think Chris articulated it well in the episode yeah and and I guess um my other Insight I I'll go like on the more philosophical B bent and then the last one is more on the technical side the economics of this get really interesting too um and this goes into like Chris's bit about the platform economics um so this is maybe less about interop but maybe more about our multi-chain endgame and where value ends up acur um yeah so so like the Chris was really articulate with saying hey um the ethereum platform if you think of it like a web two business okay we're

going to use the ethereum platform write transactions that ethereum executes build applications on ethereum it uses ethereum asset like eth e eth eth go go go go go and by the way I don't actually think ethereum is ethos is this way I'm just plainting this as like the the mental model if you're trying to frame ethereum as a web to business and the idea of this being platform economics where the whole all the incentives are aligned to Drive the user to eth and use eth and then then it makes sense that you can kind of have an extractive like you can charge rent to use that system and when Chris starts talking about some of this intense stuff where where I see it going is he's saying hey actually let's define what the user

wants to do at the level above and then you effectively can have all of these platforms compete to fill you so ethereum is going to compete to do the thing you want to do salana is going to compete Cosmos is going to compete whatever else and in a sense you're disintermediating the power of these platforms um and I'm not sure I buy this or believe this yet this is like a new thought for me but like if you play it out to its logical extension he's saying okay users are now in control block space is competing to fill them uh Hey block space compete to fil me best and uh you kind of remove a lot of the power that the platforms have um and put it

back to the user who's actually creating the intent and that's like again a pretty wild thought for me yeah I have so much to say on this because I thought about it quite a bit and it was I had this unlock a little while ago when you just talked it was during the cosmo season of the show when there was a perfect 180 degree reorientation of my worldview because when you when you hear this is kind of like this was articulated in the fat protocol thesis which is extremely popular at one point in time but even just the user experience that you have as a crypto person you're like okay this is the the flow as I see it probably the canonical flow how most people get onboarded I buy eth or Bitcoin let's say

eth because there's stuff to do eth on coinbase I then probably want to buy a token on Unis swap so what I do is I take this eth and then I you know I put it in my metamask wallet and then I have to like go to this app and it almost it it is a very intuitively feels like I am in this platform and I in this ecosystem and most of the arguments that you hear about how value acrs in crypto is is directly descended from Bitcoin right it's like 21 million hard cap very Supply based arguments and if you the reorientation that I had was eventually there are going to be apps that have proximity to the user and you could actually think about all of the control

and leverage coming from there whereas there are these like kind of multiple different sets of you know things at the bottom of the stack which like users might have a preference for or might not and they're almost interchangeable with one another and I think I think that's why there's so much emphasis on this idea of moneyness from all of these different communities and this idea there was an idea one thing that I'd love to to get your take on is this idea of economic abstraction has been around for for a long time and I actually remember it coming up within the discussion of EIP 1559 people were really concerned about the idea of economic abstraction and what I mean by

that is there was an understanding that in order to use the ethereum ecosystem you had to pay validators in eth and the reason the B part of the reason that the burn was implemented was to get around offchain payments and part of that was to make it incentive compatible right you didn't want uh different miners at the time colluding but another concern was that people could make offchain payments in something other than eth and so if you are making offchain payments in something other than eth you have de facto removed the restrict like the gating mechanism for eth and what I would point out in service of Chris's idea of these things already becoming more separate this economic abstraction is happening it's just happening at the

layer of the level of layer twos right now so like a layer two to like ZK sync that has embraced economic uh account abstraction you can pay gas in whatever you want and so it is just I think there actually is an understanding on in these different blockchains that what my protocol does I'm using my I'm losing my lockin you know I'm losing the ability to gate my service with my token uh in one way shape or form or the other and therefore I my business model is to export this token it's just I'm so interested in that as an idea I find it fascinating Mike you are saying some sacrilege things like these are taboo topics man I know I I I Feel Like These

are going to get labeled as unaligned but I I personally am a believer in like just say just just describe the system and then root for the best result right because yeah that probably is what everyone's trying to do I think no I agree and like the ethereum community they are intellectually honest so I I do think that's true um but I I think that this is again super fascinating and this is not what I thought when you know you told me we were going to do a season on the multi-chain endgame I didn't think we were gonna get into like eth economics exactly but um I I do agree with you that in this multi-chain world like and if you really zoom out and and think about it um you're saying hey uh multiple chains work seamlessly and they

work for the benefit of the user not say for the benefit of ethereum right um and if chains are working for the benefit of the user and not for the benefit of ethereum ethereum in a sense is losing its lockin um and it becomes more commoditized um in a sense now readers listen to this uh or listeners please realize I'm a huge eth bag holder and I like ethereum all that go ahead yeah actually I think there's a really bullish argument for ethereum here and I I still don't know where I fall on this but I I think there are two groups of people that are sort of talking past one another there's the

protocol comp competition people which is largely the salana camp hey it's really cheap to execute transactions here and by the way problems with salana like Jupiter launched today bunch of failed transactions like there's still a ton of work to be done in all of these ecosystems so none of this is in all of this is intended as just you know constructive criticism I'm rooting for everyone but I will say I think think the the eth people for a long time have been very concerned with the vectors of competition for eth the asset and I think they're further along in this realization than the salana camp where they're just like look I'm going to do all this great stuff for the protocol and value will ACR to the token and I think the fair push back from the ethereum camp has always been I don't think those things are as correlated as

you might think they are and what I heard from Chris is that that's exactly right and I don't know so I feel like if that becomes the the business model of all L ones is to export the token then probably ethereum I mean I just think ethereum is in the best position for this personally I agree with that and like you know I it's important to say the scary part out loud right like what if blockchains are more Commodities than we think they are right um fine um like still let's we're still here to build cool end of the day right and um uh if blockchains should be in service of what users want to do their coordination mechanisms and uh I do think that's like

what ethereum is really good at where they're intellectually honest and we want to build um the most useful smart contract platform out there um and then this multi-chain thesis is multi-chain endgame is like there is going to be competition between blockchains and that's healthy and good and everybody you have to be pretty intellectually dishonest to dislike that like there will be competition between blockchains uh to run applications on behalf of users um that's good um and the multi-chain world is going to make it easier for those chains to compete in a way that like you said really kind of a new thought there it's reducing the vendor lock in um that one uh ecosystem

has yeah all right oh yeah Sor go well I was just gonna say I got one more insight but Mike where were you gonna go I was actually going to go to I wanted to go to zero knowledge uh the proofs as a way of unifying interupt that's where my mind was going that was where my mind was going say say more about that because I I feel like this is a blind spot for me zero knowledge and proofs in general and I'm slowly learning that I need to really up my understanding of this and I'm almost wondering it's worth dedicating an entire episode to because it feels like very interesting area but what were some of your takeaways from that part of the discussion so the zero knowledge stuff is always worth dedicating time to but man it makes your head hurt right like there's just some

of these concepts are just hard it's hard to wrap your head around and I'm like so I'm not there yeah it's true moon maath but even not not even understand the math just even some the comes of concepts are like wait how is that possible um uh there's like uh this fun uh interactive uh zero knowledge proof with like hats like black hats and red hats that we could maybe have somebody talk us through uh to that it's um I think it was the algrant founder that actually came up with this proof but anyways there's really brilliant out there um and I'm probably going to say some stuff that's wrong and if I am wrong people on Twitter should correct me um but my unlock from what Chris said and

this I think is pretty cool um let's go back to like defining a protocol like a protocol is just a set of rules for how you do things okay that's like one definition for it here's my set of rules for how I do things uh there are real world protocols which are like Loosely defined you know like the protocol for uh how you order dinner right you know but and then blockchain systems um and computer systems in general Define protocols with deterministic rule sets cool um you can imagine that if your protocol is a tur complete computation environment theoretically um you can run

a computer program on your protocol to interpret some other protocol that is possible um but it might be a lot a lot a lot of work right so specifically I could run a protocol to verify the state of ethereum by understanding the ethereum execution environment its Fork Choice Rule and then running through the entire history of ethereum from Genesis to today be like here you go right and the the thing for I think the listeners to kind of Gro is like yeah you can do that and that is so much work that is just a ton of computation a ton of work to do right and then the unlock is zero knowledge proofs have two properties

they have the zero knowledge property ignore that they have the succinct property which is the cool part here so I saying what I can also do is run a zero knowledge proof proving the current state of ethereum so I generate this proof and today it's still pretty complicated to generate this proof proof but I can generate this proof and it compresses I think of the succinct property as a compression property it compresses all that work of like running through the rules of ethereum through all of its history to understand the current state of ethereum it compresses that into this one proof and then I can go and like verify the proof I can look at the proof and be

like through Moon maath it proves all of this work through this kind of like crazy compression concept and so the unlock I had with Chris was like holy if you can do these zero knowledge proofs that compress all the work required to verify the state of another blockchain right you can have protocols that would otherwise be really really hard to like plug in together you can kind of glue them together with the succinct property of zero knowledge proofs and that was like a conceptual unlock for me where I'm like wow yeah that makes sense so how uh maybe Hart could you

because I understand that at a at a high level and I actually saw I feel like I'm starting to see I know that you know zero knowledge accounts in Celestia is a massive part of their road map I saw polygon recently introduced this thing called the AG layer which relies on ZK proofs which is I I I'm I'm assuming that multiple different camps are starting to align on this idea idea how specific you know when you say you're compressing there's a it's a compression technology and you're getting a view of the state like how specific is that like could you look at this one proof and somehow understand something as granular as like how much eth I have in my wallet on ethereum like H how much is how much information is

expressed in that line of code I mean a lot right um so let's go into like AUM um is like they coin the term ZK Pro co-processor right so it's specific to your example um I want Mike you give me your ens address like mike. ether or ens or whatever um and you want to know you want to say hey prove to me how much eth I have at this block height from a week ago in the ethereum state machine right and Axiom has a ZK circuit that can generate a proof that shows exactly that and that is that is nuts that is crazy to me that that's possible that is so cool yeah and again I I do think and we

really do I I really hope I'm not saying some of the ZK stuff horribly wrong um but I do think we need uh like a ZK person to get us even if we do this outside of the of the Season just to to to up our own knowledge here right but um um do think of it like compression so I mean in my example of the axium proof the way could do that without the ZK proof is I just recreate I go from Genesis block and know the fork Choice Rule and I go up to the current state of ethereum to that block height and I look at your account balance right like you get how that's possible and so for me the mental unlock is well compress that compress that with math and this math it's it's statistical in nature it's kind of like a hash function right like

it's um but it's so statistically accurate you like can take it as true um and uh uh I can prove that statement um so so yeah when you start thinking about um you know another protocol uh succinct uh succinct is a company that uses the succinct property of ZK proofs to um communicate like the current state of ethereum to other evm chains and they do that in a way that is like totally trustless once the proof is generated anybody can verify the proof and be like this is true or not so anyways we're not going too down on ZK stuff there's like lots of other content there but ZK in

the context of interop and specifically this idea of protocols that could be very incompatible in how they operate like they have going back to the processor example they just have like different op codes like the way you program on them is just say very very different um you can glue them together efficiently with like ZK proofs um and that that I thought was a cool concept I thought it was cool too I think you're going to hear more about this as well I I know vitalik loves this idea of a bulletin board for ethereum and he's obviously super excited about proofs in general proof aggregation I heard him talk at ethcc uh this this past July about

exactly this and I feel like that analogy of like the bulletin board where there's an enormous amount of work that gets done in these more computationally accommodative environments but then get sort of posted to this chain and I don't know I had this it's I guess one open question that I have and maybe we can end on that idea that Chris gave us about what a successful you know you're gonna get I didn't know where Chris was going to go when he's like what is a successful outcome and he just had this super interesting idea of essentially removing monopolies around money and sort of control what a great way of expressing that and uh that is you know it's been a it's been sort of a question for me in general as you know if you have this

mental model of the world where there will be these zones call it salonik zones maybe where there's a lot of some offchain comput like either offchain or in uh environments where there's onchain computation but like higher Hardware requirements for computers being done but then kind of rolling down to this extremely secure it's kind of like security and interoperability are the two those are the two important things from the perspective of Basel protocols and I think you're starting to see that design in ethereum uh you're starting to see that design in Celestia Celestia is even less performant than something like ethereum you can't even it's lazy Ledger right you can't even do smart contracts on the base layer of Celestia and so I think that this idea

of leveraging ZK proofs and having the Bas layer be this like very unopinionated very nonperforming kind of bulletin board aggregator is probably it seems like that's the way that design is being that the design is going for major chains um but what what did you think about that idea of it's actually going to lead perfectly into our next episode as well so I love that he ended with that but what did you think about that idea of um you know M basically making it so easy so everyone can express their currency preferences and it doesn't have to you know we don't have to make this trade-off in between interop and autonomy I I love it I think it's a wild idea and kind of like a big idea too so um if I'm going to regurgitate it and

think about it differently it's like uh Chris's the world Chris wants is a world where communities or groups of people um have complete autonomy over how they choose to transact how they how they choose to coordinate financially um and you can think about that in different vectors you can think about what currencies they they choose but it can even be broader than that like just like what Financial products and systems and services they offer like just how they organize their financial ecosystem um and so you give everybody autonomy fine every group of people can do what they want that part seems easy um but then the hard part is like then you connect them all so they're actually

interoperable um and what that means is that the user has complete choice and complete Freedom over how they want their financial I mean it's actually it's double Freedom I'd say you have freedom to pick your community and like your uh how you want to interact financially um but you have freedom to not constrain you to only use that you have interoperability with other U Financial Stacks here and uh that is like a cool concept and I do see how it directly applies to his intent like framework here too where you're

giving each crypto Community each ecosystem complete autonomy over how they design their stack how their blockchain Works what their design choices are and you are defining at the layer above um a method to in orop through this like intent layer um so I just thought it was a super cool answer super cool answer and it's going to lead perfectly into this next episode that we're doing and the title of this next episode It's All About chain abstraction and a lot of what we discussed today is under the would be under the umbrella of chain abstraction with Ilia Pushkin of near and Sam Hart of skip protocol and actually to give you maybe an analog I

think Sam and Chris think about the world very similarly he gave me this example and I'm sure we'll probably discuss it on the episode but imagine in a future you're buying an orange and the Orange is quoted in dollars but you have Bitcoin and I have euros and I just like we can transact with one another I only ever see the Bitcoin kind of quoted price you see the Euro quoted price and you know the vendor thinks about the world in dollars and the way that I would describe that actually Sam would describe that as um localizing the store of value unit of account like the localizing the the unit of account basically so instead of a global accepted yard stick everyone has their

own yard stick so maybe that's a good place to end it big concept and uh yeah next episode's going to be great this was a really fun one heart and yeah can't wait for next week yeah thanks Mike it was a [Music] blast