Leveraging Intents for Seamless Chain Abstraction

Captions as YouTube publishes them, usually machine-generated, so product and people names can be misheard.

1. 0:08

   all right check this out last minute audible Marcus uh who was going to be joining us on the panel if you don't know Marcus propeller heads is a really dope shop check them out they're really cool so I want to plug Marcus and not just totally rug this opportunity from him but we got jacqu from OS who's coming in uh filling in last minute thank you so much jacqu and thank you all for being here and coming around to uh hang out and talk about intense with us um if you were in the earlier keynote

2. 0:33

   that Dean gave there was a question by an audience member really really good question about what is the relationship between intense and chain abstraction Dean gave a really really good answer that I'm not smart enough to summarize but we've got this awesome panel of experts and jacqu here who are going thanks Jack we got an awesome panel here we're going to dive into intense and how they can actually help enable uh enable chain abstraction and in general what

3. 0:57

   intents are and how they are impacting web 3 landscape overall all right uh so just to kind of get everyone on the same page I'm going to put out at least what I think is a pretty solid generic definition of an intent and everyone is welcome to refute that or not after I Define that I'm going to let each panelist introduce themselves and their project and then tell me how close my definition of intent was um so generally an intent as

4. 1:22

   it applies to crypto and application development it's a highlevel userdefined goal um that's specifies a desired outcome as opposed to a description of technical steps so buy an nft is closer to an intent than send tokens via cctp for example nobody wakes up and says I can't wait to use cctp today maybe I'm wrong and with that said panelists please introduce yourselves let us know

5. 1:46

   what project you're working with and how wrong my definition is J start by let's go um whoa hi I'm arjin I'm one of the founders of Everclear um Everclear is the first clearing layer it's a new primitive that we've introduced um it's very fundamental to intense you may have known Us by our previous name however which is connect um we are uh we are a project that has been working with intents for a while we were actually the first like intent based Bridge back in 20121 um and then with Everclear we're working to solve some of the core

6. 2:19

   problems that are associated with intent-based protocols which is now just about every bridge um uh and really one of the biggest problems being how do intent solvers manage inventory so how can you make accessible to everybody awesome oh and then your definition I think it's a great definition um I think another I think another helpful analogy could be um is that I I tend to use a lot and seems to resonate is Uber um

7. 2:44

   Uber is a fantastic intent protocol you basically tell Uber I want to get to this place and then Uber is like okay I'm going to figure out how to get you there as fast and cheap as quickly and cheaply as possible and you're not necessarily specifying I want to use this driver or I want my car to be like this or I want to to take this pathway awesome um hey everyone my name is kushy uh I lead research at Athos where we're building a programmable

8. 3:08

   policy layer and so basically what that means is that application developers can create policies which is a set of rules constraints conditional Expressions whatever you want to call it for their application and that determines what kinds of wallet addresses can interact with their application we mostly focus on risk management so that includes like exploit prevention anomaly detection um ofac compliance and things of that sort in terms of the intent definition I do think I agree with it for the most part

9. 3:33

   the one thing is like we focus on constraints as a company and so I think one really really important part of intents that's underexplored is how users can express constraints on their intents in a way such that it isn't an additional uh pressure on the user but still allows them to have some control over the actual execution of the intent those are uh very good uh introductions um my name's uh aior I work on aoma uh anoma is a distributed operating system for intents in simple

10. 4:05

    terms it allows users to express what they want uh in terms of intents instead of transactions and it allows developers to write applications in terms of intents instead of transactions protocol provides different affordances like heterogeneous trusts information flow control and intent level composability among other things um in terms of the definition that you gave I think that was quite reasonable I mean uh how we

11. 4:29

    think about intents um intents are credible commitments to preferences and constraints over possible State transitions uh maybe a little bit more technical but um in terms of like how we Define them in protocol I think it's a bit more precise or Atomic information flow constraints could be another one or one that I heard yesterday was uh predicates over State again like I think these are all kind of like going in the same direction

12. 4:56

    so hi Jack uh not an expert on intense uh work at fos BD so we do a lot of chain abstraction what we want to do is provide evm interface into the cosmos space that's one of the barrier trying to solve because a lot of developers you know RI the cbas is based on solidity to rewrite your c-base in another language you inherit a security uh you don't inherit a security guarantees with a battle test to coas so what we try to do is um again in in the

13. 5:29

    cosmos space place where there's so many different blockchains and to develop one application that works for all these chains is a difficulty so we abstract away in the form of evm so developer can deploy this EDM smart contracts and have it work across the cosmos chains so that's more chain abstraction but obviously we also look at the user experience size of things and for me intent is basically like what does the

14. 5:54

    want user want to do it it shouldn't require the user to think about what kind of wallet what kind of maybe even what kind of application maybe I just want to lend my all my assets and if that if I could just do that and have the system find the best interest rate for my assets that's great but I don't want to have to worry about you know like metam Mass I don't want to have to worry about bridging I don't want to have to worry about these things and that's what for me intent solves is what

15. 6:28

    the user wants to do and how we get there uh as easily as possible for a random audience poll that was pretty good right um awesome well thank you guys so much for that um I know I wanted to dive into right off the bat what is the relationship between intense and chain abstraction but I think there's one additional piece of context that could actually be pretty valuable here um and why are why are intents even a

16. 6:51

    thing like this is these are sort of this is like a new emergent property in in web 3 and there's a lot of momentum and interest behind it uh what was sort of the I guess the conditions that primed us to be receptive to intense Space Systems and why is there so much motivation what does success here mean for us I feel like that's a question for you yeah so there's a lot of questions

17. 7:17

    in there um how did intense start I mean I can I there's there's a lot of history right I mean I think like even if you look at something like the agoric papers or you look at the wyvern protocol um if you look at uh some of the early bridging protocols like um connects I mean like people have been thinking about this concept for some time um I'd say from our perspective we've been thinking about it since about you know 2020 um in terms of like concrete protocol for intents um if you think of

18. 7:49

    like what an intent is like the OG intent system is like language um you kind of say what you want and the person responds to you based off of that um there's also like body language involved d as well um in terms of what does success for these systems look like I mean I think we're seeing a lot of success already today uh with swap-based intents if you look at things like cow swap uh again if you look at bridging protocols like across I mean these are

19. 8:14

    um you know very simple types of intents but still um you can see how there could be a progression to something like modular Logistics like an Uber uh where a user can just express what they want I mean and even if you look at like Apple's new architecture um they have like an intent module in there so clearly like there's some memetics around the term as well which is helpful beyond our industry I think um going back to like

20. 8:41

    why I first gotten too intense um it was probably with the idea of a user focus in mind which is what a lot of people look at it at now um like crypto user experience was very difficult um you had to know a lot before you could do even a simple transaction on chain and I think the idea is to make it a little bit easier for users to onboard onto crypto so that they don't have to know as much

21. 9:03

    in order to get to the same end State and so it's supposed to help um improve user experience but I do think we still struggle with the Chicken and the Egg problem where we don't really have users or applications yet so we're solving the infrastructure side but we still need to solve the other side of that equation before we get anywhere um yeah these are these are fantastic answers um I guess to try to

22. 9:27

    like build on top of this a little bit like um crypto had its roots in like peer-to-peer systems uh you know like negotiating relationships with other people on a direct one-onone basis um and then we sort of transitioned away from that towards like like blockchains as consensus engines where you have like an end party relationship and everybody's agreeing on the same state because of things like reliability um and I I think like it made a lot of sense right like blockchains are a truly

23. 9:52

    incredible Innovation um but then I think we've overfit a little bit as well where we we were like okay let's just use a blockchain for everything right like let's put everything on chain and um and the reality is that it the the right kind of topology the right outcome is like a combination of the two where you need to have um you know as as as you guys mentioned like there's there's like a lot of complexity inherent to what the chain onchain world today looks

24. 10:16

    like um a really simple example of this is unis swap like Unis swap V3 Unis swap V2 was like onchain contracts you called like a simple swap function you didn't need to provide any data and you were able to swap for tokens Unis swap V3 requires off chain routing data um that's a that's a total Paradigm Shift right if you're building if you're trying to compose into building with Unis swap V3 it's you you can't

25. 10:38

    fundamentally as an application you have to like go and fetch offchain data to do that who is going to go and fetch that offchain data um so what's basically happened is that the space is just like the amount of as the SPAC has become more sophisticated the amount of context you need to interact with applications the amount of context you need about the chain state for both the chain you're on and then also other chains has just grown and so we need need a way to resolve that we need we we sort of like

26. 11:01

    need to add a gap between what the user is doing which is like wanting to use an app and what the chain needs which is all of this other offchain contextual information and that offchain peer-to-peer layer that is intense that makes perfect sense and I mean I think you hit the nailed on the head overfitting and trying to put everything on a blockchain I mean that doesn't sound like our industry at all that's Nobody Does that but we we've

27. 11:24

    talked a lot about how intents are really good for the users and sort of the the end result of what happens when we have these intent based systems um but how much does it make things easier or more difficult for the developers like from a developer perspective are these applications harder to build yeah one one way to think about intents is like what you're effectively doing is offloading like you're kind of

28. 11:48

    offloading all the complexity to a service Rider right and like the great thing about offloading complexity to service Riders is that they can instead of having to do like domain specific stuff for like a custom thing you want to have to do every single time you do it it the service provider can just do that at scale right it's the same as like you know why you go and like fetch

29. 12:04

    data from an API instead of having to like like a a weather API instead of having to like go and like figure it out from like lower level GE geological information all the time right like it's just it's just easier there's there's like economies of scale you get by just doing the same thing over and over again as a business um and that's effectively what like intents are doing is like um you are offloading this complexity to solvers who can go and be domain specialized businesses on one specific

30. 12:27

    topic or one specific route or one specific thing um and and so for for developers this is a form of abstraction it's a form of like complexity reduction right where as a Dev you're not having to go and figure out okay how do I optimally route through Unis swap how do I optimally route through like these aggregators you're just like I just want to get there someone else figure it out and he brought it back to

31. 12:50

    abstraction it's almost like we planned this uh but that really brought us right to the relationship between intense and chain abstraction and in particular um like Kushi I'd love to hear what you have to say about this because you're um a very specific form of an intense implementation focused on policy which is uh seems very much more on like the business logic side of things so how did your like how did you guys approach intents and sort of arrive at that as

32. 13:15

    the conclusion for uh like a model for you yeah absolutely so I think um going back to a little bit of your previous question as well like does like using intent-based systems make it harder for the developer um and I think it does to a certain extent like I absolutely think we're abstracting away the complexity of execution which is convenient um because you have all these other elements of the system and that part makes it easier for

33. 13:39

    the developer but at the same time developers now have to be a lot more opinionated about the way that they're building their system so if they're abstracting things away from the user they're now making a lot more decisions on their end about how they want their application to function and I think we'll see a lot of front-end fragmentation because we'll see different front-end specialized different things because those developers are all taking opinionated

34. 14:01

    choices about what their front end enables so you'll see front ends that are very very focused on security and they'll make sure that their system abstracts away the actual details of that security but they ensure that those intents are executed in a secure fashion whereas other front-ends will be very very focused on having the best price execution or the fastest execution and so you'll allow you'll see users able to make choices of their own but it'll be based on the application that they

35. 14:27

    choose to interact with and those applic will be making the actual transaction details and those decisions so I think it's like developer complexity increases slightly but it's not in terms of like a ton of coding complexity once we have nice standards for intents but I think it's more around the idea that now you have to be opinionated about the application you're building and how you want to build it and the types of users you want to serve and so that's why we focus very much on those constraints

36. 14:51

    because those actually Define the types of execution that users will receive at the end of the day you you mentioned um standards are there is there any work being done around developing some intent standards uh like who's working on that is it seem promising yeah I mean I one standard that I know of is um the one being developed by across and Unis swap in collaboration like that that's a

37. 15:16

    standard being developed for cross chain um intents which I think is is a standard that a lot of people are looking into but I think there's other standards out there as well um if anyone wants to speak on those so I think um in terms of Standards they're uh always a little bit tricky um because you can make a standard and it doesn't necessarily mean folks will adopt it um so I'll give you a concrete example of this if you look at the in blockchain communication protocol it's a great standard for

38. 15:49

    interoperability um it has some drawbacks um but at the same time it is like quite neutral and it does give you trust minimization uh when you're moving funds between chains and unfortunately or fortunately for for lack of a better term um not everybody in the industry has adopted this standard because there are certain needs um that different protocols and uh different architectures

39. 16:14

    have um and these are like path dependent decisions right so like ethereum block times are 12 seconds and it takes 13 minutes for uh the uh Gasper protocol to finalize a block so you're not going to get this single slot uh finality where you can relay IBC headers right um and roll up sequencers for the most part um can't give you this guarantee either because you have to also wait for the optimistic fraud prooof window to close or you need to wait for a uh succin zero knowledge proof to be generated to get finality um

40. 16:45

    or at least a confirmation rule that that is acceptable for you um and so I think there will be a lot of competing standards um just like there are a lot of competing Stacks that you're seeing like in the rollup space whether it's like the op stack or Sovereign SDK or something like this um but ultimately I think that the standards that are the most interoperable and compatible with

41. 17:08

    others are going to be the ones that probably wind up having some staying power over time it's very hard to say right now what that's going to look like I think things like uh ERC 7638 if if I'm quoting it right uh is very interesting and I think it definitely has some support within the community uh so i' be looking forward to seeing folks actually like adopting that and um uh coordinating around that

42. 17:34

    standard um yeah I mean I've spent some time thinking about like what makes a standard adoptable in this space um and it's difficult because like you know like unfortun the unfortunate reality of the space is that like people's intentions have sort of ended up becoming offis skated because of bags and like even if even if your intentions are actually good like like there's always the risk of bag bias and so there's like people just it's assume

43. 17:59

    that there are potentially malicious intentions with with things like standards um and so I feel like something that I've I've sort of started to see is that the things that seem to be getting adoption are when you can find a standard that is like effectively helping to uh like where you can take like a a problem space like chain abstraction right and you can identify what are the different problems associated with that

44. 18:21

    problem space so for example um you know unifying balances and like you know the permissioning aspect of chain abstraction is is one part of the problem then there's like solving and like who is actually running those solvers how are they operating things like that is another part of the problem how do you select solvers is another part of the problem right um and so I think I think what I've I've started to

45. 18:38

    see happening and and I I really like this is like differentiating problem spaces and like differentiating as as an organization towards like what specific problem you're solving um and having people self-identify into like a layers of a stack um like the chain abstraction stack for instance um seems to work really well with standards if you can make the standards just the interfaces between the layers of the stack right because those interfaces everybody's aligned on on adopting because like if

46. 19:03

    you are building a solver Network right you you don't want to have to go and do 15 Integrations into a bunch of like orderflow sources because they all have different interfaces um if you are a you know underlying transport protocol you don't want to have to go and like you don't have to go and work with independent solvers just like one by one solvers don't want to have to go and build in-house aggregators to work with every single like like bridge and mechanism under the hood um and so I

47. 19:27

    think that's that's maybe like one of the promis areas that I hope more standards can get built in um and yeah as you guys mentioned like 7 6 83 or whatever that one is uh is is falls into that category so I'm pretty pretty optimistic about it so you heard it here there's a free good idea in case anyone wants to go build a standard um just we'll we'll let you guys have that one for free um so it seems like you know implementing intent based systems will ultimately do a lot to help facilitate a

48. 19:57

    lot of this like interoperability between different applications and chains and everything what are some of the technical challenges that are preventing people now or teams now projects who have applications who aren't currently leveraging any sort of intent-based system who are saying like yes I do believe this is the direction things are going how do they start to Pivot that way and start thinking about that for cross chain specifically there's one very interesting Paradigm

49. 20:23

    that has emerged over the course of the last few years which is like the absolutely idiotic strategy of like copy pying your application to every chain and then like bootstrapping tvl into every single one and then you end up with like 11 instances of a with like you know whatever 10 billion spread across them on like 11 different chains and like look like it makes sense that we ended up there right like there was no nobody there was no other solution

50. 20:47

    you needed to be able to expand to these different ecosystems but at the same time this is a strictly inferior outcome right you are fragmenting your liquidity you like if you're if you're a you're just not able to have as much size because you've moved you can't possibly like organize liquidity on all these different chains um I think this is changing like the the app chain narrative is pretty strong and like you know everybody wants to be a chain and it's like this is one of the weird cases

51. 21:09

    where like I'm kind of like okay like I get the like you know pumpamentals and like VC you know whatever of everybody wants to launch a chain but I think in this case it actually makes sense it's like move all of your infrastructure move all of your liquidity to its own isolated environment keep it all Consolidated and just make it easy to access from anywhere um I think that's one of the big challenges today is like applications just are not operating that

52. 21:33

    way they're not thinking about onboarding users that way they're starting to realize like hey onboarding from every ecosystem is actually a better strategy which in retrospect is very obvious but um but the but the the material aspect of how they will do that where they will onboard to especially if they're already fragmented is is an interesting and open question rip the 82 die I have stuck in av1 that cost $350 to

53. 21:55

    withdraw case in point yeah any other comments on challenges that you know teams are going to face if they're trying to move an into an intent based system um either from like a business perspective or from just a technology perspective just want to pick it back right off you know like a for example right they do have to deploy on 16 different chains so butra that's not even like the issue the issue is that your wallet doesn't know which chain you want to interact with that's it's like

54. 22:25

    it's a very basic problem so like trying to talk to a like hey do you want to deploy another chain say yes and also for them it's their Core Business is e 94% of their revenue is from E and the other 5% is from stable coins so like that's 99% of their revenue for them to move out of that ecosystem you need to do a lot more and I don't know like chain abstraction I think is better for building more or bow applications like truly cross-chain

55. 23:02

    applications like you can have I later as a security layer and then you could you could do like data availabil on Celestia and then you want your applications to work on a or something I don't know some weird stuff where you have like multiple systems working together and then intent for me is abstracting that way for developers and the users because if you don't have the tool sets for building multi chain which we don't have today think about it like what kind of SDK can you use to build

56. 23:33

    like a modular chain it's it's nothing right so you have to solve for that and you have to solve for the user experience we just don't have wallets that just like knows how to figure out what chain you want to connect to which is ridiculous it yeah it's I don't know I don't know it's crazy we live in a crazy user experience space yeah I don't know if anyone here remembers like what

57. 23:56

    using the internet was like and like the early '90s um but basically like before really the worldwide web became a thing the internet was a bit of a show like you had to manually tell net to other networks which means you have to manually basically figure out where your network is in the internet your like home network and you have to connect to someone else's home network to like route to their Network manually and then you have to go and get files from their Network and then bring them back and like the web was just an abstraction

58. 24:21

    over all this right HTTP was an abstraction over all of this where it was like okay let's have a version of the internet where you don't don't have to go and manually connect to other people's networks where all of that is is just under behind the scenes it's all off fiscated behind a domain and what you do is you go to this website and like it just serves you content it just gives you what you want um and that was just revolutionary right

59. 24:45

    like the internet went from being like like 50 to 100 people running hobbyist stuff to being suddenly like thousands and thousands and thousands of sites pretty much overnight sounds awesome you don't you don't you don't think IBC could be that though IBC I think is lower level in the stock so like yeah I think this is part of the challenge that we've run into historically is that we've we've kind of like uh We've like when we've talked

60. 25:11

    about cross chain we've been treating it as like One Singular problem um but it's it's not really right it's like there's there's separate problems here there's there's problems of how do you get data from one chain to another which is what IBC solves how do you then there's how do you secure that data which is actually different depending on what chain you're on um and that a lot of different projects or a lot of different ecosystems are trying to solve and then there's the third question of like well

61. 25:33

    how do you make it cheap and fast and easy to use and because when you do make it secure and this is not as much of a problem for the IBC ecosystem just because like you know there's not as much like you're not paying as much yet for security there's not as much demand for Block space yet there but like when you're talking about like a ZK like client for instance right that is going to cost money like ZK light clients are going to cost like millions and millions of dollars to run um users are not going

62. 25:56

    to pay like $15 to $20 a transaction like that and so the the abstraction here is is similar to http like you have users interact with intents and then the intent solvers go and do One Singular batch transaction through the light client because they can you know like Encompass thousands tens of thousands of transactions in that that that they're earning fees on awesome all right we got one minute left so I'm going to give each panelist

63. 26:20

    uh one second or just a couple seconds to um give us an idea of like what's something that you're excited about in the intent space either um an opportunity that you think is reachable via intense Space Systems or something that you know your project is working on that you're coming up that you've got really excited about yes I'll take a moment to shamelessly shill um we have some uh

64. 26:45

    novel intent based applications that um things like intent Cent Kickstarter things like uh promise graph which allows you to make service commitments uh internally in organizations but also between organizations um things like multi-chat which is an encrypted slack like Network that runs with a Noma on the back end um and also things like uh time Banks or Mutual Credit networks enabled by this primitive that we call Kudos or scale-free money so I

65. 27:16

    think there's a lot of applications that you can build with intent and if you have a network that can uh take these intents uh and do stuff with them uh you can rebuild a lot of the applications that we have in web to today actually in web 3 and make them decentralized and verifiable I'm definitely excited about those applications um I think one other thing I'm excited about is like there's a lot of questions that still need to be

66. 27:41

    answered about chain abstraction broadly um things that are beyond the technical but more about the scope of what do users need out of chain abstraction what kinds of things do they still need control over or at least some ability to control and I think those are some thought spaces that have a lot of development to do um and I'm excited to see what comes out of those yeah I mean from a philosophical standpoint I'm really excited about like new ways of cordinating that open up as

67. 28:07

    a result of intense so like mutualism like mutualist types of economic systems um collectivist econom economic systems um these are economic ideas that are like not part of the current metag game of you know highly inflationary economies basically you won't find anything about this after like World War II but if you look back into like you know how uh people have coordinated before that like there's very very awesome sustainable economic paradigms

68. 28:32

    that have existed for like thousands of years um that we just don't that we can like we can now like build at a very low cost and scale um on a more practical down Earth perspective I think like for I mean I I started working in the space in like 2016 and like at that time we were talking about like okay how are we going to make it so users don't have to pay gas so that they don't have to think about accounts so that they don't have to like deal with you know like onboarding and you know many chains and

69. 28:57

    things like that and it's like it's kind of it's crazy that it's taken us this long but we are actually now very close to solving this problem and I think that that is to me extremely exciting because it means like we at least I get to finally stop doing this stuff and go build really cool stuff on top of blockchains like which we all really want to do yeah so for me I I I like we our

70. 29:23

    thesis is that there's going to be blockchains for everything and hopefully IBC will be the glue that glue together and evm is the interface so for us is we see everything could be rebuild on blockchain systems and again making these systems work together and making it abstracting away from user that's the biggest thing I want to like I want Uber to run on a blockchain I want event bright I want everything all the internet systems that are now controlled by you know a few engineers

71. 29:56

    in silan Valley like and few vs no like it's that's not what we're here for we're here to rebuild these systems to be decentralized and powered back to the people but we're a long way from that because of the fragmentation issues and technology and liquidity and intents could be the solution to that awesome well thank you all so much for this super deep de Deep dive into intense one

72. 30:21

    of the coolest new mechanisms coming up in the crypto space um definitely keep your ears out for everything coming out from these projects there's a lot of neat exciting work coming up in the intense space and thank you jacqu for hopping up on stage literally as you walked in and uh and joining the conversation this is great you did fantastic and yeah I'm going to hand it over to my better looking counterpart David who will bring us on to our next part of our programming thank you all for coming