{"id":"UBvgETfz3VY","title":"Anoma: From VM to IM - Adrian Brink","description":"Join NEAR's community:\nWebsite: https://near.org/\nReddit:   / nearprotocol  \nDiscord: https://near.chat/\nMedium:   / nearprotocol  \nBlog: https://near.org/blog/\nTwitter:   / nearprotocol  \nGitHub: https://github.com/near  https://github.com/nearprotocol\nDev Docs: https://docs.near.org/\nCreate a wallet: https://wallet.near.org/create\nApps on NEAR: https://awesomenear.com/\nLearn to Build On NEAR: https://www.near.university/\nGrants & Funding: https://near.org/grants/\n#Blockchain #FutureisNEAR #NEAR #nearprotocol\n#nft #dao","channel_name":"NEAR Protocol","published_at":"2024-04-03T00:00:00.000Z","thumbnail_url":"https://img.youtube.com/vi/UBvgETfz3VY/hqdefault.jpg","transcript_text":"all right welcome everyone thank you for making it uh I hope you're not not everyone is as exhausted as I am um with massive jet lag so I hope you can still follow along um so I'm Adrian I'm one of the co-founders of anoma and today I'm going to very briefly um talk about sort of anoma from moving from virtual machines to intent machines and sort of what the next step in the EV evolution of blockchains is and what this world will look like in the future um yes great this works uh cool so let's start with a brief historic recap of um how we got to where we are today um and really here we started sort of I would call this generation one Bitcoin Bitcoin like architectures um that had that were very transaction Centric and had Li mited scripting capabilities right so as you heard before this is starting to improve and so on but really fundamentally at the base layer Bitcoin has limited programmability has only scripting capabilities and so the kind of applications you can build on top of are limited um from there we move to this VM world where all of a sudden everyone and their grandmother started having a VM right so this was started with ethereum um and all existing architectures are pretty much actually they are all VM Centric architecture and this doesn't actually matter whether you're on L2 or you're on salana or nia or ethereum or optimism or base um these are all fundamentally transaction Centric virtual machines where end users are supposed to have fully fledged State transitions that can be immediately settled on chain um and so of today I'm going to focus on talking about intents and um so of the what I foresee to be the next generation of architectures which are going to be intent Centric architectures where users don't in fact direct interface directly with the virtual machines but rather with the intent machines as abstractions built on top of many different po potential virtual machines um and this sort of to give you very brief um conception has humongous benefits from a user perspective because instead of saying uh this is how I want to execute my compute on some virtual machine uh a user can rather say I'm current currently at State a and I'd like to be at State B someone else please figure out how to get me to State B where State B could be on a different security zone or a different blockchain altogether the user can just programmatically specify where they are and where they want and where they want to get to rather than specifying how to do this execution um if you're into zero knowledge proofs this will sound very familiar um this is one of the very fundamental reasons why you need intent Centric systems in order to actually support programmable privacy guarantees uh because without this you sort of can't really represent that I have local state and I'd like to transition this local state um to the next state um yes so lots of people talk about intents all the time um and so if you're following the modern discour you may think intents are just limit orders and in a very limited sense this is true uh limit orders are very specific kinds of intents um so like when you think of something like a cross protocol unop X and so on uh these really limit orders um what anoma is and what we've been working on for a while now is this a is a design for generalized intents where intents can be arbitrary things um they can be literally arbitrary programs um that control some amount of State on chain oh I should actually follow along here yes and so with this Anor really is a new paradigm for infrastructure um we're evolving infrastructure from being blockchain and VM Centric to being intent Centric um and really the interesting thing is many nodes may want to offer different kind of services to different users and this is way easier so like in the current model this an a node has or an infrastructure provider has to make a fundamental choice on like I'm either running this VM or that VM in an intent Centric world you can like you can be a solver you can be a PTP gossip note you can be um someone that just allows users to access the chain and another really interesting Fallout from this is that really um users are also blockchains um and this is sort of I'll go more into depth here later but currently we have this very strong abstraction between here is a note and a client right like a client interacts with a note via some amount of RPC calls um in an INT Centric world with the noral this isn't the case it just a user decides what kind of subset of node functionality they'd like to run the local device and they just another node on the gossip layer um you do not have to have the problem of like oh I have to pick my specific RPC provider to send my transaction to you rather say well I'm directly connecting as my phone to the intent gossip Network and I'm gossiping to I don't know 50 different peers that I have received via some sort of trust metric that I care about and I send them my intents uh this is also how you get information flow control going sort of at the network level where you have some metrics where you can decide I'm willing to gossip to like these five people my intents because intents will always reveal some amount of information uh but not these other five people right um the other thing is Builders um once you start thinking about these systems from an intent Centric perspective uh you can think of decentralized counter party discovering distributed solving um so one of the very very difficult problems actually is how do you do decentralized counterpart Discovery and sort of like our best answer to answer to this problem is either well we'll all put it on a single server and we all meet at that single server this is what every rollup currently doeses essentially um this is what most limit order protocols do as well um or you put it on chain this is what unop fundamentally does is what amm do right like think of amm not as liquidity things but as the place where multiple orders meet um so with but once you have intend at the Bas layer doing decentralized cter Discovery actually becomes way easier because I can decide who I want to send my intents to my intents are fully signed um if that two overlapping intents anyone in the network can take them combine them settle them somewhere um yes and also I was saying sort of on novel properties with regard to information flow control which is really uh think of this as how does how do you not disclose your entire transaction graph to the rest of the world this is what information flow control is you being in control what kind of data you want to reveal to the outside world this sort of results in one large challenge which is currently All State lives on a global mutable Ledger right this is why it's readable by everyone if you want to not do this we must move stage to Edge devices uh this is sort of the entire idea with regards to Z knowledge proof that you generate proofs over local data that is not shared with the rest of the network network however in a transaction Centric world this is very hard to implement because a transaction must be fully balanced but I may not have the state I only have the state to trade some e for near not the inverse pair right so I need to be able to um author my intent um and attach a proof over this state transition not the fully balanced one but rather my part of the intent or my part of the state transition and then someone else can sort of take the inverse or takes some combination um and then as any solver can pick this up and go ah I have one zero knowledge proof here that authorizes e to near this price point I have some other intent here from uh near to e um let me combine these two things and then settle them actually on chain um yeah and then really users uh I was going on a little bit about this earlier one of the big problems is that as a user right now now you think of execution traces really this is what you do when you sign a transaction as a user you have to think of an execution Trace you have to figure out the optimal path that you want to take through the network or you have to Outsource this trust somewhere else but this generally means you Outsource also uh control over your assets to some other party um and with intense you get to this very nice declarative Paradigm around instead of the specifying computation you specify State updates um and again you don't care this is also way easier from a or safer from a user perspective because my parents will never be able to audit evm bite code this is idiotic to assume that this will ever be the case right but in the sort of like in intent Centric model this declarative model uh uis can just L just show these are the keys that you're signing over to be updated you don't have to understand how the evm works you don't actually this may not NE actually be executed on the evm because we actually split here intent split State and execution so that the intent machine only defines a state layout effectively and many different intents may have different execute like VMS under the hood so but as a user you don't have to care as long as you have something that's State compatible yes so anoma is an intent machine this is really if someone asks you what anoma is say this anoma is an intent machine um and in all lik we will standardize at the intent layer not the chain layer or the VM layer uh because it's sort of along with a chain abstraction it is really this is fundamentally chain abstraction this is how you abstract different chains away from the user this is intents or the intent machine a know as a universal intent machine gives users a single programmable API or developers of single API to build applications against that are independent of specific VM specific execution environment ments so um I'm going to talk a little bit about multi-chain Atomic settlement um so one of the other very nice properties is that with the intent machine you can get to a homogeneous architecture in a heterogeneous security model so think of this security model as Nia and ethereum have two different security models right um a proof of authority chain run by the South Korean government will have a different security model than ethereum um users may want like essentially run things like again the example of my parents my parents will want local sovereignity run by their municipality effectively um and with the intent machine abstraction we can get to the point where our applications can roam very freely between these different security models because I am not bound to a specific CPU architecture to a specific execution envirment rather I'm writing an application and everywhere the intent machine is available on many different execution Ms I now can run the same application and I can do uh composable things across many different execution environments um this also comes with ic2 I will go a little bit more into the technical detail um think of ic2 as uh synchronous IBC across heterogeneous trust domains um and sort of when you come from sort of the old research way uh you can also think of this as to some extent as heterogeneous charting the ability to have uh many different different security models opt into a um into shared blocks effectively so um how many oh I have three minutes damn um all right some components here uh the P2P layer this is sort of uh the first component in a normal which is a thing that no one in the space spends enough time on which is networking research it's like not the sexiest thing uh everyone wants to work on fancy cryptography but this is sort of what fundamentally enables a lot of the distributed systems properties um and so in anoma we have a globally uh Spar network of many different nodes these nodes can live in different trust domains these nodes can run and offer different functionality so for example 35 down here uh may just be a client actually they may just be a phone that's connected via the P2P stack to the rest of network 38 here may actually be a very beefy validat note that is validating three different trust domains at the same time think of these trust domains as valida domain two is near domain three is etherum and domain one is Sana think of it this way um many different notes May opt into being connected to many different trust domains um yes and this is sort of how do you get this heterogenity or The Sovereign uh security domain setup where you may also have a domain here that's like the New York Stock Exchange run as a pro of authority chain uh by whatever Financial regulator exists in New York I have no idea um Plus London Stock Exchange so have similar example right um yes how does if you have this kind of intent gossip layer how do you actually uh achieve consensus over this and this is tyen and uh you may have guessed I am not a native English speaker and as a result we never realized that this is not an English word um we thought I thought typhoon was spelled this way uh so this is Typhon um which is effectively um if you are aware of tenm consensus for example this is Typhon is uh pbft um and it comes with a uh system called Comm chains which allows you to have Atomic cross chain communication and I won't go into the entire thing in detail here but essentially think of you have uh chain a in blue or chain a in blue chain B in green and both of these chains can opt into a shared synchronous block between them um for this Typhon partitions the state on chain B into s of state that's um composable with uh chain B um and these blocks are just normal blocks they are you get full synchronous calling composability for example so like I can have a contract from Blue call into green and get a synchronous call back within the same execution or within the same block it's not an asynchronous message passing style system if you don't want it um and then probably the heart of AA is the a normal resource machine or arm um and everyone always asks you is it an account model or is it utxo model it's neither um it's a resource model this is very important to understand um and resource models or the resource model is what allows us to have generalized intents and really at a high level you can think of a resource model as you have some persistent State and a arbitrary uh programmable resource logic attached to how you can update and modify that piece of State um and in that sense actually similar to utxo model well in that sense it's similar to an account model um but in the sense it's also similar to a uh utxo model because resources are always um spent so they never you never modify them you just go I have the resource that has some smart contract logic and I modify this to I get a new pointer to new resource though um yes and with this really we should be thinking more about chain abstraction most users will not want to be on any of the systems that are currently being built uh because most users in most places one like local sovereignity they don't we've been living in this world world for the last 20 years where we' outsourced everything to like three AWS servers run by three large US companies right like most users at some point will want local sovereignity over their data again and this is really what chain abstraction allows without breaking composability across the entire stack um the last thing I would like to uh sort of highlight is if you follow ether if you don't follow ether research you will have not noticed this but if you do and then if you don't you should uh Chris my co-founder just published a uh new post on eth Research earlier today if this didn't make sense there is a very succinct explanation of all parts of anoma with links to more underlying research and anoma research topics uh you can also find the Research Forum there or the aot uh topics there um yeah and with that thank you very much","transcript_synced_at":"2026-11-03T06:05:24.164Z","segments":[{"start_seconds":6,"text":"all right welcome everyone thank you for making it uh I hope you're not not everyone is as exhausted as I am um with","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=6s"},{"start_seconds":13,"text":"massive jet lag so I hope you can still follow along um so I'm Adrian I'm one of the co-founders of anoma and today I'm","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=13s"},{"start_seconds":22,"text":"going to very briefly um talk about sort of anoma from moving from virtual machines to intent machines and sort of","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=22s"},{"start_seconds":30,"text":"what the next step in the EV evolution of blockchains is and what this world will look like in the future","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=30s"},{"start_seconds":39,"text":"um yes great this works uh cool so let's start with a brief historic recap of um","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=39s"},{"start_seconds":47,"text":"how we got to where we are today um and really here we started sort of I would call this generation one Bitcoin Bitcoin","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=47s"},{"start_seconds":55,"text":"like architectures um that had that were very transaction Centric and had Li mited scripting capabilities right so as you heard before this is starting to","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=55s"},{"start_seconds":63,"text":"improve and so on but really fundamentally at the base layer Bitcoin has limited programmability has only scripting capabilities and so the kind","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=63s"},{"start_seconds":71,"text":"of applications you can build on top of are limited um from there we move to this VM world where all of a sudden everyone and","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=71s"},{"start_seconds":79,"text":"their grandmother started having a VM right so this was started with ethereum um and all existing architectures are","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=79s"},{"start_seconds":86,"text":"pretty much actually they are all VM Centric architecture and this doesn't actually matter whether you're on L2 or you're on salana or nia","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=86s"},{"start_seconds":94,"text":"or ethereum or optimism or base um these are all fundamentally transaction Centric virtual machines where end users","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=94s"},{"start_seconds":102,"text":"are supposed to have fully fledged State transitions that can be immediately settled on chain","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=102s"},{"start_seconds":110,"text":"um and so of today I'm going to focus on talking about intents and um so of the what I foresee to be the next generation","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=110s"},{"start_seconds":118,"text":"of architectures which are going to be intent Centric architectures where users don't in fact direct interface directly with the virtual machines but rather","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=118s"},{"start_seconds":126,"text":"with the intent machines as abstractions built on top of many different po potential virtual machines um and this","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=126s"},{"start_seconds":134,"text":"sort of to give you very brief um conception has humongous benefits from a user perspective because instead of","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=134s"},{"start_seconds":142,"text":"saying uh this is how I want to execute my compute on some virtual machine uh a user can rather say I'm current","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=142s"},{"start_seconds":149,"text":"currently at State a and I'd like to be at State B someone else please figure out how to get me to State B where State B could be on a different security zone","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=149s"},{"start_seconds":157,"text":"or a different blockchain altogether the user can just programmatically specify where they are and where they want and where they want to get to rather than","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=157s"},{"start_seconds":165,"text":"specifying how to do this execution um if you're into zero knowledge proofs this will sound very familiar um this is","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=165s"},{"start_seconds":173,"text":"one of the very fundamental reasons why you need intent Centric systems in order to actually support programmable privacy guarantees uh because without this you","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=173s"},{"start_seconds":181,"text":"sort of can't really represent that I have local state and I'd like to transition this local state um to the next","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=181s"},{"start_seconds":188,"text":"state um yes so lots of people talk about intents all the time um and so if you're following the modern discour you","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=188s"},{"start_seconds":196,"text":"may think intents are just limit orders and in a very limited sense this is true uh limit orders are very specific kinds","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=196s"},{"start_seconds":203,"text":"of intents um so like when you think of something like a cross protocol unop X and so on uh these really limit orders","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=203s"},{"start_seconds":211,"text":"um what anoma is and what we've been working on for a while now is this a is a design for generalized intents where","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=211s"},{"start_seconds":220,"text":"intents can be arbitrary things um they can be literally arbitrary programs um that control some amount of State on","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=220s"},{"start_seconds":227,"text":"chain oh I should actually follow along here yes and so with this Anor really is","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=227s"},{"start_seconds":237,"text":"a new paradigm for infrastructure um we're evolving infrastructure from being blockchain and VM Centric to being","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=237s"},{"start_seconds":244,"text":"intent Centric um and really the interesting thing is many nodes may want to offer","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=244s"},{"start_seconds":251,"text":"different kind of services to different users and this is way easier so like in the current model this an a node has or","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=251s"},{"start_seconds":258,"text":"an infrastructure provider has to make a fundamental choice on like I'm either running this VM or that VM in an intent Centric world you can like you can be a","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=258s"},{"start_seconds":266,"text":"solver you can be a PTP gossip note you can be um someone that just allows users to access the chain and another really","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=266s"},{"start_seconds":275,"text":"interesting Fallout from this is that really um users are also blockchains um","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=275s"},{"start_seconds":283,"text":"and this is sort of I'll go more into depth here later but currently we have this very strong abstraction between here is a note and a client right like a","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=283s"},{"start_seconds":292,"text":"client interacts with a note via some amount of RPC calls um in an INT Centric world with the noral this isn't the case","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=292s"},{"start_seconds":299,"text":"it just a user decides what kind of subset of node functionality they'd like to run the local device and they just","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=299s"},{"start_seconds":306,"text":"another node on the gossip layer um you do not have to have the problem of like oh I have to pick my specific RPC","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=306s"},{"start_seconds":313,"text":"provider to send my transaction to you rather say well I'm directly connecting as my phone to the intent gossip Network","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=313s"},{"start_seconds":319,"text":"and I'm gossiping to I don't know 50 different peers that I have received via some sort of trust metric that I care about and I send them my intents uh this","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=319s"},{"start_seconds":328,"text":"is also how you get information flow control going sort of at the network level where you have some metrics where you can decide I'm willing to gossip to","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=328s"},{"start_seconds":337,"text":"like these five people my intents because intents will always reveal some amount of information uh but not these other five people right","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=337s"},{"start_seconds":346,"text":"um the other thing is Builders um once you start thinking about these systems from an intent Centric","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=346s"},{"start_seconds":354,"text":"perspective uh you can think of decentralized counter party discovering distributed solving um so one of the","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=354s"},{"start_seconds":361,"text":"very very difficult problems actually is how do you do decentralized counterpart Discovery and sort of like our best answer to answer to this problem is","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=361s"},{"start_seconds":369,"text":"either well we'll all put it on a single server and we all meet at that single server this is what every rollup currently doeses essentially um this is","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=369s"},{"start_seconds":377,"text":"what most limit order protocols do as well um or you put it on chain this is what unop fundamentally does is what amm","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=377s"},{"start_seconds":385,"text":"do right like think of amm not as liquidity things but as the place where multiple orders meet um","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=385s"},{"start_seconds":394,"text":"so with but once you have intend at the Bas layer doing decentralized cter Discovery actually becomes way easier because I can decide who I want to send","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=394s"},{"start_seconds":402,"text":"my intents to my intents are fully signed um if that two overlapping intents anyone in the network can take them combine them settle them somewhere","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=402s"},{"start_seconds":410,"text":"um yes and also I was saying sort of on novel properties with regard to information flow control which is really uh think of this as how does how do you","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=410s"},{"start_seconds":419,"text":"not disclose your entire transaction graph to the rest of the world this is what information flow control is you being in control what kind of data you","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=419s"},{"start_seconds":426,"text":"want to reveal to the outside world this sort of results in one large challenge which is currently All State","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=426s"},{"start_seconds":435,"text":"lives on a global mutable Ledger right this is why it's readable by everyone if you want to not do this we must move","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=435s"},{"start_seconds":442,"text":"stage to Edge devices uh this is sort of the entire idea with regards to Z knowledge proof that you generate proofs over local data that is not shared with","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=442s"},{"start_seconds":451,"text":"the rest of the network network however in a transaction Centric world this is very hard to implement because a transaction must be fully balanced but I","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=451s"},{"start_seconds":459,"text":"may not have the state I only have the state to trade some e for near not the inverse pair right so I need to be able","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=459s"},{"start_seconds":466,"text":"to um author my intent um and attach a proof over this state transition not the fully balanced one but rather my part of","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=466s"},{"start_seconds":474,"text":"the intent or my part of the state transition and then someone else can sort of take the inverse or takes some combination um and then as any solver","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=474s"},{"start_seconds":482,"text":"can pick this up and go ah I have one zero knowledge proof here that authorizes e to near this price point I have some other intent here from uh near","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=482s"},{"start_seconds":491,"text":"to e um let me combine these two things and then settle them actually on","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=491s"},{"start_seconds":500,"text":"chain um yeah and then really users uh I was going on a little bit about this earlier one of the big problems is that","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=500s"},{"start_seconds":508,"text":"as a user right now now you think of execution traces really this is what you do when you sign a transaction as a user","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=508s"},{"start_seconds":515,"text":"you have to think of an execution Trace you have to figure out the optimal path that you want to take through the network or you have to Outsource this trust somewhere else but this generally","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=515s"},{"start_seconds":523,"text":"means you Outsource also uh control over your assets to some other party um and with intense you get to this very nice","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=523s"},{"start_seconds":531,"text":"declarative Paradigm around instead of the specifying computation you specify State updates um and again you don't","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=531s"},{"start_seconds":540,"text":"care this is also way easier from a or safer from a user perspective because my parents will never be able to","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=540s"},{"start_seconds":547,"text":"audit evm bite code this is idiotic to assume that this will ever be the case right but in the sort of like in intent","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=547s"},{"start_seconds":554,"text":"Centric model this declarative model uh uis can just L just show these are the keys that you're signing over to be","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=554s"},{"start_seconds":562,"text":"updated you don't have to understand how the evm works you don't actually this may not NE actually be executed on the evm because we actually split here","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=562s"},{"start_seconds":569,"text":"intent split State and execution so that the intent machine only defines a state layout effectively and many different","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=569s"},{"start_seconds":579,"text":"intents may have different execute like VMS under the hood so but as a user you don't have to care as long as you have something that's State","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=579s"},{"start_seconds":588,"text":"compatible yes so anoma is an intent machine this is really if someone asks you what anoma is say this anoma is an","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=588s"},{"start_seconds":595,"text":"intent machine um and in all lik we will standardize at the intent layer not the chain layer or the VM layer uh","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=595s"},{"start_seconds":604,"text":"because it's sort of along with a chain abstraction it is really this is fundamentally chain abstraction this is","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=604s"},{"start_seconds":611,"text":"how you abstract different chains away from the user this is intents or the intent machine a know as a universal","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=611s"},{"start_seconds":619,"text":"intent machine gives users a single programmable API or developers of single API to build applications against that","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=619s"},{"start_seconds":626,"text":"are independent of specific VM specific execution environment ments so um I'm going to talk a little","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=626s"},{"start_seconds":633,"text":"bit about multi-chain Atomic settlement um so one of the other very nice properties is that with the intent","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=633s"},{"start_seconds":640,"text":"machine you can get to a homogeneous architecture in a heterogeneous security model so think of this security model as Nia and ethereum have two different","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=640s"},{"start_seconds":648,"text":"security models right um a proof of authority chain run by the South Korean government will have a different","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=648s"},{"start_seconds":655,"text":"security model than ethereum um users may want like essentially run things like again the example of my","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=655s"},{"start_seconds":663,"text":"parents my parents will want local sovereignity run by their municipality effectively um and with the intent","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=663s"},{"start_seconds":671,"text":"machine abstraction we can get to the point where our applications can roam very freely between these different security models because I am not bound","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=671s"},{"start_seconds":679,"text":"to a specific CPU architecture to a specific execution envirment rather I'm writing an application and everywhere the intent machine is available on many","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=679s"},{"start_seconds":688,"text":"different execution Ms I now can run the same application and I can do uh composable things across many different","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=688s"},{"start_seconds":695,"text":"execution environments um this also comes with ic2 I will go a little bit more into the technical detail um think","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=695s"},{"start_seconds":702,"text":"of ic2 as uh synchronous IBC across heterogeneous trust domains um and sort of when you come","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=702s"},{"start_seconds":711,"text":"from sort of the old research way uh you can also think of this as to some extent as heterogeneous charting the ability to have uh many different different","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=711s"},{"start_seconds":719,"text":"security models opt into a um into shared blocks effectively so um how many oh I have","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=719s"},{"start_seconds":726,"text":"three minutes damn um all right some components here uh the P2P layer this is sort of uh the first component in a","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=726s"},{"start_seconds":734,"text":"normal which is a thing that no one in the space spends enough time on which is networking research it's like not the","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=734s"},{"start_seconds":741,"text":"sexiest thing uh everyone wants to work on fancy cryptography but this is sort of what fundamentally enables a lot of the distributed systems properties um","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=741s"},{"start_seconds":749,"text":"and so in anoma we have a globally uh Spar network of many different nodes these nodes can live in different trust domains these nodes can run and offer","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=749s"},{"start_seconds":758,"text":"different functionality so for example 35 down here uh may just be a client actually they may just be a phone that's","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=758s"},{"start_seconds":766,"text":"connected via the P2P stack to the rest of network 38 here may actually be a very beefy validat note that is validating three different trust domains","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=766s"},{"start_seconds":775,"text":"at the same time think of these trust domains as valida domain two is near domain three is etherum and domain one is Sana think of","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=775s"},{"start_seconds":784,"text":"it this way um many different notes May opt into being connected to many different trust domains um yes and this is sort of how","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=784s"},{"start_seconds":792,"text":"do you get this heterogenity or The Sovereign uh security domain setup where you may also have a domain here that's like the New York Stock Exchange run as","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=792s"},{"start_seconds":800,"text":"a pro of authority chain uh","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=800s"},{"start_seconds":804,"text":"by whatever Financial regulator exists in New York I have no idea um Plus London Stock Exchange so have similar example right","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=804s"},{"start_seconds":813,"text":"um yes how does if you have this kind of intent gossip layer how do you actually uh","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=813s"},{"start_seconds":822,"text":"achieve consensus over this and this is tyen and uh you may have guessed I am not a native English speaker and as a","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=822s"},{"start_seconds":830,"text":"result we never realized that this is not an English word um we thought I thought typhoon was spelled this way uh","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=830s"},{"start_seconds":837,"text":"so this is Typhon um which is effectively um if you are aware of tenm","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=837s"},{"start_seconds":844,"text":"consensus for example this is Typhon is uh pbft um and it comes with a uh system","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=844s"},{"start_seconds":851,"text":"called Comm chains which allows you to have Atomic cross chain communication and I won't go into the entire thing in detail here but essentially think of you","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=851s"},{"start_seconds":861,"text":"have uh chain a in blue or chain a in blue chain B in green and both of these","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=861s"},{"start_seconds":868,"text":"chains can opt into a shared synchronous block between them um for this Typhon partitions the state on chain B into s","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=868s"},{"start_seconds":877,"text":"of state that's um composable with uh chain B um and these blocks are just","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=877s"},{"start_seconds":884,"text":"normal blocks they are you get full synchronous calling composability for example so like I can have a contract","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=884s"},{"start_seconds":892,"text":"from Blue call into green and get a synchronous call back within the same execution or within the same block it's","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=892s"},{"start_seconds":899,"text":"not an asynchronous message passing style system if you don't want it um and then probably the heart of AA","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=899s"},{"start_seconds":906,"text":"is the a normal resource machine or arm um and everyone always asks you is it an account model or is it utxo model it's","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=906s"},{"start_seconds":915,"text":"neither um it's a resource model this is very important to understand um and resource models or the resource model is","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=915s"},{"start_seconds":923,"text":"what allows us to have generalized intents and really at a high level you can think of a resource model as you have some persistent State and a","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=923s"},{"start_seconds":932,"text":"arbitrary uh programmable resource logic attached to how you can update and modify that piece of State um and in","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=932s"},{"start_seconds":941,"text":"that sense actually similar to utxo model well in that sense it's similar to an account model um but in the sense it's also similar to a uh utxo model","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=941s"},{"start_seconds":949,"text":"because resources are always um spent so they never you never modify them you just go I have the resource that has","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=949s"},{"start_seconds":959,"text":"some smart contract logic and I modify this to I get a new pointer to new resource though","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=959s"},{"start_seconds":966,"text":"um yes and with this really we should be thinking more","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=966s"},{"start_seconds":974,"text":"about chain abstraction most users will not want to be on any of the systems that are currently being built uh because most","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=974s"},{"start_seconds":983,"text":"users in most places one like local sovereignity they don't we've been living in this world world for the last 20 years where we' outsourced everything","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=983s"},{"start_seconds":991,"text":"to like three AWS servers run by three large US companies right like most users at some point will want local sovereignity over their data again and","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=991s"},{"start_seconds":1000,"text":"this is really what chain abstraction allows without breaking composability across the entire stack um the last thing I would like to","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=1000s"},{"start_seconds":1008,"text":"uh sort of highlight is if you follow ether if you don't follow ether research you will have not noticed this but if you do and then if you don't you should","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=1008s"},{"start_seconds":1017,"text":"uh Chris my co-founder just published a uh new post on eth Research earlier today if this didn't make sense there is","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=1017s"},{"start_seconds":1024,"text":"a very succinct explanation of all parts of anoma with links to more underlying research and anoma research topics uh","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=1024s"},{"start_seconds":1033,"text":"you can also find the Research Forum there or the aot uh topics there um yeah and with that thank you very","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=1033s"},{"start_seconds":1046,"text":"much","url":"https://www.youtube.com/watch?v=UBvgETfz3VY&t=1046s"}],"youtube_url":"https://youtube.com/watch?v=UBvgETfz3VY","url":"https://near-pedia.xyz/videos/UBvgETfz3VY"}