Modular March EP 3: The Next Modular DA Launch & Avail Trinity

Transcript

welcome to modular March I'm Andy this is Robbie and we are the rollup modular March is an educational experience about all things Moder [Music] enjoy [Music] welcome back to modular march today we

have honor Arjun presenting ail the accelerating the unification of web 3 welcome thanks thanks Andy thanks Robbie thanks for having me uh on this episode of model March uh yeah uh welcome everybody who will watch this once this is uh pushed U online but yeah today I mean it's um you know like a pretty important uh um you know like deck that we I'm going to take you across in general so we are finally you know like uh unveiling the um uh final aail thesis uh what we call

call as accelerating the unification of web3 and we will you know kind of go through what that really entails what that means uh what products that we have been uh building uh to kind of enable this uh new um function functionality for example and so on right like so we'll go through these slides and understand more um and so before we get into uh the meat of the presentation in general just to set some context on the on on the problem as such right like I mean uh what has happened over the last three years probably four is that uh rollups have

really um taken Center Stage uh you know like rollups were uh derived or in the first prototypes were implemented back in 201 19 2020 and in uh 2020 ethereum also pivoted to the rollup IC road map in general and so rollups as a construct uh started getting really popular uh 3 years back just for some context uh um you know like uh I um before like like currently I'm co-founded aale but before I founded polygon back in 2017 uh built up the a lot of the stack there in general and in

2020 you know like when when we we not some people know this that polygon used to work on this thing called plasma uh that was the predor to rollups uh and so when rollups came into the picture you know like we uh at polygon of course pivoted to ZK rollups and within the same time um AA uh we started aail within polygon back in late 2020 so uh that was the The Core reasoning behind that was that we were very confident in the rollup Centric future per se because uh we knew that rollups are a very nice construct uh in general uh and this changes the

fundamental way we view uh blockchains would scale because before rollups what used to happen was uh that if you wanted to spin up a chain you had to spin up a brand new chain bootst a set of validators come up with a new token ensure that token is valuable so that it has enough cryptoeconomic security for example and so on right like so so that's why you'll see number of chains out there but only some of few of them have enough cryptoeconomic security and so there's really a lot of friction to creating a new chain and the rollup construct essentially solves that problem uh so once this has uh come now we are are

able to kind of um experiment with variety of execution environments variety of uh State model models for example and so on right like so uh historically I mean there's been a lot of evm uh chains that have really stood out of course uh there's now bunch of other exhibition environments such as the Solana virtual machine and so on but essentially it still entails experimenting with uh you know like uh uh setting up a new chain entails a lot of effort right like and that's why we have seen uh not that varied set of execution environment and so essentially what I'm trying to say is that rollups

are uh are a a very big shift in how blockchain infrastructure uh will be built going forward and so we we certainly believe that they we are moving into a world where there will be hundreds and thousands of these rollups uh some of these will be uh of course evm rollups but we anticipate the vast majority of these will be app specific rollups um and uh in in the sense that we feel that um uh app specific rollup tooling um will mature uh to such a point that you know it will be as easy as to deploy a app specific rollup chain as it is to kind of deploy a smart

contract today right like that's the kind of majority that these systems will reach pretty quickly uh in general so just to set the stage right like I mean uh essentially rollups um as a construct have taken um uh over uh we already see a lot of rollups on ethereum uh in general L2 beat actively uh tracks maybe 4050 rollups at least uh now then you know like these coming years we are going to see a big big uh increase in these numbers so the point being that the rollup Centric future is is is going to be the preferred way to scale execution on

blockchains and so rollups we really need to scale rollups right we need to solve for rollups scalability and so that's that's like the first leg of the problem that we started AA with in general right like when we started out right like we we said that we will need to solve for rooll of scalability via building out a scalable da uh layer um and uh so that's that's where a Da comes in and a DA is this state-of-the-art um validity proof driven um data availability layer which implements data ability sampling uh uh I think we are only only the only project

that implements uh datab sampling with validity proofs and and this is some of the same technology that is going to be uh implemented in ethereum uh in its dang sharding road map uh um in general so so so that's the that's the rough background uh about you know like the rup Centric future in general right like but that's just the first uh um uh problem uh in general that comes from you know like how we view the rollup Centric future where it is going right like so once we agree that rollups are going to be the future and there going to be these tens and thousands of rollups uh so you have to think out a little uh in Into the Future uh when

you're building uh you know like a project or a company you have to think 12 months 24 months 36 months into the future and in general what is already happening though is that as the number of rollups increases the user experience is also getting fragmented right like I mean users now have to deal with uh multiple chains they will know they have to figure out how to bridge uh from one chain to another for example if they want to do something cross chain they have to go through a variety of third party uh uh solutions for example and so on right like so it's of course a lot of stuff of this is in development but in general uh you cannot deny the fact that

the user experience uh fragmentation is real uh and we can see that already playing out uh um so in general uh that's that's the that's really the second leg of the problem that you know like uh we at aale have been uh building for right like so initially the first foundational layer we build is the a DA component which is the more uh uh component which solves for the RO of scalability part but once we enable these hundreds and thousands of rollups then you have to deal with this user fragmentation user experience fragmentation problem uh in general and so uh you really need this sort of

unification framework uh to kind of sit between these rollups uh and sort of unify them uh so that you know like they can talk to each other and so that you know like you can create this um framework or on on top so that uh users can then interact with that uh layer directly and uh they do not need to deal with the underlying chain complexity uh so that that is the second uh leg of the problem that aail uh really sols for and so this is this is something um you know like uh we've been building over the last one year for example we've uh deliberately kept it um out of not talked about it too much but

we are you know kind of now uh announcing this uh in general so the second component is really aale Nexus which does uh uh this unification of rups as we call it uh via a combination of proof aggregation execution proof aggregation and validity proof based data vality sampling um in general and and to round that up right like I mean so once we have this foundational deal here with a lot of rups on top and you know like we these OLS can talk to each other for example what we also have to optimize for is this this whole coordination is to have the base layer to have high

cryptoeconomic security and so uh so there we um kind of uh introduced something called aail Fusion security uh which essentially U uses a combination of the AIL native token itself of course for securing the base layer but we uh have been working on on on a construction where we are able to um allow other high value tokens tokens that already have value like think like BTC or eth for example to kind of participate in in the consensus of the AA aale uh Network itself for example so this is this is roughly uh the three things uh that that you know like we we

have been working on and we are kind of now making it public so we call this the AIL Trinity uh so the AIL da is is is the base foundational D layer uh on top of it comes a Nexus which is the cordination framework or the unification framework for rollups and uh and because this because of this unifying framework requires unified security so we build a whale Fusion to augment the cryptoeconomic security in general I'll just stop uh at the uh for now before delving into more slides if you have any questions yeah I didn't realize that I was on mutes but so

basically the the the iteration here came from okay we're gonna build a veil da to come to scale rollups that we we see this this vision of you know abic rollups thousands of rollups we need to scale these rollups outside of just ethereum da okay a veil da okay then that the second order consequence comes of like all right now we got to figure out to kind of put these all together and make them interoperate okay so that's kind of like the next iteration if we get um from where we are today to h a world in which modular da is um vastly accepted by thousands of of rollups then it's like connecting them

and then it's like okay we got to do all of that in the most secure way possible and kind of solve that next order so through the iterations here and we'll dive into each specific part um but but really it's it's it's striking to me that um you know you guys are thinking multiple kind of steps ahead and really trying to like push uh a vision forward and kind of aligning your entire project on this on this vision and kind of taking a bet on AB Civic rollups and this modular future um yeah absolutely ready to dive into each of the specific ones as well

awesome uh so yeah so in general we introduced uh the three different uh layers so I'll just go through one one one by one for each so a DA is is something you know like which we've already uh put out in the open we have been you know kind of building this for the last three years uh it's a validity proof based data availability layer that implements uh data vality sampling as I said before this is the only uh project that I know we know that has implemented data sampling uh in a validity proof based uh uh implementation uh very similar to what U you know ethereum wants to implement on its dang sharing road map for example uh

uh and in general um uh so we use this combination of um kzg poal commitments uh and uh Vol proofs uh to build out the implementation the uh some of the characteristics of of how we built out uh the base D was uh that we we we used the substrate framework uh in general to build the node uh because our tool stack is primarily in Rust but essentially uh we also uh because of that we also you know kind of implement this consensus called grandpa and babe so this is like a

hybrid Ledger uh consensus that powers Avail uh there are couple of reasons why we opted for that um uh Grandpa allows for a big Valor Set uh at the moment it allows for about a th000 validators um and next year we plan to upgrade uh the signature mechanism to a BLS one where post that we should be able to uh scale that number up to 10K validators or more um so uh so so that's one of the reasons but essentially the the aail based D itself is is

pretty uh lightweight in the sense that uh it it uh acts as a uh data Ava scalable data availability layer it doesn't have any smart contracts it has a minimal runtime for things like staking transfers and data submission and such uh so but you know like the execution of programs is is delegated over to rollups that run on top of a whale U so essentially you know like what a DA really does is it it kind of scales rollups because uh blob space is abundant uh in the in the a deair uh the core reason is because uh what is different here from a

blockchain that combines both execution and data availability is that because we are uh implemented as a uh that a is a pure data availability Focus um blockchain we are able to increase block sizes uh so in general when demand grows we are able to grow the size of the block uh without changing um uh too much on the you know like the verification verifiability side for example right like so so again some of these Concepts um um uh lot of education has been done over the last one year uh in in the industry on the data problem but just to just to summarize right like so rollups really need to put their transaction

ordering order order data on on a on a Bas layer right like and of course it started with putting uh that like etherum rollups put the transaction order on ethereum but rollups are not uh uh specific to ethereum alone right Loops can be built on any any blockchain in general uh they just need uh you know some characteristics on the Bas layer uh and one of the primary thing is to put the transaction ordering there so AA does that I mean but ail also supports a lot of hybrid constructions where you know like transaction ordering data is sent to ail but let's say you know like a rollup sends the proof to ethereum for

example and so on so but in general you know like rollups are pretty hungry for data availability um like in general um uh at the moment like uh in the in the way current rollups are constructed uh 90% of the uh running like of the rollup for example like there are various estimates but 70 to 90% of rollup costs goes to you know kind of da da cost for example uh of course it will change a little bit but in general you know like rollups are can be built anywhere in general but rollups the main bottleneck for rops is

that they need uh very they're very hungry for data availability and so that's the core problem that ail da really solves and there is this component of data availability sampling uh that is uh slightly less uh understood like a lot of people focus on easy verifiability in this in the sense Maybe users can verify the transaction order very s but yeah of course that is a good uh side effect in general that users can verify but in general uh it's it has much more powerful uh use cases which we will you know kind of get into later and that's one of the core reasons why we are able to use this to implement uh the AIL Nexus uh as well uh but in

general um you know like uh so we we've been uh building aale da uh for some time now we've been running test nets for the last one one year or so uh and we are on our final incentivized test net uh and uh minnet for a DA is is is coming uh pretty soon uh in early H1 um in general uh so that's the bit around uh the aail da there's a lot of technical detail which we can get into but essentially uh you know like uh the the the main customer for a DA is primarily rollups for now there are other uh

projects uh that also use it for more in Creative innovative ways uh but uh as we have been uh working with a lot of Partners and customers in general right like so the primary demand is from from rollups in general and these could be app specific rollups uh evm rollups uh nonm rollups uh in general hybrid construction such as validium optimums and much much more uh so that's really uh the backdrop of uh a DA uh yeah any questions on a happy to answer definitely have one I'm sure Rob

does as well before we move on when we were speaking with uh pral on one of our recent podcasts we we kind of a little bit deeper um about the um the the ZK proofs and the validity proofs with your with your um da layer and one of the things that we talked about was how each user's wallet that's connected to a veil da can actually run its own light node um or or or something to that effect the question is um how does using validity proofs for data availability offer more customizability um and better verifiability for uh the end user and the and and um the rollup as a

whole yeah yeah no I think that's a great question in general uh so what happens is so you have to look at the uh kind of um the life cycle of of of the rollup transaction in general right like so in general when users send transactions to rollups so like there are various ways to have a rollup can be implemented like I'm I'm I'll concentrate on the most popular approach which is that of a sequencer based U separate sequencer based model there can be base sequencing uh setups as well but let's focus on the separate sequencer model right like so when a user sends a transaction it goes to the

sequencer typically you know like most sequencers today also have an Executor function uh which also execute this transaction as it comes does some validation for example and posts uh transaction ordered data onto the The Da layer or the base layer in general right like uh and if depending upon whether it's a optimistic rollup or a ZK rollup assuming it's a ZK rollup it will also post like a ZK proof for example optimistic rollup sends a little uh like slightly different data but it's essentially metadata that allows to run the fraud proof process for example uh essentially uh what happens

is that once the data is sent to the D right like so the why it is so important is that the ordering is the main guarantee that you require right like I mean because if the order which you kind of execute a transaction is changed it changes the meaning of the state transitions uh so if you have let's say 10 transactions in a batch or maybe 100 transactions or 10,000 transactions right like so the order is very important if the order is Switched or or changed for example then the state transition can change pretty significantly right like so it's very important to have an ordering guarantee so the word data availability is is you know it's a very it's a

slightly um you know misunderstood sometimes right like because just of the way it is named right like I mean some some people also try to call this data publishing certainly is a data publishing is a slightly better uh nomenclature for example but the the point is that you know for a rollup like the ordering is important like so the orop has to commit to an order and the base layer has to provide some kind of guarantee that this ordering will not be reverted uh uh and so that's why you know like there's a lot of emphasis on on on on on on the da layer in general right like so um now uh from now once this is done right like I mean

so what are these like when we say validity proofs or you know like fraud proofs in the D what does what do they really mean right like so essentially you know you have this um functionality where you know like the transaction data has to be uh um encoded in a certain manner right like and and that encoding has to be kind of uh you know has to have there has to be some proof whether the encoding was done correctly for example whether it was uh constructed incorrectly for example and this can be done in two or maybe more three ways as well like I mean you can actually rerun the all the entire thing and check whether everything is uh right for example you can also do this in a more fraud prooof

based manner where you have like a challenge window which could be in the order of minutes and hours or even days uh and uh so you can play this U uh fraud proof challenge game uh that kind of kicks in when when an incorrect encoding is detected for example and so on right like so so there's a more cryptoeconomic way to do this a DA does this in in a very Val proof based manner in the sense that so we we create these kzg polinomial commitments in the um and you know post them to the uh to the header of our chain and so there are these openings um again I'll not get into too much of the

technicals just to spare you guys but in general like what happens is on the uh like on the full node or on the light lines you can actually create uh proofs of certain openings in the commitments for example and so on and the beauty of this is that rather than um indulging in a cryptoeconomic game in the fra prooof window which might take probably minutes and hours uh uh here as soon as our you know like the our a Da Block is finalized uh which is you know like most blocks get finalized within two block times which is like 40 seconds currently like 20 second block times into do uh and from that moment on once finality is

reached on the base D layer you have this separate P2P lightland layer so we actually so the D layer is actually not a single blockchain it actually two networks so there is a validator run network and then there is a separate P2P network of light clients so these you can think of this as a very torrent like Network almost right like where where you have blocks being created on the main D layer and these light lines are fetching some data initially from the uh from the finalized block and then over time once there are more more light LS so when a new light L comes it does not

have to fetch data from from the from the base D it it just connects to another light client and takes fetches the data so if if you're aware of the seeding seeds and peers in a torrent Network like this is similar to that right like so so it basically over time creates a copy of the a uh like of the data on on the a chain onto the lightland layer we we Implement something called a gamli DHT uh like it's a uh distribut has table implementation to store so the each light Clan stores small fragments of the like the data on the Avil blockchain for a uh for for a period of time for

example and so on right like so so okay coming back to the question that you asked right like I mean essentially what happens is that uh so when when these light CLS you know like these can uh these can get create proofs uh for from the commitments and we have this um you know like uh eraser coding again not getting to a lot of technical complexity but essentially the main difference is that this process happens in in in a in in basically very less time right like I mean by the time the block is finalized you can actually immediately uh start creating proofs and verifying as opposed to uh optimistic implementations which would require you

to wait for the challenge window period to finish right like and why that is important is um when you want to verify this order um or the data availability guarantee in in other systems uh so of course users are able to verify which is a good side effect but the more important part is to uh you know if you want to verify this in other systems which are listening to these to a particular rollup right like the the fact that you can get this ordering guarantee uh in other systems pretty fast that enables us to kind of you know also build something like Nexus where you know like we combine that with let's say

the execution proofs to then build out something like Nexus so so sorry for the longwind answer uh but you know just wanted to provide all the context yeah yeah I think uh yeah that's um it the the the differentiation between the the two kind of chains is very helpful uh as well as kind of the uh the life cycle and the core reasoning behind you know the verifiability and the and the design choice so I think I think um we can we can move to as you kind of push towards the next slide we can uh now kind of explain H how that brings this all together yeah yeah so so yeah that brings us to the AIL Nexus

which is the second part right like so as we explained in a DA right like we went through the implementation and architecture a little bit but essentially it provides this scalable da capability and so we can as rollup demand increases we can increase the blob space and so we can enable these hundred hundreds and thousands of these rollups on top right like but then what happens is you know like you have like as I said before it you know like leads to user experience fragmentation like it's a lot of coordination work that is left to like other projects to implement for example and so on so there are VAR iy of projects kind of you know like handling this and of course a Nexus is

uh you know like one of the so we actually went really deep down to in order to architect this so essentially something like an a Nexus is only possible if you have something like a DA uh Datt sampling uh because it's a validity proof based approach we are able to kind of now embed it into uh this layer on top right like so what aail Nexus really is it's basically a you can think of this as a coordination rollup ZK coordination rollup on top of aale um and it basically provides uh two major uh functionalities so this is not like a uh regular smart contract or

application rollup it's a more system sort of a uh rollup and it provides coordination capabilities so one is of course uh you know like a proof aggregation uh capability uh so what we do is basically you know like we are able to kind of uh take proofs from various rollups uh uh that that you know kind of operate uh independently for example and then you know like compose these proofs together to create you know like like single proof for example and so because we are able to do this within Nexus then we are able to have one rollup um have trust minimized access to

the other rollup right like because of the uh proof aggregation uh because essentially when we verify these proofs and aggregate them together what we are really doing is you know you take one proof from one rollup and maybe another proof from another rollup and when you verify this together into one aggregate proof it essentially means that this this this single proof then has trustless um or trust minimized access to or you know like guarantees to the correctness of the state transitions on each of the these rollups right like so it's so ZK proofs in that sense are are you know like conventionally they have been used or in blockchain

conventionally they've been used in more of a offchain computation sort of a way but uh validity proofs also have this wonderful problem uh attribute of composability so where you can actually uh know the uh state of the other blockchain and can you can you know kind of um rely on the mathematical guarantees of the validity proof in that sense but in general proof aggregation itself is not enough right like so you need to combine this with the ordering guarantees so so you need to know what was the order of this rollup like what was the transaction order of this rollup and the the the state proof of the

execution proof of this zup and when you combine these together then you are able to kind of now coordinate uh communication uh between rups and to be to be very sure we uh focus on asynchronous messaging uh between rollups uh to start with right like I the synchronous messaging primitive is slightly harder uh to do I mean it is possible uh with you know things like blog based timeouts and such but you know like essenti what we feel is that uh similar to how apps on the internet uh scaled via microservices you know like I mean so if you look at let's say any uh app on the internet right like if you look at Amazon for example

uh and so if you when you make a purchase on Amazon uh and it sends you to their payment page right like again so so what Isen really happening in the background is Amazon does not handle the payment stuff right like it basically there's an Amazon microservice uh sending a request to a Visa microservice which then sends back the payment page for example and you send put in your details it gets validated Again by the Visa microservice and it sends a response that you know like the uh the payment is done for example or it failed for example and so on so all of this happens very asynchronously uh but it happens in such low latency speeds that the user hardly

um you know like like uh knows about this for example and so like today you know like that that latency is something that we have not reached in blockchain till now I mean that's a focal point of a a lot of Solutions in general but what we postulate uh and we believe very fervently that what is going to happen is that uh ZK proving systems have been you know kind of maturing pretty significantly over the last two three years and there's been so lot of investment in the ZK system that the industry has been building and we believe in in the next few years the latency for things like proof generation uh in general are going to become uh

less and less right like and so what you will get to a point is that you know like latency will become very low and so then then you then you can posit that you know like like there will be these there will still be these big rollups like evm rollups for example and so on which s you know we have a lot of liquidity for example and so on but we think that most uh use cases will be uh on on these app specific rollups which will have a lot of computation within within their rollup but whenever it whenever they require the services of anything on the other rollups they will just do a asynchronous message call right like so it's similar to how for example when you want to load up like

generally when you pay on venmo for example right like so payments on venmo are pretty seamless for example but when you want to top up venmo you have to talk to another Network for example right like so so we we think it's it's going to evolve in that direction and so essentially uh in this kind of a world right like AA Nexus is going to be key in general even today um there's a lot of fragmentation already that we are seeing between various ecosystems so there are um you know like rollup ecosystems that are um uh getting uh popular uh in general day by day but like everyone's if you kind of look at

the big picture everyone is looking uh to build their own ecosystem and so now there are and and the fact of the matter is that there are so so many great teams working on this problem and so you have so many ecosystems right like and if you're a user in one of these ecosystems it's the experience within the ecosystem is pretty good but when you when you when you have let's say funds outside of the ecosystem and then then that user experience can be u subar in general right like and so what we really feel is you know like aail is that kind of credible third party which can you know kind of uh work with these ecosystems and to kind of create this

permissionless verification Hub that that can you know kind of unify these ecosystems together right like uh and so in a sense that a NEX is complimentary uh to to that in general and of course there is also a sequence of selection mechanism uh that we introduce in a Nexus this is you know as we said there's going to be lot of rollups and today uh rollups sequences are uh run by teams for example and so on but once we get to you know like figure about you know like let whatever 500 rups or whatever right like I mean you'll see that you know rollups sequence are not run by the teams themselves right like they have being of uh they're being run

by let's say roll service providers and such for example right like uh uh and in general uh so you need some sort of Select so this is so we don't run any sequencers as part of Nexus for example we don't run any provs as well we just basically from a sequence selection point of view like for a particular rollup if we you can think of this as a more decentralized Marketplace for whoever wants to run a rollup right like so you can think of this as a auction mechanism uh that that chooses a sequencer and anyone can come uh come and bid in this auction to be become part of a sequencer so again this is basically that part of the Nexus again

I'll just pause before getting to the last bit awesome so similar to the venmo analogy all of the rollups that plug into a veil Nexus layer will exist and interoperate like they're in the same network that's the end goal yes that's the end goal like in general um like um essentially we want like we acknowledge that there will be a lot of ecosystems a lot of rollups different rollups and they everyone has their own um you know like ecosystem going on but uh you know like between this rollups in this ecosystem and

between different ecosystems what you want is this this framework on top so that you know like you can you can facilitate users to move from one one one rollup to another or funds or liquidity to move from one funds to another from one chain to another and the end State how we uh think of this is that you know like think of aail Nexus as this layer and on top you know like you have let's say users interacting with their wallets for example and the end state where we feel is that the user expresses their intent to do something on on a on a particular chain uh and maybe they have assets on on multiple chains for example right like and and we

assume that you know like wallets will get more smarter they will have track of you know like assets on multiple chains and then then they can the wallet itself can talk to a whale Nexus for example and then figure out what is the best path uh to do what the user really wants right like so uh of course we will not we will not build out this entire stack on our own there will be smarter wallets for example there will be intent layers on top for example to coordinate that but we provide the the the substrate for for for for these rollups to talk to each other got it and earlier in the last slide you kind of hinted at a a bullish kind of thesis

towards other virtual machine structures and other kind of rollups with different settlement layers different virtual machines you mentioned svm for example is a veil Nexus in its state that it's planned to launch will that be able to interoperate with every type of virtual machine in terms of customized app specific rollups general purpose rollups K optimistic as well as different settlement layers or is this just for ethereum settlement layers how flexible can Avail Nexus spe to accommodate all the various types of rollups that we'll see live no it's a it's it's a uh good question uh valid question so let me try

to frame it in uh like two two answers so one the first answer is that uh our Focus currently is is uh primarily because it's a proof aggregation based mechanism so the focus is initially on the more z uh rollups um to start with right like so that's the design U we we started with um the the next uh focus is like once once we have you know we roll out ZK rollup support for example we are also experimenting a little bit with how do we do this in a more optimistic rollup sense right like so so optimistic rollups have have to be handled slightly

separately uh but you know like we are experimenting with things like consensus proofs cons ZK consensus proofs and storage proofs to also uh add optimistic rollups to support to the to the Nexus um in general from a uh like a non rollup from L1 perspective like a um non rollup chain perspective uh the ingredients that we really require are things like uh you know like like consensus verifiers for example uh so consens verifiers and and state verifiers um and so we we've for the

last one year we have been you know kind of working with u you know like with with uh internally and with other partners to you know like enable these different sort of consensus verifiers so we use this consensus mechanism called grandpa so our you know like we built something called Vector X which is a grandpa verifier ZK Grandpa verifier um for for uh our chain uh but this the same concern is also used by let's say pout for example uh and our partner sassin Labs also previously built out the ethereum uh sys committee consensus verifier which is which when we we are

able to kind of verify e consensus at least the sync committee consensus so so there we are able to add ethereum rollup support for example and you know like they also built out a tender verifier ZK tender verifier so you can think of you know like using that to connect to Cosmos CH so there are a few elements that you require uh so for example to um answer the question around the svm uh essentially you know like the theoretically the framework should support um any sort of L1 but it also depends on the complexity of building out the consensus verifier the state verifiers and so on and Solana particularly or or the svm right like so

there there's the svm element but there's also the Solana consensus element so uh so you we need to have built out the Solana consensus verifier for example and so there are efforts in the Solana ecosystem for example that work on you know building out light CL verifiers for example for the consensus uh in general and so we will we will monitor those developments uh but essentially so the answer in is in two parts right like theoretically yes it might be possible practically then has to be a susten practical implementation of of some of these Primitives uh which uh which I today are not there it might happen in the

future thank you for that um I'm certainly excited it feels like with the ZK realm it's like every every year it gets closer and closer and closer so yeah yeah makes sense I also you know like mindful of the time I will just kind of also get into the final uh part of the AIL Trinity which is uh called aale fusion and so basically because we you know like I already explained in terms of because a DA uh enables these lot of rollups and a next then kind of unifies them into uh uh into a coherent Hole uh in general to power this ecosystem we also require this uh the

the shared security layer in general to be uh pretty robust in terms of crypto Economic Security and so uh so we ripped on a lot of designs uh talked to you know like a lot of teams who have been working on similar problems and arrived at you know kind of this design in in general but but the point is uh the the core idea is that like currently the AA da chain uses the aail native token to power its consensus and so it depends upon the value of the AIL token uh uh to Define its cryptoeconomic security and so of course uh you know like we hope it'll be enough in general but uh we also have

the ability to leverage other uh tokens from other ecosystems to you know kind of add to the cryptoeconomic security and so we kind of um some of some these ideas uh you know like kind of inspired from uh you know like the mes security system from osmosis uh for example in the cosmos ecosystem uh some from you know like like I layer for example about baby la chain also is doing um something in the BTC and so we uh so these are all project projects and friends who we talk to for advice and um sounding board and you know like in general um like so so we we kind of got inspired from the designs and but this

is slightly different from from from like let's say like just uh IG layer reaking so it's basically we we are able to take any foreign asset like BTC or eth and power the aail consensus and you instead of slashing happening in a contract you have slashing happening in protocol right like so uh so there's a that's one of the major differences and it's a slightly different design than other approaches but this has been you know like inspired from primarily from you know I would say Mass security uh in general uh and this is this is because we want this unified layer to have higher security and what best uh then to kind of you know like take some tokens

which are already valuable and then you know like uh put them in in in a game theoretical manner on to power the consensus so that's um that's one of the uh uh core insights behind a will fusion um there is another small element that is also pretty important but you know like we will we will expand on that later is that that you know like along with other Val valuable assets we also want to add in a small provision for new rollup tokens uh to be added to it because when you think that you know like you have these rollup tokens coming uh new rollups coming in and uh uh and

tokens rollups releasing their own token uh the utility of the token itself is is a little weak uh in general and so if we if we are able to kind of accomplish this and to allow new rollup tokens to also participate in a will Fusion you know like that's added utility for the rollup token itself so in general so that's that's one uh one thing that we have been um you know like we plan to put in general so yeah in general um just to sum it up right like I mean so these are the three uh elements of of the aail system that we wanted to talk about um uh and in general you know like we strongly believe in the the ROP centry future being the future of

blockchains and so we solve the problems uh strive to solve the problems of what constitutes that future in in these three parts of the AIL stack which we call the aail Trinity uh yeah I'll just pause and stop for at this moment yeah Andy you're on mute I'll just quickly read this uh yeah Avail accelerates the unification of web 3 by enabling modular execution layers to to scale and interoperate in a secure and Trust minimized way which I think is very powerful given the fragmentation of

rollups now and and as new rollups come out the the greater fragmentation uh it it calls for unification more and more as that happens um I have a quick question on fusion um as as new rollup tokens are added into uh Grandpa consensus I if you know let's say there's a there's an illiquid token because it's a brand new rollup and it has some some really significant volatility would that volatility perhaps to the downside affect Grandpa consensus or I imagine because Bitcoin and eth are also considered uh crypto Economic Security for this consensus mechanism

that provides the bulk of the security and a single one of these new rollup tokens you know going down and and going bust does not impact the the consensus uh to significant measure yeah so I mean of course you know like we have to tread very carefully uh you know when when it comes to you know like adjusting that so in general you know like you can anticipate that new rollup tokens will constitute only a very small percentage of the total Economic Security because of course because these can be very highly uh volatile and El liquid in general and so you can think of the AIL Fusion as this set of separate quorums uh of various tokens so there will be the AIL

token having a quorum um you know like BTC Quorum is different um and E Quorum might be different and these rollup tokens will also have a separate Quorum right like and but together like you know like um it will we we of course um there will be a lot of training LS on this functionality for example so you we you can think of this very much similar to let's say like like U like a gradual onboarding window where we kind of you know wet these assets that can come in for example and for certainly for the new Ro of tokens we will we'll keep it very uh you know like Risk adjusted and very small portion of of that right like

unless like you can think of there being like a very like it being a very small part and if it grows big then it grows U like gets a much bigger uh percentage of the pie that kind of a thing so I mean again we we are still doing a lot of economic modeling and you know like what if scenarios for example and so on so we of course uh um you know like we we will we are going to be very careful in U on uh rolling this out yeah yeah I want to be mindful of the time as well I appreciate you coming on I just want to ask one last thing about Fusion as well which is going to be short what is GR

Grandpa consensus stand for where did Grandpa come from yeah so grandpa is basically you know like this uh consensus uh protocol uh that was created by the poot uh folks in general uh so um so P like substrate or poot in general has this hybrid Ledger consensus uh where uh they have this uh so babe is the block production um protocol and then there is a grandpa which is a finality protocol so essentially uh it's a it's like a hybrid uh finality Ledger in the sense that you have a block

production protocol that goes on and in general Grandpa kind of will come in and finalize a set of blocks uh at some interval right like mostly it will kind of uh immediately finalize blocks but it's so why why this distinction is different because it's slightly different from instant finality protocols so for example instant finality protocols like tenderman you know if there's a you know like finality like if there's a problem the chain hals right like I mean essentially uh the the block production also halts for example uh and so I mean there are different trade-offs to doing which uh uh

different kinds of so basically uh in in uh in Grand like in uh aale for example you have Grandpa and babe where babe block produ handles a block production and Grandpa then is like a finality gadget on on top of it um in general yeah all right it's just quite quite a funny name yeah yeah yeah I mean that uh yeah I mean I also put out a uh tweet said you know like grandp babe memes are uh um like a very uh like under underutilized yeah coming soon on the roll up thank you ineg for coming on and

sharing some of the uh broader vision of Avil the Avil Trinity um and generally accelerating the unification of web 3 really enjoyed it yep thank you very much for having me guys and very wonderful to meet you as well likewise thanks