Aggregation Day Keynote

[Music] good morning everyone nice to have everyone here uh actually I should rephrase we're having this live around the world good morning good evening General ZM GM uh I'm Hudson VP of uh community at polygon Labs also voted uh handsomest uh L2 Rascal by Time Magazine this year very big honor didn't even know Time Magazine cared about layer 2 Technologies so we're having uh today the aggregation day uh and we are also going to uh be talking to people internal to polygon Labs we're going to be talking to people who have built this talking to people external to this about a lot of cool stuff we are very very excited about today uh so we'll we're going to go ahead and get started uh I'm going to see if the screen share works and it does the magic of live streaming so um yeah what we're going to do is go ahead and invite uh people on stage here see if they made it in send people over yeah there we go all right Bron up here first uh hello hello Brendan farmer is a co-founder of polygon uh and a uh uh works on Pro polygon zero so Brendan go ahead and take it away great thanks Hudson um can we uh can we get slides up are there slides here hey Jord oh [Music] hey okay so slides I I'm actually not able to see thees but all right so hey everyone I'm Brendon on the polyon ZK team and I just wanted to talk a little bit about so a couple weeks ago this post and in it uh we made two claims so the first is that no single blockchain is capable of supporting the throughput required for an internet scale crypto ecosystem so uh this is sort of a challenge to the monolithic view of scaling the second claim is that uh even if we add block space and scale hor ially if we do that while fragmenting liquidity and state and users um then we failed in some sense so like when we talk about scaling in crypto we really mean scaling access to liquidity and to Shared State and so this forms the basis for the aggregated blockchain thesis which is we need a horizontally scalable multi-chain ecosystem um but we need to be able to have that with while preserving Unified liquidity and shared State and composability um so the uh sort of the the upshot of this is that polygon has released this vision for the aggregation layer which we view as like this foundational piece that will allow us to provide this so a multi-chain ecosystem that still has unified liquidity and state and composability um and so next slide I I can't see the slide so I hope that it's the next slide but um and so when we talk about unified liquidity um in in in my mind and in polygon's mind that has two components so it the first is asset fungibility like if I'm in the polygon ecosystem or this multi-chain ecosystem around the aggregation layer I should be able to take eth on zkm and I should be able to bridge it to like the okx chain or um uh like AAR or Manta or Kanto and I should get eth on that chain I shouldn't get rap eth I shouldn't have to rely on liquidity providers to swap me back into the native token for for that uh chain and so the second component the second requirement is fast cross-chain transactions so uh right now if you think about how rollups uh and cross-chain rollup transactions work on ethereum um I need to in order to send a message from rollup a to rollup B I need to to first uh submit my transaction to R Bay generate a proof wait for that proof to be submitted to ethereum wait for that block to be finalized on ethereum and then finally rollup B can accept my cross chain message and so this uh doesn't work for unifying liquidity and Che State um and so the aggregation layer and next slide is uh is proposed as a solution for this so right now we have this fragmented L2 ecosystem and if we continue along this path then even with a million rollups even with like as much block space as we could possibly use uh in the world we still will run into this issue where we have liquidity in state that's fragmented across all these rollups we need a way to unify uh that environment and make it feel like a using a single chain for users and developers um so fundamentally the aggregation layer is a decentralized service that sits uh sort of beneath the the uh the L2 chains in the polygon ecosystem so it allows for uh safety for asset fundability and super fast cross chain transactions and uh it allows for um all L2 and L1 native assets to be deposited in a single unified Bridge so that uh I can have um like L1 native eth on zkm and I can move it to zkp or to uh the okx chain and I can get back L1 native eth on that chain similarly for any L2 native asset I can take that asset and I can move it between chains really really fast um and fundamentally I can do so in a safe way that doesn't jeopardize uh the assets that are deposited in the bridge and I can do so without requiring uh an L1 interaction and so for us like the aggregation layer is kind of the this Foundation that provides safety and then we envision sort of an emergent ecosystem of coordination mechanisms like shared sequencers like relays like Builders um that allow for this cross chain interaction to happen but fundamentally they can do so in a safe way users and chains don't have to trust uh these actors uh to not misbehave they don't have to trust the aggregation layer um and so they can transact in like a super low latency way without additional trust assumptions um and so next slide please so today the sort of upshot of all this is that the first component of the aggregation layer which is the unified bridge is live today on mainnet the LXI Bridge contracts have been deployed and this is sort of the basis for additional cdk chains to plug into this emerging ecosystem um the second milestone is going to be proof aggregation uh and I am really glad to have Jordy uh here to talk in detail about about proof aggregation yes uh thank you Brandon yeah in the core of these aggregation layer is the the this proof aggregation I mean proof aggregation is very much uh uh exploring the recursive as snarks uh technology uh recursive as snark has been theorically set for a while but practically uh has not been so we have we the space has not been putting this uh technology in practice uh until the zkm launched actually in the zkm we are using a lot of this aggregations when we are taking many batches and we are aggregating together to a a single proof that proof a a segment of a blockchain that includes includes different batches so mainly what we are doing is we are taking a batch of aof another batch of another proof another bch another proof and we're aggregating them together okay so next slide when we what we do in the in the aggregation mainly is using this technology of aggreg of aggregating proof to to aggregate many proofs of many change you can go to the next slide please yeah so the the the the here is we are taking a proof of a chain another proof of chain and we are aggregating them together to a single proof but we do that recursively we take another proof and we agregate them together and then we take two sets of proof and we get it in together so at the end maybe once per let's say once per five minutes we are taking all the proofs of all the states of all the chains we are competing a single proof that proves all the chains and this is the the the chain that we are setting that uh uh on on on Main and because it's a z proof we have the warranty that uh the state transitions happens in the right way that the that the security of the full ecosystem preserves and that uh every everything works but here the important the important Point here is is this uh uh proof this ZK recursion this proof aration that we are using extensively uh uh inside inside polygon why is important this in aggregation in the aggregation layer because you cannot here all the point is that you cannot start using the fonts in another chain until you are sure that those that those chains has been uh exited that has been happens in the first chain and the proof is the the generating this proof is what rules it do okay this compared for example with optimistic approaches you don't need to White for gamings on that you just have the proof you know that this is correct you can go right so here is the the the main topic and yeah that's very much what I wanted to explain about the the the proof the proof aggregation awesome uh Mark I I uh interrupted and stepped on your intro talk because you were not on stage do you want to to give it now yeah I know I think maybe setting the stage for some of this would be would be really good right when when we really think about um well first first I just want to say like I think today's going to be a day that we are going to have uh some of the best ZK content that you can possibly uh get and that's not just coming from our internal teams but really awesome external speakers as well which is really exciting um but what you saw right is like the core of everything that we do is at is really ZK Tech at at polygon um and when we were evaluating like paths forward from like a scaling perspective what we realize is that ZK Tech is going to be at the core of that and when you take that one step further what you realize is that there is a way to improve on kind of blockchain approaches that have been taken so far right and so like the two popular approaches right is like monolithic blockchains and modular blockchains and we started thinking about that we're like okay monolithic chains have problems that everyone recognizes modular blockchains have been viewed as like an improvement on that but the the downside of them is this fragmentation that you get and so what can we do to bring them together right and and I think we're uniquely positioned um with the ZK expertise that we have to bring them together into aggregated blockchains um and you've heard a lot about like polygon cdk if you've kind of followed polygon at all um what you've heard a lot less about is kind of the stuff that we're jumping in today um Brendan gave a good intro on kind of that first part um of the agat going live um and that's a very core kind of first component of it um but there's going to be multiple stages to this and and Jordy talked about a very important kind of second stage of that um which really has to do with with proof aggregation um and we're going to need to go even further than that to get to that low latency AG layer that we all want and that is going to create that amazing user experience that people have probably heard us talking about um and so I'm kind of excited for us to just continue to explore that um I think uh kind of setting the stage with what the AG layer is and and what that next stage would look like with proof aggregation it's kind of like a really good start and I think maybe we could jump in now to talking about what some of those um kind of benefits to the AG layer are are going to be um Brendan if you want to kind of take that away uh moderating it I think that'd be a good place to to had now yeah sure happy to Mark um so I guess the first question I'll direct this to sandep and Maho what is sort of the endgame Vision like when when you imagine this like internet scale crypto ecosystem with a ton of chains and and like something that looks different from what we have today uh like what what do you yeah uh maybe I can I can start in uh then you can take up uh so I mean for me it goes back to the roots of uh when we started I have multiple times stated also that given a choice between like two architectures which you know kind of takes us to a path where there will be some midterm gr scalability but then it is not the long-term solution versus the solution which is uh uh you know in the long run is going to be it is the solution which can be the next uh you know this trustless internet or web3 that we call we'll always choose that and aggregation layer is or the aggregation mechanism and with this polygon Ang I I think like all of us first time we believe that this is one architecture which can actually scale blockchain networks to the scale of the internet without compromising the security uh of the system and then you know there have been multiple uh you know kind of definitions of aggregation that has been given uh to me I feel that you know it's very simple that all of these chains these are all individual State machines and then aggregation is basically Aggregates all of those State machines one single security zone so we have also multiple times said that this becomes like a single looks like a single blockchain uh where uh you know people are transacting the experience is much more streamless L and all that and other important part is uh you capitalistic in nature that means now you know you don't need to be on a shared blockchain if you are a developer you know we whether you are a large Enterprise or a or a crypto Deen uh you know Builder or you are a college student uh who is building some game or some experimental application you can simply SP spin up your your you know kind of chain or I call them sometimes trustless servers and all that and connect it provide your zq proofs into the aggregation layer your chin is as secured as any other um you know chain uh in the ecosystem right so that's like having that same same uh you know security Zone across big large small application that's what like internet has provided today whether there is's a large scale website or web application or there's a small application on the internet they interact Sly together although there is the security uh there is base security element but this Val Val security element is not there with these blockchains the the security of the value uh element is also there and I think it it for me it's like it looks very similar to the structure of the current Internet which is the trustful internet and this architecture can provide the trustless internet future so that's why AG for Meo over to you yeah I can pretty much second what sandip said first of all I'm very excited about the the whole aggregation thesis as well as all of you guys and the rest of the team I think in our firm belief it's the next kind of evolutionary phase of blockchain design or blockchain infrastructure uh and it's definitely going to take us step closer to that end game basically which is a global globally scalable internet scale basically network of value which is seamlessly interconnected and feels like a single Network just like the internet as as sandip said so um yeah not to go into the the through the Genesis kind of and retrospective of of blockchain design and approaches that we have taken so far as the industry so from mon monolitic to modular we really believe again that aggregation is or aggregated blockchains are are what is going to take the industry uh mainstream basically and the the end game or the goal here is this as I have been saying multiple times This Global Network of value so with with the internet over the past couple of decades we have as as the society we have been able to establish this Global Network of information which has as we all know transformed the society in very profound ways right and it kind of democratized access to to to information and knowledge and and uh um made the entire Society more efficient more interconnected and feels like this really network of equivalent network but for value is really the missing part is it's almost like how come that we still don't have this right and and uh with the blockchains specifically with aggregated blockchain approach and and and design direction we really believe that we're really kind of uh seeing the kind of the clear path to to to actually build that Global Network of value which will in in our firm belief transform the society probably in even more profound ways than the internet has so that is kind of the the end game and um yeah very excited about yeah I think that's a really nice way to put it um and I should note that uh in response to the chat s deep is not a fugitive On The Run he's just on his way to eth Denver and is currently I believe outside the Eiffel Tower uh in Paris so not for not for fun I at NF Paris and I had I'm freezing here give us since we literally can't hear a word that you were saying can we at least get a shot of the Eiffel Tower uh in the background peration you anything uh I I still can't hear you um Mark uh over to you so what we're what we're announcing today is um that the lxo bridge contracts are live and so this is like the very first step toward this aggregation layer Vision um but there's also a really cool primitive that's being exposed with with these contracts it's called Uh bridge and call and it enables uh really cool cross chain interactions um across l2s and the polygon ecosystem could could you just describe uh I believe we lost Mark um uh Brendan in the while let me just point a comment because I think it's for this just just for for for this uh I want to go back uh in 2015 in at the beginning of uh at the beginning of ethereum uh when there was not this division between data availability and execution ler was exactly the same and we're talking about sharding but sharding was a like a kind of mix of roll up with that availability because it was everything together okay and at that point uh uh uh we already knew this test is that I mean that individual blockchain will not scale and we will need many shards we will need many execution environments or many shards uh By the time and I remember the work done by the uh Foundation by the time I remember some vitalics talks talking about uh composability talking about uh a synchronous calls between shards or between that oability this agregation layer actually is the the is is Devolution of of of of of that time I mean actually we're building that part of course here we have that distinction of that availability we have the ZK we have much much much more bigger advantages for example by the time the shards had to be very um homogeneous at some point here with ZK we can have like a we can give chain much much much more Sovereign on each of the the individual chains but all the I mean all the tag of this composability between chains of calls and this refers to this I mean this message and call but I mean we can even talk about further about this uh I mean the synchronous calls or uh even Atomic transactions I mean all these uh uh what of interconnecting chains and this is the I mean this is the beginning of this uh era we right now we are start having different chains uh connected uh together and actually is when this uh I think that today the the historical part of today is that today it's possible that all that thing that was kind of theorical today is uh start being possible and all these interchain uh Layer Two communication composability uh will start happen and it's definitively possible and here we have a huge space to to to grow and and and to build uh on on that if you want taking the simuls I mean uh if we have processors one thing is doing a fast processor which what is a chain and now it's going have a multicore processor I mean you have this idea of multicore processors and then of course the communication between processors and all that becomes a lot of importance so this is very much the where we are from the technology perspective uh uh uh in the historical in historical manner in historical Point yeah I I I think that's a really good point I I think one way to look at this is like we get all of the nice guarantees of of sharting uh where we have like asynchronous and synchronous cross chain transactions we have low latency we we we have like really nice trust guarantees but fundamentally like the uh like the requirements and restrictions of sharding are not imposed on chains we we can have a heterogeneous environment where uh chains can coordinate with shared sequencers or relays or Builders but we have this fundamental guarantee of safety uh and asset fungibility that's guaranteed by ZK and so it's like a permissionless uh environment rather than one that's like very sort of like centrally managed and monolithic um I think that's a really powerful Vision because it allows like so many different participants to like come together to build this ecosystem uh and so I I I think it's like really exciting I I think a lot of people are are right to be skeptical of like of sharting but I I think this is like a fundamentally different Vision that preserves the same guarantees and has some really nice properties in terms of like who the participants are who's building it um and how it comes together uh but Mark before uh you were bridged to another stream involuntarily uh we were talking about bridg and call so if you could uh if you could just go into like before say something like can you hear me now like not only I was sitting in open it started raining it started raining also that's the first thing and second thing is BR Brandon you said that you couldn't hear a word Ando spoke to me and he said I second what said yeah yeah I can feel what you're seeing maha's answer was very good uh but but Mark uh so maybe you could go into like what bridg and call is what behaviors and functionalities it exposes and and we can kind of go from there yeah yeah and I think it it's worth kind of take a step back right and talk to like this this first component like the lxly bridge this unified Bridge like what what does it mean right and like for now right what it what it means is like relatively slow relatively expensive like transactions that that still like hit kind of ethereum right and that's like not what the a layer in its endv vision is at all um but what we thought was there is actually a way that mean that that actually uses this unified bridge in a way that is going to create an amazing experience for like developers and users in ter terms of a unique feature and this is where brid and call comes into play right and so if you think of like your typical experience right now when you bridge onto an L2 um you know you submit a transaction to actually Bridge once you get and you've actually like bridged across you then submit a transaction on whatever application that you want to use and you use that application and frankly if you wanted to then leave that L2 and go to another one you would Bridge back to ethereum and then bridge over and then start using an application again what bridge and call allows you to do is in one transaction bridge to uh an L2 that is connected to this unified bridge that is the first component of the AG layer and once you've done that in that same transaction you could let's use an example you could take eth that you just bridged and you could exchange it for madic um in that exact same transaction you could also take that mtic onto another chain and exchange and that for a skin in a game and and and purchase the skin in the game and so that is something that can all be done in one transaction with bridge and call that otherwise would probably take four or five separate transactions today and that is all because of both this kind of unified bridge and the bridge and call function and it kind of gives you just this little bit of insight into what the a layer might look like from a from a reduction of transactions perspective and cross-chain transactions what it misses is that whole low latency feel and like one kind of unified experience which is what's going to come in in kind of the coming months yeah that uh that makes sense I think um the way that I describe it often is like um like like right now in a multi-chain ecosystem you're sort of forced to choose between uh deploying your application or your nft or your game uh on the same chain where there's liquidity um or trying to deploy it elsewhere and like actually bootstrap liquidity on that chain trying to like bootstrap an nft Marketplace or an amm or or and and and that's actually really difficult to do and it's something that I think is unnecessary for Builders and so like part of what makes this Vision really powerful in my opinion is um the fact that like an nft like like you said a game that's selling skins or or an nft collection or defi primitive can deploy into the polygon ecosystem and instantly access the liquidity that already exists and the shared State and the users that already exist in the ecosystem and so like we're moving out of this mode where uh chains need to like play games and and and uh engage liquidity mining incentives to to build liquidity on their chains like as a developer you can directly focus on adding economic value in the form of a game or a defi primitive or an nft collection and and you don't have to worry about bootstrapping liquidity and so I I think that this is like sort of a step change for um for the developer and the chain Builder experience and I I think it's you know I'm very biased but I think it's it's really exciting and so one of the things there Brendon right which is like the idea that like you as a developer actually get to focus on like what it is that you're here to do and what you're good at and you could focus on like how do I actually just create the best user experience possible not how do I have to worry about like uh attracting liquidity and attracting like all of these different pieces of infrastructure that you might need around that liquidity you're now like I'm just going to build for example the best game possible um and then I'm going to abstract as much away from that user as possible and that's where I'm going to spend my time and energy and that is much better for the developers who actually want to develop and don't need to worry about it and it trickles down to the end users in a very meaningful way because they're going to end up getting a better experience as a result of that yeah yeah exactly and I think the other half of this and here just give me one sec s I'm segueing it into a question for you in Maho like the the other part of this is uh like chains aren't giving up anything to participate in this Vision like there are no restrictions on how you do sequencing on how like what your token is for staking or for gas or for governance like like like there's no Revenue share that's imposed on chains and so this is fundamentally like an open and permissionless vision that's not a rent seeking one and so I I think like Maho and sandep when you think about like building the value layer of the internet and what that looks like and and why we aren't uh focused on restricting how chains behave in this ecosystem like why why do you think that that's powerful what's what's sort of the benefit of that you want to start so before before we go and I think his internet access was cut off yeah Mah to you let me take it and he will he will definitely uh rejoin yeah so um it's it's um it's a very transformative again approach that approach of unified liquidity first of all is I would say transformative both for web 3 or defi to be more specific standards and traditional Finance standards so so we have a completely new primitive complet something completely new the the the opportunity to Leverage uh uh other chains other application other projects liquidity right it's a completely kind of New Concept and as you as you said uh currently in the in web3 we have this multiple layer tools competing for finite finite basically amount of capital having to employ different kind of incentives and whatnot to attract uh uh liquidity and liquidity by the default ends up being fragmented right here we have a completely different Paradigm where change are working with each other instead of against each other so it's a it's a proper Paradigm Shift right and and as I said both for and for traditional Finance so it's it's really exciting to see how that will play out and how projects will leverage that and and you know what's going to come out uh of that that being said to answer your your question um we have understood first of all as you know very well and as a lot of and a lot of work has been invested on the technical side of things designing the egg layer and and cryptography and how to make the whole thing efficient and scalable enough and all of that but we have also as as you as you guys know we have put a lot of thinking into how to really make this appealing or attractive to everyone to join so we understood very early I would say also from the kind of business uh Viewpoint that this really has to be a public good this really has to be a non-extractive uh um service non-extractive basically layer that is just going to bring benefits and bring value to all the chains that decide to participate and not extract value not limit sovereignty not impose ideally any type of restrictions that you have to use this token or that token or whatever you're completely completely free to do whatever you want to do in your ecosystem there's just only this added benefit of shared liquidity simulus interoperability with the entire ecosystem so it's pretty pretty exciting and I would again say of course we we are biased but I think we have done a good work um defining and designing the the egg layer in in a sense that it's really a public good and really a clear value ad for every chain that decides to participate without uh uh extracting anything from them or or limiting their sovereignty or anything like that yeah to you absolutely like you know the sovereignity BR Brandon can you hear me clearly by the way yeah so sovereignity basically has you know always been a very important element of that and a testament to that is that you might have seen uh people who are active in the polygon community that even layer ones which are like you know near blockchain and uh you know then Kanto and many other ecosystems like now which are completely separate they're not even layer to they also want to join the agar and you know I really loved uh you know vitalik's recent post regarding uh like you know make ethereum Cipher Punk again and in that one you know he mentioned that build tools not Empires right and I think like a a layer is actually you know is actually in its in its own inspiration is like that like we want to build a you know tool a layer which everybody can use it doesn't you know kind of restrict anybody holds holds anybody it just provides a great service and benefits so if people want to use it on their own accord they can use it they are not like logged into anything or things like that like you know we we we don't want to do that and then because it's a great service and there are so so many advantages like interoperability and shared liquidity and all that uh people would work over there another big Advantage which I wanted to you know say what uh what Mark was also mentioning bridge and call like many of the small uh features for the developers let's say on ramping right in like how chains can work with each other like if one chain solves on- ramping then you can like you know that on- ramping solution if they have a very you know kind of robust on ramping solution that can be used by every other chain in the connected in the agay right you know so you can just do a bridge and call on ramp and come there one good chain sols a you know withdrawal and deposits from and exchange especially the withdrawals all exchanges all all chains can get like you know withdrawals integrated using bridge and call you know from from The Exchange and all that so like as mialo was saying that this is like one architecture which also allows all the chains to work with each other anybody solves one problem whether it's for liquidity exchange Integrations on ramping and so many other things everybody else gets benefited also with that and fortunately we have one of the largest exchanges in the world the world that's launching a cdk chain in in okx and so I think that vision is something that's like very immediate and accessible um Jordy I I apologize that we've not gotten to you uh as much but maybe as as one of the last questions uh you could talk a little bit about why this vision is enabled with ZK like you you alluded to it a little bit earlier in in your uh talk but maybe just like the magic of ZK and why that's what's enabling this this technology to exist the point the full point at the end is the I mean the ZK give this immediateness I mean you can prove something with a single proof and you don't need any interaction I mean the ZK is this non interactive part I mean you prove something and that's there you don't need to White and here you avoid all kind of gaming I mean of collaterals uh gaming going back and forth and it gives you instant uh or I mean or or if you want the timey text to generate the proof but which is every time by the way every time it's faster and faster okay so this is give what gives you the immediateness okay and this when we are talking about uh we're talking about the share liquidity but the share liquidity when you have many change mean that you need this immediateness I mean you go to one chain you do execute something there you go back and this immediateness is what ZK can gives you because I mean you proof you're there you proof you I mean it's like a clock you proof and you go and and and this is this is I mean this is the importance uh the big importance of this Tech actually you could do that in let's say on chain but would be not practical because I mean you meaning that executing things in the consensus layer is with doing something in solidity you could I mean you could build that but uh consensus should execute so many um so many pieces of code that makes it not practical and Z case you can I mean you can compute that aite you you prove that aite and in the consensus you just give the proof and the proof moves forward and this is what enables uh uh everything I mean the big secret of this aggregation is the of this aggregation and because this is possible is because the is because this uh Z that's awesome so I I I think that's a good place to end this panel uh thanks to to everyone and yeah let's uh over to Hudson [Music]