How To Connect Ethereum's Rollup Clusters

Transcript

GM what is off and welcome back to the rollup today's episode is with Noah from nkit and bow from polymer these guys are two absolute gigabrain that are working on interoperability trying to solve the same problem with two different approaches we have B representing the IBC Hub and spoke canonical bridging solution we have Noah representing a more Innovative newer we'll see if it plays out super Builder approach and these guys are at the top of their game they're talking about interoperability ultimately what we find you guys are going to have to watch to find out before we get into it first a word from our sponsors enjoy Everclear is pioneering the clearing layer the

clearing layer can almost be thought of as splitwise when you go on a trip with friends but for the blockchain what I mean by that is when users are sending cross Ching messages there's often times fillers and solvers who have balances left on different chains that they need to rebalance and reset their positions this is where ever comes in ever is pioneering a collaborative approach to the chain abstraction future and they are willing to work with interoperability Partners solvers and other providers in the space to build a interoperable and chain abstraction future check out Everclear ever clear.org and welcome back to theup going on guys welcome back to the rollup

today we have the superb Builder versus IBC or super Builder and IBC conversation uh yeah we're going to talk about cross rollup interoperability again but we're going to go deep into the technicals today and kind of really understand some of the how um I think if you guys watched the recent episode with Jill from espresso we really unpacked some of like the history of web 2 into web 3 and how these ecosystems kind of came about and why this is so important and I think the key takeaways there were really trying to build open systems um and really trying to not kind of build wall Garden so I think today we're going to understand how this is done so GM rub um how you doing GM I I have a feeling

this is gonna be one of those podcasts where we ask like one or two questions but basically we're just listening to like Bo and Noah the entire time and you guys are really just like pioneering and so I'm just going to do just that like we were talking before this started we were like all right let's just hit the button and so now the button hit is hit feel free to just pick up where you guys left off if you want to do a little intro otherwise uh yeah we'll just we'll get right into it I I'm no proit I'm one of the co-founders and CEO of nkit and yeah glad to talk about interop ability among Ro clubs and chains in general today so I'll let Bo take over yeah hey guys I'm Bo co-founder at polymer Labs we work on

IBC and generally bringing various interchain or Cosmos Technologies to the ethereum ecosystem um you know love love talking about my favorite topic which is a mixture of IBC and now like L2 to L2 interoperability B and why why don't you start you gave a a chat um about this stuff and then Noah maybe we'll get into like more of the how and and kind of like these few different uh ways of going about it but Bo why don't you start kind of at the high level and we'll we'll get a little bit deeper at the high level do you want to talk about the the crashing the quy side or just like the high level of like the crashing messaging side well I guess you had given a chat um and I didn't get a chance to see it so I just want to make

sure that I'm fully caught up and as well as our our audience yeah yeah I I gave a talk at at modular I was just the Fring that I was using was like the evolution of the internet and the evolution of blockchain networking uh interoperability to me is like a sort of like a met meta networking thing where you're trying to connect these distinct distinct chains to each other the same way that on the internet you connect these distinct not so I kind of walk through the evolution internet from the early days where you had arpanet you have the evolution of all these like local area networks which kind of form clusters and devices that are densely connected and many different companies where they're working on different kinds of projects will uh adopt these land or local area Network

Technologies in order to connect their devices for their own ecosystems or their own ecosystem of devices and you're starting to see that in the theum ecosystem where you have a lot of these rollup Technologies and these rollup ecosystem which are also creating these densely connected rollup clusters which form these uh type of virtual lands or virtual local area networks I'm totally making up these definitions they're not real definitions and then I came up with like these terms of like virtual distance versus physical distance since uh if you have a density connected Network or a local area network you have may one to two degrees of physical separation and on a like connected virtual local area network you have one to two degrees of

uh virtual separation so either you're connected directly or you're connected via one intermediary and you start to see that with not just the role of ecosystems you start to see that with all the interop protocols themselves hyper L Zer all these other protocols require densely connected or one degree of virtual separation maybe the aars of the world the wormholes of the World require two degrees of virtual separation and then within the say the inter calls within each R ecosystem also have these requirements and to me IBC is the first protocol to enable a wide area network Behavior which is how the internet works stay where two nodes can be separated by these massive physical distances um maybe like many many degrees of separation and still be able to

communicate directly with one another uh so we did some work with multihop IC channels you can kind of have like any in degrees of virtual distance between them and I think in the long term we'll start to see like Contin to see these local clusters of devices but uh if you you know are at home today or you you know working out of an office you have some router uh in your office you have some local IP addresses and your router exposes a public IP for which uh you can access your network over these uh long long distances so I think we'll gradually get there we can kind of talk about how this evolution is happening and and and what things look like today awesome and no you have a few ways that you're describing I you can build on top

of BC to make it a little bit better um and kind of interchange some of these tools uh so maybe you can get into like how you're viewing this some of the options that are out there now and how you see best fit combining these different tools for the best experience yeah I think fundamentally the issue right now like with rollup interop in my mind is in general there's there's a lot of different solutions that are trying to go in a lot of different directions I don't necessarily think that getting everybody to define a canonical way of doing it where everybody has to use the same messaging everybody has to use the same way of verifying everybody has to use the same Bridge um makes a ton of sense I do think that fundamentally like because of the way crypto has developed from the ground up whether it's how

arbitrum started being very much so its own thing versus optimism coming out of kind of the research side of things as well I think fundamentally crypto has these communities in these cultures that sometimes mesh well together and sometimes don't and I think because of that we have kind of gotten to this landscape where there's many different ways of doing in our it so the current have different ways of doing it is it depends on how you want your tokens to move so you can do lock and mint which is what hyperlane and some other messaging standards as well do you can do mint and burn which you need either xcr c20s or layer zos ofs or you need a shared Bridge like the plon aglar to do mint and burn or you can do intense

based bridging such as Solutions like a cross so there's kind of three different ways where you can move assets and I believe IBC is compatible with all all three in that case as well um whether you're doing any other those different things on that spectrum and I think what really needs to be figured out is how to move data quicker rather than the ease of use on it because I do think speed kills over time and that's kind of where we focus on on nkit is we're focusing on the topic execution using Javelin which is a builder and we're able to allow for single block communication and single block bridging across different chains that into using this super Builder and they don't necessarily have to exist on the same sh sequencer or they need to pot into like a decentralized sequence

or any capacity I think fundamentally like rule up interop over time is only going to get more fragmented and because of that I do think that like solutions that can plug in to rollups without requiring large scale changes to the rups themselves will kind of win out over time because I think the coordination issue of getting everybody to adopt the same standard in ethereum is just so much harder than for example on Cosmos just because Cosmos all derived from the same Cosmos as DEC originally I think IBC was easier to get proliferation for it on that side but first time I think IBC is a great standard I just I I'm personally I personally think that messaging standards will be relatively different

in between different rups for a while at least I don't think any rup is going to work with each other to define a common meing standard but they will work with each other to figure out how we can move assets as quick as possible and that's kind of where we see ourselves fitting in stack is by providing atom execution regardless of messaging standard execution environment or sequence yeah that that makes a lot of sense I think if you look at all the RO of ecosystems they pretty much are building their own intern solution they're adopting no standard um very similar to how a lot of these large companies in like the 80s even into the 90s their own networking standards and like the way I like to think about IC in like kind of like an abstract sense here is perhaps within a

cluster of rollups you have a way of defining like your local IP address and you have a way of handling local device-to-device communication and then what IBC can do from my perspective is offer like a public IP address where this is much more commonly understood it's going be cross ecosystems so I think IBC has a lot a lot of value in Cross cluster communication um and maybe over time when like folks don't want to maintain their own networking CER they'll even adopt this like wide area networking protocol virtual wide area networking protocol for their local cluster communication as well um but that's going to take a lot of time I think it's going to happen overnight I think it took over 20 years or many decades for systems to gradually transition over and

even then like certain protocols like even Apple talk had their own networking standard that took into like the 2000s to be duplicated so uh there's going to be a lot of overlap in in these protocols and no no make a good point on Atomic composability to me I I find Atomic composability having a lot of value and in the rollup setting or just in in in a setting when you have applications that will benefit from close proximity and execution and then the mental model that I have for this and I I know I don't remember I think on a panel before I might I might have mentioned this but this concept of like Wall Street today and when I think about Wall Street today you have companies that pay for like proximity execution

proximity to where the value is being created like these exchanges so like I'm going to pay millions of dollars more than my competitor because I'm gonna I want my server blade to be in the same rack as your server Blade the closer I can get my rack to yours I can be like like a a hair length of a like a fraction of a second faster and and and be able to benefit from extracting more value or like um Malin would be a little more efficient than my competitor so I think you start to you see that anyways with like how people price blocks if there's a lot of liquidations happening the price of a block will be bit up um if there's nothing happening at all then the price of a block just comes down and to to noest point I think uh the the densest the densest areas of

economic activity will attract additional folks to to come in so to me it's like a snowball thing where like there's some like minimal minimal viable defi that needs to be uh within a certain block space and now that and now Atomic composability with that block space becomes that much more interesting and like until you get over like minimum value defi atomicos doesn't matter uh as much and then I think in the laun H of applications Atomic composability might not be as important for like things that are lower value where there's not as much value to to to extract like just like gaming or like social media or stuff that just doesn't benefit from from that yeah I think that you touched on a great point there where I do think you will see people wanting to like as

fast as possible within kind of this like minimal valuable defy ecosystem I do think like cross ecosystem Communications it's going to be interesting to see kind of how it plays off all these different clusters because at least something we've seen in kind of developing jamlin as well as like our other products at noit um is we've seen that like the most value for people that are within our ecosystem has been when we have other products that are similar to them or complimentary to them and they can work together and I think that's where you get a lot of the value I don't necessarily think that you're going to have like a defi chain communica uh gaming chain ifin a single block and that's not necessarily going to provide as much value but if I'm a gaming studio for example like Ubisoft that's GNA be extremely valuable for me

if I can communicate across all my different games very quickly because games particularly they just want to have a better ux because most their users aren't knowing that it's crypto but I believe Andy you've tweeted about this recently isn't it like 250 second M 250 millisecond in a r wasn't that what you've been twe about lately I've been catching up on that yeah I mean that's been my idea like so two things there and I'll kind of hit on that and then I want to kind of just come to a consensus on what we're thinking about the key priorities are for intra cluster kind of inup and then inter but like my vision is like okay if we can have this idea of of like sub 200 really sub 200 millisecond interrupt between um any

chain you know I think that's like good enough for now to at least give an experience such that I can have like you know a a universal account or a credible account as smart contract wallet that is accessible across all rollups app chains l2s doesn't really matter as long as they're basically evm rollup or zkm Etc and then eventually hopefully kind of like other execution environments as well but that kind of is a different conversation I think within the evm 200 millisecond kind of asynchronous interoperability on the back end and then on the front end this kind of like chain abstracted experiences like that would take us from where we are right now which is like ethereum is completely fragmented and isn't composable to like okay ethereum feels like ethereum in

early 2020 again so that's like my my like minimal viable like that's where we need to like push for but I understand like these kind of synchronous composability and these different solutions are even a step further than that but I also am kind of attributing those to be a little bit more uh in their earlier in their development cycle so definitely curious to to kind of understand if you guys align with that and then we can Circle back to uh kind of coming to consensus on on the best practices for these uh clusters yeah I can share some thoughts around like maybe taking a step back from the technical details from a product perspective like what does a user expect I think we're just strictly talking

about defy and value transfer 250 milliseconds is like like fast enough for for most people um but if you start talking about like other types of applications I know this is a little bit more forward looking these kind of applications don't exist as in in kind of today's world like imagine you're losing a game uh if you like need to like make a microtransaction you want to pay using a credit card to purchase something in the game maybe you're like okay waiting like a few hundred milliseconds but when you're committing game actions and your character is like doing certain things like you're doing like FPS 250 milliseconds between when you click a button and someone gets shot is probably not okay you're you're you're going to lose um that's that's a really bad ping so I I think to Noah's point about like

what can be what can be collocated what's interesting like a game Studio would want like maybe all of their gaming related things that require like faster interactions maybe to be automically composable with with each other and create like a a cluster and then outside of that cluster there may be like like a payments like say this maybe they have the user has like actual funds on a different chain uh maybe there's uh an nft Marketplace for the in-game items that can also be a separate thing and now you have this like ecosystem of an application that's also like spread over different uh parts of the spit over different block space that and that block space itself has different properties yeah I tend to agree on that I think that over time we're we're going to see this rise of like really

application specific and like use case specific like having a chain and internally what we we've kind of called it is like a bank chain or like a um a hub chain where users leave their money on it and exchanges on board those chains but if we think there's going to be hundreds and potentially thousands of chains in the future I don't think you're going to have binance or coinbase or any of these large um centralized entities on board every single one of these chains that's why I do believe that like having extremely fast composability between these different chains will allow people to kind of string together these different actions that previously you could only string together on a single singular chain so you can onboard quicker you build a better experience for example if you're building a consumer or F Tech app you want to potentially be able to get usdc

yield in your app via treasury bills but if you have your own app chain you might get siloed away and then a t rwba tbell product might not want to deploy to your own consumer app chain so being able to have an rwba chain where you're able to unlock T yields and have your user funds be able to get deposited in there while keeping them in your same user experience think it's really important I think this idea of like users will know every single chain I don't necessarily think users will every single chain I do think we'll see a rise of like utility first chains that developers know but users don't necessarily need to know because users don't know every single piece of software that they use in the dayto day like Zoom could use a completely different video processing

software under the hood and none of us would really notice um that much of a difference unless it was worse or better we' just notice well zoom's working isn't as intended I think this idea of everything is known in crypto to every single person every single part of the stack is something that's like slightly holding us back a bit because I think applications are the sum of their parts they aren't the parts themselves and I think that by enabling better applications that's where we're going to see the real like Leaps and Bounds thing in crypto because right now when you try and on board something that's not in crypto to like an app chain or even just onboard them to a chain that's not like ethereum or salana I think that user experience can really suffer and I think that making that user experience better

um by being able to like have these faster actions across chins I think will really help with like this maer future of actually like proliferating rather than I think I think the biggest failure of the ma ecosystem could be if it's only us for um on this live stream that actually all understand how it works I would view that as a failure of it so Bo no I think you made a good point like one Insight that I felt like you were hinting at is that um per apps it's computers that care about like lower lat lower latencies or latencies beneath a certain limit and humans cuz we we our CPU like our like processing units like

we can only process things so fast um I think there's some studies done around like what's the I guess like shortest like page low times for user to feel like it's instant um and I think that's like the hundreds of milliseconds um I I might be like totally off I I just trying to remember from from recollection but I think computers care about atomic composability Compu computers care about lower latencies um to use no as like gaming example U maybe like if you had a nonchain game and then let's say someone were to develop like a bing platform on the game for human bers you know we're observing the game you can say oh yeah we'll respond with certain uh you know latencies or human latencies but with if you have a trading algorithm that's trading on the outcome of the game if it can be colocated in

the same block space then these algorithms can be highly highly efficient because they can react to every single move and like with every single move in a game you know like the probabilities of win or lose change so now you can have this like hftt game B this is dystopian world you have this like hft like game betting system that's pretty much kind of like Wall Street like with all these like hft algorithms and um and and and that that's going to be something that folks would want to build and have collocated next to this like gaming block space yeah the like uh humans like read totally like anything below that is

fast enough for human users but what about algorithmic or computer users as Bo is alluding to like yeah maybe I go and I I I make a trade and and something feels instant but if the price moves like there's still discrepancy that that some computer is able to act in between that that time frame and so there's like user experience and then there's like application or computer experience and I think you know that's where composability gets like extra important and so and I know you had a followup on like yeah Rob I'm just trying to not Bridge anymore I mean to be honest with you I'm tired of switching networks you know that's like that's my goal here like you know I just want the basics but it's like it's like

if you can if you I think both of these Solutions the synchronous composability and the asynchronous can both remove the need for Bridges it's like yeah I can move to another app and it feels instant that I'm on that app but what if my bridge takes like an extra basis point or 10 basis points because the price changed while I was going from app to app that's the difference between like the 250 second millisecond latency versus the atomic composability yeah and so I think it's important to kind of like to kind of bring us to a foundation here based on what we've talked about with kind of some of these gaming examples this betting example and then also kind of how we're thinking about these clusters and so it seems like from inside of the cluster so intra cluster

inside of these R clusters what is most important is effectively maximum speed maximum uh atomicity same block transactions for like the chains or rollups Within These clusters for specialized use cases or just for example for like you know super chain and for elastic chain Etc just that ability to have like um all of these rollups to be able to synchronously uh share the same state at for every block and always kind of be on the same page and so perhaps there's a security tradeoff there or some sort of trade off where we can kind of dive into that if that's so but then for the intercluster so from between you know different clusters a gaming cluster a d51 super

chain arbitr orbit Etc what seems to be more important um than atomicity is the security of these messages and of these transfers such that perhaps we'd be willing to accept a little bit more latency to ensure that when I'm in you know when I'm in Kansas and I'm sending a message to Florida that my my USPS is getting there and I'm paying for a little extra um so is that kind of like the right way of thinking about this inoperability is that how you guys are thinking about it what did I miss and kind of um what can you add there yeah the way having kind of thinking about it is for like the longer hop thing is I think at a certain point um async will

be necessary I think synchronous um is really something that you can really leverage I think at first it'll be mostly within a single ecosystem but we have been exploring how to make it go beyond multiple ecos systems one of the things it's actually coming out um relatively shortly so I can give some Alpha to the listeners on the Pod today is starting now Javelin will support chains that aren't on seek so chains that have not opted into our shared sequencer will be able to leverage Javelin so they can be centralized sequencer decentral sequenc so they can be on another shared sequencer using base pre comps even L1 potentially could leverage it and we're diving into that and I think that synchronous Atomic synchronous compos buity will really

help drive down the cost of bridging I think that at a certain point ZK proofs become kind of the endgame of bridging but I always think that TK proofs will need some way of coordinating I think that if you can build a coordinating mechanism that works much faster than them currently and then later on leverages them I think that's kind the true end state for interrupt so I think stuff like the polygon agare plus a super Builder is really where I see the end state of interrupt being between most chains um not all chains necessarily but I think most chains I always think that there will exist kind of the solanos and the S and the Aptos of the world that are kind of out there doing their own thing but I also think that's one of the superpowers of crypto because you can kind of say we're going

to do our own thing and that doesn't necessarily mean that you end up with like a worse necessary product I think that like in traditional web 2 if you say I'm not going to use any SAS providers to run my program I'm not going to use it ABS or Azure or anything you might cap yourself at like how much you can do without spending an absorbing amount of money and in crypto the nice thing is you you really don't have those limits like you can say I'm going to go off on a whim and do this and you can do it which is a really powerful thing and really helps the censorship resistance of the entire industry I I'll add on to this uh topic since we're on the topic of kind of safety and validity in intra and Inter cluster communication uh I think ZK proofs are valuable for

validity unfortunately with like even intra and both intra and intercluster communication on ethereum we also have this issue of real resistance um and what are the guarantees around that since safety is just one dimension of sorry validity is just one dimension of safety I think the formal def definition of safety involves a number of things like data availability is included reor resistance is included uh there's also like sensorship resistance like liveness like other different properties I just want to talk about re resistance and validity here so my uh perspective on or how I think about what you get with like a single like inter intercluster over single

sequencer ecosystem is that you have sequencer level guarantees or that ecosystem will guarantee you safety beneath the like da threshold so let's say you post rollups in this ecosystem post da every 2 minutes so sub two minutes the safety guarantees of the system are guaranteed by the sequencer and however that is implemented and however that is designed uh and then when you go across uh ecosystems or across clusters you can go faster than 2 minutes but you're making assumption on both of these sequencer guarantees being are playing nice with each other essentially and uh Andy to your point on you know are people willing to wait a little bit longer for better guarantees

going from cluster to Cluster um maybe they are if they're willing to wait like a little bit longer than two minutes then you can have ethereum level ordering guarantees like once DA has been posted these are much stronger than secreter level guarantees so there is like a this balance between like like what guarantees do you get at what level of latency and who is making those guarantees so there's like a clear distinction of how this all works but I think to no this point I think there's going to be like really people really going to favor that like really fast low latency or maybe an atomic composability intercluster orry intra cluster and then intercluster is going to be maybe they'll wait a little bit longer uh to have these like ethereum level guarantees um and then from like a maybe I'll just use like a reward example

here my uh analogy is like when you want to enter an ecosystem to use a particular application like let's say I want to trade on Robin Hood I have to transfer money to Robin Hood and generally that actually takes some time I don't know if anyone here on on the call uses Robin Hood but I I've transferred money to Robin Hood Robin Hood does not get my money right away it's not like I'm transfer in like 250 millisecs later like Robin Hood's like boom got your funds like like like let's trade I have to wait a few days sometimes they even have to like ask me for like extra information like it's it's it's a whole process but that is normal and I think it feels like in in in web end to your point on like being annoy that having to like transfer your

funds all over the place it's actually really shitty in the real world too if you want to go between like different different trading platforms I I want to go to tdmr trade to like this trading platform okay I have to like I'm going to pull my funds back from from Robin Hood okay I got to wait for AC a transfer it's like a few days I'd wait a few days um and then I have to like Bridge like Bridge into like you know TDM trade or this other trading app and the experiences is just as baded and I think we would have the same complaints about the world world if like Robin Hood TD Merit trade every single one of these like like uh trading um applications or platforms were offering these like huge incentives to like hey trade trade on an app you're going to get $10,000 free so

like if if all them are offering these these these incentives and everyone's kind of like like hustling to like move their like like funds around to all these different apps to take advantage of this people were complaining about the same thing like can we can we like have uh like trading app abstraction in the real world across all these different trading apps yeah sometimes you forget that uh we're we're like we're competing amongst the the the minds within the space who would just want extreme speed extreme security just everything to to the extent that like but then you you compare and you're like okay yeah like in order to go from TD I have to go to my bank then I have to go out and that whole process takes a little while um you know as you're alluding to and no I think that you had

something to add there as [Music] well CTE is your home for app specific rollups at the Forefront of developing new tools and platforms that enhance the capabilities of blockchain Technology carty enables applications that are flexible and built to scale build on a full Linux environment and use traditional software libraries but in blockchain development go to cars.io that's cy. to learn more about building your own app chain today yeah I think fundamentally crypto like ends up the crypto industry itself

ends up drifting towards solutions that kind of match how crypto attention is done as well like in times of big hacks people drift towards Security in times of slow bridging time and people needing to go to different chains give different rewards whether that's like an incentivized Tes or stuff like that you see people drive towards it very quickly and I do think that's how crypto it's just how crypto as an industry is kind of done and I think it's a part of it's because it's such an open industry where like in most other Industries it's like if was an issue um like let's say there was an issue with Robin Hood like a button BR um on the Robin Hood back end you'd see some posts on the Robin Hood subreddit Robin Hood itself would know that there's an internal issue and they

might have a a slack ticket that gets opened up like the rest of the industry doesn't call out Robin Hood and say oh my gosh they had a broken button versus in crypto everything's so quick to react on everything I think that because of that you kind of end up having crypto I think has this large swinging effect where opinion can shift very quickly based off of if users aren't happy with their current option that's why they really drive towards like like the bridges like the bridges right now are quite slow and users are heavily driving towards we need faster interupt all it and I do think having that like last product Market feedback um is great in crypto but I also think sometimes you end up with like too much so on the extreme so I think that like I think better guarantees around interop ability

across chains is crucial but I do think that like we also won't get to realtime ZK for a long time I think crypto has kind of said we're going to get to realtime ZK relatively soon I don't think the financial incentives already to get there as soon because I we forget that there's only so many applications that can leverage these things and there's only so many companies that will pay the tens of millions of dollars per year to integrate things like a real-time ZKA special ship for the computer so because of that you don't end up having I think crypto sometimes like says we want something really far in the distance and it almost we get too excited and then the stuff that's here today that was a leap in Bound from the prior experience kind of gets lost

because we're looking for the stars and we forget what's here so so maybe this is a a certain perspective coming from like m 2 that as an engineer I love it so the fact that people celebrate these like technical things these like technical achievements is amazing like super interesting like I I love ZK I love all these Technologies but then I think about it from like a like a product and user perspective I'm like actually users don't really care about like like all this stuff they they just have certain things that they want to do at certain costs at certain latencies you know they have certain expectations around like using a product and to no point like not every company like wanted to pay for this sort of thing like sure like ZK may make sense for you know if you're doing like really secure def things there's

like a ton of value being transacted but maybe someone just wants to build like a poker game between like him and his friends and sure like maybe you know some of the noes would be malicious May cheap his friends out of like $20 bet but like it's not creating enough value for you to be able to like you know for every state transition have to pay like a few dollars to generate a ZK proof or even like tens of cents to generate a ZK proof just not worth it for that for that particular application yeah that's like I feel like just this these this rollup cluster idea was is like a super amazing like uh idea and it's it's an incentive structure thing that these that these rollups have created for themselves just because that's what they've what they've wanted to do but it's like an engineering feet to be able

to have uh this really good interoperability in ux Within These clusters but like now like people like me are just going on Twitter and I'm just like on on these these clusters for saying wal Gardens and like oh calling out these L2 for not decentralizing their sequencers and you know I'd like to get into that uh as well but first just kind of like understanding the understanding the kind of state of these clusters now like and differentiating them like would you guys consider the idea of of Ager as similar to like super chain and um elastic chain and um uh you know like like arbit from orbit or these like two separate kind of

categories in your your mind as far as what they are so I can probably touch on that better we've worked close to polygon on this side of things polygon plans to have agare support multiple different ecosystems multiple different chains multiple different BMS recently they came out of an integration movement part of that we announced integration of them a couple months back um so I view aglar as being much more so of like an open open situation compared to some of the other W Gardens I don't necessarily think that all the wall Gardens are going to be the worst thing ever I do think wild Gardens in general need to be avoided but I I I understand where they're coming from and I think fundamentally it comes down to an incentive structure that works the way that we've thought about it is we went

well actually I can tell the story of how we built no so in May of 2023 we started developing out the first version of noit we wanted to decentralize the sequence and what we realized once we were able to build on a decentralized sequencer was it's fundamentally an incentive issue solutions that just say like Well's let's decentralize it and we're all going to share it and we're all going to sing kumaya they just don't work and it's like a harsh reality of it and the reason why is because rollups need to earn money from decentralizing their sequence and that cannot be through some like um shortterm benefit it cannot be like okay we'll pay you a million dollars in a grant um and we'll decentralize your sequencer which I have

heard of some teams doing where they're giving these large grants um and saying like well you're going to join our Network or our Network and it we'll just figure it out over time and I think that the way that we thought about it internally is you need to have a revenue split so whenever there is a crossr transaction we do our transactions in the shared top of block and our bundles the fees are distributed Pro to the rups involved in the Bumble so if we for example have polymer and another rollup and polymer is in 70% of all of the bundles and there's only one other rup that's also on the network let's say there's two rops on the network then polymer deserves a greater split of this of the fees because polymer is in 70% of them and the other rollup and let's say

there's three rollups in polymers in 70% of them rollup two is in let's say 20% and rollup three um is in the rest so like some breakdown of it we view it as you need to allow for cross Revenue redistribution on these Atomic bonders across chains and we believe that's how you get people like adopt these standards I don't necessarily think teams will decentralize and go into a shared sequenc or structure unless it meaningfully impacts them in a way where they make more money and that's why we've seen majority of our customers came for Javelin and want that single block communication and want that Atomic executioner cost chance I think that that's I think that's one of the major issues in crypto right now is people think it's more so about like not the financial aspects of it and I think

Financial aspects is how a lot of these different like webs of networks make sense rep a longterm a short-term Grant doesn't necessarily help you if longterm you'll make less money of the protocol and we've really prioritized making rollups more money when they join our Network rather than if they did it themselves yeah that's how we think about it yeah just that like it's funny because we kind of framed this initially as like superb Builders versus IBC like those are two fundamentally different things like trying to solve two different problems like super Builders go on top of IBC and make it better like these are two complimentary Solutions IBC is great for async interop superb Builders come in and they can synchronize a lot of that and make it

Atomic and so I think at a fundamental level like both of these are great solutions for their respective problem it's uh I like like from ibc's perspective it's just as no mention just a transfer layer protocol just messaging protocol so it's the advantages there be like you know as these traditional Works started adopting IBC TCP IP not iced started in TCP IP for even local area communication it started to kind of lead them to be okay now we can it's easier for the developers they don't have to develop over all these different interfaces the way you communicate locally which has which is highly performant and the way you communicate remotely which is somewhat less performant because the you know things

are just further away um now becomes similar and um maybe I'll take a step back back into the web free space um I'm not as familiar with some of these other Tex STS but I am Sim familiar with OP stack I'll say that the way they treat state is a little bit different um my mental model for like Ager and how maybe like a super Builder approach how state is treated there is that you're building a single giant block and you're trying to treat multiple rollups as part of the same state machine maybe they're all just State shards in the same state machine um and then the implementation details can vary but you're trying to treat them all the same trying to this Atomic thing and then for opack I at least my

understanding so far following along the development of their internal interet protocol is that they're not trying to treat uh all the super CHS one giant State machine they're just like acknowledging that okay there are multiple State machines they'll operate differently but will try to facilitate lower latency communication uh across them so it's just kind of like the low latency async versus Atomic uh comparison and like whether you not new ZK proofs or not is independent of these prods yeah so Rob was kind of mentioning how you guys could collaborate um I don't know if how much you guys have chatted about that but like technically speaking uh and how would how would

superb Builder collaborate with polymer's IBC or you know if you want to speak more broadly about how kind of these Atomic composability networks would interoperate with IBC and what what you guys are building at polymer show my perspective and I'm sure Noah has a lot of thoughts on this as well uh it's a little different when you have a hub and spoke model and a point-to-point model when you have a pointto point model most of them just deploy a set of smart contracts that's going to be a little bit lower lift to integrate I know know you guys have already done the integration with Hyper Lane um and you guys just kind of like run the relay run the smart contracts deploy the smart contracts configure them and so on with the Hub and smoke model it means that the this is why we design polymer

this way that it means that the Hub needs to be able to run in different environments because every single rollup cluster will have some sequencing a sequencer or a sequencer design for their particular cluster and depending on what environment you're running in if you use a hub and spoke model like polymer our Hub cannot be a single Hub it has to be multiple hubs and each hub has to be Deployable over different shared sequences or centralized sequences depending on the setup of particular environment and it it needs to be like if it needs to be plugged in t there it should be ZK provable at some point we need to be able to generate ZK proofs these all these like different kind of

requirements so it is much higher lift to deploy a hub and spoke model into these systems um but I think we're one of the only teams that are building Hub and spoke in solutions that are like IBC aside that are really thinking about this I think a lot of existing Hub and spoke Solutions are are essentially Sovereign they have some layer one blockchain it's like I have a hub it's a layer one blockchain if you have consensus you can't also be a sequence rollup like it's the two things don't don't mix and match yeah I tend to agree on that and the with it we could kind of plug in um in the past we explored around IBC is IBC you can use a multi-sig module so we can to have Javelin itself plug in and be one of the multi signers between chains enabled via

IBC so that's one of the ways we can plug in it is a larger left essentially to plug in to kind these Hub and spoke models but we built out our solution in a way where javelin's quite scalable and we're not really constrained to one machine with regards to Jain so it makes it so it's a much easier experience for us to really scale it horizontally on that front so I think that over time we plan on supporting multiple different messaging layers right now we're really focusing on being able to support more Stacks um supporting more different environments now that we're expanding Javelin out to be able to support different decentralized sequencers centralized sequencers Bas comps and that regard as well so I think right now we're really just focusing on making the best possible user experience for this

Tech um we typically like to go very much so vertical rather than horizontal we we'd rather support one one or one or a few rollup stacks and support them really well then support every single stock but not do as deep of Integrations we've just found that for our customers thus far they really prefer it when we have all of the features that we set out to have in every single one of our Integrations rather than providing some of the features but not many not all of them um in providing kind the surface level integration so we've just found like the more Deep dive Integrations typically have been better receiv so that's kind think about it no saying that Atomic inclusion is not enough these people want Atomic composability yeah yeah all of our customers are using

us for Atomic execution and I I personally believe shared sequencers failed I think that lazy shared sequencers um really don't make any sense I think atom execution is what matters um and I actually have some memes on this but Searchers can price cross chain risk your mother can't so I don't think Atomic inclusion gets used by people that aren't like a medbot or Searcher or a solver or something like that just doesn't make any sense because I I I really don't think like the long tale of it is my mom burning usdc on base and then hoping praying that somebody mint usdc for her on the destination chain because that would be

terrible I think we end up with a lot of the same issues with the traditional Financial world if that's the case so I think think Atomic execution is the is the sole thing that we need to get to Atomic synchronist composability and I I think guarantees around inclusion are much easier to get Than People initially thought and we've just really learned that by building out a shared sequencer I think there's a I think it's a hard thing to learn unless you do the thing hit your head doing the thing and have to look at it again I think that that's one of the great things that we've kind of learned at noit is that Atomic execution is really what's hard to get Atomic inclusion isn't as difficult so yeah that's kind of been our Insight coent is solving the longterm data availability problem where a lot of the

data availability providers today in the modu space when you post data to them they actually only hold that data for for about 12 to 16 days and then it's basically they can go into archive note or often times it's just kind of put away into the abyss and what coen's trying to do is build this long-term data availability solution which can plug into some this modular data availability providers and provide this longer term Perpetual data solution so what we try to think of this as is when you post transaction data onto Celestial or Vil or any of these mod da Solutions or even ethereum is able to provide a long-term data storage and data availability platform for that data

after the time period blockchains are Billboards not databases and you know we need these providers in the space to be able to kind of hold on to the longterm data that is provided from these providers and these blockchain State this is where coent comes in they just had a nice Rebrand check them out on Twitter as well as the website thank you for being here hope you enjoy the rest of the show mantle is a layer 2 blockchain with one of the biggest onchain treasuries in all of crypto and they have a 100 day campaign called metamorphosis live now what this is is it is a campaign for their liquid staked eth token M eth as well as their upcoming liquid reaking token CME where you can utilize this token on

the mantle Network in a variety of different apps some of your favorite reaking protocols and defi protocols and earn yields in the form of powder which which will eventually be turning into a governance token Mantle's been around for a while we like them because they use their own form of Ian die uh they have a good ecosystem and they're growing so check out this 100 day metamorphosis campaign and hope you guys enjoy the rest of this video I think it goes back to the like who are we building for who cares about this level of latency and I think the question folks should be asking is like is this the ux that like you know your mother wants it's more of like is this the ux that some like mbot or some trading algorithm would want to use and I think

uh to no point I think the answer is is yes like you care about Atomic execution like you don't care about Atomic inclusion as a bot you want to know that something definitely happened so that you can take an action so that you you're guarantee that your algorithm is is making profit um crypto exact it's all AI agents down like in the future it's going to be all AI agents using these these calls and yeah he just going to love atomic execution oh 100% And I think that we're going to see the same thing happen that public markets had where most volume in publicly traded markets are hfg Bots they're not retail Traders and I do think we going to see the same thing in crypto um which I'm excited for it no

that that heart on your shirt looks like a an AI agent he looks a little drunk though it may be one I I got this um sure for for Christmas so I don't know it may be an AI agent I might just be a sleeper agent before we wrap up can you guys prove that you're not AI agent no comment can I uh I I need to develop a ZK VM first and it needs to be able to verify the circuit in my body so I can send a validity proof of humanity in oh my goodness guys thank you for coming on that was quite insightful uh really enjoyed your perspectives uh learned quite a bit have some some stuff to chew on after this so going to go on some

rants on Twitter later thank you yeah thanks guys