DeFi By Design EP#115: Permissionless Interop For The Modular Future with Hyperlane

Transcript

welcome back to the rollup High Roller Community as always nothing on the rollup is to be ConEd as Financial advice we are just an Educators in the Divi space toss us a quick like give us a share a repost send it to a friend who may want to watch this and finally give us a sub or a follow it means the world to help support credibly neutral educational content in D5 we're here to educate Empower and enrich you and let's jump in welcome to D5 by Design who we talk all things blockchain cryptocurrency while striving to educate power and enrich everyone's feeling bullish episode 115 of the D5 by Design podcast we're here Jay Cole yes we got the

rapper himself here today with us the crypto version Johan from hyperlane we got Rob Rob just came off the hudy penguin party you feel H you rob let's come on on brother how you doing it dude I feel like I am my penu my penu is me the fun time oh yeah I do one of my favorite uh crypto references it just that's a good sign if someone's in the note they know about being your penguin guys will get on Twitter's face and then just stream I am my penguin like penguin is me bri can look straight imagine having a job doing anything else right like I used to work

I used to work Wall Street Bank like can't go back yeah I like you're in a Wall Street Bank actually looks like you're in a nice Loft how do you uh yeah yeah nice I've always Dre the time L some reason I know why what's you were yeah go ahead no I was curious this that we got J Cole in the building you know not the rapper we got a we had a more of a a finan savy John here from hyper used to work with uh no regrets what's a what's a what's a crazy story that you got for us that's not confidential to to really show the audience what you guys have made up here all right well this one I don't think he'd get upset if I told um but so

you know I worked with Mike for three years still you know still talk to him uh fairly frequently he's a excellent excellent guy and just it was a blast working with him and there was a period of time where there's two folks two younger dudes who uh developed quite a reputation for themselves in the industry as investors Traders and they start thinking about uh creating a fund and of course right like Mike likes the back um you know young folks who are promising spend some time talking to them then at a certain point they're in New York is like come let's do dinner at my house turns out one of these guys uh

is uh you know is a Combat Sports kind of guy does a little bit of MMA you know he trains and some point through that conversation I don't know how they get to it but this could certainly happen with Mike because Mike loves loves loves wrestling like it's a massive passion of his uh you know he's super involved in like the college wrestling organization and at some point somehow it gets to like one of these young guys tells Mike no grats like you're old I could take you down and you know it's like a little Rado he's talking it up what he doesn't know Mike's a maniac Mike has a mat in his house you can't challenge Mike to

you know a wrestling match and not expect to actually wrestle so he's like [ __ ] I got a mat right here we're going to go right now they and I get a call um so all I know is this dinner is happening we get a call late at night from an excited Mike and he's like so I had the guys here for din I'm like yeah how I go and I'm like thinking you know he's excited because he's gonna tell me like oh is it well one of these like tells me the guy basically challenged me and he goes I whooped his ass in like 30 seconds he had no chance he didn't know what he was getting into and uh that's one of still one of my favorite mic stories it's absolutely true that's the reason why

I'm like not being more giving more details because like the person is absolutely wonderful guy super sharp and Brilliant it had no idea that that's how it was going to go down right like uh but that's my favorite mic story that I could share there the G with digital culture cor no wonder they're as to Kings over there and we got Jake v h that he come on yeah that was a great it was a great time I had that I had fun being there was this before he got the Luna tattoo and he up full of energy like let's go I think it was before I think it well it could

have been after but yeah I don't actually remember I mean that's an eagy one to be like you know that was a dog of mine I mean it's cool tattoo yeah um but yeah so you know I've been in crypto for like six years now um I got in I got involved really because a buddy from school you know at the time I'm working in Morgan Stanley I'm like a bond Trader and a buddy from school at a birthday party asks me it's like yo John it's like this is the spring of 2016 like I'm gonna put all my money into ethereum what do you think I'm like I don't know dude I know it's a good idea I don't know it's a bad idea I just don't I just don't know it's like you want to look into it m me and I'm

like sure and we started looking into it like we go back to is out and we end up spending like the next two days just going super deep into this and I'm like this shit's so [ __ ] cool I go back to the trading desk Monday morning I'm like we ain't gonna have a job we're gonna like [ __ ] we're all [ __ ] like you don't understand um and just gradually got more and more and more into crypto and they all thought I was nuts uh and then like late 17 early 18 I got my first job it was a fund called passport Capital was a big macro fund uh that was getting into crypto it was their first hire for that uh had a good run there and um give or take two years later joined Galaxy and eventually started um

co-leading the investment business there with another guy uh Michael Jordan was a good friend and just one of the sharpest people around it's genuinely his name you got J Cole and Michael Jordan yeah he's every been impressive is any of the three other famous Michael Jordans uh he started he recently started his own firm called DBA with John charbono those guys are definitely worth really that guy Subs amazing yeah yeah yeah so he replaced me you know one John with another John substantially better writer and now you're read hyperlane building the modular then we started hyperlane yeah you know the interupt specifically for the modular ecosystem you know we work a lot with

Celestia we try to have a lot of fun uh I'm here to tell you all about it hell yeah well we are some modular Maxis ourselves so um I was giving trouble because I can't Bridge my uh my my Kia back so what a great place to start um this is a d EDM Fanboys are dying to get their end on some of these precious Chia tokens luckily for us we got the hyperlane giving us some accessibility via Trader Joe and Mana Trader Jo arbit and man network better better uh liquidity for sure Trader Joan arbitary um and yeah I would

love to understand uh first from like a security perspective what the and for our community what the premise of that tia. n token is the the rap Tia token from hyper on Manta or or R kind of how is that secured and then yeah what are the plans to open up a two-way Bridge yeah of course so uh actually it is just a it is two-way you just can't do it in one click yet so if you want to go back so you can go one click from Celestia directly to mana arbitrum and the same will be true for any new chains that uh walk to get Tia and get uh you know deploy hyperland and get

added to this but the way back now is still two steps unfortunately because it's slightly separate work stream uh I just haven't had the bandwidth to get to it but if you want to get back to Celestia proper you would go from uh either Mana or arbitra to new first and then IBC transfer over back to Celestia right now the the app interface basically abstracts away the need to go from Celestia to Neutron in the first place it kind of lumps it all into one action for you uh unbundling that on the way back we still haven't gotten to that uh will change sometime in the next few weeks but not there yet to your other

question perhaps a much more important one is like how does sh work like what's the security premise here what I why am I not being rug why are you not being exactly right a very important question so unlike other uh interoperability providers hyperlane uh has a very distinct uh security setup so generally speaking speaking for the most I'd say this is probably true to pretty much every major interrup protocol that I'm aware of uh security is tightly coupled with the mechanism of like delivering the messages so whatever is the process by which like a developer

uses this protocol to just transfer something that and how that protocol secures the messages is very tightly coupled um and what that means is like there's usually some enshrined uh security model you know so like uh in the case of axlr you have just a super robust validator set in the case of Wormhole similarly you have like the Guardians some of the most premier you know um organizations in crypto just folks that have a lot to lose and obviously are not like in any rush to uh burn their business by like committing you know worb hole fraud hi hybr Lan's different it has a

modular security architecture and it basically decouples the product uh connecting chains the interface for connecting these chains from the ways of securing that connection specifically in hyperlane like we care about securing the outbound communication so like you're sending something from you know arbitrum to anywhere else we care about securing the message going from arbitrum we do that like that needs to happen once once that message is secured anyone can leverage that to receive it somewhere else so think of it as like something is being broadcasted from the chain like you put some security around it we'll get in a second into like the different forms that that security can

take and now anywhere where you have a hyperlane mailbox you can receive that communication you just have to be comfortable with like whatever was securing it uh so it's a really nice property it means that like once we start uh once your chain is supported and someone is running some type of security for that other chains can connect to your chain without ever having to even talk to you like that message is being broadcast and they can just pick it up anywhere else so specifically for the Tia bridge right now the Tia bridg is the messages come out of neutron and uh to secure those we've partnered with the neutron team they've

been a very active participant in building that Nexus Bridge and they've helped Galvanize uh a nice subset of the broader Neutron validators to secure those messages so in what we want to get it to is we want to get it to as close as possible to the full uh validator set on Neutron and then it basically looks like the same security guarantees as just using Neutron outright we're going to do the same with osmosis uh when hyperlane Launches on osmosis probably sometime in the next four weeks because again security isn't prescriptive we can basically uh leverage all different forms of security

and so what we like to start with is for chains that have their own validator set it's so nice to be able to get that same set uh because then as a user there's no new trust uh assumption that's added you're like when you use a chain you are explicitly trusting that that validator set is you know for a lack of better term kosher uh and so if you could bring that same set of guarantees for um interoperability we really like that but of course you're not going to be always able to do that right like sometimes you know we're going to be sending messages out of AUM and like we're not going to be able to get every single ethereum validator right there's so many um in the current construction of

ethereum that that is kind of like that's an untenable ask so what do you do so similarly you can have these validator subsets we're also working on activating staking and so it won't just be reputation based uh so you'll have these validators putting dollars on the line what'll be nice about this is that if you're a chain that wants to use hyperlane like or even if you're just an app that wants to use hyperlane between like well trodden chains you could incorporate your token into the security of the messages so you could say well uh these validators need to stay my asset uh maybe you want to bring in another asset like you want them to stay uh Lido's version of e you could do that uh

Cosmos chains could leverage their own assets so there's a lot of cool things that you can do with it really what this security architecture lets us do is it lets us bring in any mode uh of like State at the station or state proof and leverage that to secure to secure things and because of the more open architecture that we brought the hyperlane like we've even we're even seeing other teams not the core team bring uh Brak security modules to hyperlane so electron which is an awesome team that's built a ZK light client they wrap that ZK light client as a hyperlane security module and so where that ZK light client is operating you can now use that instead of any like validator TI um

there's a few other ZK related things where this is happening as well so we're very excited about that um but the truth is ZK is still like not not something that would fit most people's needs because the proving time is still a bit long like no one wants to wait for especially small transfers you know no one wants to wait like 15 25 30 minutes if they don't have to and so the way we think about this is with hyperland you can actually uh dynamically R messages based on the contents of the message so maybe if transac CS are of really meaningful size then you can move to that slower but safer ZK option and for smaller stuff

just hey just give it to me the fastest right what is going to get me here in 30 seconds oh it's this validator option let's go with that so that's high level on like how this is secured and also gives you a better picture of how anything can be secured uh for any hyper ring connection Goa um from so from the from my perspective it's it's almost like we you just explained more or less um in an aggregator of of options uh yes provides so kind of leading leading into that um I think that like uh aggregation of trust is not necessarily always the best case because even if you have two

of three or three of four you know you kind of have problems uh you know in some cases where you know you can choose three best Bridges but but also um you know those if some of them or if one of them is you know is a problem or there's some sort of uh I guess a collaborative effort to to uh you know to act against uh you know this the best interest of who's using it you know you kind of have some some problems there so you guys tackle this by giving users optionality from what I understand as well as uh are also putting assets at stake to uh allow uh you know protocols to also uh incentiv or or to incentivize prot not to you know have some sort of negative outcome

because they've as at stake I guess what what I'm curious about is yeah like this agation how do you guys tackle the uh the objectivity of which which Bridges you choose how do you kind of incentivize them and the users get to choose but you know do you guys pick a subset of what you guys think is the best Bridge options or yeah so you know generally we leave it up so hyper we built it to really be the uh we wanted to create something that would be a permissionless option for interoperability you just like what is going to be the permissionless layer that anyone can use to connect any chain and so primarily like hyperline is actually used by other people Nexus is the first time where Us in conjunction

with uh the neutron team like we had a goal we wanted to get Tia into uh into new rollups that are using Celestia and so that's one where like we uh you know together with them operate um and so there we're using like I mentioned right the subset of neutron validators there's no real Bridge aggregation there but as a framework hyperlane lets you do the bridge aggregation so one area where this is happening is uh I can't mention too much about like the specific product project but uh it's Neutron related and um they're going to be aggregating a

hyperlane validator set a uh a Axl and Wormhole and so uh for those messages to come through for that app what it means is like you're going to need to get the same information from all three if that doesn't come through the messages won't get processed and so um and then similarly like they're not using this project is using um hyperlane as the main aggregation framework meaning like they're you know when they're contract interfaces with something it interfaces all with hyperlane contracts and uh then routes using hyperlane routes to the other Bridges uh similar thing was built

also by a but a used like a little bit more of their own stuff where uh for uh for like governance now for sorry for interchain governance a has a proposal of uh or sorry like the work needs to be implemented hasn't been done yet they have uh aggregated hyperlane layer zero and chain link um but unlike the other project they built that aggregation layer themselves uh because you know it's a bigger project they wanted to SN their teeth into it and so it's not to tell you that like aggregation is used everywhere where hyperlane is but

everywhere hyperlane exists the option for aggregation exists and the the thing that you kind of started off the podcast with is like essentially the when we're using these chains we're trusting that these chains are Kosher happy Hanukkah by the way save D so so when we when a project implements hyperlane uh is this like oh better would have asked this question is when hyperlane is aggregating these messaging networks and bridges uh are you essentially aggregating like the the validators themselves from these networks and that is kind of like where

the messages are pinged across because I'm I'm I guess like you mentioned the the validators this is why it's important for validator sets to be decentralized so how do you think about which validator sets to aggregate as part of hyperlane right that's a great question so generally we try to do like less less thinking about that um you know and first what we did was like build a thing build hyber line in such a way where like it could integrate with an agregate any of these but then you actually need to go ahead and do some

work so that the developer that wants to use the aggregation feature is like they don't need to you know do so much by themselves and and so there we did exercise some thinking and so the first two that we've worked to like have that aggregation feature with were AEL and Wormhole simply because those sets while you know yeah maybe some people might not think that they're perfect they're they're pretty good right like both in terms of chain coverage and the uh you know the who is actually in the set right you know like we know everyone who is in the Wormhole Guardians those are very reputable just for the people

who run like really strong organizations that set is as good as it gets you know the XLR set is an expansion on the Wormhole set it doesn't have complete overlap but it has a lot of the same characters it's larger generally really strong um and so that's how we thought about like well you have to decide what to do first we decided to do those two first for those reasons uh and just like the the robustness of those sets but again like you know one thing is don't think of hyper as like simply you know think of it as a framework for interoperability not so much as like the fact that it can aggregate like is very appealing to to

some users but uh you can also use it kind of Standalone where you as the chain you can operate it yourself you don't need to depend on anyone else and so like if you're you know an upand CER just a type of chain that like can't get the you know if you're the type of team that's still small like you can't get the time of day you're not hyped yet and so no one will the other people won't take you seriously like hyper Lane's for you you could just do everything yourself with it you could leverage your own validator set any type of security you want you could run your own relay and now you can get connected to any other chain that has hyperlane and again you do it in a completely self- Sovereign way like you ain't got to uh and that to me is like yes it's not the

most uh active place where epin is being used because simply there's just not as much usage for those guys by definition but it's to me the most exciting or the most the thing I enjoy the most is knowing that we give that option to anyone yeah I mean that's what I was getting at as well it's not just not an agregation layer doesn't Define what hyperlane is it's like if I'm a young hung hungry Dent was trying to get on this modular Vibe and I'm like oh man modular narrative I'm going Moon I gotta I got to spin up a modular roll up using gelato I got to use Celestia from My da ATM so I'm going to do something else I'm going to use CES V but I need a interrupt solution to what what

framework do I use bang I can just plug hyperlane as the framework for my modular rollup so what does that look like I am trying to build a modular rollup I'm choosing and again I'm trying to establish this narrative where you need to choose a interrup solution as part part of your modular stack it's not just da VM and settlement it's also interrupt and it's it's even also sequencing which is kind of being developed as well with shared sequencers decentral so what does that look like for I'm building a module rollup I'm chosing hyperlane as my interop of my step like yeah that's like right to me very very good question in fact U you know one of the earlier writings I I had um sometime this year was just about

this notion that like in our ability is a part of the stack it has become one uh and so to you know to talk about modular chains without talking about you know permissionless inoperability especially is kind of just it's a dead end um so I glad to hear you say that that to really answer your question let's talk quickly I'll give like a high level overview what does it even mean to have hyperline ey in your chain like what do you need what are the moving pieces and so to integrate hyperline there there are three big things that needs to that need to happen first uh there's what we call the mailbox and some assorted contracts the mailbox does exactly what it sounds

like it sends and receives stuff between Chains It's been it can be implemented pretty much in uh for anym it's obviously live for evm chains uh also for also exists uh as kmm also in sea level the salana VM and uh basically near done for move thanks to the amazing work by the movement Labs team so you want hyperlane uh you want an integration that means those contracts need to be deployed on that chain you want to uh be working with the second thing is we just spend all this time talking about security earlier in hyperland we have like we have what we call isms or interchange security modules and these modules can

take the form of anything the simplest one to get set up for like a new chain integration is an authority based one where you say hey here is a validator that looks at the mailbox we have open sourced all that software make it U very easily and accessible all it is truly is it's think of it like a bot that looks at uh the mailbox and then sees oh Indie is trying to send something to Robbie's chain okay looks everything checks out here if everything checks out uh the bot signs off on that uh you can have as many validators as you'd like uh each one needs to run the same butt the nice

thing is they don't even need to talk to each other and so when these signatures are being used there's not any like interactive element rather it's just each validator on its own looks signs and then later on all of those signatures are taken to the destination by the third piece uh of hyperlane integration that's the relayer also an offchain box also something that we've opened source and uh anyone can run for any chain that relayer basically needs to be able to read the mailboxes uh that it's connecting between so it sees up on and's chain I got Andy we send in a message it looks by based on the looks of this I can tell there need to be

three signatures I see that there are these three signatures here all exist from these three uh different validators looks good to me now I'm going to try and process this message meaning I'm taking it to the mailbox on Robbie's chain and I initiate the process function this is where now we're going to check against those signatures see that they're valid signatures uh this is also where if you'd had any other logic like oh uh if the messages of this size do this if it's that size do that if you want it to like you know have a certain blacklist for certain addresses this is all where all would come in and we passed through all that logic if

everything looks good we arrive at our destination which would usually be either a eoa or a contract on the you know on robis chain and so we need a mailbox we need at least one security module you can use multiple ones and we need a relayer uh now to deploy all this we've created a c so like just a nice command line interface that helps you do all of that um cool enough is that a company called curtosis which is like a cloud management uh company where the announcement will be coming out soon but

it's already live they've actually built um you know like a hosted an environment for hyperlink where you could like you could run all this stuff uh on their infra it's like two three minutes type of thing uh now do you really need to be able to do it in two three minutes I don't know but it's it's a really really really nice tool highly recommend uh checking it out if you're going to be experimenting with hyperlane uh they're they've done tremendous tremendous work they've made it so smooth and so that's really what it means like you need those you need to deploy those contracts you need to operate uh at least one security mechanism and uh you need to run the relayer now more

and more like the core team here at at at our company Abacus Abacus works is happy to provide those second Services right of like deployment of um helping with validation helping with relaying but you don't need to you could always do it yourself even if we were to help in the beginning uh we could very easily transition to a mode where like again you retain your sovereignty you do it yourself so that's what it means to like integrated and then uh once that's done you can have nice experience like the the Nexus bridge that you use yeah and so that kind of from a from a main chain to a

rollup uh you had these things called canonical Bridges where it's you know let's say for example in an arbit it's the uh it's the op it's the optimistic rollup Bridge with a s day froad proof it's the bridge that arbitr made themselves um basically that that is like the the the bridge with the least amount of trust assumptions implementing hyperline into your modular rollup I'm sort of thinking about it in a similar mental model where you kind of natively inherit a interrupt layer that is similar to a canonical bridge in that you get the uh the level of security from hyperlane with the mailboxes in the ism um natively embedded into your into your stack uh on the rup that you're

creating and then that allows you uh to have that that kind of made it interrupt eded without needing to seek out you know other thirdparty breach Partners or you know the yeah I'll just Stu there yeah no no no that's a that's a really good point now what I'll tell you is yes that is uh Al true what you said what in practice what we've seen is especially with the VM Roll-Ups is folks still want to use uh something like the the canonical Bridge like a standard roll of bridge for I think you know in optimism they call it portal I forget what it's called um in orbit in um you know arbitrum stack but use that for the gas token and

then use something like hyperland for all the other assets primary reason is uh like very few people want to move around the gas token once it's on the Chain right like but stable coins uh you know other assets something like you might want to move between different rollups and like send to other places and so that's we we thought too that people would be like oh actually no I'll just use this for everything there are cases where it's being used for everything but increasingly we're seeing folks like ah well you know what let's for the gas token let's still have like that optimistic bridge that 7day uh exit game it's not an asset that people are

going to want to transfer generally so like fine um but that's what we're seeing in practice and uh once you get hyper on your rollup it's very easy then to like I mentioned before have connectivity with any other chain that has hyperline so it's um you don't need to then create like these you know bidirectional channels for each chain all you need to is just like make sure that that relay you're you're running uh is aware of the hyperlane deployments on the chains you you want to connect and you're Off to the Races so in that sense it's uh it's been pretty cool for people who are picking this up now yeah hence the the permissionless of

it um which is so important as there's hundreds if not thousands of rollups to be seen on the horizon it it is important for those to get connected so that they're not they're not liquidity and users in silos um and I'm curious a little bit more about like the solution itself like uh Andy brought up canonical Bridges and and kind of like the similarities and differences there another uh communication solution that we're seeing being used to communicate across blockchains is IBC like can you just kind of like quick you know like compare and contrast like what is what is the similarities how do you relate to build

the perspective of like what hyper is could you kind of like relate it it to IBC and and talk about where it's similar and where it's different for sure it is i' say like it's a lot more similar uh it's much more similar than not think of you know the way that uh I said it is I believe very uh very very much heavily inspired by IBC some of the people who you know helped us build hyperlane uh were there to build IBC right and most notably in Zaki uh so it's fair to say that like without IBC hyperline doesn't look the way that it looks today

uh it was rather that you know once you start with something when you already have the knowledge of like okay what did we learn from IBC what are where does it like where areas where it leaves something to be desired and where we felt that the most is it it provided like very good guarantees but also had some rigidity it uh clearly was we thought you know the most elegantly designed protocol how can we take that design and then just make it much much more extensible now to make it more extensible we wanted it to be able to kind of be agnostic to anym so that if someone took the time to convert

it and make it suitable for uh for a VM it should generally work out of the box we wanted something that could uh and IBC actually does this too it just uh it's a lot more work to enable but IBC can support many different security settings it's not you know it doesn't have to be the like client verification it could accept other modes but a lot of work together there so we wanted to make it such that adding a new security setting is considerably easier than it is now than like going out and creating like new IPC clients so for us it's something that is mostly done in contracts with some minor modification like at the agent level because ultimately right like the

relayer which is what takes your message it needs to know what it's reading uh so every time you add a new module like if it does something special with the message contents there needs to be some real modification but that is uh pretty light and so that's really where the big changes are it's like we try to make something that's more extensible and less rigid and easier to extend uh even in areas where IBC C already extend today um beyond that very very similar protocols um and that's the primary reason why like we are not in any way pushing for like two IBC chains to use hyperlane in you know like it's you'll

even notice right like in the bridge that you use in Nexus like we're running IBC between uh you know on the first leg of that transfer we're just going on over IBC yeah yeah makes sense I mean it again it's just it's the it's the notion of C of um going with what is best and and kind of the the most reputable options kind of Switching gears a bit um and I appreciate that uh that that distinction there and also the recognition I think you have the IBC maxies will appreciate that um and even further kind of diving into the more of the modular subject um start

with like a fun question like what what part of the stack are you most bullish on for disruption you know are you most bullish on on evm or and kind of vm's displacing it or kind of ethereum on the da layer displacing that into R is is is an obvious one sequencing like what what part of the modular stack are you most bullish on o well you know what I'll do you different I'll tell you what um now uh most bearish on or like not bearish but just the one that I think needs the most work exactly where if that work is done and like done well well like everyone should be like gig abolish right and um I forget who it was I think it was Fubar someone had uh a great

tweet that like really not that many people understand like Hands-On because not that many teams do it but like something that our my team was like oh yeah super super familiar with this one yeah something like uh everyone is like you know modular Maxi until you try and like connect to 17 different rpcs and like oh only 17 what that that's a low number it's a we rookie numbers you got to pop up those number uh but that's probably like I'd say right now the biggest problems is there is not uh there isn't like a unified place there's not great coverage for or rpcs for all of this uh I think some of the folks at

uh modular Cloud are working on this and I think their work is is very much needed and quite exciting but something is got to give you know this is currently now like the biggest operational uh sticking point to to these connections it's uh nine times out of 10 when we have an issue or where one of our users has an issue it's RPC related um and so if that part of the stack list this like boring infrastructure you know you could say it ties to the VM because obviously like the client that you're using plays a role in how you can extend uh the RPC but that part it's it's a total [ __ ]

mess right now as very honest what does that mean what does that mean like if I have a modular role that's using movement for BM and hyping for inter and uh Celestia for da and then I had a separate mod R that's using um a veil for da it's using uh you know um a different VM it's using fuel VM and it's using hyperin for interupt and then I've got a third that's using I it's using you know solum and it's using hyp for interop do I need three separate rpcs for that is is that what you're saying I need three separate rpcs no not even so each one of those Services needs to consume uh one RPC what you'll need is a distinct RPC for each one of those

rollups but uh oh yeah that's that that's yeah yeah so so that is where things get a little like hairy right because you you're this chain operator like either you're building it all yourself or you're using one of the rollup service providers there's just like everything here is still like pretty nent it's kind of like 2018 I remember um the folks from block Damon like pitching uh pitching my team on like hey come invest in Us and Them saying like listen it's imperative that spinning up an ethereum node is like a you know four click process that you can do it in like a few minutes if this industry you're betting on like is going to grow pretty much

every app developer you were all talking about like developers developers developers well they're going to need access to nodes like they can't get data from the CH without it they can't do all these things that you think that they're going to be doing and lo and behal uh Alchemy and fur BL Damon all of these became really big businesses uh because their services are so needed right like we depend on quick Noe but quick Noe Alchemy all of these folks right now when you're you know like either they enter this game and kind of create like an OnDemand rollup Service uh or service for these rollups are basically like either they're running the sequencer or

something but someone needs to run a node for your rollup and then that access to that node needs to be extended through like as a you know as RPC endpoint and that part is right now like it's the most boring part but it is the most critical uh see because none of this stuff can even happen without it so you're not even so you're just standing like forget da forget VMS we can't even have a world with all these different modular rups if we don't have a damn RPC provider who who can just very quickly make new rpcs for all these rollups totally then their L is G add all these rpcs to their freaking metamask or rabby or whomever right and so that part that

one like as so like this is the the call it like the the the the foundational issue is like an RPC one and then you have everything else like it's like adding all of those and so you're I think um one thing that we're seeing more and more is like where apps can uh prompt you to make that switch so you don't have to do it yourself because like that what we were dealing with you know uh a year ago two years ago where you had to manually add all the rpcs and you had to do the switches every time like now where you get many more like prompts to add a new network and it just like click prompts to switch and just

like click that's a much better state of the world you know than uh just like a year and a half ago but that can't happen if the [ __ ] rup doesn't have a reliable the first place so so that is one thing we're like I'm most bullish on changes in this area because it is foundational like when I was an investor I used to always think like what what's so I'm going to I'm looking at like this project they want to do this thing what's a necessary assumption what's like a base thing where if that doesn't happen none of what these people are telling me could even happen and then like lever up on those like

foundational on the ones that are needed like 2018 it was pretty clear like wait a second okay so we're all super mega bullish crypto but if there's going to be more crypto than just Bitcoin then there has to be like these proof of stake networks it's not all going to just run on you know on like work and like these mining uh operations like I was a smalltime miner and it [ __ ] sucked I hate mining I hate everything about it there's the alt L1 thesis right there that's where you had to pit yeah so then you're like wait a second there's no way that crypto is like a 10 trillion dollar industry if this doesn't happen so of course we're going to bit along this line next thing

we know Cardon is in the top five jez things been pumping apparently cardano defi is popping off funny you mention it there's actually a team that built uh there is a team that's buil or is building I should say hyperlane to run on cardano I bet there are and they're G and we start we we not that we want to but even if we stop them we could even if right so no power to the people baby yeah and so similarly I think of like you know the same for for the modular ecosystem like this I it's that's not a sexy answer it's not no I know but once you do that

like you incorporate that into all the other bets you're making like if that doesn't have it's hard to be like Mega bullish on anything so how we that on this you also said that your team has thought a lot about this yeah there's there's like a half discussion here like every other week is like so when are we pivoting to that thing because it's such a [ __ ] ISS issue like it's I'm telling you if you did like uh if you collected all the alerts that we get you know like all the you know issues with like a hyper on any chain anywhere I bet you over 80% are related to like unstable rpcs are the the role up as a service provider is not thinking

about this as well is are are they are absolutely thinking about it and I think that's one area where like they could start develop an edge um we work with a bunch of them and like these are very very short people Everyone is everyone who's like in the weeds is aware of this issue like this is but isn't it like a it's a Manpower problem right it's a data indexing problem and you have to the some ex of Manpower problems for sure there's also like uh distributed systems like how do we you know solve this like with better software so that's a part of it too it's obviously not easy to do but it's the you know if all of this was being done by like a single operator so like let's say if you had

one company and that one company was going to take in clientele for all of these new rollups and then like so imagine a world where there's all there's uh only two options either like you do it all by yourself or like you use this one rollup service provider and they did everything themselves then this would that part is like we that's just trade man power like they know how to do that that's very easy it's uh there is still like some engineering work to figure out like how to do this completely selfs serve at scale uh but I don't from our conversations it doesn't seem like you

know there's a lot of unknown uh unknown unknowns there like it's stuff that people know how to do just someone needs to take the time to do it and it gets harder the more providers there are because like uh anyways we need a standard you know like a standard base uh build Bob builder rc.com but there's you know I'm not a I'm not an engineer by training you I'm uh got walk your gig this uh but there's the X XKCD Comics that the engineers all love and there's one of them that's really

relevant for this and it's basically for like any type of developer product and it's like this guy goes we need we need a standard for this thing and he see the two the two people they're working and it's like we made a new standard and then the the last one is like now there are 15 standards for this yeah welcome to welcome to ethereum standards right there so okay so we're gonna Place aund million do uh we're g to take a Lo from the bank and place it all the next RPC provider we hit 100x on that you're okay all right we'll save some of that for uh that's right you got it for for the for the next hypo rays but then once we're

done with the RPC bet and it's freaking gone to the m and we are set for life you know we're trying to invest in the modular in the modular uh thesis the the typical stack um where do you see the most upside in disruption for the typical stack I'd say once you've done that the yeah I would have bets lining across the less explored areas of the deack like you said and again obviously uh there's a bias here but most folks don't think of inoperability as being part of the stack right they think of it as a completely like orthogonal thing you know at best they think of it as orthogonal like at

worse they don't think about it at all so definitely feels like this is an underexposed area and you know what is when you talk about it from like what's the price of an asset the price of an asset to me was always something that we viewed as like how do people think about and feel about this thing today and like what's a price change price change is in dollar terms one of two things happening are there more people uh changing their mind about the thing you know they like it then it goes up they dislike it it goes down or more dollars so either more people or more dollars sometimes it's both right like more people and more dollars sometimes

it's uh the same amount of people just less dollars right but those are this is basically what's happening and so if we both agree that hey this is an underexposed underexplored area and people are not connecting it uh so vividly to the modular faes then you can expect that at some point when they will there will be a price reaction because clearly like the number of people that feel this way is going to change and the amount of dollars that feel this way is going to change and you want to be there uh to benefit from that so certainly interoperability one of those areas but uh and I don't know how to capture this part yet but I reckon that there might be

something along the lines of um like maybe it's part of the da maybe it is part of like the rollup framework itself but like what's the base thing that lets you even spin up a rollup so mechanically like the rpcs are a part of it but you know right now people are using the op stack so does that mean like maybe you should get exposed to like optimism because of what the the super chain like that it's still you know there's still a lot of work to do in like fleshing out of that Vision but that is something I also think because almost no one seems to understand it I'm pretty sure that's an underexposed area that uh I'm feeling quite bullish on so I'm not telling you

like op like optim yeah no just general though just in general which framework do you think is going to get yeah which one is going to get the most traction like and then how are they trying to how are they thinking about you know uh value within that sphere and the super chain I think is like the most you know again even though it still has so much stuff to be figured out about it it feels right now like the most coherent story uh around like a framework having cturing some value Rob there's only one question here when token you know Jake says fool me once shame on you fool me twice Let The Shopper rain went milk and hyperlane we got to Cru some value in this up in

here you know uh gr just waiting on the answer um so I I don't know is the answer we certainly thought about uh building one and I think it's uh very interesting like you know I think using it in such a way where you kind of create an alliance or you know some type of like affiliation between all of the chains and rollups that are using hyline uh and all the developers that are pairing them right because like you know a developer tool is really only as valuable as the developers that use it and so how do you bring them into the mix um but that's

kind of the the extent of it I'm afraid we just haven't had the time to give something like this to what it deserves not joking with not was the was a funny question and I I would I'd shame you if you spent more time on tokenomics rather than building secure interrupt for the modular World um and Rob Rob I I'd like to dive into that article that we uh were uh perusing through with regards to the restating as a means for securing interoperability but I feel like you've got something that you want to ask J Cole about his previous lyrics that he fit to us today and go let M modad was thinking about when we're like when we're thinking about

these these stacks these Deployable Stacks that kind of like are able to harness a bunch of developer like like brain power is it is it kind of like uh accurate to think about hyperland is like this interoperability stack if we spoke about the modular thesis and then we also spoke about kind of the aggregation of of like these protocols so it almost has like these inner components and then it has like this this blossoming like permissionless like place where you can plug into it is is this kind of how you think about like your mental model for interoperability or do you think of hyperlane interoperability not really in

a stack format but more in so of like a like a set of tools how do you think about you know that's a that's a very very good question uh we do think of it a little bit more like I guess somewhere in between those two where you know like I guess the only reason I'm avoiding the the stack terminology is because we don't want people to think of it as like that there are two separate ones right whereas like hey here's the hyperline one and here's the Modular One and like you can kind of use both it's like no this is you know so within each layer there might be you know sub layers right so uh certainly true for the modular

there uh and I want to think of hyperlane specifically this concept of permissionless interoperability is like you have da you have execution you have settlement then you have permissionless interoperability and like that's the role that hyperlink plays now within it yeah there are you know there there's levels to this game and so uh bars n you don't want to separate man yeah exactly thought of as the stink because I just there's gonna be just more and more layers being added on like you know but end of next year it's going to be like there's eight layers to this

stck I think that like a Evolution like a maturity of the ecosystem because as it is able to like evolve it tends to specialize so we we end up like there was this monolithic and it had all of these components that are now modular baked in it had execution settlement everything da all in one stack now the industry is able to evolve into a place where it's starting to specialize these different pieces we're just kind of like getting a clear picture of each part yeah and now we're they're a to like plug in plug out of different parts with one another you can combine these pieces because now they're standalone you're able to combine them into different flavors for different use cases yeah and I think you know the big

opportunity really will be uh like it's the SE like the the experience when using a rollup that uh is obviously part of the modular deack it needs to be on par with like when I just go and I use uh something like salana right like I can't be you know and the you talk but like how the [ __ ] am I gonna manage all these rpcs like it can't be that it can't be like oh I was gonna do this thing but I guess my oh you know I want to go to the banana store but like [ __ ] banana store only accepts banana coins and [ __ ] I'm like it's all everything

it's just apples I only got Apple coing today [ __ ] what do I do like so I go back home and then I go and I go get my banana tokens like [ __ ] I don't want to do that uh it needs to be like I go the store it's like hey here's the thing and like we're done uh because that's why that's how something like uh any you know any standard long chain works right there and so what is going to the processes or the Technologies by which you know like the modular experience can be made seamless is uh you know if you want to be like an area to to invest on like whatever is going to make that happen

maybe it be us maybe it'll be someone else that to me is like the multi-billion dollar opportunity just because like it's foundational you know just like the rpc's bits that's foundational too so is this thing like without something like that it the modular experience is just not going to be competitive um that's big thing if you can't compete with Salon you're done yeah why would use it exactly and so that's you know it's it's coming along right the mzu ecosystem is still uh very very new you know but I'm pretty excited U glad you are you you you wrote a great

piece about um um that was you that wrote piece right about the uh no actually the the ABS thing no no that was one of brains I'm not smart it was your was just uh an amazing guy who works here Mr Rose what a great guy that guy is H he really took his time to take a big steamy poop on ZK light clim so before we jump into why and perhaps I'm being a little bit ere exaggerated here but could you explain to us uh evened in PR audience what is a a light client and why are light clients useful in in uh General message passing or just kind of cross [ __ ] interrupt so I guess all we need to know for the context uh like lines for messaging is a like line is just an uh

so a client or a node is something that allows us to not just run the state be uh to access the state on a given chain but because we are participating in the act of consensus we have a good sense of like well we would lie to ourselves so what my node what my client spits out is generally like a true form of the consensus but that can be quite um unpleasant to run right like there's a big lift that's involved there a light client is just a lighter version of that uh being able to you know like this is a gross gross gross over simplification so uh I'll have to people have to excuse me but you if we

had to use like full you know full nodes for everything we wanted to do from an interoperability perspective that would make the operations of that system very very difficult so like what are we doing when we send the message from uh one chain to another we need to get some level of comfort with this state that we are transferring like if you want to tell like if you want to send something from your chain to robbit then the people on Robbie need to feel good about what you sent over that like this jit like that there is State on Andy's chain that reflects what you want to happen on Robbie and so a life client is like one

way to get a sneak peek into that state and feel very good about its Providence right you know another way is like a group of people call validators tell you this is what happened it's like you know bro trust me bro like this is you know I'm telling you this is like it's true like this is I just was on Andy Shan this is legit this is the thing that's happening um a like client is a way to do that where you're just basically trusting that the chains consensus is valid uh now if you're uh if you were able to get uh if you have like a light client or at the same time you have like a validator system

but it's all the same validators the people who do the consensus and the people who validate the message are all the same then effectively like you're you know you're ending up with the same result but that is rarely the case right because like for chains like ethereum you have a massive massive validator set it would be very very difficult to get every single one of them and so that's why I like clients are very useful uh for messaging but they're not available everywhere yeah uh this lack of availab like if like lines were available everywhere then things like XLR uh or hyperlane or Wormhole really wouldn't exist because uh something like IBC could just

leverage those like llies everywhere and it would have really taken over um just kind of like the entire connectivity landscape and so part of the impetus to make Hy was like wait a second we don't have this technology everywhere we're probably going to get it more and more and more as uh you know there's like a whole move of ZK light clients still early but it's happening and then like let's build hyper in a way that it could absorb this as it comes on in the future and then if your chains that already use hyper like instead of saying [ __ ] I got to pick up you know ZK light client number 33 now instead of and I gotta abandon everything we did with hyperlane like with hyperlane it's hey I was using this security mechanism over here now there's a new one over there I'm just

like point to that one super bullish super bullish on hyperlane basically unre Irreplaceable from a interrupt perspective and and one thing that I'm realizing and I say that because Susanna Evans from IBC came on a space with a bunch of other interrupt people more or less in the evm modular space and she was pushing like clients art Rob and I also have talked to uh the union build team building ZK clients obious SE evm a lot and so you could be kind of to to be able to underpin them or I guess in this sense be the layer on top of them and and absorb them is super super bullish again kind of the aggregation thesis and the other point on the um on the on the light client

thought that I'm having and that I'm realizing with this modular stack is that a lot of this is rather than sacrificing uh and creating a trade-off with things like like clients and data availability data availability sampling you're basically not making the trade-off but you're getting the better effect of lower cost or higher speed without necessarily forgiving that security just by kind of analyzing a a cryptographic amount of proof in that in the state you can then you know reasonably justify that hey this is correct um and so it's not necessarily like a full trade-off it's like based on the proof of you know of of the code we can assess this in in just you know for lack of better understanding a third of

the total uh you know storage or a third of the total uh consensus and and you know spread it out in a randomized order you know change it up each time or however you know that kind of works on the tech and on the cryptography and then boom you kind of get the solution without the tradeoff and so that's I'm s to understand that a lot of this modular stuff is is is is getting better results by by improving the implementation of proving and of of like verification without necessarily giving the or accepting the trade-off that would typically come with it in a modular step or in a in a monolithic step uh that was really interesting yeah and um one thing I I haven't uh gone into it yet but you

wanted to talk a little bit about the reaking piece like from our perspective like the end game of uh of the reaking work is like how can we get to a standpoint where so day one when we have reaking live we'll be able to use that to secure everything that goes out from ethereum and so if you're a new rollup you want to connect with ethereum you could use hyperin and know that what's coming to you from ethereum is secured by igen layer rakers but how can we take that one step further so that your rollup can actually use it for the stuff it's sending back to ethereum and that

uh that's going to be not just an engineering challenge it's also going to be like a little bit of hearts and Minds kind of because we'll need to convince those resters that we can create a system that uh basically tells them like yeah you know what you're you should put your stuff at risk like taking you know like accepting uh you know a messages from like Andy exchange or you know basically staking your assets uh on the security of messages from uh Andy's chain because like this will work out for you long work this will work out for you and there's that's not going to be an easy thing but engineering they're

already taking the risks of rest yeah exactly so there are that's the end game we left to get to so so is is using reaking is that building on top of um the already preet validator networks like clients Etc then you add another level of consensus with recking or are you suggesting that you know using recking for eth somehow is like is like a replacement for using these other validator sets um or kind of can these both replacement or supplement so uh you know like one thing uh so the nice thing with he is like using what security module doesn't mean that you can't use others and so you could have like uh you

could use the uh reaking one which might be a more expensive option because the resters got to get paid maybe you only want to use that for really big occasions uh right say like transfers only of like five million and up should go through this because like yeah you want something more secure and you're going to pay up for that uh like don't know we don't know yet what the cost difference will be like that Marketplace is not yet uh functioning but I imagine that market fores will lead to that price uh Landing somewhere is probably going to be higher than the price that like your run-of-the-mill uh validator would want to get paid um because they're putting less at risk in that case uh so not necessarily

a replacement but in a lot of places uh it will probably act as a replacement for some of the traffic at least like the higher value traffic and in the piece it said that um you know and this is kind of general for reaking in that you know by by reaking you then get basically more um call it capital efficiency you call it consensus mechanisms for different abss actively active actively validated Services you get to open up more use cases for your for your reate e and then you know earlier you kind of mentioned that some of these uh different chains in different uh projects using iping to also put something at stake as well like

Lio um Etc what what's the what's the uh yeah what's the is the is the reason behind utilizing reaking is it because the proof of stake model of the the incentivization design is that why you guys are so bullish on it or is there is it a is it a combination of that and security and yield that you can offer people or yeah no it's it's it's more the latter sorry it's more the the more the former um that uh part of it is like people feel very good about uh you know stake the the as a mode like providing economic

security you know like people generally like the way people think about Economic Security is that it's good up until when it's not right like no one would feel good with a million dollars securing a hundred but they feel pretty good with a million dollars securing less than a million and so the biggest pool of capital right now available in crypto is uh I guess it's technically Bitcoin but if not it's the amount of uh ethereum stake and so being able to tap into that just immediately opens you up for the largest pool of Economic Security now I'm not going to tell you that it's like

um it's not going to be a slam duck because one thing this ises not really get thought of when folks uh think about Economic Security through reaking is that like it becomes less useful the more people use it uh so it has like a you know an inverse uh Network effect like as if you're the first one to use it you're like this is great I pay whatever uh you know like if there's $10 worth of it maybe I'll play a dollar a year in premiums but then you come in and you're like oh okay I want it too well what should you pay for it because like now both of us are relying on it so we could have an event like you could start

paying for it now you're we're both paying a dollar this reaker [ __ ] St they're you know they're taking in $2 of premiums your 20% premiums their Capital basees life's amazing um then like something happens on my side and now I just got half of it slashed you've been paying $2 or so you've been paying a dollar same as as am as me and now you're kind of out of luck right because what's happened is that you've paid the same you didn't do anything and at the same time the uh available budget for you like the security budget just got reduced um and so this is like a very interesting Dynamic I don't think we'll

actually hit it because I doubt there'll be like many slashing events but like this is I like probably the thing I'm most interested just from like a personal perspective like seeing how that part of the market develops uh but without this you know uh we are not going to be able to tap into the awesome awesomely large pool of Economic Security and it's a useful thing to be able to tap into it and so it's gonna happen one way or another yeah this isn't necessarily on on interop but since we're on theing point and I realized we're we're kind of drawing towards uh towards time but this has been like my my like sticking point

on reaking ever since I kind of like like really I think like like caught what was going on with the concept because my understanding of reaking is basically like these networks are taking out like essentially they're using Security on margin it's like we're taking the like proof of stake of ethereum and now we're applying it and we're reh hypothecating it to another ruup and like what happens when there's margin in reification it's like you could get margin called and what happens from that put proof of stake underlying the security of your rollup gets liquidated what like what happens what happens to all the other rollups that are also relying on the same pool of collateral

to secure their rollup this this like because something happens on J Cole chain to like get SL like does that mean something's GNA like affect Annie and Obby chain like these are the like cons those are the hard questions what was that those are the hard questions and uh it's it's one like the I think the best answer answers I have for it are come from actually like from the bond world uh where like you think of just you know I think in general like if you looked at like the credit risk uh like ethereum stakers as a whole Triple A it's very good but now

some of the things they're going to start securing they're going to look like Triple C's they're going to look like you know uh really not great company oh we know we know don't you worry so these abs pitch toan it's so uh yeah there's going to be like who's going to be the credit agency and who's going to be like you know patrolling the beat to make sure that the credit quality of the reaking pool stays AAA because that's what generally when people think of it like from my conversation so far people do think of this as like oh this is a triaa credit that that I'm buying here uh but

if it's permissionless if it's permissionless rest right I could well that's I don't think CRA i f on top of that thing yeah I don't think uh reaking can ever be uh truly permissionless unless like you know it'll be permissioned by like some programmatic means where it's like you know it's like oh we have you know like our investment strategy here is not like a person you know it's like Renaissance it just it runs you know it's just the model runs so like you feed it some inputs and it decides like yes this is an acceptable addition or not uh but there has to be some type of like permissioning it might like the way that it's handed about might not have to be

like oh well you have to go to the the reaking office and talk to the reaking officer and see if they'll agree to let you restake but uh someone's going to have to decide and like the best way or the most scalable way it might for be for it to be a machine but otherwise I don't know uh how and like the the like I think the D layer team is very very adapted this like they totally understand it they have really cool ideas about how to uh deal with these issues and that's why like you know you said like those pitch scks for those abss like not everybody uh gets through goodness gracious man goodness gracious uh by the time we have permission reaking we're going to have ai police

and uh you're not going to be able to take your dog out for a walk past 9:00 p.m. so make sure you guys take your dogs out for a walk T night and uh thank you so much jcole for coming on and rapping to us for an hour it's been a pleasure man it's been a lot of fun thank you for having me