How Fast Confirmations Will Improve Rollup UX with Ben Fisch and Matt Pearring

Transcript

how does working with espresso and their design help arbitrum move closer to stage two because you don't need multiple rounds of messages to pass between the rups within the span of one block time is changing the sequencer critical to achieving synchronous composability across l2s and it's good if we have multiple Sovereign networks plug into the same messaging interface because that means they'll all have more usage right so loaded question there yeah welcome back to the rollup your home for technical talks and open conversations that shape the modular future nothing on today's show is to be construed as Financial advice you can see our full disclosures list at the roll up.co Ventures sit back relax and enjoy today's episode after a quick word from some of our awesome

Partners safe is moving the world GDP onchain with the introduction of safe net safe is taking this to the next step very soon check out net. safe. Global to learn more about how safe is moving the world's GDP on chain all right we're live at RB versus in Bangkok um hope everyone's stayed clear of food poisoning and has survived the bikes I think the bikes are the only way to get around I'm to for two so far knock on wood um how we all doing today pretty good yeah I'm three for three so uh hopefully we can keep it going you know I think I'm two for two yeah how do you hold on to the bike you do you hold on at all yeah I'm a white I'm going just grab onto the back White Knuckles full full squeeze yeah I've been be hugging

the driver but as I get more comfortable he starts to pull off pull you know I think there's a logical progression from uh taking a car which is like transacting on Main net during an nft mint then until like going on to the bike which is like a smooth arbitrum experience or something like that we got the the T tkes in the middle there which is like an alternative layer one a little like oh what's security all right no helmet I see well guys welcome on um Matt it's definitely your first time Ben we've tried it before this might be your first time actually on a podcast so not sure I feel like I've been on the rollup before but you know what I'm not sure maybe have um quick intro would be fantastic guys and then we can jump right in um so Matt

what you start yeah so my name is Matt I run the product department at offchain laes been here for around three years so kind of seen a lot of the journey the arbitrum both chains and the ecosystem have taken over the last few years I'm really excited before crypto I was uh working at Google doing traditional Tech stuff kind of saw the uh issues you see with centralization of power wanted to make a move so here I am welcome what's up Ben hey how's it going yeah I'm so I'm Ben Fish I'm a co-founder and CEO of espresso um espresso is sort of made its mission to make all ethereum rollups um composable um you know bringing better interoperability to the L2 space we see sort of the tension between composability and scalability as like

the key thing that the space needs to solve um and that you know starts with kind of what we want launch today which is um sort of a fast confirmation system that can confirm transactions um you know with strong security but pre- L1 finality time and um it helps you know various operators in the space be able to confirm the information that's on different chains sweet and congratulations on mayet Ben for espresso a big accomplishment I know you guys are working very closely with arbitrum in a in a myriad of ways which we'll touch on but I think to kind of set the stage guys we we're at this place where we've unbundled ethereum

this kind of Monolithic L1 that that did all the properties of a blockchain in one and kind of really shifted execution to to rups that's like kind of like the rup Centric road map at its core um you know we've heard this a lot that the a consequence of this has been the fragmentation liquidity and users amongst other things and so now we're kind of as a community coming together to try to figure out how to build interoperability and so one notion that I've been pushing forward and that I think a lot of people in the community have been a agreeing with is that the future of interoperability for ethereum uh is going to be on neutral grounds uh some sort of standardization some sort of neutral uh capacity where a team uh can come in or teams and provide the

best Solutions right now there's different incentives at play for teams to build interoperability solutions that best suit them I'm starting to view this as a collaborative kind of mission amongst many teams and so just with that framing in mind I'm curious how you guys are thinking about broadening ethereum interoperability and like what is the right way that we and the community should be thinking about interoperability within the ethereum ecosystem B maybe you can start and then we can hear from you mat sure yeah that's um it's a loaded question there yeah um ethereum I mean and I'll get into why is so you know interoperability first of all uh and sort of composability or

interoperability uh can take on various forms right everything from just simp bridging of assets between different chains to uh asynchronous interactions between chains where chains will actually pass each other messages and um you know wait for some response so enter some intermediary state of an application that then waits for a message back to full-on synchronous composability which is the kind of composability that you have between applications today on ethereum um and you know each of these levels of composability or interoperability um unlocks you know different you know um capabilities for applications but also presents its own challenges in terms of um you and and and trade-offs for for l2s if they were to say you know be synchronously composable with each other

versus just being able to bridge assets um so I think the right way to think about um space which is your your question right is to um be able to you know explore the full spectrum of possibilities cleanly understand um you know the best way to achieve these different types of uh of interoperability with um the least amount of tradeoffs possible uh and to provide infrastructure that enables chains to then kind of make their own choices anything yeah sure I mean I would say maybe starting on the point of neutrality I mean one of the defining characteristics of I think blockchains that make them useful to people and especially ethereum is this notion of credible neutrality uh which is

arbitrary right like something that is decided by the user but I think that you know to Ben's point on ethereum today you don't think about interoperability because everything just works together you can call contracts in real time you can make Atomic swaps in real time um and so now that we've seen a fragmentation I think for good reason of execution environments with different rollups with many orbit chains in the arbitrum ecosystem different application teams want sovereignty over their users they want sovereignty over their application customizations uh they want to be able to maybe uh capture value or use that value in a different way than maybe a larger protocol team might want to um or a dow uh maybe they want maybe they have their own Dow so because of all these things because of all these desire for customization uh you you have

a problem where you lose some of that kind of synchronous composability you've seen in ethereum and so now we need to find a way to make interoperability you know possible I hate to run it back but that's kind of you know at least my perspective I think the importance of neutrality um it will remain but uh to V Point all there are a lot of Solutions in the space um each one has their own set of trade-offs right you have maybe more centralized solutions that are faster um or decentralized solutions that are slower but obviously more secure for some reasons and so the uh you know the importance of neutrality I think remains to be seen in the in the interoperability space cuz fundamentally you're just kind of relaying messages um and it's only a component of many of the pieces of an execution environment or a

smart contract system so it kind of remains to be seen but I think it's important yeah and I I think you know we we've came a long way there's obviously still a lot of work to do but from my understanding now with espresso on Main net this is no longer an engineering problem this is more so an incentive or a social coordination thing to solve is that the right way to think about this or well I I can see your face your face well espresso yeah I mean espresso is on mainnet but there's still a lot to you know espresso's main that today is a fast confirmation layer and that um you know in my view is a backbone of interoperability protocols that can be built on top from asynchronous to synchronous but um there are still a lot of things that need to happen to say for

example get l2s to be synchronously composable we actually just um wrote a um we're in the process of writing a standard for something we call coordinated neural communication which is a standard that can handle both synchronous and asynchronous communication I me there's other standards being written as well there's a lot of work to be done still um uh you know to be built on top of things like espresso uh so I wouldn't say that just because espresso is on Main net today that you know we've solved interoperability yeah I think like the the the neutrality position there and what Rob was getting at is more of this idea of there's been this notion that why should arbitrum opt out of their single sequencer model which for various reasons the you know ARB working towards

stage two and espresso is a part of kind of this decentralizing the sequencer model if you're giving up ownership or revenue or uh attribution or these kind of various properties you get by having a single sequencer model and so like the neutrality part of this I think is more like uh is the in are the incentives aligned for rollups to opt into something like espresso and I think part of that is why espresso is branding amongst other kind of shared sequencers have kind of like repositioned a little bit because uh again like like if if if a rollup is going to uh opt into a model there has to be in like at least some sort of large gain if there's going to be a a economic uh loss or some sort of like other kind of loss that is

currently handled by a single sequencer um yeah so yeah so the espresso Network today is not a shared sequencer actually right um the espresso Network today is a fast uh confirmation layer as we call it um some people think of this as fast finality layer as well and it doesn't replace sequencers so um we're integrating a bunch of arbitrum orbit chains for example in the coming months and they will each have their own calized sequencer still when it comes to decentralizing a sequencer you can think of it as in two parts one is to um diversify the party that's actually proposing transactions right the other is to um create a more robust protocol for confirming the transactions that

have been sequenced before they're published to the L1 right and so it's that second part that espresso Network provides today um you can have a centralized sequencer that publishes blocks to Espresso espresso reaches consensus on them and ensures that the blocks that are eventually published to the L1 are consistent um and what this enables is it just it enables any you know anyone to be able to uh you know quickly confirm what transactions have completed on a rollup that's connected to Espresso and we can get into it more but that's the backbone for interoperability protocol so there really isn't sort of an economic issue or question of like replacing the sequencer it's not really

about the sequencer itself we can get get into like what rollups might want to do if they want to use shared sequencers which are necessary for synchronous composability but just for you know being able to asyncronously compose um or or just like enable solvers to bridge faster integrating with espresso um you know is a is a is a great way to enhance that and it doesn't touch at all the economics of the chain I appreciate the clarity I think that's a good segue into this uh into this standards conversation as as well did you want to add something to that well I guess I I I don't think either of us are trying to pitch in whole espresso here it's it's like I I think um the the the goal at the end of the day is synchronous composability

across the ethereum ecosystem and I just just want to like really clarify that like even my own just technical understanding are is changing the sequencer uh critical to achieving synchronous composability across l2s yeah so Ian synchronous composability it's a I mean it's it's a you know it's an amazing capability and it's something that we have in the in the evm so I just want to start out by saying that I don't know that all rollups everywhere are going to be synchronously composable but you may just have synchronously composable clusters but yes um while in theory it is um possible to achieve synchronous composability

without a shared sequencer you do need some form of coordination okay and the most straightforward I don't want to get like we could go down a huge Rabbit Hole of how you could have coordination between sequencers without sharing sequencer and what the difference is between them but the main point is that in order to synchronously compose you need to construct the blocks of the two rollups in a way that they can sort of pass multiple rounds of messages between each other or do something that effectively emulates that um uh you know one way of doing that is through the protocol that we propose called Sur where you have basically mailboxes on the two different chains and this coordinator or shared sequencer um basically populates the inboxes of the

two chains with multiple rounds of messages that they should be receiving and so the chains just sort of carry out the transcript of the conversation without actually sending messages to each other reliant on the shared sequencer for doing that um uh yeah so that that is why shared sequencers or some kind of coordinator in the in the block construction process is necessary for synchronous composability for asynchronous you know you you don't need that because you don't need multiple rounds of messages to pass between the rups within the span of one block time yeah got it I appr appreciate that Clarity I think it it it also ties back to the previous point with with regards to how espresso's fast finality of layer if you will also further en enhances things like intense based inter it

provides solvers with a much faster way of basically being able to take less inventory risk so like there's a really popular standard that actually arbit has just opted into and like endorse which is ERC 7683 um that is an intense based interop standard and so uh my my my original question was if this was like at all competitive with what is happening here but it really seems like it's actually a perfect match for something like espresso's Fast finale so H H how would that kind of like interoperate in this ecosystem sure I mean I'll start from our end I think 7683 is just one of many examples of improving the developer experience of working with blockchains I mean you started out with a language and

a node software now we have different types of chains right and so this is I would say kind of a standard interface for using solver networks there's so many me you look at the different types of solutions available today in the market which is great uh for chains to be able to communicate to each other particularly transfer tokens between each other quickly have someone else take on that risk and be able to you know rebalance quickly is something that they'll be able to use expresso for but I think this is just a standard I think standards are good good um more standards can be better if there is a meaningful tradeoff between them but for now you know we're seeing particularly with calling these solver networks if there's a a rally around this one or A1 which we are seeing with 76 A3 that's good that makes it easier for developers

to use kind of the same types of contracts across different chains you know it's easy to kind of debug new systems that we build um and it's good if we have multiple Sovereign networks plug into the same messaging interface because that means they'll all have more usage right and that means liquidity is kind of moving the same way hopefully being managed by a more diverse party set because you don't have fragmentation of how you call them I think it's good um I think it doesn't solve everything you know we have a lot of more work to be done but it's exciting and yeah you know arbitr I think came forward with support for this over the last couple days so we excited about that yeah awesome and to add to that um I think one thing that's like when it comes to talking about standards I think one

people one thing that maybe people miss a bit in the conversation is that uh you know there's a difference between proposing a standard and proposing a protocol these protocols actually already exist there's many intent based Bridges today and it's the most common way of bridging between chains and so the main point of a standard is asking the question why should they all conform to like the same format for example and when it comes to airc you know 7683 the answer there is well solvers these liquidity providers are working with all these different protocols for the ecosystem to work you know in the smoothest way possible you want to enable solvers to work with as many of these protocols in as seamless a way as possible you also want wallets to be able to integrate them um you know as

as many of them is in the in the easiest way possible and so by by standardizing you know the apis um that the solvers need to integrate with that the that the wallets need to integrate with um you make all of this work much better um you know when it comes to whether solver based Bridges solve all our interoperability problems right you know then the answer there is no that doesn't mean you C 7683 isn't a great thing or that the solar based Bridges aren't a great thing they are they're just it's just that they're only scratching the service in terms of what you can do to solve fragmentation do do you see them as potentially going down like a Divergent path to that end goal or do you could we use those as a stepping stone to to some of our like bigger picture yeah it's all

complimentary I mean solver based bridges are built on top of they're also built on top of message passing right um I mean solver based bridges are a great way to move assets from an end user perspective you're not actually well you're not actually moving the assets between the change wer but from an end user perspective you're you are giving them assets on one chain and so for for relatively low volume you know bridging um for uh use cases that aren't really you know trying to solve liquidity fragmentation or things like that solver based bridging works great right what it doesn't solve is uh if you want to to move you know huge sums of assets

between chains then it's extremely capitally inefficient because the solvers need to hold enough um uh sufficiently High reserves on the different chains and it doesn't address liquidity fragmentation um which is one of the reasons why we talk about bridging being so important in the first place is that if you have a DEX in every different chain and you're splitting liquidity between on the different chains then defi doesn't work as well if you want to enable chains to specialize and to have one chain that concentrates liquidity and another chain that concentrates the nfts and another chain that does something else then you need to solve bridging in a way that doesn't rely on solvers that also split their liquidity across the different chains yeah I appreciate that perspective kind of to to hone in on the espresso and arbitrum collaboration that has been in the works

and research for some time and uh now coming more and more publicly and just more and more imp implementation one one key subject for all of the rollup teams is basically moving from stage zero on L2 be to stage one stage one to stage two uh arbit from being uh very very notable in the fraud proof research design um and looking to make the jump to stage two at some point in the future uh obviously it's kind of like a a slow and steady process with that kind of framing how how does working with espresso and their design help arbitrum move closer to stage two um and kind of what does that look like from your perspective now Redstone modular oracles are available on 50 plus chains trusted

by over 100 clients and over 6 billion in total value secured visit redstone. Finance to find out more welcome aboard on our journey into the singularity this is the endgame road map for frax and fractal the new L2 out of the frax ecosystem that is totally based on fractal scaling FRA the modular L2 blockchain built by fra with modular and fractal scaling in mind fractal scaling is based on the mandal BR set of fractal images and it explains how l2s become l3s and the entire app chain life cycle goes through its Evolution the frax

ecosystem is made of some of the top builders in the space and has stemmed from its OG days of the frax stable coin it is now in the LST space the L RT space and this frax L2 joins the existing Suite of fra products including fra Swap and fra lend you can see frax faery the bridge to go from any chain arbitrum optimism or main net into the frax main net Bridge your tokens and bring your FRA assets into the fractal ecosystem join today to get the best out of fraxo you are still ear early join

today take part in our Expedition into the singularity it's a good question so again I represent offchain laugh so obviously one of the early developers of arbitrum but um there's a whole Dow there's a lot of stakeholders I think for many chains working with a protocol like espresso is is really good actually um I can't comment on how much it will affect stage two just because I don't have the rubric right next to me and I have I have memorize it to be transparent but I do think that you know a critical aspect of what at least the new main implementation is enabling is having kind of another set of parties look at what the sequencer set it ordered and is going to post and whether or not it actually did post what it said it was going to right and so

this is kind of a way of adding one of the properties that folks are you know curious about and with something like a distributed sequencer U but without actually doing that and adding all the latency effects that something like that would have in a current version that we would be able to ship today so now you have this other set of parts parties kind of looking at what the ordering uh commitment was kind of doing their own manual validation making sure that things were actually correct signing an attestation posting that on chain and having proof that they you know did their job that's actually good there is a cost right but to to to have you know this set of parties actually run these pieces of software there's there's a cost that comes about that but you get this ni's privative which is okay now

you don't just have this one party ordering in Block building and having to wait all the way till L1 final to have confirmation and have you know a good feeling about this actually being final you have another set of intermediary node operators that are running essentially uh a different version of the same protocol right and giving you additional confirmation uh over that block and so it it allows parties to just like solvers can if they want to move more quickly with a better guarantee and that that's really um cool right and I think I don't again I don't know exactly how this contributes to stage two but I would argue that having a gadget like this is actually a benefit property for a lot of chains it is good from a decentralization perspective and should be considered as part of that

rubric yeah maybe just add one thing we often talk about um you know things through the perspective of oh is it stage one is it stage two like all these kind of categories that we right it feels like you know that sort of the space is kind of religious in that way but um it's good brand it's good branding for l2b know it's effective yeah no but but what I want what I wanted to say is that sometimes it's important to look at things from just a uh you know a pragmatic perspective oh what does this do differently for the chains okay and so to illustrate you know what Matt is saying in a very concrete way um think about a real concrete risk that centralized sequencers face today which is that even

if the centralized sequencer is perfectly honest right everyone can get hacked right and so what can happen if a sequencer gets hacked what is the damage that can happen if the sequenc is being relied upon for its confirmations Say by these solver based Bridges right and there are millions of dollars of eth that are being bridged from from you know say arbitrum to some other ecosystem because of this then now someone who hacks into the you know arbit from sequencer and causes it to kind of reverse transactions that it previously confirmed before actually posted them to the L1 can now steal a lot of money and so it

not only creates damage but it also creates an incentive for somebody to hack the system um if that arbit from sequencer is now you know connected to the espresso Network and the hacker would have to hack the entire espresso Network in order to carry out this attack um you know now you're a lot safer right so it's not just about being able to say you're stage one or stage two it's also about trying to defend against concrete attacks that actually do happen in practice look at the bronin bridge hack or any of these hacks it's always you know one or two or three or four parties getting hacked and um causing damage usually in some kind of bridging protocol and so then to even further

extrapolate that for espresso H how does the how does the espresso's technical design kind of come into this and maybe you could explain that a bit deeper I know that that we touched on it uh with regards to this arbit uh collaboration yeah so the way that espresso works today is the sequencer of a chain whether that sequencer is decentralized or centralized will uh post um its transaction blocks to publish them on espresso and when that sequencer then goes to publish those blocks on etherum um it has to publish the same thing that it published to Espresso and that's enforced by the L1 itself right by the settlement contract of the

chain um and this can be done at the own level can be done at the L2 level so today for example um a number of orbit chains are going to be using espresso on top of arbitrum one so there L3 is on arbitrum one but there there's a contract in arbitrum one that ensures that whatever the uh sequencers published to Espresso has to be what gets settled to arbitrum one uh and so that's how it works that how it how how it enables these fast confirmations um on top of which faster messaging protocols by the way could be built um and it's something that we are are even building for orbit chains within within the arbitrum ecosystem um but we have a second very important collaboration which is on sequence or

decentralization of um you know of of course pending a dow a vote and everything the arbit from one sequencer uh and this is the time boost collaboration where you know time boost is a protocol that's run today by the centralize sequencer of arbitrum one uh this will be a decentralized version of that uh which will involve um you know not hundreds but just tens of nodes that are running a protocol that has some really really nice properties um in terms of you know me protection um and uh while preserving some of the um uh you know very Key properties of that arbitrum has today which is being able to confirm transactions really quickly

and move it high throughput and for the Searchers listening uh time boost is actually not an operation just yet but the the snapshot has passed SN on Chan vote is coming soon uh but yeah I know I think uh a lot of interesting stuff to come in distributed ordering policies I think this is something that we could probably have a whole another conversation on but I'm really excited for that I'm holding back like on on asking and like diving down these rabbit holes cuz like I just adding noes to the network is going to protect security I want to know like how that how you're able to provide faster confirmations by adding more nodes to the network because maybe I mean maybe that's a simpler question than I I would have imagined oh yeah I I'm I'm not sure if I've said that I I didn't mean to say that um so

espresso Network today maybe expressive Network today is it doesn't do the sequencing for any of the orbit chains that use it and it doesn't yet uh it's not yet integrated with arbitrum 1 although we will be proposing to the DA to uh integrate arbitrum 1 as well with this freso network um that will likely do that in conjunction with a proposal to move to decentralized time boost right but decentralizing the set of nodes that are proposing transactions um and making and doing that in a way that's you um you know has good me protection and other nice properties and makes things work efficiently is a separate problem from the uh V espresso network does which is ensuring that whatever was sequenced by this

relatively small set of nodes uh can't be changed once it's confirmed uh and so espresso Network today um you know runs also just on you know around a 100 nodes we're going to try to expand that to um uh around a thousand noes but there'll be kind of a longer timeline for that to happen um more importantly we're moving to a proof of stake model where there will be Economic Security Behind These confirmations um and so to your question of does more nodes you know make things happen faster no but more nodes may add security to the fast consensus that's being reached on these blocks is it fair to to say then as like a summation of this that espresso so is acting as this

surveillance layer for the Integrity of arbitrum one uh block posting to the L1 is that like a fair way to think about it maybe yeah I mean when when a chain integrates espresso it's making changes to the way that its settlement contract works okay so that um like you know you can think of the sequencer of a chain as sort of you know embodying two different roles right right and you're now kind of splitting out one of those roles and so you are taking some power away from the sequencer you're taking away its power to change its mind and so which is a good thing right they shouldn't be changing their mind

right so when what what you are the caveat of course is that you're C creating a liveness risk right because now the sequencer has to publish the block to Espresso before it publishes it to ethereum right so if espresso were to lose liveness that can create a temporary stall in the liveness of the chain but you got said taking measures to prevent kind of I mean our whole thing is building consensus that works well um confirms transactions quickly can scale to many many nodes and have high Economic Security uh without losing labess and and so how have you guys been able to go on Main net without economic uh stake uh now and then transition to that to that later but like what is what is the mechanism here here and now and

then yeah you know with a proof of the stake model that I think is more well understood like that that you'll TR what we're doing now is well understood as well it's just it's called it's a permission Network okay so um you know we have we work with about um you know 10 or so sort of professional node operators running 100 U distributed nodes around the world um we're going to both open up this operator set to be non- permission and also transition to proof of stake model to involve Economic Security just to kind of close this out here Matt from the off Labs perspective when looking uh at arbitrament and doing a lot of research and development that contributes to arbitrament and the Brer

Dow um how important are these measures to offchain labs to continuously iterate on the core ethoses of crypto decentralization censorship resistance these types of of properties when thinking about building out the future of arbitr they're the number one priority right like our mission is to scale ethereum and if you look at what ethereum is about it's all of the things you just listed right um I think security is Paramount at the current moment right I mean you see across not just the arbit of one permission validator set that is actually soon to become permissionless um but across a lot of protocols in the industry there's still guard rails in in place right because fundamentally you want a distributed system but you also got real

money on the line right now and so you have to make sure there are people who know what they're doing running these very novel systems right I mean the fact that there's onchain executable governance for arage from one that's crazy right I mean this is the first of its kind it's amazing it's it's super cool is like a super power it's and it's it's new and you know hopefully it'll continue to evolve it will but um I think for us we care a lot about taking the time to iteratively improve on each of the things that you mentioned um it's a slow process but I mean we talked about all the you know not all of them but a lot of the idiosyncrasies with with the fast finality Network in espresso today um that's just one component not only of the many things that expresso team is working on but

also of the whole P right of the whole set of modular components that make up a blockchain and so yeah for us it's it's it's obviously the most important thing to make progress everywhere and and I think it helps you guys strike a balance between what is idealistic and what is pragmatic here here yes I think a great example of this is the sequencing technology on arbitrum chains we've talked for years about decentralized sequencing I mean Ben is probably one of the loudest folks in the ecosystem about this you know especially a couple years ago right I mean it is a fascinating concept and will at some point in my view be the way that these systems are run but if you look at not only consumer great Hardware but particularly commercial great Hardware that chain securing tens of billions of dollars

need to run at a performance scale that we have become accustomed to in terms of fast block times and gas cost for users if you want to make that system distributed today users are going to feel pain and is that a trade-off that D wants to make today um I would argue probably not um I think we know no because you know a lot of proposals are are not focused on that right and So eventually when it is cheap enough and easy enough to deploy a distributed system that can maintain the type of performance we see um in a centralized sequencer there will be a move it is a simply better option but until it becomes that I think uh the focus is on adding better guarantees to the system

right to make sure that the single party is held accountable yeah absolutely and I I mean I I think that is you know kudos to not just arbitrum but like our entire industry about uh open sourness and the open-mindedness to accept new technologies as they come up even though they challenge the status quo yeah yeah I like to think this as like a really solid stepping stone for arbitrum as a whole and then this next iteration of espresso kind of brings a lot more to the table as you're explaining Ben which can then even further take the next important step aligned with this coure that we laid out so gentlemen thank you so much for uh coming out I hope you had a great time in Bangkok and uh keep that two for two strong all

right Ben we're trying to grill you too hard we're just trying to understand what the hell is going on yeah yeah no it was great I didn't feel like you were grilling me too hard