DeFi By Design EP#119: The Rise of altVMs in The Modular Stack

Transcript

welcome back to the rollup High Roller Community as always nothing on the rollup is to be construed as Financial advice we are just an Educators in the defi 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 def5 we're here to educate Empower and enrich you and let's jump in welcome to D5 by Design we talk all things blockchain and cryptocurrency while striving to educate Empower and enrich yo welcome back to the D5 by Design podcast here today of course with my boy Rob got my scroll shirt on I

think it's the best shirt that I C that Devcon it's uh it's kind of like a custom fit Rob what do you think is it is it wearable on a day to now debatable I uh I I remember seeing the scroll shirt I thought the quality was amazing the only thing is the collar it's like a real big collar so I don't know it's like a half turtleneck kind of thing on a t-shirt it's different it's Unique it's good to see you again Andy and today we got rousi King of the how you doing guys not too bad not too bad I would probably not wear that shirt out um in public unless it's a crypto conference um maybe put a planel over that uh it does look like a turn like a little bit so that is my issue um

but I love scroll so a good swag we have swag we will have yeah it'll be ones you can wear them date I promise that is the litness test of crypto swag I promise and then we need a real movement party that's a real Party Movement party at East Denver you didn't hear from me you didn't hear from me so ruy why don't you introduce yourself a bit about um kind of movement of obviously you uh you have an interesting past in crypto um kind of different programming backgrounds and um yeah why you guys are building movement Labs yeah so movement Labs is building

the first move layer to ethereum powered by Celestia for da um our kind of background comes from the app testing system um my background personally is engineer to start off at by Care Systems datas security United Health Group um kind of entering that space over there um then got into the crypto bug pretty early on on engineering side um take solity km was and rust and that space um and then around like August of last year kind of went stumble across move I was like Facebook the biggest consumer app and social network in the world looked at the Curren on blockchains was like this is awful um we need a better VM we need better paraliz we need better formal verification built into the VM um so that we can actually build technology

that can billions of users on Facebook could use um it was like literally I would say two years ago I want to say that decision was made um and then that was a DM project from Facebook um foring to regulation it couldn't never see fruition and got split up in both abdos and sui um personally I built the first decks in abdos um my co-founder built the first yield dager an opos so um we're some of the earliest builders in move space and recognized that um there's a lot of issues that we couldn't directly handle um on the infrastructure side and that was around the time when the mod thesis kind of picked up um and realize the real Market fit for movm to exist within the Majic system um so that's led us to build movement the first move Layer Two on ethereum how did

you meet CP sorry sorry sorry that thought just like where did that come in we met at a conference um me and CP actually went to the same University um he was a little bit older than me but um we're just like your bua us and you go to um the same University it was a cool like jamming session um we kind of work worked together on like some move ideas um and like he buil yield aggregator so naturally like we were Partners kind of um but around like post FDX um that's kind of when we kind of linked up and decided to go all in on this very cool and like I on the on the note of like Partnerships like and also speaking about Andy scroll shirt like you guys have found a way to position yourself so

that your friends with essentially like potentially everyone in the industry Y and that's like that's also like our thesis and our perspective on the industry is like if you can if you can come up with an Innovative positioning where you you're able to like befriend everyone rather than compete with one another yeah then you are probably building something valuable in the space so like how how when when you're looking at the space like as a whole maybe we'll use scroll as an example like how do you partner with someone like scroll and then how do you keep an open enough mind to partner with essentially like all these new tools and chains that are coming up you know on a daily basis I think when we looked at the mod thesis

there was like a clear market for da right you had like three big dealers and like now near gain of space you had a bunch of Ras providers in space a bunch of like rollup Stacks you had like AP DK op stack CK stack I don't think scroll is a stack yet but I'm sure they're coming out with it soon um but there's a lot of competition but everyone was saying like we'll do all these customizations but leave the evm as it is um when most people recognize that the evm was actually the biggest bottleneck um execution after data availability is the biggest bottleneck on ethereum a from security side there $4 billion lost every hacker tax and B on the throughput side like pretty much every evm chain tops at 100 TPS um which if you ever trying and get a consumer

app on chain um is pretty problematic so that's kind of how we kind of have this Unique Edge in the market where everyone is are kind of friend because no one's really competing on the execution layer level and even if you are like there's fuel obviously there on VM and in that manner it's not tribalistic at all because the overall Market wants dethrown the evm so even though there are other style virtual machines and you can argue whether the svm is better the fuel VM is better or the mood VM better um the real challenge is deing the evm and generating mind share um for overarching virtual machines um so the current state of the market is like everyone's kind of competing to race to the bottom but execution layers are where there's kind of innovation on that side um and kind of how we partner with

scroll is um and these other like lyc to Frameworks is optimism obviously has optim the chain and then they have op stack arbitr as orbit um ZK Sy ZK stack um so we can essentially bring the move VM um to these ecosystems we have like a move stack where anyone can launch move rollup use whatever stack they want use whatever D they want use whatever sequencer they want we not competing at that end we just give them the VM um it's the first M evm so it's fully even compatible um so Works talks and looks like evm but has a move VM running under the Hood so it's a paralyzed ver machine you have formal verification built into the BM so we stop like 9 95% of all attack vectors we're seeing slowy which I'll get more into later I'm sure for anyone who's like

unfamiliar uh does Dethrone the evm mean Dethrone ethereum or like after you take out the evm what else is left of ethereum yeah so like that's not at all what it means at all like ethereum is still the best place to settle transactions um by tvl it like kind of takes up all of D5 not pretty much everything like you have eum salana with the recent Rise um but the ethereum is not evm in fact uh most people in the etherum space agree that em is a hot pile of garbage um It's s threaded solidity is pretty broken it can't scale um but theum itself is a great settlement layer it's very decentralized um has a lot of tvl and mind Insurance security um or it's like Secure Storage

of value um so the way we kind of look at is bring this next gen VM bring the superior technology to where users are um in the past of mon chains you always had next gen VMS you always had like cool Technologies but it was really difficult to get adoption because everyone is still on ethereum despite whether you like it or not um so modularity allows you to bring in experimentation so you can experiment on ethereum without leaving ethereum yeah um even further um going down that kind of rabbit hole when it comes to the VM um like what is a VM what is a virtual machine I think like your computer has a

virtual machine it's like an operating system where like in traditional web 2 you can write um code and we compile down like every app you're using on your laptop Mac is connected to the chromium or connected to the Mac OS um so you can have like a Windows virtu machine you can have like a Mac virtu machine and blockchain specifically it's how um smart contracts can interface with underlying consensus mechanism so like ethereum consensus is not itself able to be accessed by a developer user but a the virtu machine can be access so you can think of the valador set which is like the network and the virtu machine is on top of that communicates a user or developer to the underlying settlement layer um a great analogy is like Swift

for example communicates between different banks but the actual s layer um is underneath that um which is like the communication between Banks and the financial system that exists today um so ethereum just like a better Financial system but there better ways to communicate with ethereum than the evm yeah awesome just uh wanted to throw that one out there before we dive into the weeks here um because so I'm super curious as to as to why you believe that uh alternative VMS have a place um and let's just scrap evbm from our thought here let's let's look at a world of of just well known as alternative BMS but but what are some of the benefits of

some of these Tech Stacks that have been around for a while um you know what kind of capabilities could could we expect or innovation possibilities I suppose could be endless but what kind of capabilities can we expect from um nonm or just other VMS um you know and why why why do is this happening why is there why do these alternate VMS offer better capabilities than um some of the rest yeah I'll break this down to two sections uh performance and security I think performance and descriptions was a great example of what chains can scale what chains can't um we saw a lot of if not every eum chain pretty much go down with the rise inscriptions um some

chains completely halted while other chains had like extremely high gas fees although inscriptions in Edge case a lot of people didn't take account it kind of highlights like in the world where you actually have a thousand 10,000 users on chain at the same time can the network handle it and when you look at what networks handle it it was salana handled it pretty well appu handle it pretty well um anything near and say handle it pretty well the common denominator there is like alternative VMS in the space where a you have paralyzation built into runtime so instead of um the evm which is single third of processing which you have to wait for transaction to finish each time you can have parallelization in that um transaction can execute concurrently um you have paralyzed databases um what manad is working on

and say's working on um so just able to handle users on chain is a huge benefit from a performance point of view because the eralization um specifically move has a block STM pattern which was designed by the DM project um Superior way to do paraliz and handle transaction threading on second part performance specifically move is the concept of dnic nfts um we're seeing this in gaming where s gaming has really taken off um where in the past the say like D5 kingdoms on the chains they were on the past they kind of broke the chains if not halted the chains because you're constantly minting and burning new entities so let's say for example a game actually gets a million users on chain in that world you have million users on chain you're going to have tens of millions of traits being

mint and burned in because there's no way to swap out traits at well with move and this virtu machine that can take these op codes um you can swap out traits that will so say for example you kill a thousand zombies um and you to upgrade the caliber on the gun or upgrade the skin on like for your individual you can do that within the smart contract itself instead of needing to um constantly mint and burn a new nft each time you update State um that's on the performance side so you have high throughput chains as well as customizations on Smart contracts um which enable um low fee and better ux purposes on security side is kind of where move Takes the Cake it was built with formal verification in mind for those who don't really get it formal verification is kind of what Auditors do

it's their job they forly verify each lineup code to make sure there's no bug there there's no rri attack there's no Inger overflow and even then it's not 100% U we saw that recently with the kyber attack the curve attack which are two blue chip protocols both re bugs and every week um it seems like there's a new attack the total number is like four and a half billion to the date um every year lost in attacks so if you're buying the thesis that crypto is going to be adopted a mass scale um it can't be done on a solidity or evm because it can't scale Bank of America will never put their assets on the evm because it just won't house their user fund safely even talking liquid funds they can't get insurance to completely back their like kind of investment because there's a

high smart contract risk so move introduces the concept of formal verification where every smart contract that's deployed goes through a formal check at runtime so checks for resource type and memory safety um blocks for in tax blocks flash loans um and some of other major scale attacks we're seeing so next nvms are fastest and most secure way to scale ethereum gotcha um rob you go ahead because after this I'd like to take I'd like to do a little bit of a uh build your own modular rollups and get R's opinion as to how we CFT yeah that's always a fun exercise um I did have like a quick uh like a quick tangent

um which is that you mentioned parallel parallelization as kind of the thing that enables High throughput chains um I'm curious how parallelization which is a very difficult word to say how is that like similar or different from Atomic composability um and I'll just kind of like say my understanding of it and then maybe you could like correct me where I'm wrong and then fill in any gaps I heard Charles from espresso speak about Atomic composability how it arrives at a more efficient prices much quicker um and this is done by essentially putting transactions into the same sequencer and then any kind of like uh coincidences

that happen to use cow swaps term uh cancel out and then those are able to uh eliminate any kind of settlement uh bloating if you will um parallelization seems to be very similar because you can run transactions in parallel um if those transactions have Coincidence of wants is there a way to like let those settle uh atomically or is uh trans actions in parallel not necessarily uh composable with one another yeah so the best way to explain this is atomically composable refers to multiple different blockchains so let's say you have like eight different rollups or 10 different rollups composability refers to like can you do swap between rob a Rob B Rob C

Rob d r f between different networks penalization refers to State between one specific Network mostly refers to database so it looks at read read and write um calls between a specific chain so let's look at um salana for example salana is not automically composable with other chains because you need a wormhole you a third party bridge to bridge assets between Solana and ethereum but salana is paralyzed and can execute on state in a paralyzed runtime because um you can have like a thousand transactions all occur the same time um and it's composable within a specific Network so Atomic composable refers to multiple Network while paration refers to state within a specific chain that makes sense um is is

paralyzation is that the end off be all like this been the the hot turn but like is it is it is it is it really that powerful I mean and if so like is it just matter of of like the local fee markets or I know SE is different than monad than other here other BMS but is it just local fee markets or is there other aspects that make paraliz uh better other than just a cost or basic through which are ultim kind of the same thing lower cost get more through same thing is that just it or is there more to it yeah so paralyzation itself doesn't do much I would say it's like a 5x

increase or 6X increase which if you're doing 10 TPS bring you like 50 TPS um the real adjustments and modifications um Destin push is the DB so um on the apos and suicide they use like rocks DB which a modifi version designed for the block STM um pattern um and is also like how you integrate with consensus mechanisms um so paration is just one piece of the pie it refers more to paralyzed like consensus paralyzed database um paralyzation as a whole is like a system um paration the VM itself it's just one small small piece um which isn't that difficult to do it just refers to having multiple threads within Rich machine um the real bottom neck is on the database um which monad

specifically focuses on in that's why um they're able to do like the paralyzed EPM um the way we look at it is we kind of inherit a lot of the benefits of appos in the DM virtual machine um so we have our own modified version of Rox DB um which can handle the High throughput um and can have all the benefits that mon DB does so it's like equivalent we handle evm equivalency through evm runtime while monad and say kind of build it directly into the VM um so it's kind of two different implementations on the hand I got one more question before we go into the build your own modular roll up exercise and this is a question I got from a programmer friend of mine a couple days ago at a bar and he was asking me like first principles of of blockchain he's a very smart guy but I

think he was actually testing like why blockchain even exists and you mentioned database and like from a programmer's perspective like first off like how is a blockchain because okay but how is a blockchain different from these databases like you mentioned that databases are baked into some of these chains like how is a chain different from a database and then why is that like that onchain component important and necessary yeah so databases store more local memory so they store like read and write calls um specific transaction um the actual like transaction history is to the call data um which is stored directly to like

Celestial or like ethereum main net so the kind of difference here is like you can use an Excel spreadsheet instead of Celestia um for like da but it's very expensive to make read and write requests it's very slow it's inefficient it's not designed for financial applications blockchain's from a financial point of view is the best distributor Ledger in that you can send money send transactions um at the speed of light essentially in certain networks um and do so in a manner that's open source um you can trace history of it um there's no like ball neck in terms of how many transactions you can do like in theory ethereum can go on for decad is um and it's not like a centralized server that Google is running or Amazon's running um and I think the database is a very small part of it so

you not only do you have like a database but you also have ethereum has like thousands of validators like salana has thousands of validators you have a better da layer that can um handle like sampling in a way that like self spreadsheet couldn't um so long story answer here is a database is like a short-term solution while blockchain is the best way to do um payments or store um data on chain um as quickly as possible cool um to Tech some technical understandings there which is which is always good coming back to Rob's First maybe second question which is this idea of being friends with all these different potential partners for movement so before we start building a modular rollup what I'd like to do is

just understand a couple things um so I guess first can the move VM be uh integrated with every Ras provider currently yep we can be integrated to pretty much every Ras provider we designed the VM to be complete monitor um we first bring it to Avalanche um which was M1 which I'll get more into that later and then brought it to the the um so it redesign it so it can be picked up and brought by any Ras by any dealer um we have Partnerships with all the biggest dealers and we're announcing some pretty good role staff providers as well so it doesn't matter if it's like CK roll up up if it's whatever it is the BM layer can be yep you just have to like kind of build a that like

support for it so you have to build like clients or you need to build out like the rpcs um but it's not a too difficult process and same with the of course uh and then what about settlement layer could could we have move rollups settle to saana or to Avalanche or to near or to bitcoin yep I've seen actually seen a move Bitcoin impementation recently um so yeah any sement layer Works uh I would actually be interested to see actually move up on bitcoin because that could be good from security point of view but um haven't seen much work been done there um but yeah any s lay works okay cool so let's assume that I'm a I'm I'm a game developer and I'm set

on using the move VM uh yep neutrally right assuming you're you have no private interest in anything other than just building the best environment to build a game what what uh how are you building a modular rollup for a game using what settlement layer using what ta what what type of roll up how are you going to build this game B up well I think the first thing that's more important is that you build a good game first uh like build up Community like build up actual game before you worry about what blockchain it's on um so that's where I'd start and then once you have that game and ecosystem and now you're ready to use blockchain um and

you want to use move which we would encourage for gaming um you probably want to use etherum for settlement um just the best storage of value at the moment um and you can have liquidity for your NFD marketplaces and actually have users on chain um on the da side it's kind of where gets more interesting um because Celestia is obviously the only one that's live right now you 7ye da which in theory is cheaper um and then you have like IG and da coming out um I think IG and da is particularly interesting because in theory can hand one gigabyte per second um which is more designed for high through rich machines so we're doing a lot of benchmarking internally on comparing Celestia versus a implementation versus an Avail implementation um and it seems like if a can hit the benchmarks um proposing to be a great candidate for gaming

applications um you also get the security from native Ethan it's not a different consens mechanism on the RO stack side I think a ZK up would be the best because you want inent withdrawal times you don't want to wait long time um so that kind of like narrows it down to the polyon CK um sovere Decay um and I'm sure like a few other Stacks are coming up so a bet on the new BM would theoretically be a bet on somebody to build a great game for example and thus to put it onto the move VM or build it with the new BM under it as as the BM layer which would then a ton of value to

basically the this the idea of the technology behind the mm like like like from from your perspective you should be incentivizing and encouraging as many Builders to build great on Mia yep so gaming is one aspect I think we're focused mainly defi right now um because in order to have a successful chain you need liquidity on chain you need users that kind D out um so we build our core def system which we're announcing like every day basically and gaming is a huge Focus I think one of the benefits that we have is that we have like two giants behind our shoulders um that both have substantial amount of capital um and I know both E Systems are pulling a lot of money into gaming specifically in Southeast Asia Korea um s is done a great job andap is done a great job of

of reaching Korea adoption um unfortunately we get the benefit of like those fruits where U we work pretty closely with L systems um it's generally pretty friendly um and we can kind of work together and move adoption yeah yeah and I asked just this one on the base on the back of that one specifically for move it's interesting that AOS andu are obviously involved at that at that level it's quite it's quite a nice uh Advantage but with regards to um value of cruel in the modular stack like how are you thinking about uh like obviously celestius price number go up right but like how are you thinking about like I sent a transaction onto this modular rope that we just built on

this game like each part of the stack is somehow AC crewing some sort of value which also then theoretically gets ACR back to the user because they're getting a better experience better lower fees better ux but like like what else what what are we missing like where is the value for the mo stack what's the business model for the modular stack INF pieces like how are we thinking about like comparatively valuing these things yes so like the business model for like traditions like arbit optimism has always been sequencer level so you charge sequencer fees for the evm um and that's say make money it's a pretty decent business model like they're printing money essentially um but the problem there is when you now you choose

next chbms and the thesis is low fees so with the salov VM with the move VM you have low fees because you have paraliz run times you have localized fee markets um so the fees are actually pretty much next to zero which means the sequencer fees are pretty much nothing um so the value proposal actually goes away for next gen VMS in the ma thesis um a as da becomes more cheaper um sequencer fees actually get Pro progressively lower um SL you can look like manta's fees um their sequence of fees went down a lot in a good way because it's a better user experience for users you're not paying $5 for transactions but the business model slowly diminishes what's kind of my thesis and how we're positioning is the value actually occurs to the Token at a sequency level so for context um we

have M1 which is um kind of ler one built on the sow consensus that turns into a decentralized sequencer a decentralized Shar sequencer to power or theum rollup um so what that means is if I'm a validor I can now stake a native token to participate in sequencing um and actually get a percentage of the fees so yeah even in this world where next gen VMS dominate um on ethereum which is the thes behind movement and the sequence of fees are pretty low those fees get distributed to all the valers and it's actually a value cool mechanism back to the Token because valers need to State the native token to participate in sequencing and you actually can make the decentralized sequencer permissionless eventually um so long story short is the decentralized sequencer is how I see the market

trending over the next two years as sequencer fees diminish is actually economic incentive for arbitrum optimism these rups to transition to a decentralized sequencer because now you can require thetive token to be sted to participate in sequencing so Stephen didn't want to tell us that Rob no he didn't he said it was a research task or problem he we were we were grilling Stephen on uh on arbitrum staking yeah and he said you know we were we were asking him like when when ARB staking and I my hypothesis at the time was that ARB staking if it were to occur it wouldn't occur at the protocol level it would occur above the protocol

level at maybe the application Level and that wouldn't be a concern of AR arbitrum Dow or offchain laps but what you're saying is that as sequencer fees uh are driven to zero then arbitrum staking actually would occur at the protocol level and arbitrum would transition into this uh proof of stake sequencer model if you will yeah I think arbitrum is a educated like arbit optimism obviously are very successful um it would need to take a lot for them to move it and I don't think it's on the road map I think have other priorities like EV Plus+ um and they're doing a good job on that end I think this more applies for next gen VMS like we're

talking about the fuels of the world the moves of the world sbms in the world they're bringing the next gen one time to ethereum or the fees are like zero basically so Arbitron fees will always have fees because evm um in the next two years they'll probably have fees so they don't have much of incentive to the centralized sequencers while movm svm they have low fees so it's more imperative that we use a decentralized sequence or staking model or have like a staking Hub so we actually have business um for a company okay I got one more one more question on the arbitrum uh the arbitrum framework specifically because you were on a panel in New York and I think I texted you after the panel and I was like hey uh arbitrum actually has this announcement with espresso that

they're going to use some share to can at sometime and your response was basically like that's never gonna happen so could you like just explain that a little bit more as far as like why arbitrum would would not make that transition to a shared sequence you're going to get me in trouble going to get me in trouble um but yeah I think it's just like it's it's about economic incentives right um arbitr is printing money Bas is printing money like there's no economic incentive for them to use a decentralized sequencer in the short to midterm um which is the frame of reference in a in a 20- year landscape 30 year landscape sure like I buy centralized sequencers or PLA decentralized sequencers and share sequencers will dominate the evm market as well as next gen VM markets but the

fact of the matters today is that it's so profitable to run a caliz sequencer in then near to short term or short to midterm it doesn't make sense for arbit optimism or any evm chain really to use a decentralized sequencer it does make a lot of sense for next gen VMS that prioritizing low fee markets to use share sequencers because you have business models baked into it um because you have token staking um and it's also like you need a share sequencer usually to handle the paraliz high through benefits um you can't rely on a single point of the um so my thesis is that arbit optimism even if they advertise it I think that was in reference arbitrum orbit or it's probably some kind of road map things we're probably not going to see that adopted arbitrum mainnet for a

very very long time I hope the expressor community doesn't kill me after this that's our next podcast so we'll we'll make sure uh the from them too yeah so back to the back to the the where we got going on this which is really good is that Valu crew so so so let's just get are fact straight so is Da the this rise of alt da is this causing a race to the bottom for sequencer fees is that's kind of what's happening is this gonna yeah significantly reduce all sequ fees yeah I mean you're seeing it already on like Twitter so you see near competing with celesti on sequence on

like um the da so right now near is expens like exponentially cheaper um than Celestia for da even though like Celestia can customize it and decrease if they want want to but they're not choosing to um so in this like two-year threeyear Horizon where you have four Live Da layers A they're all going to compete for these big Ro Brands Deploy on top of them so they'll change their fee models at so like attract um more users and then B it's going to be like okay if Celestia decreases their fees how does IA respond if IA responds how does a veil respond so it's going to be this kind of like jockeying between different deers I'm sure there's like five of the deers in the pipeline ready to Launch um and I've heard of a few um so in this

world where like da is all decreasing the cost now you're like okay what happens to the sequence ifuse if a transaction like let's say it was a $10 to a transaction e the main net Celestia makes that $2 because of their great da sampling and then it's a progressively lower fee structure now the amount of cost per transactions from $2 to150 to Dollar um it probably hits a lower bound somewhere um which makes a sequence of fees which are percentage of that dollar um exponentially less important um so you kind of it's kind of raised to the bottom in the da level as well as like the last level as well um which kind of renders the sequencer fees um longterm race to the bottom as well and also centralized sequencers

aren't the best as we've seen with downtime issues and whatnot okay and then so with the alternative DM thesis how is this similar to the DA situation or and and how is this different I understand the biding level with with the token and the stat for sequencing aspect so you maybe if you tie that into the answer how is the race to the bottom of ta costs similar or different and how is that affecting the way you're thinking about like the business model yeah the whole fees behind next BMS and if you go on like slon or opon swe the fees are literally like 0.001 um the thesis behind X BMS is localized fee markets um for like

consistent fee structures execution so there's no bottleneck and throughput um so you don't have crowded fee markets a great example is like during high peaks of like on use arbitrum GMX goes crazy which drives arbitrum gas fees goes crazy and a sequence or prints that doesn't happen in move orvm because um if there's an app that goes crazy it's the VM itself can handle the state float um so it doesn't pass it on um to the fee markets it just stays very cheap so you're always have consistent like next to nothing fee um which means the sequence itself is pretty unprofitable of a centralized sequencer um so I wouldn't say the VM space is a race to the bottom is is probably going to compete on like devs um it's going to compete on like Security on like speed

um that's s of competition it's not a race to the bottom because like move svm fuel they all are like pretty much next to nothing gas fees the token incentive model for the sequencer is kind of how you actually have business model and value apprel back to the Token because traditional Layer Two tokens are just governance models um not really any staking or gas fees um so we'll see a lot more exp experimentation coming there I think Stark Net's doing something interesting where they have like Stark tokens gas fees um and they have some kind of share sequencing Hub not too familiar with it but um heard a lot of great things yeah they got in trouble for uh not being a them aligned apparently which was uh kind of an interesting one

because they're very ethereum align I think what they are like the Fest out ZK rollup stack that allows them to have CYO VM which is actually a sleeper alt VM that nobody talks about for some reason maybe they're not like super involved in modular conversation but I think Stark's going to do well um well thank you for for answering that I'd like to take some time to dive into the move framework um what you guys are doing with Cosmos and IBC some of the road map items so maybe just take us away as as you see fit and we'll kind of jump in as we have questions yeah I think I bis of what we want to build is always permission us interop ability like I was building on apdus building onu um and like I wanted to try new Bridges I wanted to try new

oracles um but I couldn't do that in like a moic environment so with M2 and the move stack um we kind of have enable anyone to like customize their own infrastructure so for example we built the first move hyp Lane module so anyone let's say you're on Manta you're an eclipse you connect to M2 which is our ethereum rollup or connect to any move rollups um pretty permissionless you don't need like a third party bridge to do integration for you you can do it um in under an hour probably um and second thing is on IBC I think Cosmos is something that has been slept on for a while um it's kind of app chain thesis and the Hub has kind of discontinued it uh but now it's celesa has brought a new Vigor to the ecosystem um and with that I believe IBC is one of the most

powerful Technologies in space um connecting not only to the cosmos Hub but as well as players like noble which for usdc onchain so we're announcing the first move IBC module that connects um the broader moving system to the co Cosmic system um so we'll be powered by Union um who'll be building out um the IBC compatibility for a vir machine um which allows us a to be fully compatible with the cosic system B um have Noble's usdc onchain um on testet and Main net and then C can compatible with like players like stri which are liquid staking um Kepler leap and some of all the fun Cosmos stuff um that we GRE to love okay okay so you guys have like the M1 M2 maybe we can get get to that kind

of like separation in a moment um as we talked about the move framework um I think we briefly touched on this earlier we already kind of like unbundled the entire rollup stack and and this leads us to modularity so when it you know on once we accept like the the promise of like alternative VMS and what they do to roll up throughput why did you guys choose the move language specifically like what what is it about move that separates and empowers greater throughput more security than you know some of the other like options for VMS that you guys could have chosen I mean goes back to Facebook's stage right Facebook was going to build stable coin to power their billion users on Facebook

and they looked at evm I was like okay if I have a billion users with like tens of billions of dollars and it's living on this evm it's going to get hacked inevitably so they took a step back built a complete VM and language from scratch it was rust based language um so move is built like with all the security mind to handle a Facebook Style app on chain um so think of it like salana of virtu machine and your favorite auditor combined into 1bm um so the concept of formal verification is beaked into the rich machine you have built-in B code verifier and compiler so traditionally in smart contracts when you execute a smart contract there's no one's saying like stopping it if it's a flawed line so for example rein attacks are a huge thing the KY attack was like a attack

Vector where 0.0000 1% probability of being picked up which means it's just human error and fundamentally impossible for anyone to like pick up on it um that would have been completely stopped by move because the move runtime would have flagged it as like hey there's a re bug here um we don't allow re re you machine flag a compile time and don't execute it um so it kind of audits the code in real time because you have form verification built into the VM and the evm and other vers machines you have to go to auditor each time you want to update your code um which can work the first time it can work the second time but the 73rd time when you have a 73rd update maybe there's a line that you missed which just cost you $10 million um and just lost your entire user base I think Stars

Arena is a great example it was a social f app on AV lych um it was a good good team just wanted to push the product main net um and they got extreme Traction in three days and then ra stre bug crippled them and now they're nothing right um and they're trying to work on it again but the point is the BL the the velocity production in crypto is very slow in that if I have a killer idea if I want to build the next Facebook in my dorm it takes me really long time I need to build the code out spend months auditing it pay 100K to my favorite auditor and even then there's a good chance to get me hacked I think the numbers like 90% of every evm protocol is at risk of being attacked right now like Lio is pretty much the like Edge

case but even that's like unsure so move is designed to increase the velocity production you can deploy code to main net 95% of of all bugs are pretty much handled um and we haven't knock on with SE an attack yet uh we'll see how the thesis plays out over the next few years but move was built with security in mind yep totally makes sense there's you know there's SE several project out there building alternative VMS um makes total sense why you guys are are embedding formal verification into the alternative VM logic that'll H hopefully uh increase the velocity of of development um and speed up like how quickly these projects can go to market

and Deploy on mainnet which kind of brings me back to a thought I had a little bit earlier in this conversation uh which was that by uh enabling uh like essentially modularity by enabling modularity and also kind of like bringing all these things together you're able to open up like this design space for for developers that maybe aren't blockchain native so all they have to do like because because all of the onchain work is becoming much much easier it means that a lot of the work is now going offchain and that's where the Innovation is going to happen and that means that not blockchain uh like non-natives are able

to come in and develop because now a lot of that a lot of that creative work is happening offchain and that that's because onchain work is becoming much much easier yeah I think you have like much more customizations um like if I want to build a game like you touched on and I want a pariz run time in the past I couldn't just like Fork slana right uh but now I can go to my favorite R do one click rollup and now I have a move roll up for my own use case um which wasn't possible a year ago um so you have much more experimentation the space I do think modularity introduces more like complexities but it's going to be unraveling of technology and then like kind of coupling once like the winners are chosen yeah that's what we see almost every site is like everything

kind of like unravels and then there's like there's the bundlers and the unbundling on whether or not you want customization or you want convenience uh and these are trends that we see in like several Industries almost every cycle as things kind of dilate and compress and and we see those Trends and Rob you bring up a really good point about the onboarding of developers I think um I think that's one of the biggest you know unique uh aspects of movement and alternative DMS R I'd love to hear more about how that's going for you guys um and then we can dive back into like the M1 and M2 and kind of what's what what's what our kind of community and users can expect but

let's touch on how alternative BMS are onboarding devs um you know move fluent other other VMS kind of what's the overall thought process there why is it easier what's your thoughts on onboarding uavs how are you doing it um yeah I think our thesis is more line with like eclipses and SL thesis where we have like these two huge monic environments behind this like OPM we combined like1 15 billion do Network um it's like number six developer count move is growing exponentially in Korea sou East Asia Taiwan um and the bet is okay over the next three or four years we're going to see a lot of move development because of the Firepower of these two chains um a lot of users are going to want to live on ethereum still

live on Celestia be part of the ma thesis um OB compatible all these different benefits that you couldn't get on monic chains so that's how we kind of work with app and Su in that we're extending the umbrella move to environments that were typically untouched um and simly clips ex extending the umbrella svm to ethereum and whatnot fluent is like a interesting play because they're bringing new developers to the space they're bringing JavaScript developers typescript developers um and doing a good job of kind of increasing the programming language Horizon we're focused more on like a inheriting the work that two big giants have done and then B kind of focusing on increasing security mind share within the eum system um so converting rust developers move

developers um as well as move is a pretty easy language to pick up it's like I like to say like move is to rust as reactor to JavaScript um so it's much more intuitive um as design specific use case so we do a lot of University hackathons and um have been seeing a lot of good trajectory there nice yeah developing uh developer experience huge developer relationships is huge and love to see you guys go on the University route um that's where like a lot of crypto curious people are um and it seems like that's where the AL VM thesis fits uh is to on board these like crypto curious people that aren't necessarily natives um yep so I wanna I want to push into this M1 M2

conversation uh kind of like unpack these things for us un unbundle them what is M1 what is M2 and why did you decide to kind of like separate these two pieces of the stack yeah so this comes into like November of last year um when we first started thinking about modularity at the time like this is post FX and everyone was like down bad like every developer in the space kind of left pretty much especially movie system um at the time there's only two systems where you could build a rust Bas VM um it was like polka dot Cosmos and aych um and then we Avalanche it was a great time and we kind of like really realized this is where we wanted to boost sh liquidity um create decentralized Valor set um and then kind of gotten the modular space where it

made sense to like laun to move eel to and then as we kind of established road map we realize like all the sequencer stuff we discussed right in this world we have next CH VMS um coming to ethereum it's very difficult um to have centralized sequencers because the fees are going be next to nothing um centralized sequencers have large downtimes so if we're doing 10,000 transactions per second um which is different than the evm um use cases and the sequencer won't be able to handle that because it's a one no essentially sequencing all these transactions so we need a decentralized sequencer that could scale and specifically we need a a sequencer that had low hard requirements instant finality to handle the essential through that we're having and see have native um alternative token taking so that pretty much brings us down to like

tender and snowman consensus um and S consensus was the right place to do so M1 is a subnet on aage consensus we don't say the word subnet so we call L1 um and M2 is the theum rollup that Ed to move your machine M1 um then transitions in to M2 by being a decentralized sequencer so um every transaction that occurs at M2 main net um is sequenced by the validor as part of M1 so it's a decentralized Ser sequencer powering the the roll up um so it's kind of a one Mega Frankenstein experiment um and if it works it's going to be like probably the most decentralized and highest uptime next gen Ro here kind of visual is like you have a mov and for execution that's even

compatible you have a them for settlement you have M1 for sequencing um and you have SL for GA then anybody can build a top so so M1 is is like sequencing so it's not like like the settlement layer you can't build a movement roll up with M1 as a settlement layer can you so you can launch your own move rollup with move stack which means you can connect to whatever Ro framework you use UK stack op stack whatever use whatever dealer and you connect to M1 as a share sequencer so let's say you have M2 which the main chain and you want to launch your own gaming rollup you can launch your own gaming rollup with move stack use move VM use whatever other parts you want use whatever R connect to the decentralized sequencer so you have composability that we touched on um so

if you're doing a swap between M2 and you have your own order book chain um it's Atomic composability between different rollups it's actually more atomically composable because it's a paraly chain um and you're not waiting for a state to finish on each different chains um but yes if your launcher moveable up you can connect for a sequencing purposes but it wouldn't be for settlement purposes um you'd probably still settle on ethereum yeah makes sense and then interrupt is the other part of mon stack that is under discussed but will soon be discussed more as people are getting more involved you mentioned something to do with the Chads over at Union for IBC uh move interrupt could you take this take this a little bit deeper as to what

that looks like as well as some of the other uh interesting items on the road map which you've subtly hinted at today right before our lovely podcast yeah on the IBC side um the movie system has never been IC compatible really um so we're really looking to bring a the apos and suaps um bring them to the C atmosphere so let's say for example your deck and abdos now you can use osmosis assets you have Stak Tia into your decks um which a gives you a lot more volume a lot more users um building the first decks in appos was pretty painful and that there was like two assets on chain which is AP and like one other token that had next to no use users um this was after FTX of course um

and then as that scaled we're like okay the bottom neck for new chains and mo environments is very difficult to boost trap liquidity and boost trap assets that users want so with Mod modu chains um and IBC enablement now if you're launch your move decks you can have your move assets and whatever you want in that ecosystem but now you can have IBC assets natively you can have uscc via Noble natively um you can R off Rams from kado natively um and then where like you have native East because you're living on ethereum um so that's kind of the thesis like having permiss interoperability you can customize your oracles with the move stack you can customize whatever bridging Solutions you want customize or rollup stack um and ask for road map where we're at today we have deite Fully live you can check out doc Ms XYZ we have

incentivized test net coming up next month um and then main net shortly after that um and then we'll be rewarding community members if you know what I mean yes we do know what you mean r she rewarding community members indeed very H memes of course with memes with memes community members I own a py penguin that makes me a Community member I like being a pudy pig we get a lot of you guys haven't realized no comment we rob well RI thank you so much for uh coming on and uh and having a good good vibe putting up with this

panaman Wi-Fi I think the uh recording be be quite nice covered a lot here um and it's just exciting man I think Rob's comment at the very beginning just set the tone and I think it just continues to set the tone of the modular Vision the modular idea where it's just like not everybody can be friends but it's all just this composability and this kind of like interactive nature of just building things together and opening your mind to what's possible and reducing this Max maximal istic monolithic mindset and I'm super happy to continue to see how move plays uh movement plays the part in that and look forward to uh to the launch yeah appreciate it man thanks you

guys for having me on and if you e movement will be around the r will be around um good Prett say hi yes sir yes sir thanks for coming on rushi this was a blast thanks guys appreciate the time