Modular March EP 6: Modular Security with Movement Labs

Transcript

welcome to modular March I'm Andy this is Robbie and we are the rollup modular March is an educational experience about all things Moder [Music] enjoy [Music] [Applause] [Music] welcome back to modular March hosted by

the rollup in modular media I'm here with Robbie as well as rousi from movement Labs welcome to modul March rushi what's up guys pleasure to having me um how you guys doing good man finally recovered from Denver and you're just back on the conference circuit I just I I can't imagine so perhaps to you yeah it's always conferencing um yeah thanks for having me I think I'm G to go through kind of deck quickly kind explain what we're doing movement and then um kind of open it up for more discussion that works that sounds great let's do it um yeah so movement building M2 move Bas Z knowledge role for safe blockchains um

each kind agenda we'll walk through the glossery give a context problem solution architecture and the can of where the future is headed so M1 is our main decentralized sequencer Network basically the valers take the token to participate in sequencing for layer 2 M2 is the main product which is the move layer 2 on ethereum a VM is essentially operating system where smart contracts can run and exists on a blockchain the evm is piece of um the movement SD is the software kit to execute smart contracts the move VM um for those who aren't familiar it was designed by Facebook in the DM days um to essentially power uh the Facebook blockchain which is DM and Libra project it's SP off in ab and sui which were the

two main DM programing languages and a DM um projects and then our thesis is we wanted to bring the move VM to ethereum um I'll touch more on the benefits of move shortly fractal is our even compiler so we remain end to end even compatible and then zero knowledge proof zero knowledge proof um a DA layer is essentially a layer to call there too which cuts down um gas fees that we're seeing with main ethereum today um so so context every year there's four and half billion dollar loss of smart contract hacks to date there's been 7.6 billion US Dollars hacked so far um this increases the bearers entry because large learning curves $100,000 auditing fees the velocity production today is very very slow and that if I want to

build a consumer app in web 2 I could do that in a dorm like Mark Zuckerberg did um and build a new Facebook if I wanted to do that in crypto is very difficult to do so because I have onchain dollars um so if there's a issue with my code that can lead to millions of dollars being lost and losing my entire customers um so security is actually the most important thing um when designing perfect uh distributed systems here's a breakdown of kind of the source of smart contract acts um notably a lot of these are smart contract overflows tax um kind of things that can be PR prevented if you have um better resource safety type safety memory safety and smart contracts um again touch like move language both

by DM Libra um brings in resource oriented programming supports full verification um and he's some comparison between SL developers move developers I like to say that move developers is the first year salana where it took a few years to kind of kick up and kind of build up its brand I think move is actually sing faster a with the market and then B um developers especially gaming developers really love move so can we enable smart contract safe execution within ethereum can we be move to ethereum Yes um and that's how movement is designed a move based rollup on ethereum um essentially there's two main types of rollups you have optimistic rups which use fraud proofs you have challenge periods you have zero knowledge groups which use vality proof

and are fast our computation expensive um we are using a Zer r m to make some most sense because you have move execution enir safety etherum liquidity activity you have ZK proves for ability to prove on etherum settlement layers so we're using ZK fraud proofs right now and working on setting up the ZK mve BM Cas scale from there um giv the architecture so we have the movement s living on top of the main blockchain you have Pro marketplaces kind of providing the zero knowledge proofs for the rollup and you have brti Smart contracts and verifier smart contracts on L1 um so end to end um roll framewor just real quick Ria there um been hearing a lot about these decentralized proof

networks um when it comes to approving for for M2 like are you guys going to do a lot of that in house or are you planning to kind of like give kind of like succinct or jalud or some of these other teams the ability to bring in their Network as well we're working at r zero right now just because U that's the one that makes most sense for our architecture and the one that's progressed to mostly as recently um so Bonsai obviously for improving Network and then we have our um like ZK for generator from rer itself we're looking at other options as well but at the moment rer makes L sense for us gotcha so they basically have like an inhouse um Marketplace of provs and some sort of

like auction mechanism which enables the best most efficient execution yep I would say still pretty ear Bond still pretty early stage um there's a few other designs that are coming out like gute that can provide um kind Superior Pro Marketplace but I think those are a few months out um whereas we're looking for something right now gotta cool this is what SD we support both solidity and move contracts I think the biggest issue with next blockchains is they focus developers to just write their own language without providing backwards compatibility and Legacy compatibility for existing evm code um so that's why full end to end evm compatibility so we call paralyzed mm um

or we support move and evm at the same time how the evm transpire works you have a SL script you par it to abstract syntax tree which is essentially as um breakdown language and then you go to inter language called Ule through semantic analysis you map that to move op codes long story short you take City script have 145 op codes M up the 54 move op codes and then run the move by code the BM so you have S script mapping down to move by code enabling it to paral as well as how the Prov is built in so the decentral S I think most El are like kind of flaw in this network today where centralized sequencers

simply don't scale um we will have a decentralized sequencer off the bat um this is because we want robustness we don't want Central points of failure we want latency protection protection M censorship resistance um and then two key questions here how is sequence reach agreement how is sequence chosen um we used to know make sense consensus which comes from the a system which is a leaderless bft consensus mechanism um Nakamoto consensus it's a snow family consensus it's similar to Avan consensus but we have total ordering built into our consensus model and it's completely separated from execution so it's only used for sequencing yep so I think one of the main benefits is like all these L to are

kind of printing money with centralized sequencers and there's no incentive for them to ever um be centralized sequencer which is generally bearish um like decentralized sequence there a bdb play what M1 does is it's built into layer to itself like optimism is already do this with super chain arbit is working on this as well Z is working on this as well every major layer 2 is going to have their own decentralized sequencer for the sequencers the Val take the token to participate in sequencing um so that you have like just nothing at stake problem because it works like L1 where you have slashing available um while providing security for the layer two um so having economic incentives built into the lud itself yeah because the idea with arbitrum and optimism and other kind of l2s I think starnet kind of push

this a little bit now with you're able to use Stark for gas and they have staking but like basically up to this point there's been just governance tokens as L2 tokens which don't really serve much which are like meme coins um because they don't actually do much so in order to have some token utility I think staking and um eventually fees is important yeah kind of what sequence network is built up you have D so the sequence Network Celestia um you have full no compatibility for the sequencers and you have provs that can corect the sequencer so um one thing we're also working on is mol monitoring so is a world where um you can actually use sequencer to monitor me pool if there a tax coming

into the meol we can actually like reorder the transactions um and slash uh like hacker attacks or slash rebacks in the me pool itself so you have additional protection even beyond the VM itself um something ZK Marketplace obviously the pro generary proofs um the issue is that it's competition expensive um right now current architecture using p for the movm um we like the Z like the current devet in testet is going to be Z fraud proofs which is inter step but longterm we're looking at a PL implementation for the movm um which has a cursor proof Generation Um but that's more of like a R&D project in the short to midterm it's just going to be SEC FR proof impation R zero the layer One S

layers ethereum you're evaluating Bridge contract um you have the native move and E clients um walk through this so intermediate goal movement is m2's optimistic role using CK Pro proofs you have execution which is movement SDK which is mve PM you have sequencing with M1 eum settlement s da and you have ver zero for CK FR proofs where we're trying to off this with SDK we're trying to build out um more more fractal adapter so um we can keep up to date with all the ethereum standards that are being pushed every day you have M2 which be a fully end to end ZK rollup we're trying to U build a whole ZK move VM in house which is obviously competition expensive and pretty um ambitious um but something

we're dedicating some resources to you have move stack so we actually have a few rollups looking to launch your own move rollups norly even some within the barar chaining system which are pretty funny um so we'll have move rollups on ethereum as well as other chains um we game in which is a move r on B&B chain um so move will come everywhere um and the nvision is movement Network for M move base rollups and kind of where we're getting at is it's more than just infrastructure for us we believe in security that's our main narrative I think everyone today is pushing the paration narrative everyone saying speed speed speed my thesis is that's a Lamborghini without breaks um just because you have 20,000 TPS or 15,000 TPS the difference isn't 15K it's a 20K TPS really we're barely doing 2K TPS in

today's market um it's really who can provide the best user experience and who can stop the $4 and half billion dollars loss of smart contract hacks and move is the only VM on the market um that's remotely able to even address this and out of thes is we want to bring it to ethereum and stop the tax we're seeing today while providing user experience to builders yeah that's it sweet all right so just to clarify some of the kind of um design choices here so effectively the M1 is not movement trying to create its own settlement layer it's movement using snowman for shared sequencing for consensus but still settling to ethereum why did you choose ethereum instead of going like like the let's say like the

initia route or the other protocol route and design your own settlement layer as well I think one of the main benefits of or one of the reasons why we started movement is me and CER I built the first deck snaap us I co found build the first yield diator the biggest issues as building a you system is liquidity like acquiring liquidity which is why the layer two design on ethereum makes a lot of sense because you inherit decentralization you inherit liquidity of ethereum you can have native e bridge to the decks so for example the same Decks that launched on top of apas can Now launch top of us have native eth compatibility have native evm compatibility have usdc D Zero have um Tia atom IBC assets um so the decks can actually have valuable asses instead of

just like AP and test net um to um so the main is liquidity and like tokens actually matter yeah and like the infrastructure that comes with evm and of the ethereum development now uh one one thing here um as well uh with the uh kind of shared sequencing model um I guess we can just park it here um so how are you going to make money um is printing money optimism is printing money with your centralized sequencer how are you guys going to turn your sharing Network into a yeah sustainable model well I think all l2s are kind of

going to the bottom where da is getting obsessively cheaper alt da makes it really cheap so like Manta was able to cut down their fees 99% um and then like even with alt VMS that even comes down lower because now your realization so sequencer centralized sequencer fees never made sense for us because our fees is going be less than a cent um kind of the way that works is similar to layer one where you have the Valor s the token to participate in sequencing so it's actually a buy pressure um the valers need to have a token and eventually be a permissionless network so valers can save the token get rewards get sequencer fees and then use that to run more and more nodes um and as for kind of the gas level the percentage of fees will be distributed back to the baders as well

as the foundation will able to retain some P so l2s as a business model are like that that era is kind of coming to somewhat up an end as far as like this like extremely profitable like retroactive optimism grants Etc well no I think like op like retroactive optimism Grant is completely separate like I optimism done a great job instead of incentivizing Builders ining creators I think what's coming to an end is the $5 $6 uh fees that you're seeing on ler 2os um people want to have fees that less than five cents less than two cents what they're seeing on Sal and S um and our thesis is we want to bring into

etherum gotcha I think Rob's got something my just last one before Rob takes us on a math journey is um are you guys really a rollup if you're using Celestia for da yeah I think that the word L2 rollup is so like mistake So Gone um I try to say a vote out of the debate I think it doesn't really matter what you call yourself um I like to call myself a rollup if you don't want to call me a roll up you want to call me Blum you want to call me L2 call me whatever you want um I think most people are kind of coming up to the word up is more kind of universal L2 is probably just a main net like rollup on ethereum um but there's so different definitions today I don't think it makes sense to kind of

categorize anymore rushi what is what is the counterargument here like what's the argument for the evm that evm proponents are circulating at the moment yeah I think like the evm will always have place today right it's like Legacy code if you want to a if you want to use um like Lio evm is probably the best place to do it um but we're not optimizing for the Thousand users that we have in Chan today we're optimizing for the million users that we want eventually when those millions of users come when my mom uses crypto the biggest thing that like people talk about is oh doesn't crypto get hacked isn't it scammy isn't it rugy um and how we can solve against that is making it more

secure for users and Builders to both use the same time a for Builders to ship products faster and users to not get hacked um so the narrative is really if we want the million users to come it's not going to be evm if you want to still use the Blue Chip defi apps um whe there is liquidity is Legacy code um that's probably going to stay in the evm where most liquidity will be um so kind our thesis is combining best of both RS we will shill and support and encourage the growth of move over the next four to five years but the next six to 12 months we still acknowledge the evm has liquidity and network effects and the most important Network right now um so we support both of them okay let's say that we see more of these Legacy defi apps continue to graduate

into the app chain thesis like we've seen with DX and GM now GMX like you know let's say a and Lio decide to also go that route what does that mean for the evm yeah if they have custom like uh app chain if all these Blue Chip protocols I really that makes it like more bearish that the EDM actually exists uh and like major layers to still Network effects um because eventually like some of these protocols going to realize that it's better to launch like a move app chain SCM app chain um the infrastructure right now is still pretty Bare Bones but I think once us the clips and a few groups are able to really launch over the next six to 12 months um and provide Ro Frameworks and compete with OP stacks and the U Kev orbitum stacks and whatnot

will actually be able to a be complimentary op stack so you can use a movie on op stack you can use um svm on arbitrum stack um which allows these uh kind Builders to have their own Clarity I think M like maker was looking at launching their own umm Fork um so they can now do that on lay or two instead which is exciting yeah I think um another another interesting point there is another interesting thing here is from what you've mentioned is is these kind of developments of these different stacks and Frameworks I'm curious if you think that the majority of general purpose rollups will effectively be forced to launch their own rollup stack Frameworks

as a means to generate value or to remain relevant in in the marketplace or if there's another reasoning as to like why it makes sense I think there's like a few different arguments I think scroll is focused mainly on layer 2 building on D5 layer 2 um I think that's like there's G be a few more like different interest initi is focus on real forwork bringing rups to their ecosystem we're predominantly right now focused on Art layer to and building out the E system first um so it's actually very complimentary Stacks like if you want to launch your own Ro framework you probably go to Ras provider because the infrastructure is best built um with like conduit Nisha CER these different groups for us it were focused on just being like general purpose layer too really going for like the arbit optimism

Playbook um where we have TV on the Chain have like a place people Bridge assets to um and have a lot of tvl on and I have a liquidity Hub I think like there's going to be some groups that want up Stacks like scroll right now I don't think is but obviously ZK s ZK stack um so kind of it's going to be a competition of who can a be the best layer to a general purpose Ro up and some people differentiate on the RO stack yeah with that I think that the the rollup ecosystem is so nent that there's almost been like a little bit of like a forgotten Quest here which is like Ros is trying to scale ethereum to make the ethereum ecosystem as kind of Broad and wide and and impactful in the

world as can be and right now it's almost like there's like a bit of like PVP happening amongst rollups you've got like blast sucking up all this tvl and all this ethereum main net e being bridged to these rollups and for some of it being locked like in Manson's case for the campaign some of it being just sticky because of the desire for these users to use applications I'm curious about like your thoughts with regards to where the rollup ecosystem is now and is there a misalignment of like kind of incentives or ideas versus like where we might be in a in a world of kind more let's say abstraction of chain of users knowing which rollup or chain that

they're on and kind of how the incentives might kind of shift there I think like right now there's a lot of like l2s and there different playbooks you have like the Blasta Playbook where you're locking up eth um and providing like essentially yield um for Builders and users as a service while like other groups like what we're doing is providing user experience and kind like trying to change uh like the evm and gas fees that we've seen in the past so there's G be a few like categories are come up you're going to have like yield bearing l2s which like the Manta um blast of the worlds you have like the main Eudes like optimism and arbitrum um which has the most like tvl and kind of the early Pioneers you'll have the nextg lud which is like

us and eclipse you'll have like probably some gaming l2s like Z games um and like a few the gaming l2s are coming out like immutable um so there'll be a bunch of different use case specific L and bu of like app spe like app specific rups like AO the question now is like how do you bridge all those together and have like us experience so it's like going between tabs of my browser I think there's gonna be some like I know near is working at chain abstraction which is pretty interesting um like the topic of chain abstraction is going to be really important like being able to go between rollups Within one click um is easier have than done you have automic composability you have different token standards um I think like Omni is working something similar like that um but I think the next Playbook after you

have all these rollups is create an internet browser um so it' be cool to see like a team come on like a Google Chrome for the maer stack yeah yeah that's that's kind of what I was getting and I think that a lot of these are kind of like fragmented now um kind of a spicy one why would I choose to build on movement when I can build on e Clips I can build on a different stack yeah so I think the main value there's two main value props a we fullly even compatible um so if you want to have paraliz if you want to um have like high throughput and you want to keep even code um our stack is pretty much the only paralyzed L to the market at the moment um and the second version is more security side um so move kind of

differentiates from SPM such that it stops the $4 and half billion dollars loss every year um it really is FOC on security and driving home um formal verification narrative if you look at um like how long it takes to do account Checks In Salon salana takes like two pages of code s takes like 10 lines of code and the evm takes like five to six pages um so even to like write basic simple code it's much much simpler and the last bit is like move was built by Facebook it was built by um the biggest consumer app in the world if we are to bring in new web to developers are going to come to the moving system just because after some s the next six to 12 months into driving home um adoption from web Tu backgrounds um which is pretty much the

only EOS system right now that's able to do that with your um rupture ties and traditional Tech ties yep and then more on on the why as to why you decided to build using move VM I think that that we can agree that a lot of The Da U layers will kind of be commoditized in a sense all the way down to like the cheapest fees possible um and that a lot of the value in this modular stack could theoretically ACR to the best execution layers um and and the best kind of performant VMS but what's to stop this the same thing effectively happening at the at the VM level um in terms of this kind of commoditization and how you thinking about

this are you saying like what all the VMS will be like the same eventually yeah like like like you know like in the next couple years somebody might just come up with the next best one right and just keep kind of going yeah I think there's always going to be like better programming languages ever like better programming systems if you look at web 2 that's exactly what happened you JavaScript which is like Ross this analogy and you react VI angular out that's like the movm svm and other groups um and there probably be a better JavaScript framework that comes up the question is how do you drive adoption for it and how do you get a market so any like Aon wasi can like like create a language and then promote it but it takes a lot of resources it takes a big Network it takes a big Community um to

drive home the adoption for the language so I think move is like the next iteration afterm and we haven't really seen um any real competitors come up to the movie system I think it's pretty much like I wouldn't say perfect but like it's hard to beat the move pring language it's you have the speed you have the security um there's really no areas of issues um that a new program language could take over um but I'm sure there'll be like other iterations in the next two three years that'll claim toly better than move and it's going to be depending on like can they drive adoption can they bring home users there's been like obviously like you had flow blockchain which had their own program language you had um Fuel and some other groups that are trying their own programing languages it's always

difficult to drive adoption for those respective things um it really takes a a full system effort yeah I mean from my understanding sway is a bit more purpose built for blockchains than than move it's h to I I don't know if that's true um they might tell you that but a move was designed to be built for basic blockchain it was designed for the DM and Libra project I'm pretty sure that Sway and um language for FL blockchain was a derivative of the DM project and of the move language um don't quote me on that but either way I think sway is good language R based language um excited see in the launch and see what

happens with a system what are some of the applications that you think will drive MTH adoption I think social is like the main thing like for us that's like the six-year Playbook like Facebook built this language our narrative is okay Facebook built this language we should prioritize social consumer apps um in the shortterm it's Defi and building a secure def apps so a lot of our interest in TV can on chain is family offices that have a lot of e couldn't commit in the past because of smart Contra risk and now they can deploy on top of l two restake it with re Ren one of these providers earn yield on it and not have to worry about it getting hacked um which is really exciting for um those

providers so in the short term D5 security um and secure D5 strategies and longterm I think consumer and um social apps is what's the most important thing so like having stable coin that you can send money throughout um different soci shops um having um like onchain gaming is exciting um but I think that we're still a few years out that yep um I kind of was going to ask my last one before Rob might have one which again is just another spice but why do we need another general purpose L2 that's a great question um I think and like it's like the same thing Web

Two happened in internet.com bubble you like millions of websites they're like a bunch of different websites all Forks like Amazon all Forks of like Walmart um and most of them died I that's what's happening with we infrastructure right now you're going to have all these thousands of rollups this unraveling of tech saxs and it's only like five actually win and in the next bare Market probably all these RS that you're seeing are probably going to not make it or kind of phase out and there's going to be like five strong teams that are able to a manage your finances but B um manage existing Community to withstand the bear and that's actually why bare markers are good because they wipe out um the plethora of in conru AC and only the teams that Mak sense can actually

scale so if we switch gears a bit to the application layer like what are some of the applications that inspire you the most and that you can envision normies your mom grandma's using on the on chain I think going back to the thesis for the DM project it was the ability set stable coin on Facebook imagine like instead of having to wire money to from like New York to Thailand you could press on someone's um Twitter bio or Facebook bio and send money with like crypto in the back end St coins in the back end I think that's the endgame vision for blockchains that you can abstract away Banks abstract away centralized authorities um so something that get me excited is when my mom uses Facebook my family uses Facebook instead of having to use like a third party bank

wire which is expensive sending money to India sending money to third party countries and whatnot U we can now have it within the app itself so if I want to send money to the family I can press on the username wire them 10 CC but in the front it just shows like 10 bucks um nearest sentiment no fees um and just having security within the app end to end so onchain social onchain payments um and com buying that gets me really excited got it so this is like a an iteration of Facebook marketplace where someone goes to buy a couch from someone else they have to bring some cash but no they can just kind of like put almost like put funds in escrow and

then unlock it if they are happy with the couch yep exactly very cool yeah and that's a line Rob with what Chia Wang was saying about this kind of barbell where the main use case for crypto right now is stable coin payments and the other is like aggressive speculation which is pretty interesting R I guess like the last one from us is going to be about this kind of modular expansion road map I think a lot of people are only aware of like Celestia and kind of like this D a concept but I think a lot of people will be very surprised at this rise of these alternative VMS or nextg VMS or nonm rollups launching how do you see this like modular expansion happening

throughout the rest of the year and then into 2025 with regards to each part of the stat kind of blossoming I mean I called this like like three four months ago 2023 was the year of the da 2024 is the year the VM um we believe that the evm K gets dethrown starts to get dethrown with the pushes the move VM svm alt VMS people using different execution layers I think it started with the salonic Rays people even like jiging to meme coins and seeing like how easy it was um send transactions on chain I'm not worry about gas fees I think the transition the ethereum E system as well where people can use different um BMS and use different um text Stacks to scale uh user apps on top of ethereum so I think that's the direction that headed

in and that brings in new consumer use cases that probably starts up like meme coins um like random like mem coins on top of next gen on top of ethereum and people realize that instead of paying 20 30 bucks for gas on ethereum you can now pay like sents on the dollar um and have better user experience in that transition in the games um even defi um I think that's exciting for the next six to 12 months and I know I said that was my last one but I just have one last one so we move so DA has has is happening SL has happened this this nonm rise is going to happen but similar to other RS Eclipse Etc still movement settles to ethereum so um

question is is is is modularity is this rise of modular blockchains is is this like this bearish for ethereum or or is it just further solidify its it's its thesis as like the global sediment layer or like the most secure siment layer possible for for all modular chains I think it's it's bullish for a the because it's still settlement ler it's still like the safe value of money people still want it for decentralization security um so it is bearish yeah fees decrease with uh da fees um and all BMS long term that might mean ethereum itself is making less money um but the thesis is we're not optimizing for like a thousand users making a million dollars we're

optimizing for the billion users that make tens of billions of dollars trillions of dollars on a chain in that world U where we have altas and ALT BMS that you want the user experience I argue the people who made tcpip who made the the fundamentals of the internet made a lot of money and still are profitable today because everything's built on top of them um so that scale it is really important Forum to adapt to the allas got one more rushi is uh ethereum money or is ethereum the world computer I think it be both um I I would I wouldn't say it's money like if I want to send money to my mom I'm not going to send her eth I'll probably send her USC or like some structur way um so like in

like a long-term meta like 10 year 20- year landscape I don't think it is money I think it's like an investment vehicle so people will invest in eth as a index for crypto and ethereum markets um but like I'm not going to send money to I like each to my friends really unless I'm like aing into mem coins um like it it will also some stable coins and other like similar vehicles I think it becomes an internet computer where people settle on top of it um use it for settlement use it as kind of a core infrastructure layer um but like I'm not so sure about this the stable like using as money maybe as investment vehicle but um wouldn't say it's like a daily currency got it and uh last one with uh 2024 being the year of the VM any

thoughts on bit VM I'm honestly not too familiar with it I don't really understand it it's not in my space um it's cool to see programmability on bitcoin I think I'm yet to see any use cases for it that besides like generating yield on bitcoin which is a great use case um like currently Bitcoin sitting Idol if you can provide some yield to it and some def that's cool um but beyond that I haven't seen much application um I'm excited to see that but I don't know anything about it yet yes still very early thank you RI for coming on being a part March and uh best of luck in London yeah thanks man excited have a ma March