Fireside Chat with Rushi Manche, Movement Labs, and Kartik Talwar, ETHGlobal I Pragma San Fran 2024

[Music] let's get ready for our very first session and for this I'd like to bring on rousi from movement please give RI a big round of applause we'll just make sure the mic's on hopefully it's on there we go I was prematurely doing that so want to cover a few things about what movement is and how you're thinking about the future of developer Tools in just smart contracts but uh one of the interesting things to me is that you you kind of recently why recently in the last couple years everything started you dropped out you raised a lot of money for the company and uh it's been a crazy intense period of uh starting movement so tell us about how you got inspired to do this thing and what was the founding story of you getting to where you are now I mean we're in the bay in Silicon Valley it's a similar story to the other uh Dropout stories but yeah I was always interested in distribut systems I was the youngest High United Health Group um had a backround databases um then start early on like Cosmos ecosystem um wasn't actually a big ethereum guy because I was always like I wasn't good at solity I was actually really bad at writing solidity um I kind of hated writing City so I started writing rust um and I think we're trying seeing the market tend trend tours that um but yeah November 2022 I was in college of vbill I was like a broke college kid um and this article came out that said like Facebook was building a new language called move um and their thesis was like ethereum was good in theory but it couldn't scale to the point of which Facebook itself could use it uh so they built a new language of VM um called move ver machine now aberson swe are the main l1s our thesis was we believe in ethereum decentralization We Believe the principles but we believe alternative execution environments can scale um at the time you obviously had optimism in arbitrum but no one was talking about high through execution and having son like ux on ethereum security um so we started building movement um the first move VM later to ethereum so you can think of it like apas Su salana on ethereum security um having all the benefits of ethereum liquidity and e as asset class which is why we're all here um with the benefits of a high through execution layer um and I think we've come long way and super excited to be here awesome I think you kind of set the stage for all the other questions going to ask um which is was really nice because we get to immediately flow into it um I also want to make sure that anytime we see a new terminology we going of get to explain that so let's kind of cover what an alternate execution environment is and uh what are the nuances here that we have possibly heard of and how movement defines it I like to say I'm the most unaligned layer two um I was at permission list and was on the pay with all the other ler to Founders and everyone else has like use e theorem for da and uses evm um which is what everyone tells you to do when you're building ethereum rollup but we're entering this part of the market where people are kind of Fed Up per se uh with the evm um you look at what salana is doing you look what S is doing they have like tens of thousands of transaction per second in theory averaging a few thousand while ethereum layer 2 is are tapping out like 100 TPS um so it's clearly discrepancy between like what ethereum principles are and what use cases demand um if we ever are to opt for a world where we're actually having fully onchain gaming or fully onchain order books um the current state of the evm which is a single threaded machine which means let's say you have 100 transactions you're waiting for transaction A to B to C to D to finish um which is cool if you have like a and like 10 users but when you're trying to put Facebook on chain it's pretty problematic um so the two main virtu machines are execution environments which is where you deploy smart contracts um where the Salo Rich machine and the move Rich machine um Sal Rich machine obviously for the Salo ecosystem and the move VM um was designed from the DM project at Facebook um so highr execution simply means when you execute smart contracts you don't have to wait for each contract to finish or each transaction to finish you can execute transactions concurrently um that enables you to have higher throughput faster execution um and a bunch of finality guarantees which can get into awesome so we kind of know that Facebook doesn't have a blockchain project right now because we don't see it and uh a big part of that whole movement was hey well we're going to create a a new way to think about this we're going to think about scalability and safety and that's where our move came from as as you kind of brought up um but we didn't really see a great future for move from Facebook what made you excited about move specifically as a language and kind of walk us through the decision- making of like what are the nice parts that others ignored or did not or overlooked um that made you pick move as the the place to start everything on well for context uh Facebook wanted to do the language do the blockchain it was actually regulation like the SEC was like no you can't do this this is like fraud or they've had a bunch of lawsuits for the DM blockchain Zuck was like ultimately like okay we can't do this right um so like the whole DM program that like billion dollars research some of the best research encrypto today is coming from that original research group U it wasn't up to Facebook to disband the project it was like gendler For Better or For Worse I think we all have opinions but I don't want to get sued um but my point is like Facebook had a vision they couldn't execute on it but we are open source decentralized platform and we can bring that Vision to life uh the vision is high through execution the ability to bring billions of people on chain without actually having infrastructure botx but more importantly security um I think security is a concept that goes underlooked there's $4 half billion dollars lost especially in ethereum system almost daily there's a attack between arbitrum base optimism layer ones we've all probably been hacked scammed rugged at some point um the current state of blockchains is so insecure where LPS funds institutions really can't do much on chain defi action because there's risk of being hacked look at KY look at Curve look at sometimes a um there's even like tax and arbitrum pretty much every week at this point um the evm isn't designed to be secure um so the mve VM brings form verification um which is also part of the thee road map eventually um at runtime so when you execute smart contracts um the ability to trust that they're formally verified this type safety memory safety is very important um so let's dig into that more because I want to make sure that we kind of cover how security is defined here so a big criticism of solidity and what kind of seeing smart contracts is they get hacked because somebody forgot to handle an edge case and somebody found a loophole around it and they stole some money um you're saying that's not possible in move well anything is possible Right Move is not a perfect solution the what it comes down to is like I'm a bad solity developer when I was writing solidity I can't write perfect code I'm sure there's phenomenal developers here like PhD researchers you will probably write perfect code but for the average developer coming into the space it's very very difficult to write perfect code um there's always like indal over flows if you look at the kyber attack it was like a 0.001% attack Vector where they for one lineer code it was like 257 bytes instead of 256 um so simple rounding errors indg overflows can cause millions of dollars in hacks um and for the average developer and even like like products that are blue chip that have the 50th uh commit commitment of code isn't formally verified um so what move does it has aover runtime that checks each line of code for resource type and memory safety so when he exu code it'll be like hey this module has integer overflow you probably shouldn't compile this and it'll flag it whereas in slity the integer overflow gets compiled gets deploy in main net and then this North Korean hacker looks at it and like hey I can exploit this um so in the evm it's pretty easy to see when someone messed up and then go all in on it whereas with move and rust based programming these modules aren't possible um so you can think of it like an auto at runtime it's not going to be perfect but there's over $2.7 billion Nation security move and zero hacks to date um knock on wood which we're pretty proud of that's awesome and are there any other features have moved that actually simplifi this or actually help with um High throughput yeah so block STM is the main um optimistic pitalization engine um currently like manad if you're familiar and some of the other performing l1s are not pushing this research but BLM was built back in the DM days uh to bring optimistic paration look at the Sal ofm right now the main issue with their par ation is that if you're a developer and you have 10 transactions queued up you have to manage St kind of by yourself and there a lot of conflicts with optimistic paration when you're a developer you don't have to worry about how transaction manage state so you can deploy 10 10 transactions and they'll optimistically um paralyze if you will so tldr you have this paration engine within consensus and execution that enables you to have high through execution um not is of a 30,000 plus TPS but in production even if you do a th000 TPS that's records better than um etherum layer Tu today and and I guess the nice part here is that you just deploy it and that's part of the actual definition of the chain and the runtime so you don't have to worry about any of these optimizations I think move comes down to like devx right if you're a developer you shouldn't have to worry about hey is my contract going to get hacked um how is it going to manage conflicts with other transactions you should just worry about your actual app you should worry about being the biggest Tinder the biggest Uber the biggest Airbnb you shouldn't have to worry about your blockchain um so the DM Paradigm the move Paradigm is to just simplified de devx enable you go zero to one and have the Zuckerberg experience Zuckerberg was able to deploy um Facebook in two weeks and gets feedback that same app can't happen on ethereum today because you have to audited it takes like months to write proper code that's clean um and then if you do get hacked you lose your entire social capital overnight um so security is actually really important for enabling Financial applications at scale 100% agree um but I guess when you look at sort of entering this now with where we kind of have the benefit of hindsight how do you think about what does it mean to actually transition into this because we've seen solidity get more and more prominent there is anywhere from Best Practices now to just highly audited code that people get to import um and kind of now they're being presented an option to like reset everything learn a new language from scratch and hope that that ecosystem spirals or sorry like grows and doesn't spiral um is that a is that a tough choice for people what's going of been the feedback when you've presented this to a developer and how are actual developers thinking about starting a new way to think about coding well I would disagree that solidity growth is rapidly increasing um most of the apps that are popular to on salana or R based programming um I'm not it's like I'm not fighting you on this I'm I'm saying that it's the it's growing because there isn't a second option which which is still considered growth um but at the same time um it hasn't been the case that there has never been a second option we've seen a whole new plethora l1s in the last seven years introduced their own formal verification languages or kind of just new way to thinking about improving smart contract development we've kind of seen a plateau on that so uh granted move is a lot more flexible in in better ux but how have actually developers reacted to this if you look at the numbers move started off at three and a half% of monthly active developers today it's 7 and a half% and next year is going to be 13.2% svm developers have doubled since last year um while slity developers have remained stagnant I think it's a little bit increase um so across the board we're seeing this transition in global industry from rust space programming a because it's safer and better ux B obviously price action when sonana goes upu goes up people are like oh the tech gets better that's how it works um and C uh we're seeing transition from web to developers so if I'm a college kid if I'm in in the bay it's much more intuitive for me to pick up move a rust based programming a because I'm writing rust for like data science and AI models and B it's much more objectoriented it's EAS easier and looks more like JavaScript than SL does um so the pitch to web to developers is a lot easier so what we're seeing is a lot of migrations from like JavaScript and react developers to the move and svm system um where slity still bears that um C barrier to entry now that's fair because if if you're looking at exist brand new developers trying it then this is a much better and that's their default experience and and that totally um makes sense uh where does evm fit into into movement here you you said you were still a layer to um is there compatibility is there bridging what does that mean I think we all believe in ethereum as a Global Financial computer it has liquidity e as an asset class is meant to be around but I think there's a lot to be done on how do you make ethereum itself scale um I think it's proven that just slapping on the evm on ethereum doesn't provide much uxes you look at ethereum layer 2 landscape today there's not much Nuance between them and compared to salana it's losing the battle um now there's a new paradigm called alt VMS that are coming to Market which are bring alternative execution environments that can scale you have high through you have formal verification and you still leverage the security and as a class of eth um so the way we look at it is we believe in ethereum we believe as a Global Financial layer we believe this better execution of programming paradigms that can scale ethereum better than the old evm and even if you ask like the original evm folks they'll be like it wasn't designed to really scale it was designed more of the placeholder um so that's why you had alternative VMS to Market um and then I think the way we see it is like even compatibility is important we have even interpreter the best analogy I give is like evm and SL like a PHP where like early web two websites or whatever was built in PHP no one liked building in PHP not sure if anyone's done it um and then JavaScript can market and yet react and View and all these wonderful Frameworks and you could build a website in an hour compared to like phps God knows how long and difficult it was um similarly now you have rust based programming which is similar to JavaScript the two main Frameworks like react and view are move and slot L BM um so PHP or solidity will always have a place today you want to use curve you want to use a you want to use Unis swap that'll probably the best place to use it but if you ever actually want to use uh consumer apps that can scale to my mom uses it your mom uses it our family uses it it needs to be in a high through execution environment and a language that can support security um so I think both will exist but uh the mass adoption that we all claim to happen won't happen in the evm that makes L sense I just want to confirm something you you kind of mentioned when you said you are compatible with evm is that implying that you can still run an evm code contract on the Move yeah we have evm interpreter so you can write so script it's some like so laying on salana um it's not the best evm we'll say we're probably the worst evm in the market but the way we say is like if you really are stubborn you want to deploy your sludy smart contracts we'll still support you um you won't have the same security guarantee as the movm um so we urge most of developers right move but we still believe in backwards compatibility where we acknowledge that move is still nent rust is still nent um today but in a 5year bet we believe that's going to change but today it's still nent so we still want to provide some support for that awesome um we've kind of talked about scalability but do you kind of differentiate between SC scalability and modularity well I think modularity enables scalability right the ability to have alternative da for example right now if we were to do a paralyzed VM on ethereum mainnet it'd be pretty poor you'd have like 200 TPS because C is still expensive ethereum da is still slow that's why you need alternative Das like Celestia I da whatever um to enable you to have higher through execution um so we believe that modularity enables scalability you can have alternative VMS alternative Das alternative sequencers that enable you to have high throughput I think condo would just announc like G2 sequencer which enables you to have a beefier and thicker sequencer that can sequence transactions instead of having a smaller full node um but I think across the board we're seeing modularity everywhere like salana has like 50 layer twos now um where or not they agree um ethereum now has layer twos different Das um even Bitcoin has like layer twos and there's like Bitcoin Das now so across the board modularities enabling scalability but is there any direct uh Focus that you have at movement on on this yeah we're focusing on alternative VM right we're bringing the maer movm we have it on ethereum we have a team doing on bitcoin we have a team doing on to so it will be on multiple chain not just exclusive to uh to ethereum move in itself is only do ethereum but we have Team Cloud nexio that's part of the E system team clonic that's ton um so we're encouraging this language everywhere oh awesome um on Twitter recently we've seen a a lot of um fights so to speak on just kind of definition of what do it mean to be decentralized uh because you have to kind of or the arguments have been hey the higher the scalability aspect of of a chain the less possible for it to be decentralized and kind of Vice Versa and you kind of have this inverse relationship um how do you going to think about movements here on hey if it's going to be as flexible as secure and as scalable and High throughput um does that compromise on any value of decentralization I think one thing we focus on is finality like we're looking at pre-confirmation like if you look at Unis swaps un chain they have like this pre-confirmation mechanism that uses their L2 security instead of L1 security thesis is the majority of users don't need L1 security if you're playing a telegram tap game if you're like I don't know doing like a deck swap for like 10 bucks you're okay accepting L2 security um whereas like if you're moving a million dollars worth of like us you probably want L1 security so I think in the past like ethereum alignment meant you always need L1 security but that means you have like 7day withdrawals you have like 30 minute uh ZK proofs that are pretty expensive um but now there we're entering era and this what we're focusing on is like there's a lot of use cases and at already majority of use cases are okay accepting L2 security with the pre-confirmation and you can have fast finality you can have this higher through mechanisms um but if let's say for example there are users that need L1 security you can opt them in um so I think security has now become or decentralization has now be like a optionality where some users don't really care about security and decentralization economic security and decentralization they just want fast transactions um so for those users you can provide a certain service which is like these pre-confirmation and for users that are like okay I'm doing a million dollar pool on curve I need L1 security you can have them opt in that's a really interesting way of looking at it it's awesome um kind of my final question to you here is uh we kind of learn about the nuances here we learn about kind of what you're excited about what you're building uh but give us an overview of uh kind of the timelines on what's what's available right now what's upcoming how are you thinking about what people get to use and when yeah we have a test live right now test. moveing Labs XYZ um we're helping teams raise get to Market so if you're a builder and you're interested in building a high througha app that maybe not be aligned um please reach out um and we have meting that coming up end of the year amazing well RI thank you so much for uh this amazing conversation and please uh give him a big round of applause thank you guys appreciate it [Music]