This Is the Year for UX | Gregory Markou - ChainSafe

[Music] um cool my name is Greg uh today the title of my presentation uh this is the year of ux um so I'm Greg uh co-founder and CPO at chains safe previously core protocol developer on theorems consensus and variety of other Dev tools I spent a lot of my kind of like independent time uh working on standards which is super sexy I know and it's specifically in the wall space so extremely extremely loving but I think we're in a real special moment uh a really really special moment uh this year specifically uh my presentation today was going to be a lot different I was going to be focusing on how we are abstracting away a lot of like core infrastructure and Norms that we're used to day-to-day um but I spent yesterday at cake which thankfully for for those that watch the previous presentation we got a nice deep little bit of a deep dive into cake so I can gloss over that um so the chain abstraction stuff but also today I spent some time uh at the wallet area in the wallet teams working on some standards um and I got really excited um you know I made this tweet as I left the wallet standards at 12 and realized I'm going to completely scrap this entire presentation and basically give everybody as quick as a deep dive of what's changing in the wall space and how that's being affected through cake so the framework the chain abstraction framework that was done and why I fundamentally believe if we can't get it ux this year we like it's over there is no like we can't come back from it um if you know me I'm a bit you know I like to be a little bit dramatic but I I fundamentally believe like this is truly the year that that everything changes for ux uh and so I'm I'm going to start with my presentation that I did back in Dev connect which is I started with this this question this is a very common question we have at chains saafe you know we're in R&D team that focus on protocol work but we also do a lot of Consulting in our gaming and interoperability side and we always get this question that where do we deploy you know where do we build and this is like such an incorrect first decision decision that people have to make and it's because of a old old Phil philos like philosophical question we have from when we first got the evm because when we first got the evm there was only one evm there's one execution layer that we could use and this follow the 2000 early 2000 standard of how we build web apps you have a single server and you have a single web page and they communicate with each other back and forth nowadays we don't have that nowadays we have a microservice architecture where you've got gateways pushing information to different places you know you've got image processing happening on the side while your Facebook post happens in some other server and the like you make happens in a completely different one we're far past the single server relation and so why aren't we thinking about blockchains the same way and so I have this example if we have this dap called block Racers today we're screwed to choosing one single chain because that's and we have to figure out which one makes the most sense where are our users at but re in reality you know we solve this problem by you know saying let's just Bridge some stuff you know Bridge your assets over Bridge your data Bridge your identity uh and as we know from what we've heard from intense the last couple days here especially it's I think we can get a little bit better solution than that so what we were talking about in Dev connect before chain abstraction was really like a nice concept that I could use we were you know we Define this process as how do we get rid of the network tab switch how do we go multi-chain with a singular instance of your contract core contract data and that's kind of what we have here you deploy your core contracts onto one blockchain and you have adapters on other blockchains that are just dealing with like the bare minimum thing so if this is like a application that's doing wagered bets gambling whatever something in that realm you only need to know and escrow that money on the Chain where the user is and they don't even need to know what blockchain they need to be on because you just deploy that one escrow contract everywhere and then your core logic runs elsewhere this gets augmented obviously and then you have Bridges orchestrating intense solvers doing all the orchestration of the liquidity after this can be like augmented like tenfold when you start adding rollups as a service because you can then run session Keys dedicated just to that roll up similar to other chains and this is where account obstruction gets crazy but this is also where I cut off my presentation this is I just ripped out and I said not fur but we're going hints to this for the of it at the end of the day we are going to use the cake framework and a bunch of Standards coming out in the wallet space to actually let's not say solve onboarding but we're going to make it like as close to zero click as we can and I don't know if that's solving onboarding but it's getting really damn close to giving us an experience that looks like web 2 and I think that's what we actually need here and so let's talk about chain abstraction thankfully we had a nice little explanation of cake earlier there's multiple layers most people are deploying at the application layer account abstraction KMS intense this jargon which basically says like I want to do X and receive y That's there solvers the ones doing the hardwork basically Computing what to do settlement layer figuring out how to act actually you know what's the how to like finish the process that the solvers want um and that that's cake I mean I'm glossing over it because I think if you really want to understand cake properly you should go watch some of the videos from the chain abstraction day two days ago phenomenal um what I'm going to do is talk about how we're augmenting this with new standards and get a little bit closer so obviously cake is here to solve this problem uh uh you have e okay I mean there's test Nets in there but it proves a point like you have ether why am I representing it as like five different versions of ether I don't care I have five eth not like five SEO not five optimism or five arbitrum eth I have or one of each I have five that's my total balance stop giving me we don't need this fractionalization we also don't need to like understand that usdc is absolutely a disaster on arbitrum I don't care I don't care K help us solve this right because we because the solvers are doing the hard work of figuring out what you have and what to give you at the end of the day based on the protocol you want to go to and we'll get into that in a second you also were solving this problem which is wonderful right it's like gas is no longer a problem we can do that with the account abstraction thank you to the pay masters and this is an interesting one I learned this more recently but we have another problem that we're hoping to solve with some new standards this is called we all know wallet connect yeah this is called the NASCAR problem I don't know how many people are American I don't watch Nascar but this is a this is a thing so if you look at a NASCAR car awful naming there's a lot of ads I don't know what I'm looking at similarly we have too many wallets I don't know I don't care I really like I don't know where my wallet is I don't want to find my wallet half the new users coming in the space aren't going to know who deployed them their wallet they're going to have an address or an ens name or some sort of domain name and they're going to want to figure out how to get access to that so yeah the NASCAR thing is a real prom and we keep on doing this in a space please look it up and please let's learn from this um let's not do it more often that's no that's no shame to wallet connect they do great work all right let's talk about some new standards how we're like addressing a lot of this and kind of like what you know there's a small shill at the end what we're doing but like honestly at the end of the day I really want to like educate people on like there's a lot of coming down this pipeline this year and I think we can execute it on the next six to eight months and if we do when I want to open an e an eth compound position using my die we should be able to do that and there's a few key things here and it was talked about in the previous one and any intense you know you've been to will have like a very similar thing but there's something ke key here that I'm going to relate back to the ux because most of the people talking about cake that will talk to you about cake T aren't really talking about like the frontend UI componentry that's what I'm going to get into shortly and that's what a lot of my standards our standards focus on that's where I spend a lot of my time they have focus more on the architecture side of like the underlying physical side not on like what your average react developer is going to be dealing with so when I talk about this we're going to be talking more on a little bit more high level and so there's a few things you notice there's no network there's no destination Network there's no starting Network there's no destination Network and ultimately we're doing a swap in the middle the key thing here is that as a Dap You shouldn't really care where your users are coming from nor should you care where your contracts are because at the end of the day they're just people should be just able to deposit into it and so there's three things we're going to touch on today with these stand standards I'll go to them really quick standards are boring I've made this a little bit more enjoyable I promise uh it's wallet discovery which is a problem uh account abstraction which we'll touch on and permissions which are really interesting so um with wallet Discovery so H you know are we familiar with 4337 you know account account abstraction the the latest perfect smart account modules wonderful uh how many of you have tried to create a Dap that that finds somebody's pre-existing smart contract wallet have you been successful yeah it's impossible it's actually impossible like unless you were the one that deployed it and you know through which provider you deployed that smart account through so if the whole goal of account abstraction is to have smoother permissions you know so we get wrecked less defrag mantiz the fact that we have stranded EAS living everywhere with a little bit of dust floating around how are we in a situation where every app is now deploying you a brand new smart account really all we're getting is eoas with we're at the same problem death by a thousand accounts but we're slapping on some permissions onto it not very helpful because now you're you you as a de adapt developer you're going to have to eventually do this again hey do you mind briging your funds or sending your funds from your main account back into the smart contract wallet that's not what we want to do we want to get to a place where there's modules we want to get to a place where you're simply requesting to spend on behalf of the user's actual account and that's where 75 7 tri5 comes in 75's title is SSO for smart accounts it works for EAS too as we've discovered more recently but it's SSO for smart accounts what's SSO you know you've used single sign on to like use Google to log into slack or some other third party service right it's a oneclick action you get in it's kind of like this right you know you go in you select your account that you want to sign in with it redirects you to Google Google then tells you what it's trying to take from you you know in this case I think it's email and some calendar information and then kicks you back out 755 75 tries to do the same thing you know you can imagine we have a provider login you click it it redirects you to chains saafe domain we then know how we created that smart contract wallet for you we authenticate you and log you in and then we return that address back to the DAP so the dap's aware of now what that address is we can take this a step further though and this is where like it gets really magical what if we have login with name login with domain what if all you had to do was go to adap type in Greg theeg greek. and it does a reverse resolution and looks up who is this a eoa or did somebody deploy a smart contract while for the user for me and then it knows who deployed it and automatically hits the wire protocol which effectively 75 tri5 is just a wire protocol and automatically redirects the user to the provider that deployed it so in this case it might be chains safe it could be Pemco bomy I really don't care but it'll automatically know it's embedded into the web 3 modal and all I do is I walk up to a new dap and I just say yeah I'm Greg the greek. I know I can prove it you know maybe it's through Pass Keys webn maybe it's through my ledger but I know I have a signing key that I can prove I am that person just you type it in I click enter bam it takes me to the site similar to that how the Google login works it goes over there and I do it the next step if you're following me is where we add permissions into those smacks so if you remember I said I want comp compound I'm going to open up a compound position using D but I want an eth position that's an intent right that that is an intent that needs to be solved but what we need to do on top of that is adaptive developer instead of deploying a wallet you deploy metadata that describes what you're trying to get your people to do your user to do what permissions it needs to do and then what happens is the solvers act on that this is a standard that literally is coming out of today you can kind of think of this as like all of these together as like the TLs moment for wallets and what that looks like is if you remember here where we have this off page that people know really well in the web 2 world and I made this like 30 minutes ago because I deleted my entire presentation bear with me compound wants additional access to your account when you allow access a you know I don't have the save or allow button n um open a eath position on your behalf but compound also already has access to add a Max of 10,000 die per week because you've already actually done given them the allowance in a previous state now we have a web two flow that looks nearly identical to what we're used to and it's all through wallets all you had to do is type in your ens name maybe it's Greg theeg greek.com maybe it's a handshake it's clusters you know we don't have to be opinionated about this and that's all you have to do and you might be wondering okay hey but where does a user get an ens name well Unis swap hands out for free Argent hands them out for free all we need to do is just set up a resol a resolver to say how do I find the provider and these are all things that are coming out of like literally the last like 48 Hours um I got really excited and that's why I nuked my presentation why I want to show you all this um and I think the final thing that's worth mentioning today about where ux is heading is 3074 does anybody is anybody familiar with 3074 o call and O so it got shot down pretty hard two years ago this was like a way to give an eoa the ability to do something such as approve and transfer from in the same transaction scope as you know an EA like a metam Mas for example is unable to do that you have to do two transactions 3074 allows you to do that it does doesn't give you pay masters really easily it doesn't do all the fancy stuff 4 337 does but it gives you the ability to do multiple chain calls in a single transaction frame in a single dot call and I think that's really powerful and we had an allore devs call today where it looks like it might actually become a reality because we can use 3074 in conjunction with 4337 to make a smooth onboarding experience for users to get eventually into that endgame state of a smart account wallet while still maintaining the Integrity of an EA that they need um and this is really powerful because like we can try and 3074 like actually prevents people from getting wrecked you want to go like swap 10,000 die on Unis swap you typically accidentally hit unlimited 3074 will literally only approve 10,000 because it's in the same scope there's no difference the gas cost is marginal it's a really big Improvement um and I would highly advise everyone watch to listen to all cordev that happened today just the small segment on 374 you understand why like 2024 is going to be a huge year we might actually solve a lot of the onboarding issues and we're probably going to do it a lot sooner than December um on a final note a small shell super small uh this resolution nice um this is something that we have been working a lot from an R&D perspective at chains safe you know we do a lot of research especially a lot of ux research uh in the interoperability space if you're a Dap and you're looking at how can you do a deep integration with something like solvers that maybe the across team talked about earlier or how do you basically obfuscate the network tab from ever existing inside uh give us a chat we've got some we've got some people in beta with some of our products that we're working on super unopinionated uh it uses off-the-shelf tools it's not brand new proprietary stuff that we're working on we make sure to use the best solver the best Bridge the best whatever exists at the available moment for what you need to do um because at the end of the day we want to everything I talked about today like I'm not even we're not even working on half that stuff like we're working on some triple 5 everyone else is working on other stuff but we want to bring it all together and make sure that you have the best from the ux the front end perspective to really build the best applications so you can attract users from every single chain and not just the one you Deploy on because I think that's by the end of 2024 you will no longer be asking what chain you Deploy on you'll be asking how do I how do I get access you'll it won't even be a question of how do you get access to every User it's just going to be you know part of the mpm install walk through flow to just say like oh yeah like any user can use my app I don't care if they're on arbitrum optimism to be honest near salana it doesn't matter you should be able to get access to every single user and that's really where the future ux is going 2024 is the year and uh yeah thank you if you want to connect with us I'm Greg the Greek everywhere this is a link tree you can get to chain safe from there as well um and thank you I will feel the question but like I don't know it's this is kind of like philosophical it's probably not and yeah thank you