Is Account Abstraction Really Working? Ethan, Gregory, Harish, Shahrukh, Aditi

[Music] oh you're not from J I'm sorry and we also have Ethan from partical Network um Harish from killan and Sh shuk shro yeah shro from Echo and AD adti from per all right let's start and the topic is going to be about uh account abstraction all right thank you all right great thank you so much we had a little bit of a hiccup with uh our panelists but maybe to start off um since all of us are building with AA in some way um or are doing things with chain abstraction maybe let's just kick it off with uh your name an intro and then we can get into it yeah so my name is Tabasco my real name is Ethan but I use Tabasco on Twitter um I uh do developer relations I do a lot of things at particle network marketing all that kind of stuff um and kind of the TDR what we do is that we do mostly chain abstraction that is using account abstraction rails and so people use chain abstraction as you know like a narrative term all the time and so I usually tell people that what it actually means is that it's an experience where you don't need to like select An Origin chain but bridges are used for you automatically so if you have eth on arbitr and basee you can use that balance collectively to then go and buy a token on salon or on ethereum or whatever and so it removes a need to bridge or to maintain account balances on different chains and basically makes your entire experience across the entire ecosystem evm and nonm feel unified and so this is basically what we build at particle Network through something called Universal accounts um my name is sh I'm product lead at Eco and the tldr for Eco is that we want to translate every single onchain action into a one-click stable send and the technologies that we're interested with in account abstraction is combining chain abstraction and token abstraction to basically create a unified stable coin balance regardless of if you have any type of stable coin across any different chain that's it hi I'm I'm Greg or Greg the Greek online I'm from chains saafe and Sprinter we're a solving team uh kind of just focused on providing liquidity and solving basically like the intent uh World which is tightly related to um chain abstraction and C abstraction um and otherwise like I spend a lot of my time writing standards or reviewing standards in the AA space um not really affiliate to a wallet just kind of passion uh hey everyone my name is Harish I'm deel at coalin and our focus is on making access to onchain data super simple for applications uh that build in web 3 so uh I think similar to what Tabasco pointed out our focus is not so much on account abstraction but chain abstraction and and so with our tooling you're able to seamlessly pull data uh across chain without even actually specifying a particular chain uh and that's really made the uh developer experience uh quite nice for wallets accounting tools um defi uh projects building with us and hello everyone my name is Dakota and I'm the founder and CEO of the inar network um we originally focused on uh Dow tooling which then got us into chain abstraction because we wanted to have a single Dow that could operate across all change just like a business today can operate do business in all different Sovereign countries and things like that um which brought us of course to chain abstraction there's account abstraction token abstraction this and that um so we always Center on Omni chain abstraction we want to solve uh abstraction like just once and for all for all chains for all networks um it's actually like my belief that in if not five maybe 10 years uh like most browser companies will be either defunct uh or most wallet companies will be defunct or they've evolved into a browser based product first um just it just makes sense that way one day you're going to have an account embedded into your browser it's one account it'll probably use a mix of like with us we use NPC Tech um so that you can have a single master account then you can have you actually have an account on all of those different chains you can hold balances natively do anything natively on those chains but you just have this one account it's embedded you don't even need a wallet screw that you can see your one native balance it could be in USD or in Stables or your native currency and you're just going to use it and you're going to call it the internet and that's that's our vision great thank you um also realized I didn't introduce myself uh I'm a DTI I work on perah uh we're building an embedded wallet um an onboarding Suite uh for app developers so um with that uh this talk is really not meant to be a primer on uh account abstraction or uh chain abstraction but is really meant to be kind of getting all of our real takes on how the Space is really doing and so to kick it off um curious if account abstraction and chain abstraction is really actually making wallets more userfriendly or if it's just adding another layer of complexity for developers to manage to Dakota's Point there's already this like wallet piece that people need to manage but now there's also this account abstraction piece chain abstraction of course account abstraction is all on chain and that just adds even more complexity to the process so I'm curious um you know what all your takes on that yeah it's a good question I think starting with account abstraction I actually don't think account abstraction as defined in ERC 457 has been particularly successful at improving the user experience mostly because of the need need to have different contracts across different sources that use different smart account implementations and so if wallet a is using bomy and wallet B is using you know zero dev then my balances are different between these implementations which seems like a small thing but when it comes to fragmented user experience actually becomes quite an issue and we actually build on ERC 437 and we've been a major infrastructure provider for this standard for a long time um but we've ran into this issue and I think that ERC 437 can enable some incredible user experience it's a big driver for chain abstracted and chain abstracted experiences but fundamentally I think that we still have a ways to go before um it actually creates a net positive effect on most wallets because if my metamask balance isn't easily you know portable over to Phantom or something for base or ethereum I think this kind of just makes the entire issue that we're trying to solve with chain abstraction you know 10 times worse um just when it comes to fragmentation and so I think 7702 is quite a promising solution to this problem of fragmentation having the EA be the origin and having that be the Fender and then you just execute arbitrary code when needed I think this is probably better um but you know there's multiple solutions to this and I think what we've seen is that ERC 437 is good for enabling some very specific things like gen abstraction but I don't think it's been particularly successful and we have seen this in the adoption rate in terms of improving the user experience as it exists today and so I think that hopefully with 7702 and with the proliferation of chain oftion which still hasn't really been implemented yet I think chain obstruction is a very new thing that we're seeing only emerge really over the past months as a tangible solution I think as that starts to continue to be adopted and as I think as we see 772 roll out then hopefully we'll my answer will be different in the future which will be hopefully yes can can we round back to your previous comment you said that the bomy versus the Zev zero Dev example what's the real problem there though these are different contract implementations which have different no no no no what's what's the real problem as an as an app user I go to an app it's the discoverability problem thank you this is the problem yeah so like the app and the you the app and the wallet have no way of actually knowing when you connect your wallet to an application you have no way of knowing which wallet you're actually even using in the first place and because you don't have that you the the application basically doesn't even know should I actually give you a user op or should I give you a transaction and but but why because you don't have like a like a some kind of like a lookup concept so you this gets it and it it is um it is a big barrier to adoption right because um I do agree with Tobasco immediately started out with some very very hot Tex which comes very true with your handle and I do agree with everything that you've said but one thing that you said which is very important um is that it does enable certain use cases and before we talk about like all the problems that 437 has I think we just take one point um and really talk about you know what has worked with 457 what does it enable and for us personally at Eco 457 is extremely personal because before Eco is was what it is today it was actually a Neo Bank we had more than 50% state coverage in all over the us we had millions of dollars in US user funds but we were a custodial product and because we were a custodial product in the last Administration time we actually started to get deep platform notices from all our banking partners because they told us categorically that if you put anything crypto related ated or even touch stable coins we will deplatform you and you will have to tell your users to take your money elsewhere and in that environment we had to shut the app down and I lost a lot of good friends and colleagues when we had to wind that app down so when 457 launched we deployed Bend 3 weeks after it went live on mainnet for us the main thing that we were looking at was that we finally saw a way where we could create userfriendly non native asset based applications in a safe and feasible Manner and that was the one thing that 457 did ride but I do agree with a lot of takes from Tasco I do agree with Greg the discoverability problem is a pain everything else that you've said is a pain interoperability because they're embedded is a pain but it did make the self- custodial part work and for that I just want to give a shout out to the authors and every bundler and pay Master who's done the work so a story like the Eco app doesn't happen again so in that regard I think it's it's it's uh it's done as J so I just want to add to because like one of the pain points um with uh so account abstraction is one thing like a product could probably solve for account abstraction very very well when just looking at the perspective of using their one product same thing as for a chain a single chain if that was the only chain you'll ever use or on board you could probably on board to that once beautifully whether it you know it could be Bas whatever you could find a one wallet that's done it well um but there's always going to be the other ecosystem um and so so let's say you have something that's nice like Phantom so someone on board they use Phantom they're like oh this is a cool onboarding experience um but now they're restricted they could use Bitcoin they could use ethereum aie base um but they I don't know they can't use Tron or arbitrum you know what I mean um and so I feel like a lot one way like I was speaking with um uh Q over like Alliance sty he was telling me that one of the things that um he doesn't see changing which I disagree with him on is he believes that you know basically it'll just be up to wallets to choose which chains they accept which is good but there's a bandwidth problem um there's a reason why every why every wallet doesn't just go ahead and support Every Chain is because it takes some time um and they don't have the time to do that so I have to prioritize what they want to do which is why one of our focuses has been um like with what we're building we have no interest in building another wallet per se um the way that we see it is there's uh there's plenty of wallets people are going to have their preference um there will still be new wallets that come about but what we would much rather focus on is providing an st K that would allow any Wallet current or new to then be the only wallet that users need to access all of web 3 and that could be something you know you got to you got to earn trust before you reach this level of adoption but it could be something to we like yeah Phantom or metamask they integrate this and then everyone who already had metamask it would be some migration um but or Phantom or whatever they'd go through this and then they could have the one and last wallet theyd need now there's of course things like particles working on with account balances and all that to where it's it's not as simple as I as I spin it out right now um but I do think that while like the immediate easy thing to do is going for um the experience of each individual wallet um I do believe that it we won't fully solve it solve abstraction for the entire industry unless we address Solutions um at like the highest point which like comes with your account um and just having like that one account you have you know all the other accounts everywhere else to do everything um but because there's always going to be all these different wallets all these different things then if the solution is oh well then we'll build the next wallet that can and then you got to go ahead and get all the Taps to integrate and accept this wallet rather just give an SDK for those who've already been out there to then just become that wallet and service them and you know allow others to also keep working on other things to add to that wallet stack um I'll just make one more one quick comment here maybe just going even a step further than what Greg pointed out like why is this even a problem in the theum ecosystem you don't actually hear a lot about this in the salana space for example and you know you're kind of seeing the wave of um inbound flow from new users into that ecosystem not all that complaint so I think part of this is just the fragmentation that's actually happened in the space uh with the you know flourishing of of rollups uh and it's a good thing it's definitely help the the throughput but I think it's necessitated a need for something like a account abstraction and then chain abstraction because discoverability and some of these other issues given how fragmented the the space is but I think like one thing that Solana doesn't try to do like I don't I'm not that familiar with Solana I will frankly admit but like gas abstraction is one of some of the use cases that we try to do which 487 was designed for these things I don't see a lot of those activities happen in Solana is gas substraction even an issue in the salana ecosystem I think it's multi it is if you if you have a stable Point denominated experience right you you do are you going to be like dusting wallets at the back end I I guess my point is I haven't seen the requirement to purchase Sal being a barrier to entry for users in the salon ecos for people who want to create stable coin denominated experiences and they don't want the user the user is not if the if the target user is actually speculating on Meme coins then I would agree with you but if the target user is an abstracted user who doesn't even know that they're using crypto then I would disagree with you on that because then gas is actually a non-trivial pro well I think neither of you are right or wrong it's it's more so just the matter of if he's right if you're talking about salana you don't need to worry about this because you just you have your one state whatever you're doing is on this one chain you get the price from this one API on this one chain everything consistent but once you throw in a second chain it could be eth and then Bas or arbitrum whatever but once there's a second chain now it gets more complicated because you have to Route about and so that's where it gets interesting compared to I don't think so I think that's the difference no no no like that's not the difference Sal the difference in Sal is like at the account model the problem is that it's so bloody cheap to use that and the fact that they way the programs work on salana versus the way that ethereum contracts work and the and the way that tokens flow the biggest bottleneck today is the fact that mostly is the fact that we have message. sender and tx. origin as like a main blocker um on how every of how like actual movements of flows happen where in salana the programs don't have that same blocker that's why they don't have like an approved flow and because it's not that you're wrong you're not wrong I'm not saying you're wrong but I'm just saying it's over complicating the fact of like why is salana EAS use you could say to the account it's that the it's cheap and I don't need to bridge anything it's but near why is ethereum on that point with I got to bridge things we also have four minutes left so so sorry want to make sure the conversation keeps going yeah did you want to say I was just going to say like near has a lot of chains but near has abstracted the their account model from like the key model and because of that they don't have a chain abstraction problem like chain abstraction only exists because the underlying accounts don't work in like that current model we're in I me this is something that we've built so we've built we've improved on near account model so they have their username model we've added in usernames this is in substrate though so it's not uh ERC but we've made it to where you can have yeah you just put in a username you use a private key and now you've onboarded not to a wallet but to the actual chain itself yeah but like we're band-aiding interfaces instead of like Plumbing it into the evm and that's that's well that's what we plumbed it not the interface we plumed at our chain we're not an evm chain we're a substrate chain but it's just cuz it's easier to do it's easier we can fix it you know faster on that point on gas sponsorships uh at scale is that really sustain stainable because we're just kind of funding gas for users at scale is that something we want to continue doing in perpetuity or is there really something deeper at the ux level that is wor solving we're going to see a mix of we need to solve this is whether it's you have people sponsoring transactions then routing things and all that so people can just go about their way you have like some ticket system where people are buying you know gas subscription packages or anything or really whatever we'll have all those going about but I think eventually what we'll see is more and more chains that just don't have gas fees that's the end game like and if you're getting then you need to think bigger I'm a protoc like that's not possible then how come it's happening and it already exists if it's not possible scale it it is sced I think you got I think you got to hear me out I'm not trying to throw shade get out your comfort zone um I know that you have your ways and I'm not saying they're wrong but when you break outside the box you'll see that uh it's Common Sense business models if someone's using something well enough if I get security from elsewhere I know you're going to roll your eyes cuz polka dot shared security um but if I get shared security which means I can then I get that I don't need to worry about it and people can do their stuff and they're going to pay money I'll get fees off of I'll get a little fee off the actual service they're using we're we're core developers in polka dot filecoin and kind of I know you guys hav been around in a bit but yeah well no like like I know how I know how substr works I know I know but so then why you rolling eyes saying that it doesn't because it doesn't scale out but it does so the I think frequency is the eighth highest most shoes roll up okay uh in the safe to say there's a lot of problems right now with the c exraction but there's also a lot of good I think like for gas sponsorships like in Ben we're actually I think one of the first applications to implement like fixed feed models so they were we experimenting with experimenting with that one of the highest and one of the biggest problems in uh 487 is the call data that you're posting and that's actually going to be a huge blocker uh but the cool part is especially for applications like car if we're focused on stable coins you can actually and if the stable coin amounts are going to between let's say a fixed amount like between 1 and 1,000 and let's say there like most of the transfer are going to happen in like five major stable coins you can actually have compression where you compress the call data and and especially when you have ADD BLS signature aggregation on top of that I think there is a world and we're actually experimenting with that internally where we can bring the transfers if not on par and in some cases we're very excited about some future where we can actually make it just like a transfer just like a conventional U normal eway transfer so I'm pretty I'm pretty bullish on it's not there yet but there are PVS that I can uh that I see that we can actually scale this uh feasibly yeah I think you know we actually aren't that worried about the gas cost of ERC 437 I think really the important thing is giving user the ability to pay for that gas using for example stable coin denominations or any token that they're using in their order so if I'm buying a token or if I'm doing a transaction that's denominated in let's say even multiple stable coins like USC and usct and ether whatever and not to say that ether is stable coin but um um and then but and my gas is being paid through one of those tokens alongside whatever order I'm doing um and even if it's on a on some chain where that isn't the gas denomination I think that's the important thing at user experience- wise is that gas shouldn't feel like some distinct problem or or mechanism of the transaction it should just be looped in with whatever I'm paying with which often will be stable coins so I think that this is the the real user experience problem rather than just trying to continually optimize feeds which I think are going to happen naturally but anyways this is totally agree with that a good summary we're at 18 seconds so lightning round kiss marry kill the following words 7702 pass Keys solvers Tabasco you're up Mary solvers um kiss 7702 and kill Pas Keys okay oo kiss solvers okay Mary pis and kill the other thing Mary solvers kiss pass Keys kill 7702 yeah I agree on the kill 7702 um uh kind of indifferent on the other two actually to be honest okay yeah so kill 7702 um who I want to marry who I want to kiss uh Mary salvers kiss paskis great so we're in agreement there all right well that's it guys thank you