Panel I Evolution of Wallets I Pragma London 2024

[Music] let's bring on our next set of panelists so this panel which is our first panel for the day is about the history and the evolution of smart wallets we're going to talk about how things started where they're going and how can we think about better user interactions with our wallets and the three panelists to talk about this are going to be Pedro from wallet connect IAI from Dynamic and moderating this conversation will be edar from Argent so without further Ado please welcome all three of them on [Applause] stage hi everyone Pedro Tha great to have you here thank you thank you thanks let's maybe start by some intro so that uh everyone knows who we are and what we are doing I'll go first oh wow so much pressure I think uh hey everyone I'm eai uh co-founder of dynamic um what we essentially do in a nutshell so we've been at this for about two years uh what we do in a nutshell is everything that happens after you click login on the website so whether you have a wallet uh you don't have a wallet and you need an embedded wallet uh user management linking multiple wallet to the same account uh that's pretty much what we do you can see it on sites like the pudgy Penguins Marketplace or Doodles or uh the starware uh airdrop or or you know their Bridge or anything of that sort uh iyk token proof if you went to e Denver uh and you kind of clicked to log in that all kind of uses a version of dynamic so that's inherently what we do we're based uh State Side in California and Miami and that is pretty much it that was was that fast enough detailed I don't know we'll see yeah we we'll ask more questions after exactly uh my name is Pedro I'm the founder of wall con and that means many things but mostly it means the wallet conect protocol that enables remote connections between all the mobile wallets that you know today to daps and we also have our own SDK which is called webt modal which allows you to essentially have this embedded experience where you can connect multiple wallets not just mobile wallets but also browser wallets and we also recently released web3 inbox which allows wallets to receive D notifications thanks and I'm it Tamar found out Argent we started our journey as the first smart wallet on ethereum back in 2018 where we had the great idea of paying gas for users on L1 uh then we started our journey to scaling and and cheaper solution and now we are very focused on Stark net where we have more than 2 million smart account a billion dollar in assets and around 80% of user use Arin when tracting with stet um so there are a lot of topics on the on the wallet topic and what we thought we would start with is a new trend we see a lot which are embedded wallet uh I think a lot of people are talking about it are probably using it if you if you have used something like FR Tech or others you often faced with an embedded wallet let's start Maybe by around around what what people feel think what are the main issues and is that the trend will that here to stay uh am I correct it that you guys actually have an embedded wallet in your offering so maybe you can share okay so let's start with you yeah uh thank you so I think I think you made a good observation embedded wallets mean a lot of different things to to different people right it can be in it there are a tool and a set of tool kind of uh in your toolbx for for how to do one of many things whether it's to onboard users to web 3 so if you're a Dap and you want to kind of open up uh the ility to access kind of your your protocol by non-web three users that is a use case of embedded wallets there are completely other uh use cases so if you want to build a decentralized venmo that works across Argentina and Brazil you might need some sort of embedded wallet component in your um app that actually kind of lets you do stable coin to stable coin transactions right so there there are a lot of different kind of aspects and we'll get to it I think you know there are discussions around whether wall embedded wallets are siloed into an app or Global uh but inherently what we've seen is we see them being used across a very long list of use cases so again anywhere from kind of a decentralized venmo type thing to Global contractor payments where it's very siloed within the app and it's just um a way to hook up to crypto rails or crypto network uh to a more dap use case which is a Dap that had kind of wallet connection and now they want to kind of increase um the the scope of the types of users that can access it and anywhere in between so that's what we see on a day-to-day basis we can probably double click on those over this conversation Pedro I'm sure you have an opinion around unembedded wallet and how you can interact with really any any dap uh as as wallet connect I think universal access to every single dap is something you you care about or do you see uh embedded wet play a role there yeah so wall connect protocol was built on the idea that every one will have a dedicated application to manage their wallets so embedded wallets are challenging that uh but it's not coming without its costs um web 3 Model also has an embedded wallet feature just like Dynamic has but there's some disadvantages which kind of has this more like Silo experience some uh embedded wallets today if you connect to an application let's say Unis Swap and then you use that particularly embedded wallet then you get a different account if you go to openc and now we're creating this concept of a global wallet but it's not truly Global because realistically I know the dynamic has a different vendor option than uh web model so if you use an email with our STK it will get you a different account to Dynamic and I think this is a big problem because when we created the wconnect protocol we created with the idea of full interoperability and we have achieved that with browser and mobile wallets which I think is a great success but now embedded wallets are kind of fixing the user experience but also pushing back the interoperability which we have addressed before think we can go on some of the disadvantage maybe my view on the Argent side on embedded wallet on one and we we have an embedded wallet on Stark net the web wallet so clearly we we wouldn't do it if we didn't believe it can really improve on boarding but I think something we see uh with the builders is maybe they believe too much that this is the issue so I think we as an ecosystem put too much too much of an excuse around oh crypto there's friction there's an there's an issue there on onboarding the reality is that in in our research the wallet side even with a seat phrase is actually a small part when if you start needing to put money in the system that takes you know significantly more more time if you're on ramping with a card or others kyc Etc and even worse if even if you need to bridge from L1 then the cost is prohibitive so there are so many other friction Point uh the wallet is probably not the main reason you might not have enough users let's say it that way so maybe on The Silo oh it if you have something else maybe Silo versus global what should should everything just be Global wallet and and maybe to clarify to user Silo is really your your wallet only work with that dap if you go on the second dap even with the same vendor they might give you another wallet and so you you lose composability in some way uh is a place where Silo is better yeah I think first I think ped you made made this point which is it's first when we define global we have to Define it's Global within an ecosystem right so a good example is the correct me from wrong the the web wallet right on on starnet is global within Stark within Stark net and within a place where it uses uh specifically the SD your SDK right so it is when we say Global we really mean uh and I'm assuming it's the same right what what wherever web3 model is used it's Global right and same with if Dynamic does a global type offering it's wherever Dynamic model is used and so when we say Global first we really have to make sure everyone knows which which is it's Global within your ecosystem right and so one Challenge and I think you mentioned this is there needs to be some sort of interoperability across ecosystems right so uh there needs to be some way in which a web 3 modal wallet works with a dynamic wallet and vice versa which I think is a challenge but but the the you know we talk a lot about global wallets I I kind of think that a bunch of people don't really care right uh I I think it's a it's a problem we care a lot about but again there are many use cases of embedded wallets that have that make no sense for Global wallets right if you're using a Dap I think the use case of a Unis swap you go to Unis Swap and you log in and then you go to magic Eden and you log in into the same wallet is a phenomenal use case for Global wallets right because uh it is adap like interface but then if you just want to use build as a startup a a new experience for um you know Global Payments or as an example you're building a memec coin trading app on your phone right you should not care and your users do not care that these wallets are Global you're putting you're tokenizing an asset and you're putting on chain and you're letting someone buy someone in Argentina buy us treasuries they do not care about the wallet that is just an ends to a mean for them to access an asset that they couldn't access before or transfer money or do anything of that sort and so there's an entire use case of global wallets which essentially just is an easier fintech stack together with account abstraction and on-ramps Etc it's an easier and sorry for the extreme log monologue here so I'll stop in a second but uh there's a full kind of stack of use cases for embedded wallets where Global makes no sense because it's not users don't even know or should know that they have a wallet just like you shouldn't know that you're using a uh behind the scenes when transferring money from point A to point B well I have to challenge that because I don't believe that's true um I believe that users do not care about technology but I think that's what you actually trying to um emphasize here but they do they do care about the universal aspect of the internet and even if you go to more traditional paths like in Europe uh we have standards like psd2 which essentially creates a more open banking system where essentially doesn't matter which banking provider you have you essentially have this more Universal identity over all your bank accounts and if you go into you know not so open but meta is trying to create a single identity over all their meta products and this has come with great advantages for the users to have one login for everything and one profile for everything and if they choose to have different profiles they can still manage under the meta account Center so I I think that this is a very important problem to solve because users need convenience it's a convenient thing to have a single identity over all of these accounts it doesn't matter how it's going to work under the hood that's our job to fix but they definitely need this more open system other than having to remember like a thousand million passwords with a thousand million emails and now we have pasis and it's just going to it's getting more and more complex and siloed so Universal identity in the internet is the end goal I countering your counter uh just to keep it interesting to your point uh the counter to that is do we really want ma to have all our information and have there there's the counter to univers like universal access is privacy right uh or lack thereof right and to your point there's or a counter I think their ways to address it and your point if we do this in a proper way I I I struggle with that a little bit because I I I think we don't have an identity I have my gaming identity and I have my work identity and I have my personal identity and I kind of want to Silo those sometimes right and even in your example on AA um I have my login with Facebook but then I also have my login with Google right and even at that scale there are multiple ecosystems out there that that aren't necessarily don't necessarily share information it's important for consumers but it is I and let me know I you know more on on the kind of European banking but I don't think anyone stops opening a bank account or opens a bank account as a result of that interpr ability it might be better over a long term but it's not it is kind of a an Evolution versus kind of a shortterm goal type thing one way to see Silo wallet but I think there are Pros for it but to me it's also it would be the idea that you'll be topping up a balance on every website so suddenly you want to top a balance on Amazon on Apple on on Netflix and that to me is a bit strange versus no I bring my own wallet and I will pay these different parties which would be more the global wallet but I've been talking to very smart stocket developers who who are asking for this Silo wallet and I'm trying to understand why because there are so many advantage of the global one users are well is bring your own wallet bring you all your assets so in one click you can start interacting and I think there's an element around if you really want a totally invisible wallet uh as a you put a lot of trust in the D the rate is it's significantly less secure uh literally the DB could make you sign whatever they want uh you have much less security than from using an extension of mobile app and you could accept that if you are really a siloed wallet meaning I'm giving $1,000 to that dap I fully trust it and now I just use the DAP I don't see the wallet anymore you would want a silo wallet or some mechanism could be session keys out there that really really make the the wallet totally invisible and that's where maybe Silo is a bit of a shortcut that could work I mean one thing that I wanted to bring up that D Tai said we want to create this segregation between your gaming identity your financial identity and your social identity but embedded wallets right now are siloed not based on user configuration of what they believe is the correct segregation they're actually segregated by uh business relationships and this is the challenge that we need to uh address here because just because your favorite app uses web modal or dynamic that shouldn't dictate how you actually segregate your accounts we should have fully Universal accounts across businesses and then you make a decision of actually compartmentalizing your gaming identity from your social identity now how do we actually fix this I think that this is something that you know Argent has been working on for a long time which is bringing smart accounts and smart wallets into the picture because the reason these silid wallets even exist in the first place is that evm specifically has this assumption that for one key there's one account but if we actually break that assumption and we actually introduce smart accounts then we can actually have you know a dynamic key and a web modal key and an Argin key control a singular account and I think smart accounts might be the solution for the problem yeah I by the way 100% agree with that I think their use cases I kind of go back to use cases there are some use cases which let me know whether you agree that that where the embedded wallets is just a part of a financial Tech stack for taking an action it's not a web three type thing versus web crypto rails as a more efficient path to kind of completing completing an action that could be trading that could be tokenized assets or anything like that then there's within an ecosystem so let's take the Unis Swap and Magic an example can we do kind of a single wallet and there I think that the answer is absolutely we should right because you should just like you go and you see Argent on website a and you see Argent on website B right and it should it should be the same Argent wallet um that should be the case for an embedded wallet I think you're also 100% agree the a a phenomenal path for this is a count substraction where you assign multiple signers right to the same um kind of account but then I think you made a good point there's like a lot of security risks here with how do you make sure how does a magic Eden or a Unis swap ensure that the random dap developer doesn't kind of you know take advantage of their new found power of having access to kind of a wallet in maybe session Keys play a role there I'm not sure uh is there a risk that with by pushing too much unbedded wallet we we destroy self custody um you know there is they are not custodial but let's be realistic in most case if someone stop paying the bill uh your AWS uh infra you probably lose access to your wallet um is there a risk there how do you approach that I mean if we just stop development right now and we consider this the end stage State yes like we're not in a good state for recoverability for fair markets for like internet Freedom like it's really like we're kind of challenging um like going in a good user experience but compromising and we're kind of in the dance between compromising on like Tech and user experience and this dance is necessary because otherwise we can't find the balance because you either are too balanced and you don't make any moves or you just like sometimes prioritize user experience sometimes you prioritize Tech and interoperability and we definitely need to solve this and honestly there's advantages to having competition like having Dynamic build their own SDK having web modal build their own SDK there's advantages because we end up optimizing for different set of customers the problem now is that these keys are siloed and I need the web modal key to control the same account that the Dynam key and the Argent key and the only way that I believe that this will be feasible is through smart accounts uh because like you said these are pretty much custodial systems and very unlikely that they will be self- custodial if they're embedded maybe just on the custodial point just to challenge a little bit again to keep this banel interesting hopefully uh I I would argue there's a difference between custody and Trust assumption right there's more the the legal and I'm not a lawyer but they're more like the legal and compliance definitions of custody versus trust assumptions right correct me if I'm wrong but if if uh I I don't know about Argin I'm assuming if mamass goes down I'm assuming as a as an entity there's going to be some issues accessing The metam Mask extension right I don't know if you can technically take your key but how many of us have you know saved uh everything in an easy to access place there trust assumptions that you make with maam mask you make slightly less trust assumptions with the dynamic uh you make slightly more trust assumptions but in the end of the day we all you know all of us probably have an apple uh laptop or uh I guess some of us have a PC I don't know if that's a thing anymore but uh uh you trust apple as well right there are trust assumptions here right we we make them every day and so there's this balance uh between you know uh the custody aspect these are I think non-custodial uh wallets versus uh what do you have to assume about the internet uh when you kind of make these decisions just as an extreme example all of us can stay in business but if AWS goes away everyone's also Parton my language okay I'm not going to say it here but every so with an f um you know and and we're all making those trust assumptions on a day-to-day basis so that do I would just separate kind of the the custody aspect from the trust aspect ECT interestingly on the smart account point you bring Pedro smart account for uh interoperability to be honest we never thought about it that way we use Smart we've always used smart account for us it came more from the angle of upgradability so you would we see the web wallet or the embedded wallet as a first step uh probably I wouldn't advise someone with $1,000 more more to stay there and then with smart account we have a pass to move them to a mobile signer or an extension Etc while keeping the same address the same account so we saw it more as a way to evolve from this very smooth onboarding to then a fullon uh full ownership of my wallet uh experience but uh I like the interoperability angle too realistically inoperability just means shared infrastructure and it's very hard for people to agree on shared infrastructure it's actually pretty much a miracle that wall conect protocol worked and I think it just came because we had such a great EOS as a community for the last 5 to six years that kind of just trusted to actually Build Together the wall connect protocol and it got us to the day where we have like the best interoperability compared to other ecosystems wall connect today is Now supported by other ecosystems but it's not at the same level as ethereum like it just completely dominates the experience in ethereum and we kind of have to have that same moment with smart accounts now we have to have an a second wck moment it's probably not going to be called wck or maybe I hope so um that basically have full interoperability between all wallets and smart accounts such that we don't care what kind of vendor you use we don't care what kind of business relationships or SAS you use that's just like business lingo and at the end of the day the user just cares that their username works everywhere and it has always the same assets and that's their identity just by the way I think that to your point just maybe a shout out here which is it also you guys did this recently was 6963 right where uh for folks that don't know right and and you'll be able to like probably give a significantly better description than me but uh at the very basic level uh browser extension while it's pretty much pretended to be made a mask over the last couple years that was their entire kind of thesis is in order to get distribution I'll just pretend to be made a mask and I think uh you know the wall connect team has led this incredible initiative to kind of standardize that and within a very quick time frame pretty much all wallets are now adopting this ability to kind of work together right next to each other and so I very much hope that we can do the same thing for for the account abstraction layer so that's a good segue maybe to account abstraction on evm um a lot of people ask ask us how did we get adoption on stocket so on stocket we have 100% adoption of a can obstruction every account is a smart wallet every single dab supports smart wallet every single dab does multic call you you will never see a near C2 account I mean even centralized exchange like Ken of b or binance are using the arent account and people ask us how did that happen but the reality is we can't transfer those learning the reason it happened is we we went so early we started on a blank page and we worked with starware to say that's that's how it should work and EA should not exist that's the standard interface of an account and whether you arent or a competitor you will use the exact same interface and s have only one way to think about it and that's easy to do when you start from scratch how do we do that while you know while dealing with the Legacy and well and backward compatibility rather than Legacy uh on evm do you see a pass or or clear next step to to push that adoption forward I mean I don't know how many people are in this room are aware how ethereum evolves like the etherum protocol is basically decided by the C devs and cev are not some cabal like they're just devs you know that's kind of right code and they are part of different teams and then they come into agreement of what is the next step but obviously these teams are very focused on like the largest problems like scalability and now they just did the blobs upgrade and we kind of have to voice them to them that like the user experience is a burning fire that we need to put away and it's something that has been kind of like forgotten because because we had proof of stake we had like EIP 1559 with the fee mechanism and blobs and I think that now we're starting to see some awareness that we can't survive as an evm ecosystem without properly adopting like user experience focused upgrades like that's something that we have never actually done as ethereum and the reason Stark net has 100% account abstraction is because they prioritize that it's just a matter of focus so if it's in my power I will try to convince as many core devs as possible that the next upgrade is focus on user experience such that the account abstraction doesn't become this hack around like we're going to build more smart contracts no we should just make the evm more smart account friendly just to add to that my two sense is it starts with the l2s right it actually I would venture to say that I I wouldn't be I'm willing to to bet some money that uh base for instance over time would just mandate account abstraction as a default right with like I think there's RP 7560 as a first step there uh my two sense is we're going to start seeing the sub Eco systems to your point that prioritize consumer crypto that prioritize uh user experience just go all out account abstraction and completely remove these concepts of kind of gas fees or anything of that sort and then over time I think after we get a bunch of these sub ecosystems that would be how kind of we go back to to your point like the the the core devs and convince them it has to be at the evm level but I think that many manyi kind of coalitions is the way to go here I fully agree on that it's a I mean etherum cannot evolve as fast shouldn't uh L2 is the opportunity they've solved their cost problem most of the time they true put so focusing on ux is is the right opportunity should we try to get a few questions if we have a minute I see yeah I do believe we actually have a minute so like we have one question thank you guys uh what do you think about these EIP 7212 I guess I can start I I think it's a pasy p so for anyone doesn't know EIP uh 7212 essentially kind of uh works on kind of Pas key validation directly to the to the uh smart contract essentially skipping the the need using your your phone inherently as the eoa private key signer uh I think it is a phenomenal idea in theory I am a huge supporter of it I think it faces a significant amount of hurdles right and one of them is first and foremost is not Everyone likes pass keys right that is Step number one uh second is they don't necessarily uh and I say that in pain because I love paskis but uh um you know the second aspect to this is that your most chains correct me if I'm wrong I think polygon Mumbai is the only thing where it's actually deployed and so besides that it's pretty expensive gas wise to do everyone everywhere else and so it is it's going to take a long time to implement I really want to believe it's going to happen I I don't I don't I don't know yeah I don't know if you agree it's going to happen like like we just have to accept the fact that like nobody cares about this like and wants to have an actual Hardware wallet your phone literally has this technology like like it's not just paskis like 7212 introduces a new cryptographic curve for signatures which goes beyond pasis so even if you don't like paskis uh Apple securing or Android key stores all of our smartphones have secure elements that can be natively supported by the evm so if it's not the etherum Cavs it's going to be some rollup like it's going to happen Ju Just on that I think we see Pas is now being adopted on web 2 so I think that's clearly there's an acceleration once and for all I want to clarify that Pas keys on on Mac OS and on iPhone are not in your secure Enclave so people should stop thinking about it's about convenience it is not stored in your secure Enclave you can use the stic curve to sign with Hardware signer The secur Enclave but you will need a mobile app installed for that so uh pasy is really about ux if it doesn't kill the true put I would really push for for that uh as you say if it's not too heavy in computation but it's really a nice ux and um stet also working on making this much cheaper as it's quite expensive right now yeah I would just say one thing I know we're at time but to your point it's ux I would also argue it increases security mostly because it's essentially you generate an invisible password per app without the user having to remember anything totally agree but from our experience basically the key extraction is not the how people lose their money people lose their money because they they go on Unis swap.com with twps and they sign the wrong transaction that's really it's about front monitoring RIS scoring uh that's how you lose your money in 99% of the case you are the one with your own wallet with your own key sign the wrong transaction and you know everyone has will not admit but everyone has signed such transaction in their life yeah no comment I think thank you thank you everyone was actually