The Gnosis Chain Merge

Transcript

we are live I'm Seth 3.0 and today I'm speaking with felica Ernst who is the co-founder and CEO agnosis gnosis just recently merged uh I know ethereum emerged like two and a half months ago but all of a sudden nurses chain merged out of nowhere um so I wanted to talk to felica about why uh most of chain merged and why now and what it means for those who see an ecosystem and also talking about some of the new developments that are going on in this chain like account distraction uh building and privacy and all the sort of interesting stuff that's going on in the ecosystem like Dows Etc and I also want to find out why she thinks people in the blockchain space don't value decentralization NF before we get

started though please make sure to subscribe hit the notification Bell and the like button to get notified when I do live streams every single week my guest FedEx earns is coming up next right here on the interrupt [Music] hey hey it's good to be here on this side of the interview room yeah yeah it's good to have you um yeah so for for context for those for for those people who listen to this podcast but have no idea uh about the

other things I do um which you know there might be some people uh in that camp um felica is the co-host of epicenter which is a podcast I also host and um we usually interview guests together but today I'm going to be interviewing her um about nurses chain the ecosystem and and specifically like the merge the hypnosis chain merge and you know this is something that I so I I did a bunch of interviews at um at dapcon uh a few months ago and I talked to Stefan and you know this was right when ethereum was merging and he told me about the fact that like most of Shame was going to merge at some point I like totally forgot about it and then um I think you told me about it last week and I was like okay we have to do

like an interview about this um so uh yeah I mean this is a little bit outside like you know usually we cover a lot of you know IBC Cosmos interchange stuff here on this show but um I really like you know what you guys are doing over at gnosis and so um we'll we'll uh we'll step outside of the boundaries of what we usually do to uh to talk about this other important stuff outside of your comfort zone no but it's all like you know also like yeah you you have to also step out of your comfort zone and turn you know to like learn about other protocols and it's a little bit of your comfort zone though because basically Moses chain has the IBC model but for ethereum so basically it's uh it's like it's like

outside of your comfort zone but basically just barely the other side yeah yeah yeah um cool and I I prepared here like I got my magnosis swag and everything so um yeah well let's um let's dive in here like yeah you know where tell me tell tell me a bit about your background even though I know the background but like for for our listeners um what's your background and how you got involved with gnosis yeah so um my background is in physics um I was a research physicist um I dealt with low dimensional complex Quantum systems um I was firmly um on route to becoming an academic for

life and then chickened out at the last minute to co-found gnosis which at the time um was set to become a prediction Market protocol which we did actually build we also built a metric factoring of other things along the way but uh yeah so we we built lots of stuff for the early and then later ethereum ecosystem um uh yeah like uh the nose safe wallet um that's now its own company uh cow swap also its own company we build lots of Dow tooling um yeah and uh now we're here with uh nurses chain yeah I I think a lot of people maybe

have forgotten that gnosis was once you know trying to build a prediction market and it was like kind of pioneering in this space in the early days of ethereum and you guys have built like a metric [ __ ] out of [ __ ] of stuff you know and like I remember once seeing you give a talk about conditional tokens and it just seems like there's so much research and so much yeah just like so much r d that has that diagnosis has pumped out and in addition to that like actual products that people use um you know the the the the the bigger the the bigger products that most people you know associate the gnosis and user not the prediction Market it's like you know noises safe or now like gnosis chain but it's produced like enormous

amounts of value and also like standards for the industry I feel like I mean like no safe is the standard that everybody uses um and uh it's like a huge value for the ecosystem I I often wonder like I also called cow swap also remember Castro as well um how does gnosis actually make money what is the business model um well currently the the business model is that gno is a well actually knows this currently does not make money so I think maybe this is this is kind of uh what I can preface this with so uh no the dno token is a staking token on the noses Network so basically um if you stake gno um then uh the the transaction fees over

the network accrue to you just like ethereum is a staking token for the ethereum network um currently the fees um harness that way are fairly low because um blocks are not filled up all the way so the um the fee Market hasn't kicked in um but in principle that is the way of uh the way that uh the dno token would accrue value to the gno m holders including the nurses now um the way that we have financed ourselves over the last couple of years is by doing a conveniently timed Ico in 2017. just being really early being early and being good with money so

basically the the 12 and a half million that we raised back then are lasting us um to this day and we still have a couple of hundred million left over it's a good place to be in and and actually doing things with that money because there are a lot of people that raised icos are a lot of projects that raise icos that are just kind of sitting on the money or have not really delivered uh you know on on any of the things that are actually provided uh you know much value oh yeah no totally I think we've built a lot of things I think we we yeah no we've built tons of stuff we haven't Rock pulled anyone so yeah cool so yeah let's let's talk about

noses chain here and you know first I I want to maybe go back a little bit and um yeah talking about the where it came from and you know with the story with xdi and how like that chain came to be because I'm like that's kind of fuzzy to me I know like I know the history I think from the point at which it became knows a shame but I don't have much context for what happened before so like yeah what was the story there yeah so knows this chain actually started as X Die Chain in 2018. it was one of the very earliest side chains um to ethereum uh nosa's chain ID is 100 so that gives you an idea of how how early it was

um and it actually started as a proof of Authority's chain so there were 20 known validators so trusted parties who would take turns to validate blocks so it was um not decentralized in my mind but basically what what would pass for decentralized format many other chains um but we we wanted to we always wanted to kind of take it further than that so it was always kind of in terms of technology and what was possible on on exide chain um it was always a very Tech heavy chain that a lot of ogs deployed to

um but we wanted to make sure that it is seen and it actually is one of the most decentralized chains out there so sufficiently decentralized um which is why we started a gnosis beacon chain um exactly a year ago before the match we only noticed on the day that we actually emerged that it was exactly a year ago um but we started the noses Beacon chain that ran in parallel to gnosis chain exactly as ethereum did um two years ago we skipped some parts here because like x-time became gnosis chain I feel like we skipped that part oh yeah we let let me let me back up then um yeah so there was a vote in the

gnosis and xdi forums about um gnosis and xdi joining forces so xdi actually had a staking token token called stake this is this is how you know it was an early chain sake was still available um and uh basically the nurses style ended up um buying out all the stake holders and gave them gno instead so basically steak was replaced by the gno token as a staking token yeah that's like has this ever happened before where like a chain kind of buys out another chain no I think we were the first yeah yeah I remember at that time like it's Christmas and there were some

people that I think you know were not happy with this and but like overall I think it has been uh like a benefit for the ecosystem there were some there were some stakeholders who were not happy because they failed um like a lot of the other um side chains slash who's had um uh 100x or something recently and they found X die was um still liable to go that way as well so basically when they were given 50 percent more than what their nominal share was they were unhappy because they felt they had been cheated out of um like 100x which is I mean it's not

reason I mean yeah I mean you can take that stance I wouldn't say it's a reasonable one but also it turned out that in the actual vote that was a very small minority of people um who who just happened to be very loud on the Forum and but were the xdi team and the agnosis train team like how did that merging come to be were they both like in Berlin or did you know them or it's just just like how did that happen yeah so we knew them um we thought that we had um very commensurate values in terms of what we wanted for the chain and they were also very solid Engineers um they it actually turned out that

merging the teams wasn't as easy as kind of merging the tokens and and they eventually are left so basically now knows this chain is populated by gnosis people and people whom we hired for this purpose okay and so it's been uh over a year now that that merge has happened right well the the X die and knows as much well sorry like the the like merge zero uh which is the those two chains merging that's that's been a while now and so you you mentioned that like so so gnosis chain was running as a proof of authority chain with some you know like small-ish amount of validators uh but it

was a proof of stake chain right so it's like proof of authority proof of stake ethereum well yeah yeah so in principle yeah it was more of a proof of authority yeah it was approval of the proof of authority chain yeah um and so there was a there was a beacon chain that was so adjacent to the way ethereum launched its Beacon chain there was a beacon chain for uh nurses chain that was launched like some sometime about a year ago exactly what was the idea there like uh the idea was to basically kind of mirror the the merge with as ethereum has merged exactly so basically um I mean the way that

um the merge on ethereum happened um was that uh basically instead of one uh and instead of one client you actually have two client clients and one executioner client and one consensus layer client and basically the execution layer um just worked as it always has but the rules as to who is allowed to build the next block um that's set by the consensus layer and that was kind of um that was merged in um because obviously the the uh the execution layer um in ethereum test Nets has also always been able to run on its own so basically there's plenty of ethereum test Nets

that were never proof of work right so because for test Nets proof of work is terribly a bad idea it is usually a bad idea so um that's always been possible to kind of run the consensus separately from the execution um and this this is basically how ethereum merge just switched out how the consensus um layer Works um and the same happened for noses chain only that it didn't switch from proof of work to proof of stake but from proof of authority to proof of stick and so now there's uh so after the merge just like with ethereum there's like an execution layer client and there's a consensus layer client and validators

will run both of those clients but are there several or is it like I think with ethereum there's like two or three per uh like there's two or three execution layer clients okay yeah and is it the same same clients with Mr chain yeah so noses chain is actually one of the very few maybe the only side chain slash layer 2 with more than one client so basically in terms of um contentious layer clients there's um Taco prism Lighthouse Nimbus and lodestar and um on the execution uh site we currently only have never mind uh but uh Aragon um get and beso are going to come soon

okay so why why did this merge um take an extra two months why what was the the reasoning behind a waiting until after uh ethereum main chain had merged originally we said we might do it before ethereum but there were a lot of hard engineering problems that were not just copy paste from ethereum for nurses chain because we transitioned from um Aura proof of authority consensus and there were a lot of because it's very old chain um there were a lot of Legacy debts that needed to be taken into account and once we'd kind of

um uh missed uh front-running ethereum we didn't really see a uh a good reason to hurry up and risk anything so we kind of we we uh just like ethereum we went through several uh Shadow merges before we did the real thing um and uh the real thing went um off with without a hitch so I think that was the right car did you have a big party at full now to uh we did not have a big party at well actually we had we had we had we had yeah sadly we actually had a our Christmas party yet for no the next day but okay somehow it seems like literally all of Berlin is sick so this was was it

was like literally the only party that we've ever had that wasn't completely overrun actually it was there were maybe like 60 people here or something which for for usual for no party that's like it's like probably a quarter or something and um and so what were some of these technical issues like these technical challenges that you had to go through um and what did you learn from you know the post-etheria merge uh that that helped uh hypnosis chain oh there's a lot of there's a lot of moving parts right so I mean obviously you kind of you need to make sure that uh you you keep on producing blocks and that you're

finalizing and everything works on that end but also the other side so basically things like um token explorers and infrastructure and so on that also all needs to work through the merge and basically so there's a lot of moving parts and I think this is something that um uh we generally don't appreciate enough also how much for the for ethereum the merge how much of a coordination effort it truly was yeah I wonder like I mean because ethereum the ethereum ecosystem is so vast and there's like so many moving parts and applications using it and contracts and what was it as challenging or was it easier for agnosis to kind of like line

up all of the stakeholders and application developers and you know block explorers and everything like was it easier or was it like how would you qualify that yeah I think it was probably easier and harder so basically it was it was easier because there are fewer dependencies or noses chained than now on ethereum so basically if there's a [ __ ] up there's a [ __ ] up that's still bad but it's not as bad as kind of Downing uh ethereum right but on the other hand it was probably harder because there wasn't basically at the time if you if you if you think back to you know the week of the merge ethereum had the full attention of the

ecosystem everyone knew this was happening and obviously we did not get that same bandwidth of uh uh of you know just you know brain power yeah because like it's a smaller ecosystem but at the same time like and and also you you guys are like one team right so like gnosis is very much sort of I think leading development on those chains So you you're probably like sort of centrally controlling uh coordinating a lot of the a lot of the teams around like integrating with post merge yeah so actually interestingly most of the people who actually work on no sustain on the protocol level don't actually work at

gnosis proper they work at the clients on on the clients right obviously the basically the client uh the clients have teams that are dedicated for no oh we lost all right I think maybe she uh see turned off her browser there she is hey something went wrong like yeah no [ __ ] okay uh yeah so you were saying that the client teams are are the ones who are working on

yes so obviously we kind of we have the same thing as ethereum that's obviously like a cool client coordinator um but yeah so basically the the protocol development happens client-side okay so uh what what happened at the merch that it's just it just it just it just worked everything worked everything worked surprisingly literally everything whacked and we we went from 20 uh validators to a hundred and ten thousand validators like just like this it was insane it was fantastic so so I'd like to talk about yeah about the the validity of the ecosystem here because like you know on on the cosmos

side of things um typically chains have between you know like a hundred to maybe 175 I think we can push it to like 200 validators that's what tenorman consensus um kind of works with I think I think it could go up like a little bit more than that but generally we're in the you know one to two hundred range um for like instant finality tournament consensus um obviously ethereum consensus Works differently and this is what allows it to have you know many more validators but also making some compromises on finality um as I understand it um what uh yeah what what what does it mean like how many of those validators are

like in do you have any sense of how many of those validators are like individuals or are they running on like uh home uh validators or are they running on infrastructure are they delegating like give us a sense of what the validated landscape looks like yeah absolutely so I think to kind of back up a little bit um I think we kind of um I mean obviously this makes us look good right the number but um I think we need to be honest here and kind of uh compare appets with apples and oranges with oranges so um in my understanding if you look at um the tendermen consensus and the cosmos protocol that that's like

individuals or I mean individual entities that kind of run nodes right nodes for delegators yeah yeah whereas um in the ethereum ecosystem and basically that extends to gnosis as well basically ethereum has 450 000 validators but that just means that um there's uh 450 000 instances of 32 eth being staked and basically anyone can run I mean there's a limit how much you can physically run on uh on on Hardware but typically like 100 or 200 um validators per you know decent sized Hardware is totally within the ram of

the possible so basically when when we say there's a hundred and ten thousand validators or noses chain we don't we don't mean that there's a hundred and ten thousand distinct entities that run validators there's uh yeah so and I think basically this is this is this is um it always makes me a little bit queasy because obviously Cosmos is really decentralized and just because you know there's I mean having 200 distinct entities run nodes that's fantastic that's that's great um so yeah so basically nurses chain um the the way that we approach this is we try to make it as easy as possible for people to run a node and this has to a

certain extent this has also been ethereums um a goal so yeah whereas um on Cosmos you kind of you need to be a professional to kind of really run a node um it's not really something that lay people should attempt or do attempt um okay yeah yeah I mean I I mean look if you have some amount of like devops experience it's it's possible but it is not trivial like yeah really you have uptime requirements and stuff right so basically yeah yeah on ethereum

um you get penalized if you're not um up but you get penalized very little so unless everyone actually goes down at the same time um then you get penalized a lot but basically if if it's just that your computer goes down like your kid puts the plug or something yeah it's it's not it's not non-punitive so um I mean what what basically um for ethereum is um the bottleneck is that you actually need to um stake with in an increments of 32 East and 32 is that's a lot of money right so basically most people don't just have that lying around so if you want to

um stake less you can you can delegate with someone and someone else can stick for you and like any of these like liquid sticking providers yeah exactly um our noses chain the capital requirements are way lower so you actually need one gno that's currently a little bit under a hundred dollars um and you can do it I mean you can do it on dedicated Hardware like that nodes for instance um but you can also do it on your home computer provided you don't shut it off yeah and it's pretty it's fairly easy um to install um we've actually we encourage all our people to to do it and we have many people inside of noses who are very much I mean Quest mod but non-technical and

they can all do it so it's like um we you want to make it easier still ideally I'd like to get to a point where you can just um do in browser uh staking or something uh but yeah I mean it should be even easier but we try to make it as easy as possible to kind of become a node in the noses chain uh in the hypnosis chain why could you is there are there like any impediments to being able to do this like on a phone or something it's like a phone is typically always on and um other than draining the battery yeah in principle

um I think in principle down the road you should be able to do it currently it's we're still not there right so basically kind of it requires treating um running a validator kind of as a product and we try to do this and I think we've actually made it at least two orders of magnitude easier to to validate um but it's still not um at this at the point where you basically where you just run it on your phone yeah um so when you say that when you say that there you need one uh no uh Geno token to stake does that mean that like basically like one token is one validator or you know okay so if there's like a hundred thousand validators that means that there are a hundred thousand

gno tokens staked essentially yes okay do you have a sense of how many people or entities that is like any metric to um yeah it's it's difficult to say I would I would guesstimate it's on the order of um a thousand or so so basically you you kind of you can I mean you can see where these gno lie right so basically it's this do you know that are on the same addresses obviously it's the same person but often people are kind of um run validators from different uh you know addresses and so on so yeah I I would I would estimate between 500 and a thousand that's pretty good I mean like that's I think that's cool yeah

yeah basically we're trying to get to um we know that they're so basically from IPS and so on we can see that there's um gno validators in 42 countries um and uh basically we we want to get to a hundred because obviously um there's different um jurisdictional um uh influences um that can be exerted upon notes so basically having um having this be truly distributed and not just um in uh you know in the same jurisdiction to us that's a huge upside what's also important to us um is that people don't just run it

um on data centers right so basically in principle you can you can run it on on AWS uh servers or pets or elsewhere but um that's not as good as actually having people sake at home which is why very early on we partnered with um people who make and distribute Hardware that is dedicated for homesteakers so people like that note and avado and yeah because basically a network that is run by at homesteakers it's Unstoppable and I think this is where we need to get to yeah I I've I've been like wanting to buy one of these dap nodes

um yeah I think I mean I think I I met I mean we we had a we had a podcast on epicenter with um you know he was showing me that product and I think yeah I think it's like a cool product it is I think for a lot of people a bit prohibitive because it does cost upwards of you know 2500 Euros or something like that um uh but I mean you can do like a lot of things with it like you know you can you can run a Bitcoin node you can run ethereum validator you can do like ivfs and a whole bunch of other stuff so I mean basically what one should say that you could you can actually do all of these things in parallel but basically

if you have like a disused computer that you don't typically shut off or that you don't need to shut off and you can also just get all the packages from that note for free and install them on that computer so in principle it's not tied to buying the hardware yeah is there like a high memory requirement like you can't do this on like a like a Raspberry Pi or something like that you can't do it on a Raspberry Pi [Music] yeah um cool yeah so before the show you told me that you know you you felt that decentralization is something that people don't value properly especially in these ecosystem and I know that this is something you've talked about before and what we've chatted about um what's that about why why do you think we know value decentralization

decentralization enough I think it's something that's gotten lost over the cup over the years so basically I think people who actually started this ecosystem they were idealists and they were driven by the by by the idea that um drawing something in a truly um decentralized and distributed manner makes um systems incredibly resilient um and I I in a way this is what the entire blockchain ethos has been built upon but then um we lost part of that in like the the

intervening hype Cycles so obviously kind of blockchain that you know blockchain as as a tag draw in a lot of money and and people kind of forget um what what it was that Drew people in in the first place so now um we see lots of blockchain ecosystems that are as we say in in Germany Dino's so decentralized in name only [Laughter] German regulatory slang thing but um and in a way you kind of you lose so

basically building something undistributed networks is really hard it's so much harder than doing it in a centralized Fashion on a server somewhere on a service on a service service somewhere um so basically if you do that you need to make sure that you also reap the benefits but basically if um the distribution just consists of having 15 known entities or sometimes even fewer than 50 known entities between whom this um all happens that's a list of people to call if you want to shut anything down so I think that's you're getting a lot of the downsides without actually getting the

main upside um and yeah so this this was my controversial take that people don't value decentralization enough and people forget why decentralization is important until the minute that they need it until until you know the three letter agencies knock on their doors and shut down their blockchains um people forget it yeah I mean I don't know if that's such a controversial View at least like not in the circles that I run in uh you have nice circuits well you're in my circle so like

I feel like there is a part of the kind of crypto Community um that is building software and you know building like decentralized applications and where a lot of the decision-making process also includes or um you know the decision making process you know also takes it takes into account uh the amount of decentralization in not only the validator set but all in but also in governance and I know like this is something in Cosmos that you know we we talk about a lot where well for example you know many of the

top Cosmos chains validators like binance and coinbase and Kraken are you know often in the top five if not like first or second and this you know this is this causes problems because well they control uh a large stake and they also don't vote in governance and there have been instances at least one that I know of that um governance wasn't able to pass core because uh one of these validators one of these exchanges wasn't voting and so there's a lot of conversation and debate I think about like how do we um address that and yeah for me it's like also part of this this this this conversation that we

you know want to maintain this ethos of of decentralization and I feel like the ethereum ecosystem certainly and many of the changes in ethereum um also carry this ethos and certainly like Cosmos does and Bitcoin as well I think Cosmos does for sure I think we in the ethereum ecosystem don't always get this right so I mean if you look at um EVMS you know in the ethereum uh space um evm chains I mean I mean just I mean if you look at them by kind of transaction volume or so the number one is definitely Finance margin

um and uh so basically people are paying transaction fees for being included on binance Smart chain which is definitely not a decentralized chain and then other other l2s on top of ethereum they are all centralized in you know they have a centralized choke point um in the form of the sequencer so basically that there's there's you know in in very basic terms there's one entity that's allowed to build blocks that's the sequencer um and the sequencer can exclude anyone they like from from uh being included in this and uh basically the these attitudes they talk

about decentralizing the sequence and I think eventually they will um but until then there's a very centralized choke point um that can be uh used to basically shut down or sensor transactions you're thinking specifically of like ethereum l2s yeah okay yeah and I mean also at once in the ethereum ecosystem right so basically binance smart chain polygon and so on it's all fairly centralized um yeah I think this was actually this was actually something really good that actually happened to the cosmos ecosystem um that you remember a couple of years ago when the entire uh it looked like uh

like the interchain foundation and kind of all of the different entities in that it kind of it was a huge mess and it kind of fell apart but I think because there was no centralized stakeholder anymore it actually enormously it could have been it could have gone either way but it enormously benefited the cosmos ecosystem in the end because uh and yeah so that didn't happen for ethereum and I think I mean ethereum is decentralized mostly yeah I mean if you look at offic compliant blocks and kind of uh I mean yeah so it's it's kind of I think we've turned around and um notices recently uh released oh this is one of the other

things we were doing we recently released a uh a a a relayer that doesn't sense it's called agnostic relay um and basically ever since uh the number of ofac compliant blocks on ethereum has come down significantly again so yeah yeah yeah we need to do a whole episode about that um at some point the whole olfact thing um so how does so so someone here was asking in the chat um how governance works on gnosis chain so do people like people vote directly like there's no delegated voting um

it's it's like direct voting yes okay you can delegate your in principle you can delegate your votes but there's no um there's no interface which kind of shows you whom you can delegate to so basically people currently use it mostly to delegate um to an externally owned address if they have something on a smart contract wallet or something but yeah we we need to um establish um some sort of Delegation system because obviously you can't expect every every gno holder to know about other things going on in the you know in the nurses ecosystem you can't expect everyone to keep abreast of that so kind

of delegating in principle um should be uh should be fantastic to have hmm so let's um let's zoom out here a little bit and talk about some of the uh interesting innovations that are happening on gnosis chain and you know one of the things you mentioned earlier you mentioned like some some tweet that said that uh basically uh you know ethereum goes through two years of of research and development to avoid two minutes of downtime and uh I think that yeah it's like a good way to to look at it right there's like a ton of research and development and debate and and um working to find like the best Technical Solutions in order to avoid you know any uh potential downtime to

the chain with gnosis chain maybe the approach is a little bit less conservative um where there's like less at stake at least for now uh and that enables you guys to move a little bit faster and maybe move fast maybe move fast and break things is not the right way to look at it but at least um I think it's always been in the ethos of gnosis chain to try to be as like you know feature compatible with your ethereum as possible but also pushing um pushing the envelope on like some very Innovative and kind of Cutting Edge stuff so what are some of the things that you guys are working on now that like people should expect in those to change and then you know how does that like research then does it benefit

ethereum directly where you know because it's been built on like it's been done on the ocean Shane then there's already some amount of research and testing and um an experience that this can kind of flow up into ethereum is that the idea um yes I think uh that that can happen for many of the things that we're planning to implement um so if you look at uh if you look at what the onboarding experience is like for people who come into the crypto ecosystem um it's bad and it's uniformly bad between different ecosystems I believe it's the same in Cosmos as well so basically you you

um you're on board and you're told these are your 12 magic words never lose them never show them to anyone or you'll lose all your money obviously this is a terrible onboarding experience so basically things like on board with email and later change your password or change who got who's your guardian for uh for for your seed phrase and so on having social recovery to me that is something that um is absolutely necessary it's absolutely necessary to have and the way to actually go about this is to um to Institute um account abstraction um by what what is account abstraction exactly it means that every account um that uh gets created on noses chain

um is is automatically a smart contract wallet in the back end basically for us that would be an also safe and that's kind of it's it's the it's the de facto standard in the ethereum ecosystem as well so basically that allows you to replace the keys that sign for that account after the fact so basically you can say okay I'll let you I'll let you onboard your Sebastian gmail.com or at uh whatever that's mine that's not you um I'm not doxing you here so um and uh basically and that way basically I can custody your keys for you but if you want if you want to take your to take

custody back or give custody to a third party um you can rotate the keys so basically you can take away my signing pen you can put in new keys um and obviously that will facilitate onboarding because you can upgrade security after the fact um and you don't have this you know how when you download metamask for people and you know you're like oh these are the 12 words can you write them down you can't take a screenshot shot on your phone I mean basically it's it's terrible this is not how we will onboard the next hundred million people Yeah so basically yeah if instituting account abstraction this is number one for nurses Jane um in the new year um and I think so just just on that for

a second um I mean the issue with account of I mean to the extent that you know I've used things like Argent uh you know which I think was one of the earlier wallets to do this kind of social recovery while like every wallet is a smart contract um the issue that they ran into there with ethereum at least was just the exorbitant fees right I mean it's like every setting up a wallet was expensive and then also transactions are also more expensive with with this sort of account abstraction is the idea then for that to be kind of at the chain level where um like it doesn't have the same uh relationship to fees as like some other smart contract would have yeah I mean there's different ways to do

it so basically you either you do it on the account level uh on on the on the protocol level or you have like some sort of relay service and basic basically in terms of engineering all of these are feasible it is true that basically if you have very simple transactions then yes it is it is a lot cheaper to actually do it from an externally owned address wallet um but there are certain things you can actually where you can improve the user experience markedly um namely by Rolling different transactions together you know how you have to kind of on on ethereum you kind of you have to um set an allowance for your wallet for a certain token and then basically in a second transaction you then have to um to uh do the transaction itself and

basically rolling this into one so you can you can actually do this um in one transaction and you can you can do much more complicated flows as well and you can even do things like you can abstract the gas cost away from the user completely so basically you can for instance have it such that when people use your uh your adapt that you cover the transaction cost um while kind of making the money back in some other way uh and there's no need for people to actually pay for each each and each individual transaction yeah that's really cool um what what do you think of like so I I I I I I'm also of the opinion that

you know 12 words is not uh so the long-term uh way that we on board you know lots of users uh maybe 100 million is a good number but um and I also think that this kind of progressive uh securing of one's account right so where you can start with an email address and then uh move to something more secure as your account balance grows or you know over time I think this is also a great idea um what where do you stand on the like MPC versus smart contract um line like because like some people are building this as MPC uh systems where

um you know you have like basically one share of an NPC and then like some company has another one and there might be like a third one for a backup or something and then there's like the on-chain smart contract um model uh which which uh which is better in your opinion or they're like one of the trade-offs here um I think you can replicate more or less the same things with either so basically a smart contract um is um so I'm not a d I'm not an expert on the MPC implementation but basically on what you can do in a smart contract is you can um have very granular permission settings so basically you can say like for instance um I will give a key to this person they

can spend up to uh 200 a month um on these kinds of things I don't know on uh this educational store or whatever for my kids um and then you can give uh you can give uh Keys you can give out key set in principle can replace keys but can't do actually any transactions themselves you can have um like social recovery um you can have social recovery where you where you say look these are my 10 closest friends if eight of them agree that this is really me but I've lost my keys um I can just uh wipe this clean and kind of put all new keys and so on so in principle smart contracts are

extremely um uh extremely flexible in how you implement this and how you augment it after the fact yeah yeah I I like the smart contract approach I mean I think it introduces issues with with um it introduces issues when fees are um fees are high but for something like nurses chain where fees are like fairly low um I think it's a good solution I I don't know how that would work on ethereum though are there any any discussions already about like how this works with you yeah so I I think this is kind of the general discussion of what should be

on ethereum based layer and what should go um in the layer twos or basically you know what you can actually do with a dank sharding um you can have like lots of information that is in principle available for the layer tools and you can kind of roll up transactions into um into you know very pinpointed transactions that get checked into ethereum so basically everything is checked in principle and basically with scaling Solutions um I fear that for the foreseeable future we will probably be okay will it last us forever probably not but uh yeah I think in that sense

um ethereum has moved itself into this position where it kind of banks on layer 2's on top of ethereum um so basically that that uh I mean I I think basically distinguishing between a side chain and a layer two um which people sometimes use um uh you use basically people sometimes mix them up so basically a side chain is a layer one that runs in parallel to ethereum but um is um connected with ideally trustless Bridges um and then there's the layer twos which kind of intermittently um give their stay to ethereum to kind

of make sure that everything that's included in the layer 2 is also included in ethereum and does get base layer security from ethereum and uh where Cosmos has kind of gone the uh different ones that are trustlessly connected with each other out ethereum has gone the um and who's on top of ethereum Route um and I think basically you know this chain is a little bit between the two and that it is a side chain that runs in parallel to ethereum but his trust is deconnected with Bridges um so uh yeah yeah yeah I mean with Cosmos I think we also are starting to see like different

models emerge I mean not quite yet but there are like you know we can sort of see them in the distance with things like Celestia um that uh you know potentially will host a whole sort of set of applications as as Roll-Ups um and then these would also be uh either connected directly through some IBC implementation or via the Celestia uh main chain and then you know there's also uh all of the smart contract chains on um on uh in the cosmos ecosystem although these don't like typically have l2s uh you know they run applications and so then the idea is okay well how do these interoperate through IBC I don't

know of any you know sort of traditional kind of like in the traditional sense Roll-Ups uh like in the form of ZK Roll-Ups or uh optimistic Roll-Ups in the cosmos ecosystem I I don't know that like people are actually working on that other than in the context of like something like Celestia where there's like a data availability layer um but yeah perhaps at some point once once you know like block space becomes scarce enough that might become like a necessity I should add that other tools that ethereum has noses has them as well so basically you have a we have optimistic roll ups and ZK Roll-Ups and they work exactly the same way as as in ethereum

um and we we actually make sure that new depths actually deploy to them so um someone in the someone in the comments asked whether there was something like poab or noses chain and poab is actually on no sustain itself um but uh I I think in the very beginning poet was on ethereal mainnet and it just became unfeasibly expensive so they they might they migrated to a nurses chain many years ago um but uh for instance we have so you guys may know the Dark Forest games um they're deployed on noses chain but not on nose's chain directly but they're um on optimism on gnosis chain

and then there's um uh a payments Network called Oli that we're launching um that's also deployed on an optimistic roll up on top of gnosis chain so basically we try to kind of mitigate this and kind of prolong the time until block space becomes um much more expensive by already kind of pushing things to add tools or noses chain even before noses chain blocks are full yeah so there's this reminds me there's like a few more things I want to hit on before we wrap up here and I realize that you have a hard stop but um so yeah this This Ole Visa only product I think is super cool uh so what's how does this work and uh when is

it going to be live oh so it's um it's already live but it's currently improved it's in proof of concept so basically the team we have cards but that's it um so um it is a Visa card that is also a hardware wallet basically it holds your private keys to your nose's chain account uh you have your stables in there and you tap on any uh Visa terminal worldwide and it takes your uh your Stables out of your noses chain account directly um and puts them on the Visa payment rails um it is more or less instant so it takes six seconds

um and it's got the exact same user experience um as a traditional credit card so you tap and unless it's above a certain threshold like in Europe it's often 50 Euros or so you you enter a four digit PIN and that's it um that is literally all um you have to do and the super Nifty thing is if the merchant accepts Staples and that there basically if enough if enough people actually have these credit cards they are very much um incentivized to do so because the Visa payments act takes about two percent out of your purchase um

uh if the merchant accepts Staples directly visas bypassed um and you pay in-stables from your wallet to the Merchant's wallet um and they save the two percent that they would have spent on the Visa stack so when Stefan told me this is just like blew my mind so like because because you know you guys like he started getting into some of the like the technical specs of visa and apparently there are like some extra messages that you can send so you guys kind of like really took apart the Visa protocol and Leverage The you know the the sort of like very proprietary protocol in order to be able to do this but like this this idea that you can use your Visa card and it'll

uh it'll fall back to stable coins uh and send stable coins directly if the merchant is excellent it's so cool like I I've never heard of anybody else do this but it must be like very technically complex to be able to initiate that transaction from the card itself on the payment terminal actually that's um I mean not to kind of knock on our achievement here but uh that's not actually all that technically complex you kind of you just need to make sure that um you integrate well into the Visa payment rails and you actually need to work with someone who is like an actual payment processor so basically we

actually did this as a joint venture between us and soak pay um who has uh they they have millions of Visa terminals um in Europe um and uh and tangent who actually make the credit cards that are also Hardware wallet so basically it's kind of there were very disjunct pieces but um the actual signing of the message is less complex than you'd say you'd think but the what what's the beauty of it to me is that it's a form factor that people you know instinctively know how to use because many many crypto products actually maybe our crypto products we've built so far have been crypto products targeted at crypto users and this is

something that is targeted as at Nomi so basically anyone can use a freaking credit card we all know this yeah um yeah and um so so it's currently like uh running as a as a as a pilot uh what when is it going to be available to the public and how can I sign up and create circulate a Google form um shortly and for people to kind of get on the wait list um I think first half of next year is a good guess okay cool um yeah and so so we talked about account abstraction uh yeah let's let's talk about this bridging stuff a little bit because uh

yeah yeah I'm not like very plugged into the state of interest of like interoperability and ethereum and uh like the I realized that there are some very particular challenges there that are different from the challenges in the IBC world and so how are you tackling this with um these ZK like clamp bridges that you mentioned yeah I mean ABC was kind of designed for a word of interoperable blockchains right I mean this is kind of the design spec for for IVC and ethereum very much wasn't that so um what is always tricky between uh different changes you kind of you need to know what the state of the other chain is right so because you you need

to know if you send something um what what that that's actually you if you receive something from a different chain you kind of need to to verify the state of that chain in order to kind of um reproduce um that token on the on the target chain um and what in principle I mean if you look at how it's done now usually it's it's kind of mighty six um so basically kind of like uh uh like five parties or so who independently look at whether a transaction has happened and then kind of automatically sign off on the transaction on in the Target chain that's kind of Mighty sick Bridges um but what you can in principle do is

you can have a light client that verified verifies the state of the target chain um uh the uh the chain that uh funds come from um on the target chain you can have like lines on those chains directly obviously you can't really run a light client on gnosis chain or ethereum just because in terms of compute it's really expensive but what you can do is you can actually put it into a um ZK socket and use the succinctness quality of the zero knowledge circuit um in order to prove the state of the chain um on the other chain um and that's exactly what we did so we

have ZK live client bridge between noses chain and ethereum and principle can work between any um evm chains that were that have uh that are on proof of stake um because you kind of you rely on the validator sync committees um for this as well but um yeah so you have completely trusted Bridges um obviously I mean they're currently um they're around in uh there's a proof of concept um it doesn't currently operate under load um because we've been hesitant to kind of transfer everything from the current mightysig Bridge the Omni bridge that we have between noses and ethereum

because um we have seen so many Bridge hacks over last year and if within completely new technology obviously um the space to make mistakes is enormous um and what we will actually do is we will use a mighty sick of bridges so we will have two Bridges run in parallel so the old mighty six Bridge um the Omni bridge and the new CK lifetime bridge and if they both agree that there should be a transfer the transfer will happen if they don't agree um then nothing will happen and will have to be resolved um but that means in order to kind of

hack the bridge not only do you have to find a bug in the ZK like Clayton Bridge you also have to compromise more than half of the sinus to the to the Legacy multisig Bridge so basically it becomes it's kind of like putting a sieve in a sieve right so it becomes a lot less probable that people actually find two bugs at the same time and we are also cooperating with a number of other teams who build implementations of ZK like client Bridges um for instance there's one within the status community and there's one at Berkeley and um they um basically in the long run what you actually where you want to get to so

that you have several implementations of um CK like Klein Bridges between gnosis chain and ethereum and any other change that you want and but you don't operate them singly but you operate them in tandem so basically only of all three different Bridge implementations agree that something should happen this will actually happen and basically finding um vulnerabilities in three different Bridges at the same time obviously this is much much less likely than finding vulnerability one Bridge and I think this is kind of how we can um how how we can do this in the interim

until they are um you know tried and tested under load so with with uh this implementation what's the uh what what is the implementation path so do all do all uh chains need to implement this ZK like client and then all of a sudden they're just like all you can Bridge them all together or do you need to create um the two-way Bridge it's a two-way so they're so much like with the IBC like every chain has to create a bridge with another chain so you have to have these channels

um okay and um and then and in terms of like the user I think like the user experience of bridging is one that is is an issue that uh is often overlooked right so like you know you might have one experience with one bridge and with another Bridge you might have a different experience it might not be like compatible with your wallet and the nice thing about IVC is that like if you use any Cosmos chain moving assets between uh chains is it's like you know it's the you do you do it the same way across all chains right like it works in any wallet wallet applications have implemented it and so the the user experience of moving assets is fairly uh simple and it's like quite Pleasant to use

um what what does it look like with with these EK Bridges or do we expect that like the use the ux will kind of unified across all evm chains or well some of them have different implementations that may or may not be compatible with some or some other wallets like what what's the what's the complexity Matrix there yeah in principle they need not be um compatible but I expect that there will be a de facto standard that people will adhere to um the the problem that we have in the evm space is that as soon as you bridge one acid over a bridge that as it gets new deforged basically on the other side of the bridge right well no sorry yeah in IBC now that's not

the case yeah yeah and then what it is but it is but it is like a fungible asset I mean although you know if you if you IBC say like from chain a to chain B and then chain B to chain C and then back to chain a um I don't think no that that you basically have to unwind those transactions in order to come back to the main token Yeah so basically uh on on evm Chains It's like that as well so basically the the things that are bridged if the same token gets Bridge via different Bridges um it's de facto non-fungible it's not the same token so basically what you actually end up relying on is an army of Market maker who we're kind of exchange those different representations of the

same token for like a basis point or two basis points um and and then basically um the DAP developers will have to make sure that basically if someone comes with a token um that they somehow integrate this Army of market makers or like a Marketplace for exchanging these tokens so that the user doesn't have to be bothered by which token it is I think this is something that we can abstract away quite easily from the user and users shouldn't have to know um what what what bridge their tokens came over ideally they shouldn't even have to know which freaking chain they're on it's like I mean they should know which apps they want to use and then things should just work and I think

this is where we'll get to eventually um it'll take a little bit of time but uh yeah we'll get there yeah well uh hopefully yeah I think like the ux around in probability is like probably one of the biggest challenges in crypto yeah with with uh you know like accounts right like and securing once one's coins um there are inherently huge problems that don't have a single solution but um but maybe several solutions that we need to try out and you know well um I think we can I think I think we've covered a lot here and uh yeah uh thanks for thanks for doing this and thanks for

thanks for having me on allowing me to explore this ecosystem a little bit uh deeper uh it's been it's been interesting no it's been it's been super lovely to be on and it's uh it's super funny to be on the other side of uh of the interview yeah and now so now I've I've done three well three of the I don't know I don't remember I don't know how many episodes are hosts we have now but three of them have been on interrupt so now Maher news to come on uh maybe Felix and um yeah I guess that's that's mostly the main ones [Laughter] all right thanks again

thank you bye and to our listeners thanks you for tuning in uh if you want to get notified when I go live every Thursday please make sure to subscribe hit the notification Bell and the like button until next week hope you have a good one