Gnosis Chain: extending Ethereum

Transcript

hi guys i'm fredericka i'm from gnosis i'm the ceo at gnosis the company it's and someone at the dao you may know gnosis from the past uh eight years in this space we have built lots of stuff over the years starting with prediction markets frequent market platforms um gnosis multisig knows the safe uh gnosis auction um cow swap dutch x yeah lots of stuff and now we're doing something new knows this chain

so let me tell you about why we're doing this and what exactly it is so gnosis chain actually launched in 2018 as xdy it's got a proof of authority consensus with a dual token model so that means there is a staking token this token stick and the x die which is the fee token and sometime last year there was a proposal in both the gnosis dao and the x-dire forum to actually do a token merger between gnosis and exile and the proposal passed on both forms so

this actually went ahead and people who held stake could swap their stake for gno tokens at a fixed rate extire was renamed to gnosis chain and the gnosis beacon chain was launched which is a proof-of-stake chain similar to the ethereum beacon chain this was at the end of last year i think noses beacon chain launched in november now we're at more than a hundred thousand validators um and we are set to actually complete the merge

before the ethereum match the ethereum merge is now um pencilled in for september 19th so we hope to be doing our merge the week before that and gnosis chain sees itself very much as an extension of ethereum so it's not um it's not a competitor it's a companion in the sense that apps that used to live on ethereum and now no longer can because it's no longer econo economically viable and may find a new home on gnosis chain i was going to start off with a problem statement what are we solving

and i'm leaving that for later because when when i actually reviewed the slides before the talk um i actually went on to a rant just because the problem is so large so let me review the ver the technical um the technical infrastructure first and then go into the into the problem statement just after so um when looking at layer one blockchains i always find it tremendously tremendously useful to actually think of them in the blockchain blockchain trilemma framework right because it tells you where um the architecture is making trade-offs

and what what the blockchain emphasizes so as i am sure most of you all of you know is the blockchain trilamba that's on the slide you see security scalability and decentralization on each of the edges and the dilemmas pick two right you can never have three so how chains how blockchains are organized um is actually reflected really well in how they actually perform in each of these dimensions so let me go to uh the dimensions first in which we shine um decentralization so

it is um it's our goal to actually be the third most decentralized chain um so if you actually look at the stakers.info infographic on the right hand side you can see that moses beacon chain has a hundred thousand stakers that's a quarter of the ethereum number and as i said this has been since last november so how did we actually get there we actually tried to make the barrier um to entry as low as possible for validators so the minimum stake is 1.0 that's currently around 150 the market is pumping so might be 160

now the technical setup is fairly easy if you want to set up your own validator you need a you need a computer that is usually on um probably stationary somewhere and probably around an hour or so of your time even if you're non-technical so anyone can set up a validator because we have we we always intended to have lots of um non-professional stakers um we have non-punitive uptime requirements so basically if your node goes down yeah it happens whatever you're not being punished for this and to further

extend our reach we went on a call collaboration spree with dab note and avado i don't know whether you actually know these companies they make little validator boxes that basically you you plug into your own internet and power and they just work it's pretty magic so you can actually validate ethereum gnosis many other chains on these validators so there's a slight downside to having a larger proportion of hobbyist seekers you miss more blocks and this is quite apparent here so basically if you look at the top left

graph you can see that our noses chain around 10 to 15 percent of blocks are missed and on the theorem beacon chain don't don't know whether you can see it you actually have to zoom in quite a bit it's less than one percent so um the question is does this have any bad consequences probably not really at that level um so you lose a little bit of bandwidth but that's about it so basically losing 10 to 15 of blocks it's not that big a price to pay for actually having a very decentralized validator set okay so back to the blockchain trilemma let's look at the second edge security

okay so basically if you look at security in proof-of-stake networks um you're always worried about 51 attacks right so there's two attack vectors um one is the token price so if you buy up 51 of the token you're doomed right i mean so basically then that that network is taken over by you know the person or the entity that holds 51 percent and if it if it if it goes down it goes down the other vector is actually gaining control over a stakeholder validator that together have 51 of the validating power right so

i put the ether scan graphic here of depositors to the eth2 contracts contracts you can actually see so ignore the blue one that's others but if you actually look at the big ones on the left hand side the big red one is lido then the green one is kraken the light blue one is binance then there's bitcoin swiss steak sake dot us steak fish and so on so if you actually manage to compromise several of the largest validators you can also take over control of the network obviously i mean

each of these i mean even on ethereum this is kind of seen as as a problem and if you go to east research or crypto twitter people will be talking about this and whether whether um uh the liquid staking with lido and so on should be somehow reduced or whether it should self-regulate um however that may be i am sure that most of these takers all of the big stakers um have really good opsec so um actually compromising them is probably really hard hopefully impossible um

so for ethereum it may not be that big a problem but obviously if you have lots of less professional stickers it's even more important to actually have a decentralized network and by having a very decentralized validator set you at least preclude the attack vector of capturing validators worth 51 percent in total there is one caveat so currently on notice chain there are no trusters bridges but this is um this is uh this is an interim state of affairs um you thinking back over the first half of

this year there have been a number of notable bridge hacks um and these usually had to do with um trusted bridges so basically um glorified multi-sigs in a way obviously we all know we kind of we need to move away from them um there has been a proposal to no sustain uh to know this now uh gnosis improvement proposal 57 and by by the xerox park team um to build the specification for a snark enabled bridge so basically that leverages a zk light client um on knows the speaking chain and a zk light client on ethereum and that inherits the security um from

the consensus algorithm so basically you are as secure as the network it's coming soon hopefully in like two months or so okay so let's go to the last note this is uh the note uh that is a little bit um where we're making many of the same choices as ethereum um so if you look at ethereum so we see scalability is its number one problem right now right so and we have we have championed um pushing um the efficiency efficiency frontier on evm networks for many years um so you may re remember the ethereum one point x

efforts with um turbo gas now called aragon and so on we've we've funded we we've been the largest funder for aragon after the ethereum foundation pre-noses thou and after noso style um so what that actually does it is is it turbogath optimizes the data structure of the state so basically if you look at the table on the right hand side you can see the archive node size on the right hand side for the client on the left hand side you can see that gas has six terabytes opening through six terabytes nether mines five terabytes and so on aragon

one terabyte why is this so because the state is optimized for uh for efficiency um and i think this is something that doesn't get talked about enough in this ecosystem that the fact that um uh the what's actually rate limiting is the the computation itself so if you ask why can't we just um increase um the the block size limit um by a factor of 10 or so the the the simple reason is because this cannot be computed because the state is so large so aragon does something about um the state upload so basically you you make good a little bit so basically you can

push the block size limit from probably from 30 million to probably 100 million and nether mind another gnosis to and ethereum 2 client will implement the same optimized data structure this also recently passed through gnosis tower as jip-47 um and then there's a slightly shorter block time on nose's chain um compared to ethereum so five seconds versus twelve seconds so all in all that buys you around a factor of eight with respect to ethereum you lose a little bit of that because of hobbyist validators as i said 15 or so so maybe it buys you a factor of

six or so six or seven so this is nice but it's nowhere near sufficient for the scaling requirements that we need so basically gnosis chain takes the exact same approach to scalability as ethereum so you pushed apps to um layer twos on on nose's chain okay so now that we've kind of understood where gnosis chain is in terms of technology let me give you the problem statement the problem statement is this lots of things are going wrong i mean if if you look at the state the world is currently in um and what we're doing about it

um we're very good at ignoring problems right um i mean wealth inequality income inequality child poverty are up intergenerational income mobility down energy food security are down climate change don't probably don't even need to get started it's um just after six year in paris and the outside temperature is 40 degrees celsius in july so yeah i mean we have a big problem we're also killing the planet in other ways right so we see we're currently way into the sixth mass extinction so obviously this started way back when we

started killing mammoths and other big big animals whose survival tactic was you know be too big to be prey and trap you reproduce slowly and clearly this didn't really work well with the arrival of humans but we never really stopped so basically we're killing um we're killing animal species species at an enormous rate there's also other things going on so ukraine china the us may be on the brink of a civil war who knows so yeah lots of things are broken

and the tragic thing really is that lots of these things we actually in principle would know how to solve so climate change no i mean in practice it's not that it it's not that simple but in in theory you solve it by taking carbon dioxide out of the atmosphere that's it and there's different ways of doing it there's books about this but somehow we still cannot coordinate as a society to actually get that done and that is that is uh preposterous i mean clearly the consequences to each and every one of us are going to be horrendous we should be able to coordinate at you

know some sort of uh reasonable cost rising inequality clearly we don't have a resources problem we have a distribution problem somehow we don't get this done mass extinctions probably not all that easy probably should should stop uh killing the rainforest and polluting rivers and so on but i'm not an expert so there's also other problems where we may have to resort to the current solutions so you know tick-tock dances uh to help ukraine and so on um so what's the blocker why why aren't we getting our together do people not care

or is this simply a coordination problem so somehow and somehow we have this enormously complex state apparatus times 180 times 180 countries and somehow we we still can't manage to solve humanity's most pressing problems so let's find out so basically what what we really want to do is we want to provide tools for daos so tools to help humanity organize at large scales novel ways to kind of to to act against you know these problems that we're

facing um we're also trying to make a universal basic income happen so if i mean this is a really lag it's it's it's it's a big proposition right and i don't know whether daois can fix all our problems i don't know but i think um currently it's it's a shot it's maybe a good shot so i think we absolutely need to try and we are building a chain that is optimized for daos and a ubi it's a general purpose chain we'll get to who lives on it in a bit let me just talk about the implicit assumptions we are making here

for a second so our first assumption is that the evm effects are strong so basically if if you look at the developer mindset that's a mindshare that's around the evm it's huge so it's it doesn't really compare to other chains um and uh yeah then daos can make a difference and in order to actually make a ubi work on a worldwide scale you actually need to use decentralized technology those are kind of our implicit assumptions so who is on those chain today 1500 plus dollars we've also built extensive dowel tooling gnosis guild

if you haven't checked it out it's called zodiac check it out circles ubi and proof of humanity both live on gnosis chain social apps like poep giveth clr fund games like dark forest and commerce privacy so tornado cash nova lives on gnosis chain um claroson chain link and a full set of finance primitives so amm and non-amm daxes landing markets derivatives one inch curve cow swap uniswap um including v3 sushi and so on so what's uh what's up next so first up is the merge

we're targeting the week before the ethereum merge what are our current uh focuses the first one is client diversity so basically we have um aragon and nethermind has have dedicated teams working on nosy's chain clients um we have commitments from um gath prism lighthouse and tikku to make their clients compatible with nosyschain having more clients makes for more resilient networks network you all know this um and we want to stay as close to ethereum to allow maximum compatibility to kind of allow um you know passing back and forth depending on use case as much as

possible there are notable exceptions where we want to implement things that ethereum is not implementing or not yet implementing one is account abstraction so if you look at the user experience that we currently have on ethereum eoas are big problems a big problem so basically um the the seed phrase this is not something that we can bring to the next millions of users right so what we want to encourage is the use of smart contract accounts so eip 4337 that is under active discussion in the ethereum community or similar

leveraging and also safe as an account standard this would actually buy you many things first thing is flexible onboarding of users because you they can later exchange their keys for that smart contract wallet right you can have social recovery in various flavors you can have trusted guardians you can have recovery through relationships you can even have recovery by default so basically if you say i this is my account and i've lost the keys you can stake an amount maybe like 20 percent of what's on the account to regain it and if no one contests this within say three months um

you get to you get the right to reset the keys otherwise you forfeit your stake um all of this is possible with smart contract accounts batching for user experience gases transactions and so on so to us um basically if if this is something that we want to bring to the next millions of users this is a no-brainer we're talking about other things as well so things like trying out single slot finality this is also a topic under active discussion in the ethereum community or privacy built in two bridges [Music] where to find out more come to the nosy style forum

from dotnosis.io there's dev docs there's an ecosystem website chainstats and a tutorial how to become a validator we're also hiring so if you're interested in co-development in bridges in validators or similar come find me after the talk thank you guys [Applause] and i see i have one minute and nine seconds left so

if there are questions i'm happy to take them hi um yes this is really interesting to you know align with the technology and the roadmap of ethereum um how you know um how long term do you see that kind of tracking um is that something just initially or is that a goal to try to stay you know a little ahead or as compatible as

possible with ethereum two over the longer term yeah in the foreseeable future we want to stay compatible with ethereum obviously i mean i'm i'm we're not maximalists but we think ethereum has the the potential to become the global settlement layer and it will need like um a larger basically the the volume of transactions that can happen on ethereum proper are not large enough and we kind of we we see ourselves as like uh the the the uh the pool that kind of catches transactions that um for some reason don't make it into ethereum so basically yeah we do see ourselves long time

aligned with ethereum okay um but the security domain will be different right it will still be the different set of gnosis um the gnosis validators there won't be any kind of uh you know roll up style security fall back no no not at the moment so basically there's no uh no plans to be become an actual layer too on ethereum so basically it's kind of currently more like a surrounding layer um we've also thought about kind of becoming an actual layer too obviously this would um this would uh need some restructuring of um the architecture and it's not it it it could happen it's not it's not precluded okay thank you very much

yeah i think you feel really good um simple question how do you incentivize builders to build on grocer's chain this is a good question so we don't do a lot of the things that many of the other chains do so basically we think you should build or know this chain because um it's a nice place we we have a fund so you can apply for for for a grant or some sort of cash injection but it we don't have the we don't have the bags of other layer ones so it's more of a values proposition you