A Technical Deep Dive into the Evolution and Future of Gnosis Chain | Dapplion | ETHTaipei 2023

have a type lion the merger the merge coordinator of ofnosis chain so no sister is one of the first ethereum section scaling solution so he will give us a deep dive of the evolution and evolution and future of no system yeah let's give a clap to like thank you thank you everyone for having me again awesome thanks for the intro so I'm dabbling on the fork coordinator of nurses chain I just make sure that all the chain upgrades are done safely and that we're moving in the direction that we want and today I'm just gonna talk a bit about the chain history where the game is coming from where are we going and what makes Nazis different and special but unique so nurses chain came from xdi I think many of you would be familiar it's a quite old and historic chain of ethereum that was launched with the original purpose of having an native token that was stable it was done with partnership with maker now that's when it got the name the key characteristic here is that the native token is is die so the next year the the networked transition from POA Aura parity style of consensus where the consensus said was closed into a more permissioned permissionless style of governance with post now then in 2021 NOS is proposed to do the merger here I'm gonna use Fusion because we have done quite a bit of merch and then on 2022 the beacon chain was launched with the goal of eventually transitioning from post now to a fully proof of stake permission system and finally in 2022 we did the merge now nurses chain and nurses Beacon chain are the same thing it's a fully permissionless proof of snake chain that mirrors the architecture of ethereum so just to give you some numbers on the execution side the networked picked up significant interest during the last bull market 2021 2022 we have like 300K transactions per day with applications like Co-op Dark Forest azuro circles giveth one Hive and many more and then on consensus we've had a steady rise of keys of course because withdrawals are not possible so this line can only go up and now we are at about 120k keys and there has been a strong emphasis on solo validators nurses run some unique programs where they directly incentivized homesteakers to buy their own Hardware with projects like tab node which one is really cool and definitely had quite a bit of success so to point the few differences with ethereum just to put everyone on the same picture on the consensus side it states quasi equivalent so it's literally the same code base there are no changes to the source it's only some config values that have been different to align the ux to what the former XDA chain was the slot cycle is faster it's five seconds instead of 12 in ethereum and the cues for onboarding and off-boarding are also faster for for this reason you can run all the clients uh lodestar Lighthouse and Techo stock like same binary no modifications and the other two ones you have to build them we do that and it's published uh so yeah full client diversity which is really cool to have in execution now after the merge almost all the components are the same so same ebms hsrpc that has been like like this always but before the merge due to post now and all these different the consensus engine was different but now it's literally only small tiny difference on the state transition function that makes us have a slightly different code base so nethermine has been driving the chain since a long time so kudos to them edicon is now now coming online the their client is completely implemented but we are carefully rolling it to production uh just to make sure it's safe guess really exciting it's going to be nosy scapable but only post merge and beso not yet so just to touch a bit on before the merge uh the characteristic of of Exile once the set of personal contracts that had a different set of functionalities that now are covered somewhere else so before the merge we had this block reward contract that handled payouts of the proof of stake system and it also handle all the bridge logic to make sure that the die token is fully marked uh now this is disabled and it only cares about the bridge functionality we had also random audio with use rundown algorithm now we use the one from the beacon chain so this one is disabled and then we had other things like permission and stacking stuff which had some early attempts at different features that we will do later but on the merge all of this is disabled so yeah ethereum equivalence again the main component here is the fact that the die is the native token to have that this way we need modified Logic on the clients on the state transition function it's literally this one line we call the reward contract and we ask the contract hey what do the authorized features Bridges um agree that we have to send to the different reward payouts and then that result is minted to the user with support in this feature yeah having the native token you can have users participate in the chain without having Market exposure so one of these clients here is the key piece that links to the rc20 to Native token that's the special Bridge unique to nurses to support having the native token backed by something else and then everything else and any other message cross chain is used with the Omni Bridge um fun fact and we hope stays that way these two has never been hacked so either we are lucky or the the team is just really good at doing this um there has been a bit of uh drama recently if you guys remember with usdc getting a Deepak and in consequence die which was unfortunate there is a path that we are studying where we can diversify the basket of tokens that replace the eye because it doesn't technically has to be specific dye as long as it's something that shrunks the dollar and we understanding these options that's on on research and very very interesting to see where it goes cool so on execution that's the meme from The One and Only We also did a merge and I just want to recap a bit how it went so it was a big Endeavor too to finally get rid of the Legacy components and become fully permissionless so yeah we disabled the personal contracts Cape only this tiny thing on the Block rewards and to come here with it many definites many Shadow Forks we also launched a new testnet Seattle remember that one we had wide Community participation on all the phases and yeah the merge was was good we only saw a temporary drop on participation but no technical issues and eventually the happy side effect is because people were losing money due to the not obligating the nodes now we have really good participations so it's good to see so as a result of the merge testing we now have these Seattle test net which it looks a bit like sepolia on ethereum where the consensus site is permissioned it's closed now for now and only run by trusted participants the client teams and other other operators so we can keep the consensus set very small and the sync is fast we plan to have a second test net that it's much bigger and it would look like kind of patter which was fully open so validators that want to test big setups or big openators can use that and for this one we will probably open like a fraction of this loads maybe like 10 to trust that Community participants as long as they run the chain and they don't cause issues with finalization should be should be fine cool so after the merge due to this complete alignment on the specs on both consensus execution uh nurses inherits the same modern architecture than ethereum going forward it will be compatible with all the upgrades being whatever ethereum goes we can follow it can support roll ups and all that good stuff and something we're investing a lot now is on ZK breaches so probably and truthfully the criticalist component of gnosis is the bridge that both supports all the non-native value and the native value there is a team working on on ZK breaches and they have successfully put the alternate protocol the Altair like Lion protocol inside of a circuit this is live today and it's really cool to see the thing is it's I don't think it's secure enough or that we want to go so these and other teams are working on putting the entire consensus logic inside the circuit and then we could have like a truly one-to-one transless Bridge which yeah that's that would be really cool to see I actually checked with cryptographer from the AF today and he says it could be done in two years so that would be really really exciting to see so on the other front Mev remains like a huge worry of ethereum and on nauseous values protecting privacy is also really important and we cannot deny that still today front running is one of the biggest ways for Traders to lose money and it's this has been causing huge instability issues in the chain we even saw a very new type of Arbitrage a couple weeks ago were as a validator proposed an equivalation on purpose to steal mbb and that's a really worrying Trend that we don't we don't understand all the unexpected outcomes that this can provide so I think it's important that we invest on that front and we can do something at the L1 layer itself and the main effort that we are taking right now is the chatana spawn chain that essentially means that you will be forcing proposers to commit to a list without knowing the content of the list via encrypting the transactions with tkg ceremony where the key is available later for them to decrypt like the specifics are still on heavy research this has similarities to both account extraction and inclusion lists but it's a very promising option that yeah we're looking into it and just to start there is a b0 version that we are looking to kind of start to roll out this this infrastructure in a way that's still uh not secure because it doesn't have slashing conditions um it's complete opting and off protocol but kind of how it will work is uh users that want to do any action on chain they will submit this transaction to an RPC that will encrypt the transaction on on this behalf so there is a trusted link and then these these transactions are submitted to a contract where then validators that opt-in build from that transaction list and again this is an incomplete solution this is what we're going to do eventually but this will give some information of how these interact and have some buying from the community and hopefully have enough push for this to become a reality and the last point that I want to touch is because if nothing is inherits the ethereum architecture it is vulnerable to customized fluctuations like right now transaction of nurses is pretty cheap but if the network gets adoption then it will become as expensive as ethereum as any other chain and this happens everywhere so there is this fascinating initiative of reserving block space two stickers and like the idea is if you are an app that is heavily involved an international stake and then you are given a specific fraction of blog space and that would mean that either the base fee is not charged or it's fully refunded to some set of conditions that you give to the chain and I think these would be a pretty nice Game Changer going into the next Bull Run so yeah if you want to find more we have great dogs great Community please run a note it's very accessible we'll be more than happy to to help you we have download packages and everything so yeah thank you so much for your attention [Applause] do you have any question from the audience hi peace peace I I'm quite fascinated by the last slide about the reserve block space would you mind just I guess I don't have a specific question but would you mind explaining it a little bit more since maybe we have a little bit time sure so yeah this is still a research idea we don't have any specs or anything specific but I think application of noses and why it could work is keep in mind that here the native token is is die so we cannot really burn it because it's it's a claim on the token on the other side so how it works today when you burn fees on noses they don't go to the zero zero zero address they go to some fee collector address that it's controlled by the nurses Dao so it would it would be possible to Route some of these phone These funds according to a set of logic back to the users or even not charged it at all but that would involve changing a bit of the of the protocol so it doesn't hurt the security guarantees of the network and yeah what what other specific mechanism do you want me to talk about here that's great yeah now I sort of understand like where it comes from because usually we think of it as burning but you just explained the kind of the mechanism so yeah yeah I would imagine um as long as as long as you can define a set of patterns that uniquely Identify say app usage then it should be good uh say if if everyone that interacts with your app only uses a specific set of entry point contracts you could say anyone that calls these contracts the base fee is funded like we would have to understand the risks of Melissa reduces and if that can be exploited like what are different those vectors okay cool any other questions yeah so uh I wonder um so do you mean like the CL of nurses will be like more creative uh new consistent algorithm and but for the El it will be evm compatible in the near future yeah so um if I understand you say if the CL and L are compatible with ethereum yeah the Evian logic and yeah so at the execution uh where are we here yeah so the the evm of gnosis is and has always been exactly the same code AS ethereum uh so the only difference really at code today is on the execution chain the state transition function everything else on execution and consensus is the same folks of yeah and any other question from the audience hi can can you just Advance one slide to the difference of parameters for ethereum and gnosis right um okay the seconds per each one block is six okay okay seconds per slot is five right okay no just just uh thought I saw something uh I misunderstood but cool sure no problem oh they are not Power of Two but okay any other questions okay they got cool and uh again and give claps to Lions brilliant talk thank you thank you [Applause] [Music] foreign [Music]