Aave Governance v3: how to scale control & participation over a multi-chain DeFi protocol - @eboadom

Transcript

yeah yeah hello everybody so hey I'm I'm Ernesto I I am co-founder of Border developing big labs before I was on David Jones team I was a CTO on this protocol probably quite a lot of you are familiar with and yeah today I will be talking about the next iteration that will be applied on on the other protocol for for governance So currently we have apple governance V2 so this is about our governance B3 yeah first of all like for I don't know

how how familiar is everybody with a with other governance because it's a bit more hidden piece of the puzzle compared with the liquidity protocol that I think everybody knows about um but in general is quite similar to to a lot of other like governance systems on Qing so basically it's a set of smart contracts and with which the app token holders and control the every other piece of the Arab ecosystem so this is more Contours control all the instance of the liquidity protocol so the current one is rb3 but also rv2 and even rb1

um the safety module which is this where people deposit survey and it is submitted to to slashing if anything happens with the liquidity product and even the other token itself like some other minor pieces of the ecosystem um how it works is well you have other tokens and somebody creates proposals that basically are executable payloads to do certain action on the protocol and and then like holders of other on ethereum or people that receives delegation from those holders can just go to the smart

contract and vote in favor or or against the the proposals um well others as you probably know using not only on ethereum it's also like on polygon and Avalanche from optimism arbit room a lot of places so the current type of governance V2 also contemplates that so apart from like the voting that happens from ethereum um whenever like a proposal needs to be executed on on another network uh well the system just supports like what we call cross chain so like the decisions that are taking on ethereum gets breached to to whatever Network and requires the execution

now this type of governance is exists like for let's say quite some time you can consider quite some time I think around two two years more or less and one of the main problems that it has and to to which everything boils down is that they are expensive to run like if you need to vote with your tokens you need to do a transaction on ethereum as we all know ethereum has a lot of good things but still is not the cheapest Network to interact and this in the case of of assistance like governance on which you

you require as much participation as as possible well it's a barrier so when you have other holders that have maybe one ABI or tengabe spending five dollars to emit the vote and especially right now the the there is quite a lot of proposals all the time uh was spending like five dollars every time that you need to boot every week once or twice is not something that that will work like in in scale so the maybe the main question is um is there room to actually improve uh what is the fact like standard of on

chain governance systems and obviously always keeping this on chain and with no no points of centralization well the idea is that yeah it's it's possible that is this talk about uh and and we think that the we we have sold the with cyber governance V3 and this is well a bit of uh a diagram of the system that I will I will display a afterwards so as I mentioned before um Abby is basically in in all the big networks out there the liquidity protocol I'm talking about

um the execution of proposals on government governance V2 is also able to go to all these networks because somehow you need to control them and so like can we do kind of the same with with the governance system and and optimize it uh so yeah the idea is that we can the problem again is that right now we have almost 200 proposals all on chain proposals that our holders needs to vote on so and and it's getting accelerated like at the beginning of the year there were maybe 120 and in three months we have like almost 60 more so like this High load is creating a bit of of a problem that needs to be optimized

a way of optimizing this is is relatively simple to understand if we have like ethereum that is a bit of expensive environment but then we have multiple other environments on which hobby already has trust like polygon Avalanche or optimistic relapse even CK Roll-Ups that are are coming uh close uh what we should use in those those networks somehow like uh and and advantage that they they bring to the table this is also pretty important because and it was not possible like even maybe one year ago and for sure not two years ago

um because the bridging ecosystem like the brilliant technology is getting material mature more and more so like what before was really really experimental when for example Long Island we we introduced like this cross chain system and now is is becoming a reality so yeah it's the the right moment to to experiment on the on the application on the application side so which are the parts of this new system well if you we are not not just removing ethereum from the picture a ethereum is still will will still be the the core of of the other governance but it will be a bit different so um for what we want liquidity sorry

liquidity for what we want ethereum is for because of its security and because everybody is already understanding that the ethereum will become like a settlement layer and a security layer in in the future so in our case we know that the the participants on the on the governors needs to be the other token holders we know also that the the balance of of Ave um like Mayor Lee is from ethereum and will stay that way because it's the most secure place still um so we know that the core of the governance still needs to be with you but the the big change is that is not anymore or let's say a monolithic system

where everything is on ethereum but ethereum becomes like the like the governor's part of ethereum becomes like some kind of Access Control component a m validation component so like it will be a smart content on ethereum the one that will will check that everything is valid where somebody votes and it will be the one that checks that the Quorum requirements uh to understand if a proposal is approved or not and they are passed or not basically so yeah but now like before boating was in ethereum and this is the big change on other governance V3 because we know that voting is the big bottleneck so people

ideally should not spend almost anything or anything while while voting and this is what we are actually moving to other networks so uh the idea as well maybe is more or less clear in the in the picture is that the all the networks will be candidates for voting a pair proposal only one network but all of them will be candidates and this will be done by deploying this kind of voting machines on each one of the networks and then by selecting every time that somebody creates a proposal which network are going to use for this proposal so yeah and then ethereum will still be

another Network on where boating can happens but the idea is that it will be like the last resort like a voting machine or voting Network so if everything goes well and is way cheaper to vote on polygon or in avalanche let's say well proposal 200 will be voted on polygon 201 maybe on Avalanche because it becomes a bit cheaper at that point but if there will be any problem with Avalanche or polygon the system can like proposal 202 could be voting on ethereum and and basically it will mean that it will still be expensive to vote but especially the centralized and and the system doesn't get blocked or anything like that

now all of these as you can imagine like involving so much networks like this a bit of an improvement on the on our cross chain infrastructure which helps like to right now just move the decisions to other networks but in this case it will be more back and forth movement because you need to create a proposal move it to go to some other Network return it to ethereum to validate move it somewhere for execution so for that um we are also improving the the other cross chain system that we have um because right now like the reality is that it's quite dependent from the canonical breaches like so whenever we

want to execute uh something on the highway B3 polygon our system under the hood uses the polygon canonical bridge to send a message to that Network or whenever we want to do that in the same on relapse at the same with this new system we are and knowing that there is more Bridge providers apart from the canonical out there we are just building the the base layer to be able to for Abbott to define a consensus of which bridges are acceptable which not if we accept like two out of three as a valid confirmation or in some cases maybe it's just with one breach is the most secure one so we just use that that

bridge so what this like the big advantage of introducing this kind of layer is that the expansion of the protocol that we expect that will will continue during this year with this new networks that are appearing well this will be more flexible for us on the governance side on the control side of those networks to expand there and and not having any kind of centralization via any kind of multisig or anything like it will be just if an instance of fabric gets deployed there pretty easily we will be able like if exist if anybody exists to plug our cross chain infrastructure and support that that Network I'll move from day is here

yeah so now I I explain a lot of like Theory but how it actually works how is the flow of of a proposal on this new governance it's actually not too complicated I I hope and so yeah everything starts like for who is not so familiar to create the proposal long habit you need to have other tokens at the moment is eighty thousand have a token which is quite a lot um but but yeah everything starts with with some semantic that has the other tokens it's a bit more complicated than that

but let's let's assume not so like the the first step is that the proposal right now will create like what the so-called payload which is just executable code the smart contract uh on the network one or multiple payloads by the way on the network or networks uh where the action needs to be executed uh at the end if the proposal passes so in this example like let's say that we want to change some configuration on Lobby three polygon on the liquidity protocol or list an asset so um The Proposal will deploy a smart contract which is the payload on poly when the the this payload gets deployed

um The Proposal will receive from the smart content identifier just a number and of course like he will have the reference that is a payload on on polygon so with that [Music] um he will go wow he receives the identifier first of course and then he will go to ethereum which is as we recommended like the validation layer and there the the governance core system will identify okay this uh this wallet has the 80 80 000 require other tokens or delegation of other tokens if that is the case and with that identifier of The Proposal in polygon he will be able to to create let's say the in The Proposal

on ethereum here is when the the magic starts to happen so immediately after the proposal gets created on ethereum if it will be on B2 there will be like some some time delay and then voting will start on ethereum right now it's different uh once the proposal creates like there will be a bridging like to polygon in this case because it's also the voting chain and what we'll get a breach is the block hash of just before when the proposal was created I was playing a bit more why um but yeah this bridging will go through our cross chain infrastructure

and we'll kind of notify uh our polygon governance part both in part that there is a proposal to to be voted on so it gets received by the our so-called voting machine on volume and well there is some technicalities but the Boating starts on polio now if you have a token you just go to polygon and just boot like with your other talk but the main question here is like okay but I have a token from ethereum how can I vote on on polygon directly and without bleaching this is important

there is no bridging involved like so you you have your wallet without ethereum you do a transaction on polygon not on ethereum and Bridge to to polio and the answer is called storage proofs and that is the main reason of why before I mentioned that when the proposal gets created on ethereum the Block hash and gets forwarded polio so how this works is the blog of ethereum contains all the information about the balances of of the users it's just that it's like in a compressed way so like because the the block has like um the so-called state tree which contains all the information all the

data that is stored on ethereum and basically it acts like a snapshot of of balance or whatever other data is so much more contrast from ethereum so what we do is that we send the block hash to polygon and at that point we have a snapshot of ethereum on polygon so when the user votes on polygon is actually it's not the remote control is not reading the balance the user is passing its balance and is proving with the storage proof that the he had that balance on ethereum whenever the snapshot was taken so that is the way of you don't need any religion you just go with your balance and the smart contract

on polygon understand and can validate that you actually had that balance from ethereum and that is kind of the magic so people just bought from polygon and after the voting period And this is kind of the same as as before on our governance we do the decision and gas breached so the decision will be for example that proposal had a 200 000 Jazz boards and 50 000 no boats and that will get Bridge normal Bridge not storage proofs to to ethereum Via our same cross chain infrastructure a it gets forwarded to the smart contracts

from ethereum from the bridge and then like there is like some queuing time lock same as as right now but uh if for example there is like a payload also in ethereum and and on polygon something will get executed on ethereum and to execute something on polygon again we will send like some message via our cross chain infra that will be received on volume and and the change will be will be done and yeah that's it about the the flow but well I didn't explain so much about the crocheting infrastructure just about

this consensus and so on uh so God is about like this is there anything new or not on the crossing infra so yeah Crossing infrastructure was kind of this way so we have a in the left like a representation which deals with proposal right so what happens is that in in different time slides a of the life cycle of The Proposal let's say that this this is like an execution of of a payload that was already approved on ethereum uh so the system will send the proposal the governance system will send the proposal to our cross chain and there is a component on ethereum

which understands which are the let's say why listed Bridges underlying bridges that could be like challenging ccip could be layer zero it could be the canonical breach of of the network it can be whatever and then when all of these breaches or at least like a like a kind of a number of them sends the message and these gets received on the recipient chain uh we have another smart contract there that that understands which is the consensus rules so imagine that we have yeah layer series CIP and the canonical Bridge so we could configure that if two out of three uh tell us that that this message

was approved on ethereum well the the system will just accept it as correct and will forward to what we call the the executor which is a simple time lock on this side uh that after the delay will just send to the well in this case the one of the instances of the protocol which is the the polygon one so which are the main advantages of the of the system like from high level perspective so first of all what I mentioned uh it will become really really cheap to boot as you can imagine because it will be a relatively simple transaction on polygon and what right now could cost

five dollars to emit the boot potentially it could cost like one cent or two cents to a meter boat on polygon and this is pretty huge because uh IV is like for example paying on behalf of the users are pretty realistic at the end of the day like we have 100 I think 200 000 other holders more or less but realistically if um 5000 uh participate on voting so it will be five thousand uh one cent so it will cost about fifty dollars to cover the whole cost of of governance and it will be factually free uh for for the for the voters to participate will be a

some kind of system via with relayers and meta transactions another big point is that we have these dynamics of Delegation of both in delegation and proposition delegation and this the big change is that there is no change it will just work as usual it's just that from the on the polygon perspective there it will be possible for our smart contractors to understand that this is a balance of the location instead of balance of voting but for the user is just exactly the same for the delegators is exactly the same um and then another big big Advantage is that right now like with this system of models like it is fairly easy like to

implement like the Boating logic even on non-evm chains uh because at the end of the day like the ethereum part is quite agnostic of of how the voting works so within one side only sends somewhere a proposal that needs to be voted while is from where it receives and the the response and and then if it's white listed it just accepts uh yes boats and no boats and validates uh if that is is good or bad there is like some some other advantages but it's a bit more technical so for example uh those of you that are like developers like you know that the for governance tokens

usually uh you need to have snapshots of balance so every time that somebody transfer and a snapshot gets made because the governance systems read from those snapshots historic snapshots so with this system we don't need any more of that like we just can remove all the snapshoting and it will be just a completely simple erc20 token let's say yeah um then the smart contracts that are only ethereum and that hold up will be able to load exactly the same not exactly the same as all other voters because in that case like the the contrast cannot interactively with polygon obviously they are on ethereum but it will be possible through our bridging system

like the Smart Control on ethereum that hasabi will send a message that in this case will be I want to vote Yes and it will need to set it to to the voting chain which can be polygon for example then is a big change that we will be able to add new assets that the with voting power right now is not so simple because for example there's an options so like if a token if you want to add like I know some random token which makes not so much sense like but to give it both in power on ecosystem it requires that the token is implemented in a certain way which means that like tokens that are not upgradable and don't have those features are just not eligible for that then you will need

to wrap them and so on which is to complicate this way is fairly simple to just Implement a strategy that says okay we want to give like 10 of the voting power to this other flavor of Fabe let's say and and we'll just it will just work and then yeah something more on the technical side is that a it becomes way more isolated and way more modular in the the connection between the governance system and like liquidity protocol and and everything else so yeah for now everything seems like advantages like fully decentralized like cheaper like and no disadvantages which are the disadvantage well actually there

is not really many disadvantages or almost non-disadvantage is just that with some new systems like and because of how how everything goes around this storage proofs and that is a bit interesting field so one of the the problems is that like CK or Melody Roll-Ups and they are not so happy with storage proofs because of how how the hashing for example works the default hashing of the network so what it even if all the transactions kbm of polygon will be cheaper for a reason whatever you try to do storage

proofs is actually quite limiting because the the gas implications the computational implications of doing that in that environment are are high you have also a bit of more unpredictability of the boat in duration but pretty minor and this is basically because whenever you ask you need to do bridging like back and forth you will have a small delay of of that bridging obviously it's not a optimism or Arbitron optimistic relapse and that have a like big delay let's say for a fraud proofs to come back is you cannot use it unless you want to have

like a a voting a period that is really really long but both of them for example support like Fast bridging with third-party breaches and with our cross chain execution it could be perfectly possible just not using the canonical breach of optimism orbits and then like on the not so technical side it's going to be interesting also how the community is going to choose on which chain to to vote every time because as I commented before it could be a situation on where like the next proposal that gets created and now we end the proposal chooses um can be voted on polygon and potentially could be a proposal even that doesn't touch polygon afterwards it touches only Avalanche a and the next

proposal it could be voted on Avalanche next proposal could be voted on optimism so it is going to be interesting like how the community is going to decide to uh where to to vote if you just where is cheaper or or not and yeah that's it if if there is any Quest if we we obser like that there is like

some change on the Dynamics more turn out of building participation so first of all the system is not live yet so we don't know but how we thought about this is that there is not really any potential negative consequence in terms of participation so like even of course it's not realistic to to think that just because we introduce like and made cheaper voting that like all the other holders will vote because it's not a it's not a matter of being cheap or not it's just a matter of is people capable of taking decisions like do they want to do it it's not even if it's free or not so uh we don't know but pretty realistically we know that there

is a lot of users that is they say like we don't vote because it doesn't make sense if I hold one hundred dollars in other to spend five every time that the proposal happens so realistically if we have participation of let's say 1 000 or 200 2000 people right now we should expect at least multiply that for by some numbers like maybe not by 100 but having like a bunch of thousands more participating is pretty clear yeah