UMA 🛠 Decentralizing your governance with oSnap

[Music] hey everybody glad to have you all here especially the AI Bots it's really interesting to see like uh this like pattern emerging I honestly think it makes a lot of sense to be able to get notes on a ton of different cool sponsor projects that you could integrate with so I'm a smart contract engineer at Uma and I worked on this project called osnap which I'm going to dive in to today so let me share the screen cool so we're going to talk about how to use Uma and in particular oap to actually decentralize your Dow so what is uma uma is an optimistic Oracle and it basically lets you pull any kind of offchain or cross-chain information into your smart contracts it's super flexible it allows you to build all kinds of different products on top of it and we'll talk about a few of those uh you might wind up building some of those related products rather than like our governance tool with osnap on top of um and the hackathon super super flexible really fun to play with super easy to set up so let's dive in so what does an oracle actually do a lot of people think about Oracles in price terms and defi terms you know how can we bring in uh offchain price data into smart contracts and that is one of the important use cases of oracles but really what it is it's a truth machine it's a way to verify some kind of offchain Truth record that truth onto a blockchain and then use that truth to trigger some kind of action so it could be some kind of price related data like what's the price of Bitcoin right now but it could be also who won this election or like did this other thing happen on this other smart contract or did this governance proposal pass on Snapshot and so optimistic oracles like Uma enforce correctness after dispute and the way that that works is somebody's making an assertion to this Oracle saying this thing happened and here is my money and if I'm wrong take my money and then there's a challenge window where anybody can show up and they can dispute that assertion on chain and they have that Financial incentive to do so because they'll take the proposer Bond if the proposer turns out to be wrong and it's super super flexible you don't have to know in advance what you're going to be verifying you just need to know the economic incentives around it and those economic incentives are what actually secures the data that you're pulling on chain it has some kind of comp computable uh cost manipulation for actually resolving these disputes it's super super flexible you can build anything from very uh datadriven objective things like the across cross Chain Bridge or you could build something like poly Market where you have a new prediction Market being created every day where you don't really know what you want to create a prediction market around in advance it's all natural language it's actually resolved by human beings rather than bots so you have a very very wide range and so what goes into it you have somebody that's requesting some kind of data you have somebody that's proposing that data either because they have some kind of like built-in incentive there or because they have some extra protocol incentive a good example is an insurance protocol if somebody is going to get an insurance payout because some kind of insurable event happened they're going to make this proposal even without any kind of incentive to make the proposal like their payout only happens once it resolves so they'll say for instance this smart contract system was hacked and I think that the people that bought insurance against it should be paid out and then that'll get paid out to them you have to have some kind of dispute resolution process and the way that Uma deals with that is we have a commit reveal scheme so you make a secret commitment on chain as a staked Uma token holder where all of your UMAS at stake if you're incorrect you can't actually see what other people are voting you have to see uh what the actual state of the world is and the shelling point there is the truth the most likely result is that other people are voting for what is actually true in the world and you want to align with them you want to have some kind of documented methodology so that proposers disputers and voters can all determine what is correct according to what your protocol is trying to prove so if you're doing some kind of complicated technical thing you want to write out the steps to verify whether something is true or not because otherwise people won't actually know how to give a response and you want to have economic guarantees that bound the cost of manipulating the system how much would it cost to actually um control the vote results for a particular vote and then even if you could control the vote results are you able to get a payout from this protocol or not it's not necessarily the case that controlling the end result of the vote will actually trigger some multi-million dollar payout within the protocol and we'll talk about that a little bit in the o snap case so it can answer a lot of different kinds of questions a couple of examples that I brought up before poly Market uses us for prediction markets it can answer like sports or like election or world event type questions you can answer Insurance type questions which are actually very related to prediction markets very very similar so Sherlock and cozy use us uh with a cross they're actually verifying that a bunch of uh relayer refunds within their Bridge system are being calculated correctly so it's very computation f it's very bot driven it's actually Bots that are making proposals and disputes in the across system as compared to like an insurance or prediction Market where it's very human driven you can build these onchain derivatives like shapeshift use us for like a very um like defi native VC sort of fundraising where VCS get an enhanced payout if the token price goes above a certain point uh by March 1st 2024 and no enhanced payout otherwise and you can also answer things like uh like what is the tvl of this protocol so you can have tailored incentives for a protocol based around what the team is actually trying to accomplish it's not just like spray and prey here's a bunch of tokens I hope that you want the same thing as me it makes for much more interesting airdrops where all of the airdrop recipients are aligned because their payout is tied to whether or not you hit some kind of Target in the future and so these are all things that have been built uh one thing that I'm going to highlight for the rest of this presentation is optimistic governance so a typical governance flow is that you'll have a forum discussion you'll have this social layer you're reaching some kind of social consensus about whether or not this is something that a dow wants to do then you'll have a snapshot vote and all of the Dow will get together and they'll vote on Snapshot and it's gasless and great and you vote on that um but then you get to this multisig where like three out of five people are determining what actually happens and that's not really great decentralization uh and it's a a huge operational overhead huge operational risk uh a really kind of like bad point of centralization in this whole flow that makes it like did the snapshot vote actually happen so this is something that we wanted to make more effective that the snapshot vote actually has some weight to it that it can be self-executing the multisig signers who by and large by the way they don't want to be multi-sig signers it's just that they didn't see any other way they don't have to be inolved at all once the Dow makes the decision our Oracle can verify that decision and you can relay those results on chain so we called this o snap uh you can basically vote gas free on Snapshot that's good in a lot of ways obviously you're not spending gas it's actually better for decentralization because the lack of the gas cost means that more people can participate you can have wider voter distribution you don't have to delegate to like a handful of delegates that are then going to vote on everybody's behalf the transaction that are approved by the Dow and this could be 10,000 people voting on Snapshot you know uh they can be automatically executed on chain anybody can dispute any transactions that are incorrect with our optimistic Oracle the truth is economically incentivized and our Oracle has secured for many years hundreds of millions of dollars of value billions of dollars of volume and you can set it up in under a minute it's very easy to set up it fits in with existing Dow flows where you have a snapshot space you have a safe you plug in the Uma o snap module it's literally like a one minute setup modulo however long it takes to get all of the signers in place so we're seeing a lot of adoption in defi um this is a slide that I updated just today because we had a lot of dows that were added to this just since eth New York which was only a couple of weeks ago uh a lot of defi protocols in particular multisig signers do not want to be multisig sign ERS they don't want to have this point of centralization they don't want to have this point of personal risk they don't want to have this uh this kind of like weird in between step we're like okay we voted on it and then a handful of people decide what actually happens so we've got an adoption from really major projects cow swap connect uh across barnbridge shapeshift many others and many more to come so we're securing over hundred million in assets right now and that's growing and we're of hoping to unlock the next phase of dows where you want to have this kind of flexibility to do a lot of like um like offchain calculation you want to have variable voting formula you don't be locked into what a smart contract says that you can do because the smart contract deployment flow is going to take a couple of months if you want to change anything about governance Dows need to be more Nimble they need to be more Nimble more open they need to be empowering to the members of it where a lot of dows right now feel kind of disempowering and this is the basic flow uh you just deploy the module you set the parameters this is onchain you create a snapshot proposal and then if the snapshot proposal passes whatever transactions you attach to the proposal can be executed automatically if they're not disputed uh if they are disputed they can be reposed if they were actually correct um and the voting system is actually going to resolve what happens to the proposer and disputer bonds so when we talked about the optimistic Oracle earlier we were talking about you have to post a bond and say take this money if I'm wrong the disputer also has to post a bond that's matching and whoever is correct is going to get the other person's Bond more or less and we were trying to gear this towards where Dows actually are right now so uh a lot of dows use safes a lot of dows use Snapshot you can just kind of like plug o snap in the Middle with the Zodiac framework so now you can vote offchain with snapshot and then anybody can execute onchain with your safe and then we'll validate and make sure that those proposals are correct before they're actually passed and typically that challenge window with osnap is one to three days so these are just kind of steps to set up um I think that all these slides are available publicly that you can get into them um but uh we would love for you to actually set this up and if you have like a real world Dow that you're using this for even just like a hackathon project please contact us we would love to co-promote you we'd love to highlight your cool project and this is kind of the flow when you're going into your safe app so you just add the zodiac app you add that to your safe you go and deploy the OSAP module and you just set a couple of parameters you set uh what collateral type usdc or we you set what the bond amount is you set what your liveness period is which is like the challenge window that somebody could come in and dispute bad proposals and this is very bot dripping by the way so realistically a bot is probably going to dispute a proposal within a minute and these are just very conservative values and you set your snapshot space URL where this is where the Voting is going to happen what the minimum Quorum is what the minimum voting period is on Snapshot and all of these go into natural language rules that are actually interpretable by our Oracle our Oracle can support bot driven flows or human readable flows and actually every proposal that goes to the Oracle gets reviewed by three people like minimum and people are incentivized to raise anything that might be disputable and then it's open for anybody to dispute uh these proposals if they actually are bad in the o snap case because it is very easy to verify whether or not something passed on Snapshot it spots doing this realistically so there are Bots that are running 247 that will automatically propose these things that'll automatically dispute these things and so kind of going into this uh idea as a public good uh risk Labs which supports Uma and across we actually run a bot that will automatically propose and execute transactions on chain after your Dow approves them on Snapshot and will actually post the bond and pay the gas for you like our bot can look and see like okay all of this is correct we feel confident posting this two we Bond on your behalf to our Oracle the bot has no special permissions or anything if the bot is wrong the bot is going to be disputed uh but we're just wanting to make this really really easy for Dows to use we want this to be super efficient so if you add o snap to your Dow not only are you able to vote gas free on Snapshot but the execution is also gas- free risk Labs is kind of covering that as a public good and these are just kind of like extras uh so we're audited by open Zeppelin all of uma's contracts are audited by open Zeppelin uh we work really closely with the snapshot and safe teams um you can set this up on a wide range of different chains and any chain that we don't currently support we'd be happy to support if they're evm Chains It's not that hard to deploy Uma there it's really beginner friendly it's very approachable you don't have to be a technical person or understand how to build raw transactions to use osnap and uh even to create proposals we have a transaction Builder that's integrated deeply into the snapshot interface so token transfers are very easy contract interactions are very easy you don't need to be an engineer to actually create proposals and be actively involved in your now and you can just deploy and interact these things through familiar web user interfaces the rules themselves are human readable and our Oracle system can actually support that and coming very soon we actually have a native Safe app and a native snapshot plug-in that is specific to osnap and kind of going beyond that uh you know there's all kinds of different things you could build around optimistic governance we know that Dows are not stagnant we're not at like the best state that Dallas can be we're not anywhere near that I think that osnap is a really big step forward because you can have more decentralized voting systems and that's great but even if you have thousands of people voting on something and those votes are actually enforceable on chain and that's what osnap allows in the longer run not everything needs to run through a voting system necessarily and we're seeing a lot of uh roles-based kind of Dow governance systems that are coming into effect um a lot of experimentation in the space and we want to allow that and Oracle can enforce these rules in a very decentralized way if you can write down the rules of your governance system these contracts can enforce them you're not limited to what's easy to implement in smart contract code and that's kind of how we wound up in this rut where it's just token based voting and just multisig because those are the easiest things to write into smart contracts that's not necessarily how you would organize a group of human beings and every Dow is unique every Dow deserves the ability to create their own rules that they understand that they're able to tweak as needed as they learn how their Dow is actually operating and this is kind of our most advanced form of generalized Contracting these are all natural language rules at the end of the day you read through them you enforce them and every Dow can have their own so you can have maximum flexibility with minimal trust because you have this very battle tested Oracle system that can handle natural language rules so yeah uh you can learn more at um. XYZ snap DM me on Twitter at JD shut and I'm happy to answer any questions I see there are some things in the chat here let me see I guess mostly the autopilots sending letting people know that they're transcribing but uh DJ Anthony you you have your autot yeah hi um thanks for the presentation it really informative um I've just you know created a dow a programmable treasury on the juice box and the thing I'm really kind of struggling struggling with with is yeah is just the voting and the problem I have is most of the conversation that the chatter is on Discord I didn't see a kind of what I was really kind of looking for is some sort of plugin where I can it can be you know kind of coordinated to Discord because again it's I really love what you're doing it's going on another platform on another platform on another platform you know um so how do I how how do I integrate with Discord with some sort of b or application command so yeah that that's a really good question um so we do have some monitoring Bots that can connect to Discord so they could alert people of what's happening on snapshat but if you wanted to build Discord voting as a hackathon project um uma's Oracle could theoretically enforce the results of Discord votes as well um and that would actually be something that would be really interesting to do that's not something you currently do but if that was something you wanted to build out yeah yeah yeah I mean I mean either through the docs because it's just you know the because it's about really the Dows it's really about building the community and then obviously having the automation so Discord for our is just perfect because you know they're in game or whatever they're doing and they can just dip into the D um so I I mean what would be really great would be if I could DM you or your team and then maybe you know you could direct me like with the tools to build it but um you know you know so for example we built we really liked juice boox the juice box Pro protocols the way they kind of build in the Dows it's just um you know it's just it's just um comp it's just um there's no noise you know um I get what you're saying but I just I just need to get from A to B right and then maybe and then you know later on I can get from from B to Z but you know thanks for the presentation you know so yeah for sure what what is the Dow uh that you're building trying to do okay so the Dow I'm building originally from the Caribbean it's called the island Al and essentially what we're looking to do is fund Web Two web 3 projects across the um Caribbean Community Caribbean Network um born of three inflection points Resurgence of what we call like the parner system which is a community based Financial system amongst the Caribbean Community secondly obviously the blockchain and thirdly what we call the CARICOM 10-point plan for reparations um which is a mandate across uh you know 50 governments in the Caribbean about about uh technological um transference to um to the global to to to get to parity the global standard sorry it's aveng down basically yeah in N show um typically uh two out of the four world's most um what sanctioned countries are in the Caribbean Cuba and Venezuela but that as a consequence of that it kind of blights the rest of the Caribbean um and stuff like so we're we're we're trying to fund we're trying to create like a venture to fund Web Two web 3 um projects um you know basically across across the Caribbean um based on those three inflections um yeah no I think that's that makes a lot of sense um I think for so are you interested in Discord votes being used to determine what the Dow does correct yeah yeah because it's just the it's just because we're building community on Discord it's just it's just easy to kind of access and um and and and also there's not a lot of noise from you know you don't get spammed you know by all by emails all this kind of stuff and um it's just it's just easier to build it you know you know using the Discord kind of platform or you know because that's you know it's just channeled right so and for some of the other functionality that we need to use um like um for example like token gating which is something that want to get into it it just lends itself very very easily um and people can connect to us like very easily um on Discord um and um yeah I mean it's just ease of use um I like the idea behind what you guys are doing but I'm I'm just thinking about okay you know I'm G to have to spend you you know a disproportionate amount of time going through the docs all these kind of things and yeah um you know what I mean like I I just wanted to build it if I can governance the voting you know you know using the Discord platform you know if I may um I I think you definitely can um so the rules can be tweaked I think it's perfect for a hackathon project so you go through the o snap documentation all of the natural language rules are kind of geared towards snapshot because we were going for that particular use case but tweak them to gear towards like here's how you verify publicly somehow that this passed on Discord so some kind of public source of information that disputers and voters are able to see what happened on your Discord as long as they can see that uh you're able to enforce that so it doesn't necessarily have to be snapshot and I think that that's actually a fantastic integration and we' we'd love to help you with that I mean we can and like support figuring out how to do that hopefully it wouldn't take up like too much of the time because I know it's like a multi-week hackathon and like yeah too is like the voting system is like not the point the point is like what the Dow is trying to do so we want to make it like build your own voting system if you have a voting system in place that that's Discord Focus that has some kind of public source of information you can verify it with uma's Oracle yeah I i' really love to do that and I'm just conscious I I don't want to take up other kind of questions from other people so um what would be really great is if I could um DM you and and and the team is it at JD uh sh uh uh you know on Twitter and then we have a thesis and everything like we've been building it's just been like you know we've been building this like over um you know uh for quite a while but uh and then we just got the treasury kind of launched we're going to have to start a new kind of funding cycle but it's it's the governance bit is obviously as as you pointed out is it's important but also you know it needs to be um you know kind of integrated you know and for us I think you know if we could you know I don't know what the exact final solution would be but if we if we get it to kind of live on Discord you know that that would be just so much more kind of helpful you know um for our community um yeah yeah no I mean DM me but also uh in the eth global sponsor channel for oh yeah yeah like definitely drop it in there we have a lot of people watching those too so you'll get a fast response yeah yeah I'm actually taking part in the E Global so as well but this is a this was a um kind of the second project so I I you know um but yeah I'll I'll I'll do that now thank you very much yeah for sure and then uh rash me see you have a question or hand raised yeah uh hi John uh thank you thank you for the great presentation and uh I like the approach that you guys are solving towards uh the important problems and DS uh I wanted to know how do you guys resolve this dispute can you uh can you explain a bit more on that uh the dispute resolution topic yeah for sure um so there's like a 24-hour commit period And this is kind of like rotating every day where if there are any disputes you have 24 hours to put a secret commitment on chain of what you think the answer is and during that period and actually prior to that period people are posting their arguments on Discord all those arguments are kind of presented in the Oracle dap interface to make it really easy to review as a voter and see uh like what you think is the most compelling argument here like what what does the evidence actually suggest in the o snap case it's much more objective than something like poly Market where you might have somewhat ambiguous wording somewhat ambiguous outcomes very natural language very uh like I don't know like uh like prone to ambiguity with o snap if something passed on Snapshot and snapshot itself is not getting hacked or there's some kind of compromise in the snapshot admin um that is something that can pass through uh so it's kind of more similar to a cross in that case uh like cross is the cross Chain Bridge that uses Uma and has done over2 billion in volume where you can kind of objectively see like did this happen on this chain or didn't it like you can just go and look um so there's not a whole lot of room for ambiguity there uh osnap is kind of straightforward we're expecting that any disputes that come up are actually going to be pretty bot driven and it'll be like oh somebody's making a proposal to this safe that has the osnap module but this thing wasn't actually approved on Snapshot you can go look at their snapshot it wasn't approved it gets disputed automatically by a bot and those Bots are running 247 but then you also have this human verifier Network where every proposal is getting reviewed by multiple human verifiers that are flagging anything that looks like it might be disputable and uh we kind of have like those layers of Defense I suppose uh to different types of proposals okay thank you so uh when you say the verify Network do they have per to VTO or is it more like an after something happens we'll manage the crisis or do we like prevent the crisis you prevent the crisis because a proposal comes through but the proposal is not uh settled until after the challenge window is over so you only actually use that data or that assertion if it doesn't get disputed so in the o snap case that's like a one to three day challenge window because we made it super conservative for Dows realistically it could probably be a lot shorter and be totally safe but like Dows are not in a big hurry they just want to prioritize uh like safety um so if somebody is like hey like the Dow said they're paying me $10 million to this address and they didn't actually say that that gets disputed and the person that made that malicious proposal is going to lose their bond and they're going to lose that Bond realistically to a bot because Bots are looking at these things and they're they're competing to be the first ones to dispute if there is something that's disputable okay thank you so much thank you so much for explaining that c awesome are there any other questions anybody has okay and if not uh thank you all for attending thank you John for the great workshop and presentation and I hope everybody has a great rest of your Friday and a great weekend so thanks thank you everybody