 How's everyone doing today? So this has been absolutely great, hasn't it? I mean, we waited three years to come back to DevCon. The talks have been absolutely incredible. I just can't stop smiling. I'm seeing so many friendly faces, so many new faces. I'm just so happy. There have been many themes of this conference. The merge, of course, is the highlight. We have a lot of L2 development, ZKs, DeFi protocols with their financial primitives, and many more. But we can't forget about governance. After all, governance, or lack thereof, has a massive dependency on the success of our protocols. So today we're going to talk about our governance structures, why they're failing our DAOs and protocols, and how to approach fixing it. I might have a small live throughout the presentation, so I hope you can spot it. But yeah, again, I'm Charles St. Louis. I'm the Chief Operating Officer at Element Finance and head of governance. I previously worked at the Maker Foundation as the decentralized governance architect, and today I'm talking about how to fix our governance systems from the failing our DAOs. So, of course, Element is a fixed-rate protocol for variable and yield markets, but today I'm talking about a governance protocol that we also built. So our company is very focused on research and development, and governance is very dear to our hearts. So we spent a lot of time countless hours on this governance research, and we came up with something called Council, which we'll get into a bit later. But before I talk about all the challenges and things we dislike about governance, I just want to do a quick, little fun journey down history lane. So since humans have existed, there's been some form of governance and coordination. Let's say it starts with tribes and clans three million years ago, where people started grouping together to make resource gathering a lot more efficient. Give or take a couple years later, we had kingdoms, and that came with kings, taxation, and better resource defense for the resource gathering that we developed over the past thousand years. The next came empires, which introduced a large amount of hierarchies and power inequality. But the thing about empires is that they always fail. From empires, we came to democracy. It didn't come easy. There was a lot of battling and fighting to make this happen, but it truly introduced the first concept of the vote, where everyone in society, or at least some people in society, inclusivity didn't take that much time to develop through that. But each person had a say in their governance decisions and how things move forward. So democracy was great and all, but it also had some problems. Tyranny of the majority and also bad decisions getting pushed forward because people just simply didn't want to vote. I mean, I can probably count, like, 80% of you who probably don't want to vote in their local jurisdictions or in decentralized governance. So how can we expect this to be an efficient system? From democracies, we had republics, which weren't too far away from each other. As you can see here, like, they're not mutually exclusive. The only difference is that we're essentially trusting people who are paid, compensated to be experts in their area to make decisions for us. And obviously, with republics came constitutional republics and so on. It is paraphrased history. So today, many forms of these governance systems exist. They're emulated, they're improved upon. But we also have new forms of technology that have evolved our governance systems. The internet, of course, blockchain, and many more. So at this point, I hope you can kind of see some similarities between what we've done in the past and what we've learned and how they translate directly to our systems today in DeFi and crypto. So the first one, I mean, direct democracy, I think it's pretty self-explanatory here. I think the closest one that we have is the one token one vote. We have the new system of reputational voting, where you have non-transferable voting power based on your contributions to your community or your society and whatnot. And then we have an emulation of representative democracy, which is token-weighted voting with delegation. We even have republic-like dows, where teams are voted in to do their areas of expertise and basically deliver, like, finance work, operations, and all that, but we trust them to do that work on behalf of our dows to push it forward. So by now, I'm sure you're all wondering why I gave you this weird history lesson, but I swear I do have a point. And that point is governments fail. They failed and evolved over thousands of years. It takes time to find out what works and what works often changes. Nothing is permanent. So why are we building our governance systems as if we're getting it right on the first try or first deployment? It's pretty silly, isn't it? It's not realistic. It doesn't work. The rules are complex. People don't want to vote. It's just not working. So these are some of the pioneers of the governance systems that we've seen in Ethereum and generally in blockchain. And I'm not picking on these protocols at all. They've done so much for the entire ecosystem, but there definitely are some challenges in these systems around adapting to slowly or not at all. We have a lot to thank to Maker. They were the first on-chain system in Ethereum and they've done so much great work, but at the same time, they aren't easily adaptable. They have a smart contract system that was deployed once. In order to change anything, you'd have to completely redeploy the system. You'd have a token migration. You'd have a lot of technical debt, user disruption, and much more. So it's not simply easy to evolve. In a similar case with Compound, yes, they did master upgradeability of smart contracts, but that also comes with the technical debt, user disruption, every time they want to add a feature or a security feature. And that is a problem because a lot of protocols in this space have actually forked compound governance. We've gone from Alpha to Bravo. Projects like Uniswap and Gitcoin are both following that system, and they have to upgrade their systems each time, too, unless they fall victim to vulnerabilities, security, and lack of features. We also have a big favorite here, Yarn. Their protocol is absolutely awesome. They have this upgradeability via a modulary system in their core protocol. However, on the governance side, it's still a little lacking because it is relying on a lot of semi-centralized parties with multi-sig executing transactions. Lastly, not so related to DeFi, but we have Aragon. We had one of the projects who invented the term DAO and really brought it to the market, and they've done a lot of customization with their projects. So you can do a lot of different voting systems, different types of quorums, et cetera, et cetera, but because it's so customizable, the smart contract structure suffers a little. It's too upgradable in a sense of smart contract upgradability, not just modularity. All right, so here's the little lie that I told you about. So I'm not going to tell you that there's one solution fixes all our problems. There's way too many to address there, but there is something that we can do to help significantly improve the failure of all our governance systems and help them adapt way faster. And that is to build adaptable governance systems from the start. So that's vague, but what we've done at Element is to build a very minimal core structure at the foundational layer and everything else is modular. That way you can upgrade quickly, you can remove things if you don't like them, and it evolves more quickly with the political dynamics of your protocol or DAO. So the hope with Council is to reinvigorate the standard model for governance while keeping decentralization at the forefront of governance, but also allowing our DAOs to scale better and make decisions quicker. Outside of the adaptability, there are a few other goals that we strive for when building Council. And the first is to continuously strengthen inclusivity. The one token, one vote just simply can't work, especially when it's a capital burden to acquire that. There's many flaws of that model because people with tons of capital can capture tons of votes and they can sway governance, activist, investor style. What we can do is allow the actual people in the trenches on our GitHub, on our Discord, speaking on community calls to have a say in governance as well to equalize that capital burden. The second one applies to all of DFI, of course, in governance, and that is to make voting not a burden for people. Voter apathy and voter fatigue are real and actually can be detrimental to a governance system and protocol. People stop caring, bad votes come in, systems fail. So the last one is the most important point that I want to get to. And I know upgradability in a smart contract sense is a highly contentious topic, but this type of upgradability is built on a very minimal structure at the core and everything else is modules you can add on and off. So we want to design a system that can actually adapt as we change our minds throughout history instead of having to redeploy and fail and fail. So this is the core. It's very simple. That's it, that's all. All this does at the center of the governance system is to define the voting process for those with governance power in the system. It tracks proposals and counts their voting power and it retrieves user voting power from approved sources, which I'll get to in a quick second. So that's it. That's all the core system does. Everything else is a tooling or tool belt to use in current situations, future situations, and much more. The modules or tools I've selected here today are my favorite ones and we've decided to choose these for very specific reasons. But keep in mind anything you can come up with can be added to the system, whether it's a new form of delegation, rank choice, expiry. The possibilities are endless. So the first one here is kind of a take on traditional corporate governance where board of directors are trusted, voted in, compensated to make better decisions for the organization, and obviously scaling decisions making a lot faster. There is a twist here, but I'll leave that for later. The second one is my absolute favorite, voting votes. In a way this allows you to assign voting power to any person use case, position, etc. So it breeds inclusivity, but also capital efficiency which is a huge problem with D5 voting, governance voting. The next is optimistic grants and that's kind of a twist on optimistic rollups but also the literal sense of the word optimistic. Because many grants programs with those organizations are very pessimistic. They assume the grantee is going to fail at their work. So this takes the opposite approach where it assumes they're going to succeed and if they do they automatically vest their payout but if not it can be clawed back at the discretion of the governance system. The last one is Dow teams with authority and this one is really important but it's not a module itself. It's an addition to the same structure that the GSC has and that is giving your Dow teams authority over their areas of expertise. If a Dow team has to go through the regular governance process that everyone else has to go through, decisions take forever, processes are developed, kind of become outdated and it results in a lot of people just not wanting to get their daily work done because it takes a month, three weeks to pass a vote. So a little more on the governance steering council. Like I mentioned it was definitely inspired by traditional corporate governance because we can't discredit that. It's worked for generations, it's been battle tested. I get that we like inventing new things and trying new things but this stuff works so why don't we try it here? The way it works is you become a GSC member by getting delegated votes from anyone here. Once you reach a certain threshold you're then able to join the GSC and what that means by joining the GSC is like you're given authority over your work. People trust you, you should be able to do things. The authority we've baked into this is the ability to push votes on-chain automatically so without having to go through that long strino-governance process so they can make the calls for people at the top level. The second one is they can spend part of the treasury up to a certain limit but they can still do one off-brands, bug bounties, you name it. They don't have to pass that vote through like a month-long process to spend some of the treasury which is good in some senses which is why it's capped at petty cash levels. And the last one like I mentioned other councils within the system can be formed using this smart contract structure. Dow teams can have authority over their expertise so a treasury management working group could spend the treasury without a resistance or a protocol team could basically deploy things to the protocol without having to be burdened with the rest of the voting process. All right, so this is my favorite and it is voting votes. And the main reason we developed this system is that a lot of governance systems today make you lock your voting tokens in the governance contract. All you can do is vote. If you're a small minnow in a sea of whales, your vote barely matters. Why wouldn't you go use your tokens and put it in DeFi or make some capital out of it? So it's really inefficient from that standpoint. So what we did here is we allow you to assign voting power to those types of positions or reputation and those people are automatically given votes in the system. So it balances out the capital barrier of people just buying tokens. And they're also modular and upgradable. So what I mean by that is let's say you assign voting power to GitHub contributors and you find that their voting power is too little or too high. You can change that through governance. You can also upgrade these things quite easily. And I get that this is very theoretical. So I did highlight a couple of examples to make it real for everyone. The first one is probably the most interesting and that is a compound of a maker voting vault. So like I mentioned before, you have to lock your tokens in those systems to get voting power. But in this system using counsel, you can essentially borrow against your collateral in these systems but still maintain the ability to vote or delegate. So you get the best both worlds. In a similar case, you have the LP vault. So you could be in a liquidity provision and still maintain the ability to delegate or vote. And lastly, the identity verified vault is pretty cool. So you could verify your identity on-chain and get voting power for that. Or even a reputational thing where it's an aggregate score of your work in GitHub discourse, community calls, working groups, and it scores it into signs of voting power. And the last one is an NFT. So if you own an NFT, you could basically map that to voting power. In the element system when we launched, we gave a gift to the community and that was the LP versus NFTs. So right now someone in the community is building a voting vault to give anyone with an NFT voting power in the system. It really spreads out the inclusivity of the whole community, which is absolutely amazing. So dry here. All right, so the next is optimistic grants, which like I said, it's an optimistic twist on the pessimistic view of grant systems, whether it's just from an organization, a DAO, and whatnot. So the base, like the TLDR is that you can set a schedule and a time expiry. The grantee can claim that grant if the time expiry is hit without it getting clawed back. So when the grant is allocated, it contains that expiry date. If it isn't removed by that date, they can withdraw and get their grant. If the governance system or the GSC realizes that they're not making their milestones, they can claw it back. So what does this entire system look like? Like I said, it's modular, so there's many options. The first one being what we chose for element DAO. And it is a refined form of representative democracy. So with the GSC, it's basically like representative democracy. But as you can see here, this is only the core contract. It tracks the proposals, counts voting power, retrieves the voting power from the defined voting vaults, and basically the time lock up here can add new quorum or change the voting vaults as we go. So all of these things here are modules. This is the only core. I really want to emphasize that this is the base core system, and very little things can impact that. Your system could also look something like this, which is liquid democracy. So you take away the GSC because you just want simple delegation. And the last one I want to show is just inclusive governance. I also took out the time lock. The time lock is definitely a recommended security feature because it has a pause between when you make execution and calls. But if you want things to happen quicker and you don't care about that risk, you can remove it. It's like DeFi Legos, but governance Legos. So by now I've talked about the long history of failing governments and how it takes so long to improve. I've talked about our considerations with how we built this system and the core and the modular aspect of our architecture. So now I just want to talk a bit about the council protocol in action. So we launched governance for element on March 31st of this year and we were really excited to just see the traction happen. I mean, a lot of this was a big experiment, right? The GSC, the rolling delegation was very new. So we weren't sure if it was going to work. But today I'm happy to say that the GSC has nine people and they all have together pushed forward quite a few proposals to improve the system on the social layer but also on the protocol layer. We also have two voting vaults, a vesting one and a locking one. The vesting one essentially allows like the core contributors to create this protocol, investors, advisors, key partners to have voting power in the system but a lower multiple. So the community outweighs the actual core team which is a pretty cool balance of voting power. Like I mentioned earlier, we also have the LVNFT voting vault that's in the works so there will be three coming up quite soon. I've talked mostly about the on-chain and smart contract aspects of the system but the social frameworks and DAO structures have also moved really quickly and that's because we've designed that proposal framework to actually adapt really quickly. So people can make changes to like the proposal process or how we want to onboard assets. And that's already happened twice already, without friction. So we're all really excited to see how the governance theory council evolves. Right now, I mean, they've pushed proposal forward. They haven't spent any of the treasury yet. We have to see if it actually helps scaling these decision-making processes. And if it doesn't, we might go back to the direct democracy that everyone else uses where everyone just has the ability to delegate their vote or vote themselves. So what can you do with council today? Well, you can do a lot. I mean, you can fork us, you can build with us. It's all open. Any DAO or protocol can use this for their own governance system. And like I mentioned, the system is very upgradable. It's modular. It's literally built for experimentation. We want people to try things, add it to our system, create a voting vault so that your governance protocol can have vote in ours and vice versa. The possibilities are endless. So you can use this system or you can add more to it. So with that, I've gone over all the problems with governance, how we can adapt faster, and the architecture of what it could look like if you were to adopt this model. But right now, I just want to say that governance is evolving as always. It's going to fail as always. So we should build these systems to adapt quicker. And for us, governance systems are literally the make or break for our protocols. I think the best is yet to come. And thank you to everyone here for being a big part of that and for caring about governance and being here today. Thank you. Any questions? Yeah. Finally, some water. Thank you. That was super interesting. I'm really amazed by the way that you're thinking about governance and the upgradability of it. And so my question is a little bit philosophical. Sorry about that. Delegation, I see it as a way to handle this problem that people don't want to vote, right? But delegation can't always work, because if people don't care, it can very easily turn into popularity games, who do I know that I can delegate my votes to? So we have this problem that people don't want to vote, and delegation is not really a solution to it. So how in your experience can we have people engage more, be more socially responsible and all of that? And I have a second question. I don't want to monopolize, so feel free not to answer. But I also think that reputation is kind of a very key element to solving governance issues. But I think that reputation itself is not a solved problem, as far as I'm aware. So I'm wondering how do you approach reputation in element? Yeah, so for the first question, delegation can definitely be a popularity conference, especially with big people on Twitter with previous experience being delegates. I think the best way to do that is just have accountability layers. So if these delegates are just doing nothing or abstaining from votes, you have to kind of have that, like, disincentive there or transparency reports. I think Gcoin's doing a pretty cool job with their report cards on how their delegates are behaving, how they're participating, and all of that. I don't think there's a quick fix, but there are two things that I do want to see more in delegation systems, and that is rank choice delegation. So if I choose you as my first one, you my second, you my third, if you're away for a weekend with your family, it reverts to him and then to you. So it just kind of plays more to that human level of governance, like people have their lives, they're not always going to be online and so on, but also adding the time expiry of delegation. So if I delegate to you and I forget about it for a year, my votes are basically just wasted, especially if you stop voting. So if the delegation actually reverts or expires after a year, it puts the omen on me to actually find someone I align with more. As for the reputational thing, it's definitely a tough one. Mager did a really cool experiment with SourceCred, where they had a custom algorithm that took months to kind of filter in their discourse forum, but it assigned incentives for those participation people. I think the reputational layer can definitely be gamed, and whenever you add metrics to something, there's definitely risks. But if you could aggregate certain scores and for my involvement in Discord, my involvement on community calls, or the more tangible things like GitHub commits, and you aggregate into a score, I think it can kind of give you some level of their participation in the system, or if they own an NFT. So you kind of have to pull from a lot of different sources to even vet if they're an active member versus the gamable aspects. But yeah, it's definitely a tough subject. I can't give you a great answer. Charles, you talk about, here. You talk about delegation expiring after certain amount of time. I'm wondering, are you guys creating certain rituals, or let's say, re-elections every certain amount of time? And yeah, I'm asking this is because, as you said, some people use delegate, and then they forget about it. And yeah, there's no rituals. There's no way for coming back. As we do it in traditional governance, we have, let's say, governments, we have four years' re-elections or stuff like that. Are you guys thinking to do something similar amongst those lines, or what? Yeah, so when we originally came up with the governance steering council, the idea was to have election styles every six months in the beginning and then maybe every year. But there is a big problem with that, and it is the off-boarding of a member who is a big delegate or a council member. You have to create a very strung-out process, involves governance. So that's why we went with the rolling king-in-the-hill style with a delegation threshold. So if someone were like that to just kind of abstain from everything or do something malicious, they would simply just drop in delegation and then someone could kick them off the GSE. So it's a little more like dynamic. When you say someone, can you explain that? So if a GSE member were to fall below the threshold to be on the GSE, anyone in the community or anyone on the internet in a wallet can kick them by clicking a function. So it's immediate off-boarding basically. So is that kind of like a liquid democracy where you're essentially like, I don't like this delegator anymore, I pass my votes to another person once the threshold is below a certain point, basically they can get kicked out. Yeah, so it's representative democracy mixed with liquid democracy. That's a mouthful. So it's a group of people on a council being able to do things. If someone does something malicious or they're just not doing a great job, people stop delegating to them. And then once they fall below the threshold, someone kicks them, they're no longer in authority position. So it's a lot quicker than having to do an off-boarding proposal where it takes a month to off-board someone where they can still do bad things or be inactive and get compensated for it. We'll see if it works. The election style is not out of the question. I think it's just a worthy experiment to try for sure.