Futureproof Governance Frameworks Andrew Thurman mod; Sky Minnert DXdao; Dennison Bertran Tally; Jul

all right welcoming in our next panel are you curious about future proofing governance and dowse well this panel is for you we have skye from dxdow dennison from tally shannon from live pier and peyton from maker down moderated by andrew from coindesk [Applause] hi everyone thanks for coming through yeah uh as the uh announcer said we're gonna be talking about future proofing dows today uh have an excellent panel for this particular topic representatives from maker down tally um uh dx and uh the current chief decentralized growth officer perhaps the first time that title has ever been deployed uh over at live pier uh so welcome everyone um let's get right into it i think we're gonna just start off by talking about some of the problems facing current dows and then we'll get into talking about our panelists respective projects so starting off um i think a big problem we're beginning to face is that the current state of dao tooling and sophistication has been wildly outstripped by the amount of power these organizations have and the amount of money they control back envelope math between d5 tvl treasuries and market caps of governance tokens gets you within striking distance of a half a trillion dollars nonetheless governance systems are relatively rudimentary um extremely simple uh ostensibly this adds to their security but we're beginning to see some cracks in as bad actors realize that dows can in fact be uh targets for attacks um what is the current state of dow tooling in your guys's mind and um is it sufficient for where the ecosystem currently is um skye let's start with you thanks hello oh hi uh so uh daos are all about decentralization and basically the that's what gives them their strength and i would say the most the most vulnerable pieces in the system for a dao are the centralized points of of uh failure and in our current system we try we can you can try very very hard to maintain like decentralization throughout your entire flow but often you know you have to rely in this current world on centralized systems maybe for your social media like twitter forums even uh discords chat communication channels and so at dxi we spent a lot of time trying to push and think about the tools that dao's actually need and we invest in projects that are building those and i i'm calling them ddaps which is dowdapps and basically if we have a a tool that can be a dap or a decentralized product and you can log in with your wallet what dows really need is the ability for the dow to own that account and if the dao can own that account and control the actual permissions of the accounts that they've delegated to their contributors and members that solves the problem but most of these dapps that we're using in this space if they're either web 2 or they're they can't be actually owned by the dow and so we want teams and people that are building those to come reach out and we we can give you money and most dallas will give you money because we all need this if we want to actually operate in a decentralized manner yeah moving on hello hello hey hey yeah i mean i think to step back like what what did dao's do dows govern decisions over shared resources so i think there's both on and off chain governance that happens so i kind of think of tooling around those two separate things i think we're in very early days of both on and off chain tooling for dows um on the off chain side you know there's everything from the communication layer how we use discord how we use any kind of communication platform to share knowledge um develop ideas and then of course on the on chain side there's actual like token governance standards and i think the tools around just the actual token execution of how decisions are made we're still in fairly primitive days for that i think that's probably something we'll get in to discussing on this panel but um how we how we evolve towards new token governance standards beyond just liquid token governance i think those are that's an important tooling question as well hi um so we think about this question a lot at tally and one of the things we found is that the fomo always outstrips the tooling and i think that we've entered this kind of interesting phase where when we build better tooling the fomo gets bigger and outstrips the tooling even more so something that we started to realize and think about is how education is a really important component here we're primarily a company based of engineers so we have a habit of thinking the answer to most things are engineering solutions we try to engineer our way out of everything but we found is is that really people need to be educated a bit more on what dows are what they do what the norms are um before we even sort of experiment with other kind of things although i'm happy more the first person who loves to experiment people need to sort of like understand what dolls really mean if you look at the sort of wonderline finance thing where danielle put this tweet out and he said enough the dow is going to be managed by me now and and people started saying calling it like a danielle autocratic organization uh you know that's a real thing right people have to understand what is a dao right like it's not just the fomo of like i threw my money in number go up we're all in this together awesome it's it's about the responsibilities of participating right the responsibilities of the success falling on your shoulders as well so we really think there's a big educational component here that needs to be done to sort of manage folks expectations but also address their responsibilities in daos as well because some of the latest things that you've seen happen really aren't engineering things that you can engineer your way out of but you need to sort of like educate your way out of yeah very thoughtful thank you uh hi peyton from maker dow here uh so i think the tooling actually is coming along really fast and that's one of the neat thing about open source structure right is that uh new ideas tend to birth very quickly once they come to market however the problem i see with it is that we're kind of in this situation where we're laying down the train tracks while the train is chugging along and anytime you implement a new tool a new system a new voting mechanism you have to really understand the benefits and and drawbacks before you implement it or you can run into a serious danger situation um so when we look at like nation states as like a great example of people have been doing governance for a really long time we see that their bureaucracy the people that actually maintain their tools and their systems is several are several times larger than uh the people actually making the governor's decisions uh we're not quite there in d5 so i think it's actually more of a problem of getting people are more interested in governance and can experiment with these tools analyze them research them and really present plainly to communities like hey all right here's a really great thing about this and here's what you need to look out for if you are going to go forward with it yeah you guys seem a little bit more optimistic than i am about this stuff so let's stick on uh why why dows are bad for a little bit more um just in the past two months we've seen uh what i think some people would characterize as successful governance attacks on some of the most high profile dows justin sun successfully took out a flash loan and passed a ave governance proposal to have a deployment of ave onto tron and attempted to do similar shenanigans on compound and makerdao um having a stablecoin he's associated with added as collateral um it seems to me that people who are powerful or clever enough and there was uh what last week there was a dow where a single member transferred the control of the token contract to himself infinite minted the contract and destroyed the project dows are beginning to look like attack vectors and what are the tools you know how quickly do we need to be laying down the the train tracks in order to avert what's increasingly looking like looming disaster uh let's start with you there peyton sure absolutely uh security is always going to be your number one concern right a really big problem is the ability to borrow these governance tokens at such a large rate because basically a core tenant of a lot of governance systems is that people who hold the token are uh incentivized right by the performance of the protocol well if you're short a token you're uh inverse incentivized so you see these governance attacks uh potentially being quite profitable uh so i think really what it kind of comes down to is designing systems and understanding uh when you might be at risk in those systems for instance if you look at our maker voting portal our dashboard where we handle our votes we like to have a little box that says like all right here's how much available maker there is at other lending platforms here's how much maker is staked in the governance contract just so it's very clear to people that like hey if these numbers start switching one way or the other we might be in danger and we kind of need to mobilize and take action so transparency is your best friend in dao's right the greater chance that everyone can see and understand a situation means that when something starts going wrong you can have a much faster response to it because there will be attacks there's just too much money at stake for people not to try sorry please to sort of like follow on that um we can't expect the technology itself to save us right like again i'm thinking about like an educational component people also have to be paying attention right dows cannot simply exist as smart contracts and expect that you know just because it's christmas eve you don't have to be like watching the shop right like it's not enough to think that the systems alone are going to be enough because the systems the way they work are pretty clearly spelled out and people have a lot of time to sort of think about the organization of these systems and kind of find a way around it i think we also have to be very careful in what we characterize as governance attacks right there there is a fine line between people we don't like getting ideas through and people actually attacking right and and that is something we have to keep in mind like justin sun of course is like the boogeyman and we don't like him is the sort of idea that the ethereum community pushes but i think that's a dangerous idea to push right he is another contributor in the ecosystem of web 3 whether or not we like it and the systems that he were interacting with were permissible for him to do what he tried to do right there isn't in my mind an enormous difference between borrowing tokens to get a proposal through and whining and dining delegates to get them to agree in a back room that your idea is something that they want to support right so i think we have to be careful on that and we do have to double down on education and vigilance which is saying hey if you're part of this dow people have a responsibility to watch the shop and probably dao should step up and pay people to make sure that they are doing that on a regular basis yeah i'd agree with that i think yeah it's important to just be clear on what we mean by the risk and i think you know civil attacks are kind of like the number one um problem that gets identified when we think about liquid toca governance and that's when you have anonymity one anonymous actor creating multiple accounts that effectively makes them the whale i think in the case of ave we actually knew who the attacker was so you know there's even some transparency around that so i think you know there are different kinds of problems we're talking about um i do also want to just call out that the dell i think you've set up the question around the d phi dao landscape and i represent live pier which is actually not a d5 project we are a proof-of-stake protocol that's using blockchain technology to actually build core internet infrastructure we're doing decentralized video transcoding so people show up for a different reason fundamentally um when they run a node um with live pier so i think that the threats you know are still real but they're a little different than in a d5 protocol when you have purely financially incentivized actors i think you can actually design a system that is not susceptible to those things um the two examples that you gave the first is it can be attacked with a lot of money so if you have a lot of money you can attack it so you can you can solve that problem and then second is it's not decentralized so if something's not decentralized and someone has like a key to access something and they can take over and then mind everything that's also a problem so if you design a system that can't be attacked with money and you design a system that's actually distributed and decentralized and you can you can basically design a system that can't be attacked i think yeah that's these are great points dennis and i want to start with you on this one because um you touched on this a little bit um one of the ways you can defend against these types of attacks that we've seen or not attacks you've made a good point there um is through greater participation some of the most important votes in these protocols uh get two dozen addresses involved at most and only a fraction of a governance token circulating supply um you know we were just dragging danielle but the vote not to unwind wonderland's treasury is by far the most popular dow vote ever it had over 20 000 participants and it's kind of wild that the most popular ever is only 20 000 participants um so what can we do to galvanize more participation what kinds of crypto economic networks could we create around incentivizing participation or is it really you know a public good people just need to get off their ass they shouldn't be paid yeah so those are some great questions uh something i like to think about is you know when we talk about 20 000 people showed up right um a lot of people recently showed up in the ens sort of brouhaha that was happening in their community people show up when they care right when it affects them when they have an opinion um they show up when they care a lot of votes that happen in d5 protocols are extremely abstract technical things so people say oh only 20 people showed up but i would be really impressed if 20 000 people showed up to like readjust the multi-collateralization ratio across like you know defy lending pools at like you know some sort of inverse ratio to something or other right that would be a different world that we would be living in um so i think it depends on what people are voting on and certainly i think you probably can design systems that are more responsible of saying hey you know if only three people show up to this maybe the vote should be going longer or maybe like more people should have to like go through this process maybe maybe you can design systems to give the community the ability to to look over these decisions a little bit better but again if you don't understand how it works you don't understand how it works like with wonderland i think a lot of people were just promised number go up and they weren't sure if number would go down and so they got in there and it's just like oh no save us let's vote um but do they did they participate before then right like when number went up did they check in or care did they just say ah number goes up everything's fine i don't know right additionally we need better tools right we need better tools to understand what's happening we probably need more primitives that become universal that they can understand right so i think it's a combination of things but one thing that i do want to point out is a lot of these systems are very new right so when we talk about decentralization it's there's a little bit of a mismatch of thinking okay you know three women get together and they decide that they're going to build some sort of new lending protocol it's very complicated and they launch it uh you know four weeks later it's a little unreasonable to expect that everyone around the world is going to participate in how it runs right when most the knowledge of the the project exists within these you know three or four women's heads right so when we talk about decentralization we should add a time scale to it right if compound is not decentralized by the year 2340 right like yeah it's a failure right that's not decentralized but is it if it's not decentralized a couple years later right like you know i have a small child you know five years into having a small child you still have a small child right like it's not done yet um so sometimes we have to like take a step back and think where at what point are we judging the decentralization of these organizations because if they're complicated they're going to be hard to build yeah i also think that there's a misconception that unless everyone votes on everything it's not a dow and d fight doesn't have delegated governance we actually have as again as a proof of stake protocol a delegated proof of stake where token holders delegate their tokens to node operators who have you know those governance rights in our case delegators can if they wish choose to vote against their node operators but again the types of decisions we're making based on the protocol and infrastructure network that we are they are highly technical but as a node operator you're highly incentivized to want to pay attention to the changes that are being proposed to make changes um and to vote on those and and again i think it's like not everything has to proceed to an unchained vote as well i mentioned the difference between on and off chain governance like surely there are things that we should just be able to you know collectively decide on and um those who want to engage in that decision making to deliberate openly and be really transparent so that people can see ultimately what decisions were made i think that you know as a member of a community what you care about most is being informed about the decisions that are happening for me uh getting so whether you're a dow that already exists or a project that wants to decentralize and become a dao or a small dow that's just starting i think the most important part is getting governance of your ecosystem in the hands of people that will be good governors and that's not an easy thing to do but one way to do it well one way not to do it is just let people who have a lot of money buy governance power in your system one way to do it which is not perfect is to give governance power only to people that contribute value to your ecosystem so starting that from scratch is very hard there is there are people walking around eat denver this week that are actually handing out nft governance power to a dow that's about to start and that that is a very unique way to distribute and launch a dao but it's coming from a couple people's minds right another way is to only give governance power to people that actively contribute and do a proposal to your dow and add value to your community and that can be dev work that can be liquidity that can be marketing that can be community support and if you give those people governance power they're clearly interested because they've already been contributing before they got the governance power and even if you have other people that have governance power if they stopped contributing or they were initially had governance power and you are basically giving away new governance power it creates inflation and it dilutes the people that are not being active that maybe don't care anymore and with dxdow it's it's worked there was an initial distribution and we've had about 60 inflation and it's worked rather well now it's not as distributed as some big projects there are not 20 000 people that have governance power in dxdow there are about 550 wallets that hold governance power and only 100 of those maybe have voted in the last two months but you know that those 100 people that are voting earn that government governance power by actually contributing value and so we believe that that is a good a step in the right direction there are you know molecules also do share based governance where you have to actually contribute value or and sometimes contribute capital to get that governance power um but these dows are pretty open uh even if you can't buy a liquid token to get governance power if you contribute value to a community they will probably award you governance power no matter what community that is so it's something that a lot of people are interested in the tooling's not great right now we have some tools that we use but these are going to get better and better and a lot of people like this idea like even vitalik would support this kind of thing i think yeah peyton i want to hear from you on this because you're our resident paid protocol politician you're a professional governance participant um is that a viable model is that going to be do we just pay the smartest guys to uh to run governance for us yeah that's uh certainly an option right uh there's gonna be benefits and trade-offs to everything um we talk a lot about cost right and we tend to go directly to economic costs of like okay well uh layer one is expensive so what if we move things to layer two and that's a good thing to do that's something we're working on at maker dow however a much harder problem to address is how do we address the opportunity cost of actually being an informed voter right if you have 10 proposals going up this week and it takes you 40 hours to actually make an informed decision on them uh you're only going to have your insiders voting on it whether you like it or not or at least voting uh as a fully informed like a voter someone who who's kind of uh taking the interest of the dow uh fully fully in heart um so maker dow one one thing we've uh kind of come out with is is delegation um we're about six months in i don't know time's a little weird someone can dunk me on twitter if i'm significantly off there but uh we basically came up a way where we could as promiscuously as possible allow delegate delegation to take place while still putting on some kind of off chain restraints that make the larger delegates recognizable uh have them communicate their reasoning to the community that way even if you're not able to take part in a vote or you're not able to pay attention to what your delegate did last week you can catch it up really quickly in a post that way you can at least evaluate how people are making informed decisions and kind of use that liquid democracy aspect to reallocate your token weight uh if for any reason you feel that uh someone is is not doing their job of of actually evaluating these things got it yeah um well we're supposed to be talking about future proofing dows so i'd like to get into architectural trade-offs that you guys have to make um i think you know everybody's familiar with the blockchain trilemma you have choices between decentralization you know computational speed um and security i think dows face similar problems you know uh the uh danielle autocratic organization look a multi-sig of like two guys they they can sling some funds around and they can do more complex things with their treasury because of the centralization um so when you're thinking about potentially creating a system that still needs to exist hundreds of years from now and facilitate you know potential entry from institutional players from from nation states perhaps what are some of the trade-offs you make in building these governance systems and what should people you know which accesses should they be leaning towards when they have to make those decisions in order to make sure they build something that is durable and can last skye let's start with you yes from my perspective we already have centralized systems whether we're talking about the nft space or um web 2 or finance we already have centralized systems so we can already use those and so if you're making it down it's it's leaning it's sacrificing decentralization in order to achieve the other things to me that's not that interesting so the core principles which a lot of them come from the ethereum space like maximum decentralization and moving that direction is for me one of the most important things because that's what allows censorship resistance if we have a d5 project that has the front end owned by the actual company in new york city it's it is censored and it can be shut down and we are yeah that's not that interesting if you have a d5 protocol that where the front end is actually owned by a dao actually owned by a decentralized entity and even if they could come after a couple people they the no government in the world can actually go after the entire dao and shut down this front end and so things that that as an example um is something that dx dow it probably leads in the space like decentralization first and then we move a little slow because of that so we're less scalable than someone who is a smaller team and i forget what the third part is but um we probably the decentralization is the most important but other dials don't have to do that and and we'll see it only matters when it matters and it hasn't mattered right so far but i think it will at some point yeah i'd agree with that i think the premise that decentralization is what's needed to future proof governance and to make projects sustainable um and then thinking about how how you achieve that and the threats to it um it is about identification and and having as much confidence that um you do have a large number of truly distinct um distinctly identified you know nodes or participants in your network and i mean that's what we talk about when we say governance risk is when you have one actor that is maliciously pretending to be a bunch of decentralized nodes in your network so i think the question of identification is a really huge one and we're just really starting i think as a space to to tackle identity systems um in governance i'm a pretty proud practical person um and i think we're going to get there something i like to think about is how the pyramids were built to last for all time and they're still here but the people who built them are not and again you know my experience has formed a little bit for being in an organization that is very engineering heavy so we tend to think that engineering is the solution for things where maybe it's it's not right like if the people you know you could have a smart contract that is fundamentally completely decentralized or you know you know impervious to to attack but if the people leave it's still a failure right and i think it's difficult to say that we want to design something that can last forever we don't know what forever actually looks like right are we going to be doing d5 you know 150 000 years from now right maybe maybe not and while that might seem like you know kind of a silly analogy it's like okay is it 75 000 years or you know 22 000 years or three years right like what is the amount of time necessary to to say that something has succeeded and in my mind currently is if people still show up if people still care then it's still succeeding right and some of these governance attacks are fatal for for the organizations that have been building them and that that is a shame but the people are not gone uh and they they are able to rebuild it and rebuild it better so i think there's this kind of idea of like you know when you break bones and they heal back they'll yield back a little bit stronger i look forward to the next couple years of governance attacks to actually teach us where the edges are what really doesn't work and i think from that learning we'll probably be able to build much better systems definitely that's very inspiring um i guess the three things i think about most are security incentivization and transparency because i think if you can kind of work along those axes you can really build towards towards the resilient system the issue is like oftentimes you can have a really secure system without an incentive structure but then you introduce an incentive structure and suddenly you go from a situation where you have a few malicious actors right that might be motivated to take you down to an army of rational actors that can profit by taking you down and that's a much harder that's a much higher security barrier um but the nice kind of bit of transparency that adds in there is like it's actually okay to fail like it's obviously there's a lot of money at stake you need to protect people and uh you should fail softly but the nice thing about open source about being completely transparent with your systems is if something breaks you can always start again you can always go back and refork and and take your lessons learned and and apply them in a better manner so i think as long as you're thinking along those three axis like you can really build something that'll stand the test of time awesome um i think we have about two minutes left so very briefly can you guys talk about what you know new things are happening with dows at your respective organizations and um you know what are the new uh progress that your organizations are making on this front oh hey uh yeah life here just migrated to uh arbitrary layer two we went through our governance process for that um we do have liquid token governance but we are going to be exploring um identity systems in our protocol as well so just one emphasize we are not a d5 project we are building core internet infrastructure um and that's why we govern yeah so we we're we build dow products for ourselves in this weird little niche space but uh what we at dick style realize is it'll be awesome to bring this to the real world and so one thing we recently did this jersey is copenhagen flames which is an esports team in copenhagen and dick yeah dicks that was the shoulder sponsor and you know this the reason we did this is because the founders of this team love crypto are you know all about will want to learn about dao's and they actually want to dalify their community and so we have a goal in 2020 to to help their community become a dao to help them make decisions obviously it's not going to own the financial like wins of the team and stuff but if they can downify their community and that works we can do that you know you can scale that across other spaces and then start doing it with towns and states and and governments as well at tally we're building tools for making proposals easier but one thing that we found that it's a blocker to decentralization is the knowledge of making change resides in too few people and you have a situation where someone wants to do something but the person who's technically capable of doing that disagrees and just says sort of like yeah sure go do it and then people who want to do it are like oh we don't know how and they're like ah sucks to be you uh so we're trying to make tools that make parts participating easier i don't know we're iterating on delegation that i mentioned before uh one thing i'm really excited about that uh it's not me some of our awesome guys from the ducks team our development and user experience uh they're working on getting us l2 voting uh this coming up in the next few months uh just continuing to chip away at the the cost of governance yeah that's awesome well thanks guys for your time uh wonderful hearing from you all