Using DAOs to Make Decentralized Protocols Actually Decentralized by Luis Cuende (Devcon5)

we're gonna talk about dolls today a few years ago if we were talking about Dallas people were talking that we're thinking that you know you would just put money into them and your money would Michael a disappear because of the whole the doll hack the back in the day you know we were like well that's right not inherently bad so let's just focus on them that's actually making great the Dow was unfortunate but I also put this page forward by a lot in terms of security and so that's what we're doing at Eragon we've been doing dolls since November 2016 focusing on making them very high quality and very easy to use for anywhere in the world so what I want to talk about today is how we can use those to make decentralized protocols actually be centralized so I don't know if I Cohen this term but I like to use the term CFI so probably you all know defy the centralized finance actually and there's no term called C Phi which is centralized finance and a lot of this defect that we talked about today it's actually c5 which I mean I would talk about that later but I think it's it's okay sometimes to start centralized and then decentralized but we have to be conscious about it and make a conscious effort to actually get for it to be decentralized and so I want to talk about some things today starting with you know smart contracts the best invention since ramen I think we will all agree immutable smart contracts are the default state of smart contracts unleashed on blockchains like aetherium and so they are fully trackless so call this law kind of thing what you've read into a smart contract is basically what happens and but the issue is they are not upgradeable and so we learn the hard way we are kind of you know some some other issues that we hide the theorem community that it is very important for smart contracts to be upgradable for mainly two reasons one of them is do you want to be able to fix bugs and the other one is you want to be able to add features it is very hard to compete against other systems that are gradable enhanced over time if we don't have that capability or if the only way to upgrade something is via Forks we which require a lot of human coordination so each time you want to add something you have to tell your whole network yeah I deployed a new contract is the new address and you know they have to like review it again and it's really they have to reach consensus say again that that new contract is the one that this was actually use and you lose a lot of network effects as well if you want to fork every time that you want to upgrade a contract that may just break for example like you know defy protocols because they just kind of be like yeah this is v2 and you have to move all your fans from one place to the other imagine like all of that if people were like closing CD piece in like you know one system a bring the other one that would just break everything then on the other hand we have the centrally govern the smart contracts that's easy for developers right you have a smart contract and then just upgrade it and that's it but it is not really trash this because it's custodial so you have fans and then you have unilateral permission to upgrade the smart contract what is the difference with just having a centralized server in which you can like push code to and so I like to use the the ether to ramp analogy here I I made this kind of picture with like a Stephens logo and and tram because I think this is the is the worst from both worlds you have like a theorem which is you know like kind of clunky to use and it slow and all of those things and then it costs money to like click buttons as one team member from our team said but then also you have the worst things from like you know it means centralized so there is one person one sort of like dictator or entity that can just push changes right so I think centrally governed the smart contracts are use a very bad idea so that's why the data Trump I hope you you don't have nightmares about the ether Trump and things young for coining the term it'll Trump I love it and now well what I wanna propose is a different thing which is a different combination that I think takes the best from both worlds and that is community governance smart contracts and so the idea for traditional contracts is that they are upgradeable you can enhance them over time you can enhance your protocol you can enhance your apps but at the same time there's a community there's a governance process a little governance process that ensures that you cannot push an update that you know get through people's money get away with it right and so I think this is very very important to to work on and the issue here is of course you have to figure out your governance structure you have to think alright if I am gonna put my smart contracts governance behind a Dow or some sort of like community governance what is the governance structure that fits the best and so we've been working with projects for a long time to figure this out and there are a bunch of different things you can do the cool thing about Aragon and the way we thought about it is that it's very modular so instead of saying you know we have like the key ingredient like the Kissel's and there's gonna solve all your governance problems and we're gonna use focus on that instead of saying that we were like well we don't know how organizations of the future are gonna look like we don't know you know in the future which kind of Dallas people will want to deploy we think every protocol has or each community even has their own needs and so let's build this very model framework in which you can basically do everything that is here and much more so they're two axes in this in this curve map here and so one of them is speed so the speed at which developers core developers can implement changes to the protocol and there's the other one which is strapless Ness which is how trustless is actually this protocol there is from what trusses Ness is important it's not because I like visualization which I also do it's more because if he's not trustless again like if you just have like a protocol that is custodial you may have a lot of issues from not having trust from your users because you can use with rather fans or you know maliciously in to like intervene in their apps and the working systems or even because you may have legal legal issues like I love this see five projects today I think they may face some kind of legal issue if they are used basically the ones controlling the whole protocol in the background right so because they are and they're supposed to be licensed revised but they are not in practice so then it is very important the boot restlessness first where we're building transit systems if you don't want to trust the system you can always use MongoDB and use go decentralized route so now I have amazing news experience the centralized for me has like very very high speed but it's not trackless at all then you have something like a technical council so there is a a project at V fiber or actually called melon protocol and so the way they structure their governance they wanted to do something in which token holders have the ultimate world but also there was you know fairly fairly fast to push updates right and so today it is very hard to so token holders of your community smart contract and be like well do you want to approve this upgrade I mean that a crystal of coordination cost and stuff like that and there is very specific knowledge that they need to have on solidity and yet another protocol design instead of doing that they wanted a technical council so docking holders who like both for this council and then this council would take care of the of the upgrades afterwards and so for me that's that's not like super super transness that I think is like Travis enough and also guarantees a very high speed because you can iterate very fast with this council and you can ensure that these people actually know what they are talking about they are experts in their field so these are very very example of how a protocol went from centralized to decentralized using Aragon and our tool in and it's a very exciting project and right now it's been a minute for for a few months in terms of the the governance structure that they are using with Aragon and I'm super super happy that it's working out for them they you have other models for example community veto so here you may say as the core developer I still want to retain the power to put upgrades but I want to give my community like 24 hours for eight hours whatever time you want for them to check the upgrade and be able to veto it if you choose like out Reyes if I'm just gonna like you know steal their money or something like that so with this you may have this period where like you need to allow talking holders to vote no to your proposed upgrade so it is not that fast but it's more trash this right because you can have your whole community basically stopped updates at any point in time so that's cool when you have like stuff like a liquid talking occur see there is no I call it talking overseas because it is not like democracy in the sense that you have like one person one vote you can also do that with Aragon there is another doubt that is theirs in that doing some kind of like civil resistance using an app that they have in which have to like take a four or something like that it's a legal system but here I'm referring to doctor no curses so basically the scheme the game kind of kind of token are see where like you have one token you have one ball basically and so you can have something like a liquid token aa cracy where you can delegate to other people's you're talking holders don't have to read every proposal they can use delegate to other person that they they think is it's more expert in that in the field but then it that person that delegate is not performing well they can with rollerball immediately and give it to someone else so that's very cool we're working on that implementation and then you have something that is I think the probably the most Restless but also like ever slow which is direct democracy that's what we do also in in Aragon for our own governance so it's three months Torian holders get to decide on the future of the project and I think I think that's a very cool model it's again like it's super trackless like you know every three months basically they can do whatever they want you don't have to do it every three months you can do like continues as well but of course if you need all the other kind of holders to come together and take a decision maybe it's better to like concentrate that into one moment each quarter one moment each year kind of things so it's more you know it's a slower but it's much restless so again you know you can defend all of this with Aragon and you can basically see what are you trade-offs what matters to you and then you know customize your doubt to fit those needs so I think it is very important to build a community not only to ensure that the protocol remains trustless and so you can eventually move governance from centralized to decentralized but also because if you looked at like where is all about is basically network effects around communities the code can be forked of course like brand and stuff like that and community cannot so it is very very important that we focus on the right things here in communities direct thing if you look at the theorem you can you can forget theory on today you can name it a theorem cache try no one would use it because a community we have here you guys and everyone out there that cannot be forked so I think also is very important as a South South student the other day that we see organizations and governance structures as products and the users maybe in this case I was talking about like a more like a company structure like an organization but I think it's the same for a cryptic protocol or network so when you create the governance structure that is also a product that is a product and your users argue community member for example so when I think of the governance process we created for ourselves in Aragon we have a lot of these people who out for new governor's proposals we have these people who edit them we have these people who then are community members and they get affected by them so at the end of the day you have to look at your community and your governance as Houston all product if you have to care of it as if it was just another product that you're building and of course a very important one I think also when talking about 1001 vote a lot of people get outraged about you know plutocracy and capture and all of these things right I think if we look at token or se or what I call the evolution of a stakeholder governance if you look at shareholder governance for the last hundreds of years it has to work beautifully to create value now we are in this weird position where we have common goods or something that kind of resembles a common good or is that you know kind of like a gray area between a common good and also like like a parrot property because I mean it's not certainly like a hundred percent common good you have to like at quiet if or Bitcoin or whatever to a part of these networks but the same of at the same time like is not use a traditional private company safe in here with this evolution of stakeholder governance one token one vote there are certain things like cool improved but there are also things that are really great and they are very handy if you is when I like have a community bought in today so in service get in the game of course if you have participants in your network that owned a big stake or they own a mistake they're going to be incentivized for the network in the community to grow and work they are easy to deploy today like you can deploy one token one vote today very very easy allows first autonomous participants I don't like to watch it I think it works for some models governance models but I think allowing students participation is very important sample in the our own community there are people who are synonymous they pass proposals into the ballot and they actually get up to buy a whole community and we don't know who these people are and that's beautiful as a beauty of it and then also allows everyone to benefit from value creation so a key difference with like private companies decades ago is that you will only benefit from this value creation for example in Apple Facebook stuff like that if you were the first engine investor the seed investor even you know later in the kind of like states in which you could buy shares right but not all people have access to that especially not like everyone who was not a bc here everyone can just chime in and you know buy bitcoin early in the day or buy if they're buy whatever kind of token or community so they can become participate in that value creation so my message here is only honor I don't think it will last we need to improve traffic Ness and transparency with dowse especially for protocols that are supposed to be decentralized I think it's important also if you protocol is centralized today I think you know there's some trade-offs there are things that need to improved and item the angel I technology wasn't there a year ago for these protocols to actually be more decentralized but it is today and so we need to like if the way we need to think about the roadmap and we need to actually get these things into a point that they are actually trust less if you get an early on your problem we can help we're working on this one click set up template for protocols so if you go to the Oregon app you see these templates that you can just click on and then feel like a bunch of text fields in our beautiful UI click create boom done if you want something more complex there are like CLI like command line tools that are like an their second SDK with a bunch of things you can really get crazy with it but if you want to get like the basic basic basic templates is it really like graphic like user interface I don't have to do anything else and we're working on integrating this into our unlike our own client so in the future you will be able to use go to app argument org select crypto protocol and then basically you can write your smart contracts which Apple here this can be your protocol like on the on the left side of the screen you can write your smart contracts you know as much flexibility as you want then instead of like putting the only owner to your account or your company's sir let your treasure or whatever what you can do is put the only owner to the alien tap in the Dow and so basically everything to the right is the Dow so that I can have multiple apps the voting app the finance that allows your doubt to have money you can get really crazy with permissions so for example you can you can give a boring app to get money out of the finance app so you can even have like a protocol reward out in which like you know each block or the H worker like you know its period of time there's a new talking created that goes into this finance I pinned it down and then token holders can use the boring app to decide how to spend those funds from protocol development but then we created this app called agent inside Oregon an agent is basically the the dowels face to the world to the worth 3 world and so what I am tell us is furthered out to interoperate with any worklet protocols so thou may be able to open and see if your maker that I may be able to lend money on compound all of those things and the cool thing here is you don't need to get into our SDK to use this like you just basically have to create a protocol whatever way you want and then put the address of this am tab as the only owner and that will basically make your protocol be governed by now it's that easy so I think as a summary like Dow's make defy to serve the D I think I think right now we're in the process of actually making sure that if I actually deserve that I think it's more like see if I today I think everyone else are very easy to deploy you can literally blow them in like 30 seconds they are working canisters from maintenance like the one that I mentioned from from melon protocol if you go to the Argonne website there's a case live so with most depot maintenance cases today and we're working on this one click setup template that will make things way way easier so I said before like we have a bunch of those now mainly I think it year ago technology was very early I think now we are getting there so more than 700,000 minute today you can create yours a tablet argument or various for like 30 seconds we just released a new version two weeks ago or three weeks ago that makes the user experience much better I think we're gay they're like we are slowly making sure that is comparable to web to experiences in terms of user experience so I'm very happy about that and I just wanna thank everyone you can follow Aragon on Twitter you can follow me on Twitter if you're interested in these challenges the company that I worked for arguing one we're hiding much of roles especially fronting people just really care about designing we really care about bringing these two people that actually need it so you go to Aragon but one it's less jobs and also you can check out their own website so thanks [Applause] and actually have two more things for today so one of them is I reckon is gonna be released in a couple interesting news later today about a couple community developments that are going on so instead a stay tuned for the era news during the day and then a second one is we are throwing a party and so we are throwing a doll party the equidistance basically focusing on the dico model and the upcoming launch of Aragon's and racing that would allow any dau in the world to find resolution bonding parts so it's very cool we're gonna have we're gonna have actual like sushi there for the first people that arrive not like party finger food some rehabilitate and I invite everyone to use sign up you can RSVP just go into Oregon's Twitter and there is a link there so see you later and how we can enjoy some sushi and dial conversations thank you [Applause]