Directing Decentralized Autonomous Organizations with a Focus on Work - Juan | MakerDAO

[Applause] thank you all right thank you everyone for for coming it's very nice to be here um yeah I don't know how much or how many experts we have here but before we start very quick introduction this is me back in the day uh where I was in good terms with Racers and and scissors no not so much but yeah I did a bunch of Consulting before jumping into startups and scale ups and I discovered web3 by accident and here I am um some disclaimers as usual My Views are my own I don't represent any company this is not Financial advice or legal advice maybe mental health advice but that's a conversation for later and yes this is the Dilemma that I and probably we face every time that we do one of this is like should I go very introductory and try to be as welcoming as possible for for newcomers or should I go and hit that governance Niche geek section where we're super passionate and it's like I don't know 30 of us talking about this so the agenda and the objectives uh hopefully cover that today so first I'm going to do a super quick introduction to Dallas uh hopefully some of you find it useful then I'm going to share a bit more about governance design um yeah again this is quite Niche but I might get some of the newcomers passionate about it to be able to to move forward and yeah I'm going to go through a real use case on how we by flipping one of the of the concepts we we changed a lot of things and hopefully that clicks everything together for you and finally I'm going to give you like some when I was your age piece of advice if you want to join the the industry all right so that was before they were cool that was our decentralized autonomous organizations some smarter people and me out there like to call them decentralized Collective organizations because the autonomy might be might not be there entirely but before we dive deeper a couple of examples that you might be familiar with Alcoholic Anonymous is a very well-known organization it's been in existence forever and it includes thousands of groups that are helping each other pretty much anyone can start a chapter so that's how decentralized it is you just need to follow some rules have some experience they help you with with material and stuff but but it's it's relatively easy to to do that and and that's uh that's by Design um and then yeah something important you can adapt it to the local needs of uh of wherever you are so if you feel that something I don't know doing something online or or having like a huge group is it's better for for the whole Community than you're welcome to do so another one more on the for-profit department this is a Spanish company It's relatively unknown but it's uh it's quite big it's uh yeah revenues of the on the billions uh it has over 75 000 employees and there's this very clear process if you want to join this kind of like Cooperative how to do it but they do have the process online but all the way to the to the point where they have a bank that's ready to support you if they like your business plan so you might be in business much much faster uh so again that's another example of kind of like a Dao uh but yeah the the power of web3 is here and that means that right now we are able to create this organizations much much faster experiment with them take some learning see what's working see what's not and and move from there uh is there any questions that we can talk about that later or or come see me afterwards um so now when would you dial or not um again you do not always need a dollar maybe it looks cool in your name uh some of the doubts out there are not really those but uh this is a this is a fine line that you need to think about and you need to be very mindful when you when you're creating an actual dial one of the really good things is that you can push the the power toward the the edges meaning that for example you're trying to take care of a community or if you're trying to clean a river or do something you can actually give these people ownership over the project uh through these the mechanisms and I don't know you can create all sorts of incentives and there are a lot of really cool experiments out there but that's one of the of the main features if you want to keep all the power for yourself maybe a startup is is a better a better idea transparency is the obvious one everything that goes on chain can be seen by by anyone with a block Explorer and some basic skills so that's that's an obvious one then you have self-executing Smart contracts so potentially you can create something that will self-execute I don't know a thousand years from now or or automatically or by someone else so you get this programming power uh right there under the hood incentives can be integrated so that's another really really cool thing that's well that's why Bitcoin Works um and that's that's another very powerful thing that that you can choose to to have and it's globally out of the box meaning that the moment you launch the smart contracts anyone can access or almost anyone based on the jurisdiction composability is my favorite one I think so if you if you want to build something on top of another project uh the contracts right there no one's stopping you no one can stop you so I think that's the really powerful thing and talking about no one being able to stop you you have the resiliency and the perpetuity so we kind of know that the uni swap contracts will work there for eternity um all right so that was the very short introduction to those hopefully you learned something if you're new uh now about this Niche called governance design um and this is what I try to explain my parents what I do over and over again so yeah this slide might help but basically it's the process of Designing and implementing any process structure or or systems that will help any organization it could be as small or as large as you want to make the decisions to manage conflict and to allocate the resources so that's that's pretty much all we do um so it's the tldr is uh we kind of invent bylaws that then we need to follow the clear example uh or an example I like using is designing a Sandbox so you can you can think of every decision that you make as a trade-off so for example if the walls are too high the sand might not get away but then you will actually um I don't know the kids might not be able to get into it or or the the it might be dangerous for kids if you create some really crazy uh uh plays or games but then they might not have you know fun Etc uh so that's that's pretty much it like designing a Sandbox uh you you need to put some some thoughts and trade-offs in there and that's that's really how you how you can think of governance design as well um now yeah this is what I wanted to talk it's this real case uh example that we saw an issue happening and how we we flipped the the mechanism or the model but basically yeah just to set the context um this was happening with maker we created this system called the core units which were groups that were autonomously being um funded by the protocol by maker um and then yeah what we discovered is that not very surprisingly token holders want to value for the protocol so they actually want to get something we can go up four hours about rage quits and stuff like that um then something curious that not many doubts know is how much of their treasury is being used for protocol development and how much is being used for I don't know if admin is the best word but it's uh anything that has to do with governing the protocol uh so the moment this is not clear you don't know how much is being spent on bureaucracy and how much is actually adding value to the protocol so that's a very important premise so going back to what I was mentioning VM care holders ideally have a vision maybe they don't but you really want to believe that they do so the the way it kind of works or that this should work is that the work uh or this vision is kind of like deciding what work is being done that work produces the value or the or the outcome so it could be like a new project an integration a bridge I don't know any type of thing and and you move from there uh the challenge that we had with this core units model that I was describing is that because we were funding teams it was extremely hard Forum care holders to actually have a any type of Direction on on how this thing worked so the good couriers were there kind of like operating as a black box and it was uh very very hard for almost anyone to come and say hey this is not working or what I working on um what I we're using the funds that we're transferring to you uh to do what and then the projects were or were not getting done uh I guess random is not the best word but it was very very hard for M care holders to actually direct and give any type of Direction um so with with that in mind when the bull market when markets were thriving and and everything was uh Rosie the way that it would work is that it's like hey we need I don't know to go on l2s for example so any cool project would actually mean that there would be a new CU that would come and do that so it was just throwing money at the problem and potentially the CU could or could not deliver the work um so again it was very very hard to to kind of give some kind of Direction and um the main issue that we found is that when you fund a team it's a kind of the team decides what they're doing so they can do something that it's very very Mission critical and that's I don't know 20 30 of the work uh but the other 70 is some Pet Project that's that's not directly benefit the the protocol and this is not this doesn't mean that it's done in a malicious way it's more like the way it kind of works and and how human nature is um but yeah going back to the example as soon as the bull market becomes a bear Market uh the opposite happens it's like okay we should start cutting budgets everywhere so even if like a new career comes it's like hey we have this really really cool thing that you should implement we can do it for good value that might get rejected while other core units that were there before from the blue Market uh are not getting any type of checks um so it's it's a bit uh arbitrary in that sense um and yeah they're very weird conversations that were being had uh and these are some of the of the conversations that we saw uh anyone working in any uh medium-sized Dao is probably familiar with a lot of those but uh yeah it's a lack of um lack of vision for example it's one of that you hear a lot it's like we don't want micromanagement uh we cannot uh just fire teams because it's very bad for the moral uh kpis uh what are these guys doing so you get a lot of noise and a lot of very toxic conversations um that happened so that's something that you want to to look for and and potentially save so yeah what's what was the problem uh confrontation conversations in a very very very unstructured manner um to the point where it's almost a feature to stay low or to lay low and just not being seen because then they will just keep funding you uh well anyone any other on the spotlight will will probably be in a much worse position and they all yeah the All or Nothing negotiation means you could fire a core unit but then if they were doing that even if it's a 10 something that's critical you're kind of losing that so all the conversations end up in this everything or nothing which is very very bad for a lot of reasons um lack of actionable measurement tools yeah the fact that we didn't have a very strong Frameworks lead to this so maybe someone comes and starts evaluating a proposal in a very specific way two months later this other proposal comes might be better might be worse but we're not using the same parameters so that's that's something very very bad for to make it scale to make it fair for everyone and and and to make sure that we're having the right discussions and finally yeah as I was saying the vision it becomes quite unclear for for everyone um it's it's even hard for teams to come and join it's like what do you guys want uh if if no one's prioritizing the work in a in a good way so with that in mind this is what we tried doing and it's like flipping the model so again the vision is provided by the M care holders I'll get into the implementation in a bit but basically this budgets are actually funding projects and again you can call them projects you can call them work however you want to structure it it doesn't need to be in a time constraint but basically what what you end up doing is that you do have teams that might be able to work for different projects even more than one but potentially you can always change the way that these projects are being funded um so the core unit kind of can come and choose and becomes a bit the the consumer instead of the one creating the the situation on the conversation uh and yeah if you do it properly it becomes much more open for for anyone to come and join um it becomes easier for a new Korean or a new team to come and compete if you set up the right process so so that's a that's an experiment uh worth having um yeah I'm going to quickly show you an implementation example um just so you guys understand a bit how this works but basically the strategy it's a bit like the vision that I was describing and that's kind of like out of the scope of this and what happens is that there will be different dial level objectives that might be prioritized again that's part of the strategy but once that you have these objectives that you want to achieve um you can have proposals that are kind of like responding to it so an objective could be a bit like the okrs and then these proposals are a bit like the an answer to the request for proposals so you just have proposals that are answering to this objective that the Dell has set um and hopefully that I can choose whatever they they like best um so yeah again several proposals can cover different deliverables of the different objectives you can set this up custom to your to your dial based on the on the needs but that's a bit that's what we presented to to one of the to another dial that's that's not maker but those were the the custom um yeah the variables that we set for this specific down again this is just more to show you the the level of granularity but there were four requests for proposal cycles per year uh where you kind of like allowed some budget to to go for this objective and then at some point this would be cut and you need to reinstantiate it if you wanted to to keep going and yeah this is my my last section I kind of added it by by accident in one of the talks and people keep telling me that they really like it so I'm either really bad at the first part of the talk or or they appreciate this but uh for anyone that wants to join any type of web3 um the first one is an entrepreneurial mindset so if you kind of like like or if you try to see where the value is and try to create things from scratch you will have a much easier time that if you are are waiting for someone to tell you what to do that's uh that's usually not how it works in this uh in this project the second one is have lots of drive and curiosity keep doing stuff keep learning keep yeah keep coming and and and see what you can do and eventually you will find something that you like provide value and and be onboarded um and this is a bit my my last time so if you're not like a super specialist uh I'm I don't know if you're a crazy solidity developer it might be much easier uh but if you're not I think that generalists will have an easier time if they start organizing simplifying structuring any type of complex information so you kind of like start documenting it the way that you prefer then you show it you kind of reiterate and eventually you end up with a lot of knowledge you're helping a lot of people that didn't have this knowledge and um and yeah I I thought that was very useful for me and and unfortunately by now so those are the the pieces of advice and the last one of course is to to show up uh so if you keep delivering uh you might not find the the your perfect dial at first but eventually you will be at a position where you uh yeah where someone sees the value that you're generating and they will bring you on board because that's what's best for the for the organization itself so yeah these are my socials if you guys want hit me up and we can keep talking about governance and thank you happy to take any questions if you want [Applause] I think you need to raise your hand or something so the microphone guy run runs at you we have a code blue in the microphone Department okay I think this one works um are there any dials currently in existence maybe other than maker that you think are doing a good job and like other dials can take inspiration from were you yesterday in Target because I got the same exact question um but no I think there are a lot of different again that was a it's a it's a tricky word you need to Define what a Dao is and then I think decentralization is not binary it's more like a spectrum so there are a lot of organizations or foundations that are trying to decentralize um for for better and worse um and they have some features or bugs that uh that are sharing um yeah I I have a lot of uh of friends or really good acquaintances in the industry from other projects and and honestly I think that they are doing a lot of great things and we're learning a lot from from each other um I don't know if you expect like like names specifically but uh we can talk later about specific names if you want any more questions and did you really have a question or can you talk just a little bit more about conflict resolution and how you're thinking about that I am in particularly interested in like when you turn the core units into consumers rather than like you shall not pass and being a black box around like specifying the work what happens in the case that there are two different core units that take on the way I cannot hear anything I don't know if it's the same for you or it's just me that it's up here can you hear me now hello let's try again okay conflict resolution particularly in the case where you have core units that are competing for the same value for a given piece of work so say your core engineering and Oracle teams both take on a data availability problem for Mega on a layer 2 with different solutions there's only one part of value available for that they have different approaches different implementations that's probably a good thing but like how do you handle conflicts how do you direct competition amongst your core units towards really valuable outcomes so the that's that's one of the main things that we kept seeing is that the whole topic got very politicized um so if you had your core unit set up and you were working and there was someone that did something it didn't need to be 100 what you were doing but something adjacently uh yeah comparable to what you were doing and they were good you could see a lot of really bad excuses and a lot of politics being played were it's like I don't know conversations with delegates or why do we need these new projects if we have this other one that's being done um so that that was I think that was really really bad because eventually you end up with this bureaucrats that are trying to protect their jobs and and are not uh actually um yeah so you you kind of like are stopping the actual value that the token holder could have um so yeah this is one one example uh we we haven't seen it yet live but I think that this RFP um time uh would actually help with that and then yeah you want fast governance so one of the things that we thought is that if someone kind of like uh wanted to compete you would extend the period if there was no one else so I can say hey I want I don't know a new logo a graphic designer could say hey I will do it if there's no one else you get like Fast governance and you can approve it in a week or whatever uh but if someone's like hey I also would like to do it then that would automatically extend the period and then you'll go into the full RFP section where you actually compare the projects and and you can ask questions and I don't know give more um more details but all this needs to be managed it's not that you can just code it uh anyone that run a grants program know this so it's it's not as easy as it sounds and it takes a lot of work to actually evaluate set up the process and the whole um the whole story so it's uh it's it's not that easy hopefully I answered a bit of a question thanks for the talk it was very informative um a question I think every protocol right now is kind of a dilemma oh whether they want to position themselves as in something immutable right following all the smart contracts and the original vision of something that you know once is there or it cannot be touched and and then it's easy to integrate it's reliable and so on versus an organization that is flexible you know it adjusts to the market Innovation is happening and therefore requires changes and therefore a committee of people that applies Envision and then applies these changes I was wondering if you can just provide some you know an opinion on where where you stand on the Spectrum and some recommendations yeah that's a that's a tough one uh I think that every project is struggling with that uh my my main takeaway from what I've seen is that if you can create something uh that's very very simple and composable it's potentially easier to just create it and and let it run and make sure that it's there forever uh it's a bit what we've seen with uni swap V1 V2 and then V3 um but yeah you know that that V1 is going to be there running forever so you might migrate to a newer version but if you decide that something didn't work out you can you can roll back uh regarding the full um yeah the full project it really depends what you're trying to build I think that it's cool to have to build things that are going to last forever but I think that code should follow uh kind of like common sense and they need of of us humans at the layer zero so ideally we're settling into layer one uh anything that's that's good for for the humans and the layer zero so that we're using it for us um so with that what I want to say is like it's not because we coded something 15 years ago that we need to to keep doing that and keep using that and and never change or roll back anything if we if we decided that we made a mistake it's okay to go back I don't see any more hands so thank you everyone this was great [Applause]