Make your DAO sticky with Mochi! Gabriel Tumlos

i'm one of the members of the most recent seed club cohort sco4 active contributor of smaller dows like shifay logic school and in a past life i helped co-write the gnosis white paper and released some pre-nft tokens on a decentralized music platform named ujo music way back when since about 2014 i've been very interested in how you use markets incentives tokens to get people to do stuff in organizations and make them a little bit more decentralized autonomous hence daos so who's this presentation for maybe you are just dow curious you've heard of the term maybe you're new to web3 and you want to know a little bit about what to expect when you join a dao or maybe you are a little bit more experienced you consider yourself a dow contributor already maybe you're a down ninja you practice dow karate you take on bounties you make proposals you vote do all these kinds of things fantastic this will also be for you and maybe you're one step further and you've created a dao you're a dao operator you are charged with keeping the dao alive and you're looking for new ways to activate and engage and reward your community maybe you're a summoner a gardener a builder whatever you fancy yourself as this presentation will also be for you so why does this is a great meme that a good friend maggie put together for shifai ahmadao she's the if you're a millennial you'll get the joke but the point is that it seems like every single day there is a new dow cropping up on twitter um there's daos for creating cultural uh buying cultural artifacts there's daos for by excuse me buying the constitution um governing decentralized financial protocols financing people of color women lgbtqia organizations in art nfts so on and so forth there's even daos that exist now to create other daos and it's become the hottest three-letter acronym in the space since nfts and um maybe you think like me this is the future of organization um but maybe if you're also like me you have a little bit of a reservation about this being the future of human coordination maybe you've read the new frontier and coordination which was put together by git coin uh and bankless which talks a little bit about some of the challenges that daos are facing um now that they've come into the forefront and the mainstream maybe you feel a little bit of a dow fatigue maybe you were promised the organization uh the organizational kind of enlightenment of no managers no bosses that kind of thing no hierarchy and slowly but surely you're starting to see these things creep into your dow as it's starting to scale so i'm going to zoom out a little bit and talk a little bit about dows and some of the challenges that they face specifically and this is kind of like the 10 000 foot view that we'll go through first and we'll zoom in so um what is a dou a dao is just a web 3 community not just but it's a web 3 community that gathers together and to solve some common problem or towards work towards some common goal they do this in an environment that is characterized by something called information asymmetry so the tldr on that is that we live in a world of imperfect information uh and access to that information this creates two kinds of problems that daos have to figure out how to do because at their core what they need to do is figure out how to make decisions collectively the first problem is adverse selection which is a dow contributor might exploit access to information to their benefit at the expense of the greater organization and the second problem is what we would call moral hazard in the realm of information theory which is a dow contributor might make a poor decision because they do not bear the costs of the or the consequences of that decision so here's a little bit of kind of like a technical overview of some of the problems for some of the more technical people but maybe you are a busy person and you like business terms so here's another way of thinking about some of the problems um dao's have low barriers to entry which is really really awesome but also really high barriers to participation so as you know anyone can kind of join a discord or a telegram and start participating you know voting reacting going gm that kind of thing and that's awesome we want people to come and participate as much as possible but sometimes the most eager person is not necessarily the most qualified to be making decisions not only that it can be expensive to participate when the cost of transactions on ethereum specifically are you know rising as the days goes by second thing there are low in real life kind of like physical meat space but high url uh you know virtual coordination costs uh given that these are kind of web three virtual uh native communities so we're mostly online um but that comes with its own challenges you have to spend a lot more time and effort in coordinating people usually across multiple time zones usually across multiple different cultures and you experience these things like voter fatigue and it's hard to keep people aligned on what the greater mission is of this decentralized autonomous thing um and last but not least this is actually kind of like a very specific problem to uh crypto and web3 in general um we operate in kind of a high pseudonymity uh environment that is uh you know as a result a little bit low trust this is a good thing because there are new ways for now marginalized groups of people to participate in what's supposed to be the new economy um but this can also create a kind of environment that's very low trust that leads to sub-optimal decision-making because you might trust somebody that is not worthy if they're a bad actor and you know next thing you know you have some of these kind of like long cons that end up happening where people are high contributors at the beginning and they're really just trying to steal your ether which happened very recently okay maybe you're not a busy person maybe you're not super into the business terms here's for the community people maybe you like uh diagrams and circles and not thinking about things in triangles so here's a common model of of your organization your dow from a high level you have your larger audience right which is kind of like crypto twitter and web3 zooming in a little bit more you have the people that are really members of your kind of like your home they live inside of your digital home whether that's slack or discord telegram these are the people that go gm or you know they react to your posts and they join your um your town halls things like that an even smaller subset of that are the people who are really participants so they're participating in token voting maybe they're making proposals maybe they're taking bounties so on and so forth and i would also kind of like uh include contributors in here and then at the very base you have what i would call the core these are the five ten depending on how big you are like core group of people that are constantly uh participating and adding value to your organization they're they're notifying people of changes updates uh new things to participate in um so on and so forth so you as a community leader have a lot of problems in this giant space of uh trying to decentralize their decentralized things uh and if you like colors here's something for color the the color friendly people in the group um fundamentally what you're trying to do as a community leader is you're trying to move decisions out of the core group out towards the edges that's this blue line and the next thing you want to do uh simultaneously what you're trying to do is bring members off of the edges from your audience you know the the kind of like people who are less contributing closer towards that decision-making core all while you're trying to grow your community outwards and this is the whole process of progressive decentralization um very very difficult and there's a lot of missteps along the way you encounter the same problems information asymmetry moral hazard adverse selection so maybe you like memes and uh you consider yourself a meme lord and a meme lady and you like to learn by experience so here's a perspective you trying to stay zen while you're trying to decentralize the biggest thing possible you know whether that's the serial brands the music industry art nfts that kind of thing let me know if this becomes a little too triggering i'm just gonna you know maybe this is just yell stop if this becomes i'll start playing the sounds if you don't say stop okay all right i'll stop i'll stop so we all know the problems we know them like deeply we feel this kind of like sense of uh maybe coming undone a lot of chaos and freneticness in the dow space so that's where i'm going to kind of bring it back to why do we want to make things sticky and the question is can we reduce the cost of coordination as these organizations start to scale and grow enter mochi which is basically a dial coordination game that we've created to help people form teams uh set goals and stick to them for an extended period of time what we what we did when we started making this uh uh this kind of this dow coordination game was we asked ourselves a simple set of a couple questions uh before we even started to get working and the first question was could we start with a coordination protocol that's a little bit more decentralized than what we're used to in the corporate world that people are already familiar with we picked agile and scrum if you develop technology in an iterative way you might be familiar with this already the second thing that we asked ourselves is can we make information more transparent can we make meetings more asynchronous as organizations scale that's when we started developing a discord bot because most people um and and communities organized in discord uh we actually started in slack so if you're a slack community don't consider yourselves left out and then the third thing that we ask ourselves is could we use games and tokens to frame kind of the organizational problem problem and make coordination fun and sticky and moji says yes you can absolutely do that so how does mochi work here's the journey in a nutshell um you and two up to two to four players your friends can get together and for an eight week period which we call a journey and what we ask you to do is set one goal doesn't matter what it is it can be something as simple as i'm trying to sit for meditation for 10 minutes a day could be something as complex as we're trying to create an nft zine together the only thing that we ask is that it's small and measurable and the only other thing that you need to play mochi is a small amount of eath that you stake when you um set this goal and we call this bonding with mochi and basically the only thing else that happens throughout this eight week period is inside of discord our discord bot will ping you at 10 am every day and it'll ask you a set of four questions and again if you are familiar with scrum you might be familiar with this already how are you feeling today uh what's something that you did yesterday towards xyz what's something you're doing today and what's something that you learned um depending on how you answer these questions something will happen we have uh so here's a little bit of the dao of mochi the yin and yang are are basically this uh this rainbow mochi and the shadow mochi if you respond on time to your check-in request your mochi tokens very simple however if you don't respond on time or if you don't respond at all the shadow mochi quick shows up and slashes a little bit of your eath if you keep missing the eighth scales and if you keep missing your team starts to lose ether as well and this all goes into a slash pool that you can use um to fund whatever community projects that you would like to fund so what have we learned in running this uh we basically have done 375 journeys in our own community we took our community from about 20 highly engaged people to about 100 highly engaged people using this system we've awarded over 15 000 tokens we have almost 12 000 daily proof-of-work check-ins that are you know a little bit more of an evolved kind of gm we have slashes there's uh about 40 eth bonded just in our community alone and we've discovered a couple things about participation so what are the things that we discovered one incentives work so if you add addison to incentives to a framework that requires kind of human coordination about it it will increase uh their ability uh and willingness and and desire to kind of complete the task at hand so what we found is actually on average nine times out of ten our people show up to our community just by playing this game alone the other thing is everybody eventually misses nobody is perfect so some money eventually trickles back to your slash pool which again you can use to fund whatever it is that your dow is about uh we like to um organize parties so uh and and and kind of create content so we've done two nft scenes on zora we've thrown parties events in new york city uh lisbon prague uh soon to be denver osaka new york uh la all over the place we do parties so we use our slash pool basically to fund some of these activities for our members um not only that we've also discovered that once eath gets staked for whatever reason it tends to stay there and you can use that ether to go create passive uh earning opportunities for your members using defy yield farming we're all familiar with these kinds of things so while you're getting your stuff done you can also be earning money on the side so this is another kind of value-added benefit for your members if you use such a system and that's pretty much kind of how the game runs in our own organization so in summary daos i really do believe that daos are the future of human coordination but we definitely have a long way to go the problems that i presented to you are not going to be solved by one specific application but what we're trying to tackle is a very small portion of that problem which is how do you get the members at the edges of your down your organization active engaged participating in in the mission of whatever your dow is about we believe that the way to improve daos is to actually reduce the cost of coordination not add more management not add more governance we want to automate as much of it and reduce the amount of intermediation ie middle management managers that appear as organizations uh start to scale we think that we can do this actually by using a mix of web 2 and web 3 technologies and for us that's takes form of a quiet little discord companion that we called mochibot and uh kind of at a more fundamental level uh what we believe is that small repeatable patterns will lead to higher order beauty and complexity and organization so if you can train your dial collaborators and your doubt contributors to be more reflective more long-term thinking and to check in with themselves we think that this will actually lead to happier more autonomous people in whatever shape your organization might take in the future so if you are a dial contributor or you're a dao summoner or you're a dao investor or maybe you're just now curious and you want to get involved in trying to find an interesting way to start playing endows come follow our journey we're actually opening our discord to the public it's been private for the past two years we're opening it to the public today follow us on twitter bond with mochi and our tagline is basically dream big start small and stick together so hope to see you soon take any questions thank you yeah we got a good five minutes or so for questions and you in your study did you find that people game the system and just give uh bogus answers that is yeah um so that's a really good question um there are some people that uh you know something some some days it just gets busy and you don't have a ton of time to kind of do the check-in or whatever there were problems where um some people would actually support or submit kind of like bogey answers periods or you know just one word answers that kind of thing we don't have a solution form for that yet the way that we think about solving this this problem um as we kind of develop the protocol is two pronged one is to increase the incentive for producing good check-in reports and the second is to you know again penalize kind of this is the dow again the good and the bad uh carrot and the stick so the way that you incentivize better reports we believe is uh actually by making it so that the the benefit that you get from that report is greater than uh the amount of effort that it takes to produce it how do we do this so i had a really interesting conversation with somebody recently who's very very very into artificial intelligence and was telling me all about gpt4 and how he used gpt3 actually for a personal journaling experiment uh what he did was over a couple of months uh he he did some personal journaling daily where gbt3 would ask him questions and he would respond and after a couple of months he found himself in kind of like a rut a little bit of depressed and he asked gbt3 what do i do to get out of this um and it knew it had kind of like it had developed a memory as a result of fine-tuning the model uh and it kind of like repeated all the things that he just kept getting in a loop on and it helped him break out of that depression and start making progress towards his own self-betterment so the idea here is if we can create structured data for people working on any kind of goal whatever it is you're trying to achieve and create some kind of long-term memory by fine-tuning some of these uh some of these artificial intelligent models that are becoming more ubiquitous and available our hope is that you can actually create some kind of intelligence that's greater than just a bot that's checking in with you and trying to get pester you to do something hopefully that will add more value to you as somebody who's trying to accomplish anything in your life the other side of it is you create an audit system right we take an optimistic kind of uh approach to you are doing good check-ins um and maybe that's not the case all the time you can create some kind of system basically where you have somebody that goes through check-ins you can have that can be a group of members in your dow depends on how you want to solve that problem you sample a set of the check-in reports and then you penalize people who are producing bad reports so there's a little bit of a there's a good way a bad way or you know nice way and a not so nice way of solving that problem actually i'm going to go here first also if you are claire and you are going to talk to us about space and nfts please check in over to the right-hand side of the stage hey how's it going great talk um i was wondering if mochi was considering doing any kind of github integration for developer participation um it's not it currently in our roadmap but it's something that we've thought about and have definitely seen in the market if you're familiar familiar with source cred this is something that they take a very explicit kind of stance on so participation in some communities looks like how many commits you do and pull requests and that kind of thing it's not something that we've really considered for our community specifically because um our community is a little bit different we're a bit more social we're into nfts that kind of thing but if your organization is filled with developers and you want to reward people who are contributing to your your code base contributing to the core stack um and that's what really you're about come work with us we'll figure out a solution for you and we'll figure out how to incentivize um something that should work for you yeah cool thank you yeah you're welcome all right we have one more down here i think yeah i can repeat your question sure so the question was have we thought about scaling the game for larger teams teams of teams or maybe dows that have a little bit more of a complex kind of situation so the answer is yes um but the question also is what are the limits of doing such a thing so right now we've intentionally kept it very small teams of two to four uh you need at least one other person because the social dynamics are very important um keeping accountable you know we're social beings you want to perform well for the people that are around you now as as as small teams kind of grow and become more complex the cost of coordination also becomes more complex you have to hire people that are kind of just like arbitraging information around to make sure that everybody's up to date and in sync so there there's some good kind of like uh you know fortune cookie wisdom out there about like what the maximum team size should be and we're actually interested in kind of exploring expanding teams out to maybe eight possibly maybe ten there's that that popular kind of management rule of like two pizzas right so it's like if if the team is bigger than uh what two pizzas can feed it's probably too big to try and coordinate it maybe trying to to split it up so we're interested in kind of expanding in that direction and then to the extent that that is successful for whatever organization uh kind of takes that on uh we might start to create some kind of solutions that allow for inter-team communication and then i think maybe part of your question is like how do you do this at a dial level i don't know but we'll get there yeah cool all right let's give gabriel our round of applause thanks very much