Hoi Polloi: Gratitude Guild - a story of community incentivization

Transcript

Good to go. All right. Um, before I get into it, I have to say this is a project idea experiment of my own, not an official Polygon project, even though I am with Polygon, but I will not rest until it's official because it will be a roaring success. I guarantee it. Uh so welcome to the talk of the gratitude guild or undemocratic community incentivization. This was supposed to be a major project internally but it got kind of paused until Q3 and so on but I'm still working on it actively and I'll tell you about it today. So briefly about me um uh this is me on on X on Telegram. You can always reach out to me via any of these contact info. I'll be very happy to to

talk to you about development or anything else that you're building on the Polygon stack. be it katana or polygon or ag layer or anything connected to it. Um as a devril I am very passionate about making it very easy for you to have fun on our infrastructure and as chief dog fooding officer uh which is my self-proclaimed title which I'll explain a little bit later. My goal is to make sure that if you do build on our stack you will have real users. And I don't mean users who will just use your thing once and leave. I mean really the super precious users that are first time that are encountering your product for the first time. So that precious over-the-shoulder feedback you will get it uh with the doc

voting program. So keep it keep that in mind if you're considering to build on polygon. Now to to jump right into it I want to talk about why incentives generally fail in web 3. And it's often not clear from the get-go. Often you have some disillusionment like I'm building a great ecosystem. People will care about me. People will stay loyal. People will love it. It never happens and it never happens because of three main reasons because generally we are not shipping gratitude. We are shipping greed for the most part. And so you have this trifecta of problems plaguing the entire incentive mechanism which is plutoaucracy, vanity points and fatigue.

Plutoaucracy basically even if you have a DAO even if you attempt a DAO you will always end up with plutoaucracy. Somebody's going to have the bunch of tokens and make all of the decisions and others will be discouraged by this. This has been proven time and again for the most uh canonical and biggest demonstration of this. You can see the chain that is itself a DAO polka dot vanity points. We've seen a flood of those over the over the past few years. you have these uh points instead of tokens being collected and then airdrops distributed based on points. Even a recent addition to this is the kaito yaps which is basically another form of points. And then there's fatigue where people simply get tired of this because they have too much stuff to compete for, too many things to yap about and too

many discords to beg in to get in to whitelists and to get these tokens and get these airdrops. It's just discouraging and it doesn't foster loyalty or gratitude. it just builds unhealthy communities. So, uh why not handle this democratically, right? Why not have a DAO which will democratically uh reward people and just monitor their engagement and so on. This is difficult again because of some uh really entrenched problems. Problem number one is whale capture. Again, plutoaucracy. You will have these first movers who will get most of the influence, most of the money and then control most of the system. you will have a low turnout because people who do not have this much power will simply not be interested in participating. And then

you have other problems like for example rider bills where even if you have self-inforcing DAO decisions, people can include really non-transparent thingies in those decisions that kind of can derail a project. Again, a canonical example from polka dot is a runtime upgrade into which a fee bump by a factor of 10 was snuck in by the founder himself which was a really awkward situation. Now democracy also doesn't work because and I think you all know this this video where Osho says democracy is governance government by the people for the people of the people but the people are You simply cannot have the whole community vote on things because simply some people's

votes don't matter and they will never matter because when you have a telegram or a discord full of people who are asking when token when pump when moon these people simply are not value additive and you've actually uh been able to witness this particular on spaces and live live streams of any kind that support Q&A lately since the socially engaged for airdrops. era began where you will actually if you if you listen carefully to some of these spaces you'll actually find teams of uh I mean clearly it's Nigerians I think who are uh asking chat GPT produced questions in the very space they're in and chat or

Gemini in this case is live parsing the conversations before the Q&A and composing questions for them live which they then ask in Q&A a in roundrobin fashion which is why you get all of them you you you notice that they're all having the same accent and so on and they're all asking questions that they can't really understand what they're asking but that the LLM has given them why because this gives them followers because they get engagement in that space itself and it also gives them the uh yapping power that Kao will value so they will get these tokens later this is not a healthy community growth strategy uh so these governance scars in general in democracy are real. You know, we wanted decentralization. We got

plutoaucracy. We wanted merit-based governance incompetence, but we got whale capture and marketing um marketing in in the form of, you know, I don't know, blockchain logos on the sleeves of football teams and stuff like that. We wanted fair distribution, but we got first mover kings who never get dethroned, which again does not really inspire fairness or or the perception of fairness in a community, which you could argue is not always necessary, but would be good if you want your community to consider itself healthy. You get you wanted community loyalty, but you got basically mercenaries. They get their airdrop, they leave, you are left with a ghost town and um you don't really keep them around in any productive way. So you end up with airdrop hunters and you

want social evangelism, but you get an info FUD and you get info wars where people dunk on each other on social media just to get replies and get those points up. Not very healthy. So all of these negative things are piling up because we have these incentives all wrong. And I was long I was thinking about this for for a very long time since years and years ago before I was at Polka Dot and before I was later at at Polygon. Uh I've I've often thought about how to make this community healthier. How to make these incentives work? How to make DAO governance work in an ecosystem which is completely undemocratic. How to make sure that people who actually provide value feel

valued and can provide that value. And the first micro experiment where I also kind of self-awwarded myself the title of chief dog fooding officer at Polygon was with the dog fooding experiment. We ran an internal program where I argued very hard with the CEO and other high levels that um you should not be working in a company without knowing what's built on its infra. So if you go to an event and somebody else knows more than you about what's being built on your infra and you didn't actually use any of the apps that you yourself have produced, let alone that others have built in your infra, this should not be allowed. And so let's kind

of change that and let's make it an internal program to get people to dog food our own uh programs, our own apps, our own software, anything really. It doesn't have to be like if you're non-technical, you don't have to, you know, run a blockchain node, but it'd be great if you could test the on-ramp flow of a grantee in a grand ecosystem or something. So, anything at all really. And it was a really small program with a kind of small uh subset of of people participating. It was completely voluntary, but we did get 36 pieces of feed feedback in two months and uh 34 of those were rewarded. So we had an internal incentive program where if you do this based on the amount of points you are awarded, you have this gamification where certain tasks are

more complex than others and if you do those you get more points for that. And 30 of those pieces of feedback were implemented and directly benefited the projects that have submitted those tasks. Now by tasks I don't mean the typical uh galaxy quests where you you know you have like get 50 points for retweeting something. I mean literally test the swapping of Bitcoin from one network to the other and report back. You can report back in any format like markdown write up or or even a video or screencast and the project will then rate this as useful or not useful and if it's useful and it was almost always useful because of this really really valuable over-the-shoulder feedback of a

first user first encounter. very valuable because you know where your users are going to be getting lost in your product. And so we had a lot of feedback implemented directly into these candidate apps that then have dramatically improved their their entire UX flow. And this this is this is open this is still open to any application, any project, any product that's working on the Polygon stack. Just get in touch with me. We'll get you added in. You don't have to do anything. You don't have to pay anything. This does not come out of the pocket of the apps. this is just available to everybody who's building with us. And this worked really well. Why did it work really well?

Because uh it fostered loyalty through appreciation of these projects. So these projects were often communicating with these testers directly. So you had direct feedback. You have this oneonone. You don't feel disembodied. It also was really valuable because you were contributing to something tangible. some change that you have communicated has made it into the user flow of another user which is really cool. And then you also have other aspects like the gamification of it, the points, the leaderboard that we had. It kind of engages users and keeps them around.

People kept checking the dashboard based on the analytics and kept looking at where they stand and then grabbing more tasks because these tasks didn't didn't disappear once you do them. It doesn't matter if you have like test my on-ramp. Uh it's not going to go away once it's tested once. The goal is to have all 30 people tested. And so this is really valuable. You can keep accumulating this until you've tested everything. You can retest once the change has been implemented. And so this really worked well. Now lessons from this coupled with lessons from before from any kind of governance, democracy, whatever are that you need to in order to keep a community around and make them feel valuable and make a DA of any kind successful and this dog fooding thing was kind of a DA.

You need to first remove money from voting power. This is the number one mistake. If you keep money and voting power entrenched, then people will be buying voting power and that's it. This is where the entire process ends. The second one was that you need to persist people's value ad. So you need reputation and merit tracking over time based on performance. This is very important, but you cannot keep it around forever. Again, you cannot encourage the first movers to stay first movers and kings forever. The first mover kings are not allowed. Third, you have the gamification aspect. One thing that I've noticed from before when I was in Polka Dot when we were trying to make this governance system work better was that once we added collectibles, cosmetic

collectibles that people could show off after voting, turnout doubled just based on this gamification, just based on the fact that you could get this for every vote that you cast. people were very proud to collect their NFT trophies and put them on the NFT shelf and show off that they are a very active member of this uh democratic process. This worked really well. So gamification works for some people, for some people does nothing, but it did double the turnout. So it works. And number four is you enable a global item economy on top to make it fun and interesting and to keep that funnel alive. This means that the project themselves that are participating in this process can actually grow this pie with you. And I'll get to that in a in a moment. So

here's where we get to the hoy framework. This is a governance framework that I'm working on which is this this is a mouthful. Gamify decaying categorial meritocracy. I know there will be a link where you can read more about this concept but this is how it works. So uh gamified means you you've gamified it means you have some fun element that people want to compete for. Decaying means that your merit decays over time if you don't use it. This is a very crucial component of this. If you have a reputation and you haven't used it for a while, you no longer have this reputation. So this makes a system auto purifying. It makes sure that new blood that has the ambition and interest can

always get in and catch up to the first movers. This is the missing piece in most dows because you can't really decay people's money when it's bound to voting power. This does not work in regular Dows, but it does work here because you can decay merit or reputation over time because people have stopped using it. Just like I should I was a PHP developer back in 2010. I don't know much about it anymore. I should not have a vote in how it's developing its core ecosystems anymore. But if I start working in it again and build up my reputation, I should. So that's how basically it works. Then you have the reward sync. Then you have the categorial part. Sorry. Uh so categorial means that you have a category of reputation for each subsection of this

project which means that if you're very good at runtime upgrades and voting for them, it does not mean you're very good at treasury dispersement. And your expertise will be categorically bound. So you can grow your reputation in one category while staying flat in another. This is also very important because you can't have somebody who's never farmed a day in their life on an actual field decide what kind of I don't know seeds you can use. It doesn't work that way. And then you have this meritocracy where you actually accumulate this merit and reputation. So you have basically this um these four components reputation oracle a tip engine a reward sync and the item economy which work together. So the reputation oracle reads onchain and

off-chain signals. This can be GitHub commits or posts or whatever you produce and converts them into a single reputation score per wallet. This is getting aggregated across multiple sources depending on what kind of system you're building and it has these different oracles for different sources. So it can be an oracle that is completely dependent on humans grading your I don't know UX report or it can be an automatic ingestion oracle which reads your onchain activity which reads your GitHub commits and so on. You then have this part that basically lets any this tipping engine which lets anybody propose any tip for your submission. So this is at its core a tipping dow in a way where you essentially if I have a review of an app

I put it into uh some online service so people can read my review and then others will uh submit my tip my uh work for a tip. So they will propose a tip for my work. they will reward me for my work. Then the voting engine kicks in and it will either be approved or denied based on the subjective usefulness of everybody else grading it. Whoever participates in this entire process gets the merit and reputation in that category in that track. So everybody who participates grows in it. Everybody who doesn't over time decays and you always make room for new people who will be getting rewards from this ecosystem. The treasury subm module kicks in after a positive vote and just distributes a streaming payout to that tip recipient.

So for example, Alice gets 50 USDC payout and then it streams over a month and then uh Alice will get this over time. And then there's also this gamification aspect. You can issue and this is enabled for partners. You can issue soul bits and swag bits. So soul bits are pieces of this reputation that capture this merit and contribute to your merit score in that given category. These are non-transferables. These are uh modular NFTs that will go into your profile and you can equip those and visualize them and show them, but you cannot sell them. And swag bits are tradable equipables. So if you have your avatar, you can get your fiery katana if you're doing some work for the katana blockchain and then you can equip it and you can sell it as well if you want for

some extra income. The royal royalties of this also go back into the DAO uh at least a part of the royalties. And so you have these uh this tradable item economy on top of this ecosystem that makes everything fun and enables projects that participate in this to build their own swag bits on top of this. So you have a dog fooding dow for example. You have year enter this dowo. You have ern make a blue bandana with a year logo. And people the first 50 people who submit feedback for year's new vaults will get this swag bit as an equipable that they can also sell and trade. This has actually resulted in a very very significant conversion to turnout where people really like this

aspect of collectibles. It also contributes to cross ecosystem um user pollination because people notice this swag bit this equipable and say like wow this is cool. Where'd you get it? And I was like I did this and that for this and this and that. Then you kind of get this um kind of cross-pollination because multiple dows of this type are compatible with each other. So the swag bits are compatible with each project. So you can have these projects work together even if they don't directly. So if you want to sign up early for this um you can scan this. It's just hoy.app.

Uh you can sign up here for the weight list. We are looking right now for design partners in this. So you will be evolving this alongside with us in right now. I have these modules written right now. I've tested them privately, but also with some LLM bots. This was also especially fun because you can just have LLM bots pretend they're human and then just go through this process at a sped up time frame. And it's it turns out to be really interesting, but not at scale, not on real humans. So, we will need design partners to test this on real humans, on real communities.

And uh the gratitude guild is the alpha version of this. It will be the first experiment of this kind inside of polygon encouraging community participation in a productive way in a non- mercenary way. So gratitude guild is actually an instance of the hoy framework that will be using the polygon community to grow it to grow itself and see how this works out. Now uh so for tips I'm kind of out of out of time so I'll just skip through here. Uh so for dog fooding and poy how it would work a project publishes a list of desired testing scenarios and they have high merit on this category because they published it. Then a dogfooter will test this and report back. Self-submit a t tip request if they are high level enough because there's leveling aspect as well or ask a sponsor like a chief

dog footing officer who is assigned to this process and then the hoy will vote on the feedback projects uh vote is the heaviest. They have the most merit and others are still meaningfully contributing to this based on some math that we have. Uh the payout will either be denied or executed immediately on quorum and um yeah so the loyalty flywheel is self-reinforcing here. You can just do blog posts or videos if you're not technical. You can do tutorials. You can do tinkering demos and tools, IRL representation. Anything works for dog fooding and for sh for for spreading this Devril community uh and for growing it. So for the road map of Hoy now we have this weight list and early access we need these design partners. Like I said in Q3 we'll be doing a private beta for this and in Q4

we'll have a public launch with cross ecosystem template packs that will be usable by different projects in different context and categories. And that's it. I made it just in time, I think. Thank you very much. [Applause] No questions, right?