SingularityNET: Deep Funding | Community Governance – November 9, 2022
Yeah, well, there’s maybe a benefit to that. I’m not sure if any of you voted on our portal. But yeah, you did put some comments in, etc. So that might give you some bonus, which you wouldn’t have gotten when you were here to discuss the algorithm. It was actually interesting because there were two people to award a team’s, Kenrick and Robert Haas. For the first time, so it was a bit of a switch, but that was actually a good thing because they are both very numerical inclined, so to say. And so I could have a good chat with them on the quadratic voting and all that kind of stuff. Actually, what we wanted to do is to have the blog post out like somewhere in the morning or afternoon today. But yeah, maybe I was a little bit late in delivering it and people ill and things to do. So it’s not published yet, but I will show it to you. And actually, I already gave the link with the password on Discord. And for it, yeah, I couldn’t open it for some reason it says you’re not allowed to preview drugs. Okay, and didn’t you get a form to enter the password phrase? No. Well, then you’re up for a surprise. I would suggest to put the blog post on screen and then go through it and then I can introduce you to the results, which I think were quite interesting. And yeah, you can comment on it and I can maybe explain something wherever needed. So outcomes of deep funding governance around one. So as a flashback, what we set out to do is we did it. Did this experiment for three reasons. Of course, we want to know what the community was thinking and what they would like to have improved. That’s reason one. Reason two is having a real hands-on experiment with reputation ratings and then also using them. And then we do to give extra additional weight to votes. And this is possible since a month or two months or so since our portal partner has included the web three login on their portal. So we could connect activity on the portal with activity in our voting portal by means of the unique identifier of their wallet address. And then we can finally maybe not the most important but still interesting to see what would happen if we were given an incentive to people in the form of a GIX to be distributed for putting in effort and work and time to give productive comments. So let’s go directly to the results. The result is that we will have a pool for existing services. And as you as I will show you later, if we had used another calculation method, it might even have been a complete round instead of a pool. But the main thing here is that there are only very few people that would be totally against it. So I think there’s clear consensus on that thing. Now the other one, do we need around for ideation proposals in the ideation phase was a little bit less determined. It was more equal. And actually there there was impact from quadratic voting in the first part, but there was also impact from the extra weight gained by reputation by some people that were active on the portal and that voted, which eventually made the outcome of that question flip originally one token, one vote, it would have stand no chance with quadratic voting. And that was quite equal. And then it tipped off by the extra weight given for reputation. So that’s a really interesting and interesting result. At least it shows that that makes sense that this reputation actually impacts outcomes. So I think it’s a good result we can we can talk a little bit more about it later on I will show you the details also. There’s Tivo regarding goal three. Yeah, we distribute 200,000 AGI XR that’s quite a lot. Based on reputation scores, we will distribute that. Meaning that the lowest range will be something like 1,500 AGI XR and the highest would be 10,500 AGI XR for and that that consists of two parts a little bit for the act of voting. And that’s just related to the reputation score. So that is really a lot. And I hope that will motivate other people in next rounds when we will do this again, which I think we will. And add comments and add proposals. If not for the greater good of AGI X, then maybe also a little bit forgetting for getting some rewards for that. I hope that’s not the only reason, but it might help a little bit. Just give that extra push, right. So I already spilled most of the beans right now, but this is the general overview. And I’m very happy. I’m very happy with the result. Let’s let’s go down and put it on screen. So you see here. This is the official rating. So it includes quadratic rates, meaning that we take the square root of every amount of every wallets and use that as voting power. The result of that is that the extremes are flattened a little bit. So the few wallets that have really high amounts of AGI X gather lesser weight than they would have had purely based on one token, one vote. And that is also a lesson learned from phase one, around one where we actually saw and analyzed that there were a few wallets that had an uncommon large impact on the overall vote. So I think this isn’t improved. There are also risks involved to that, but we can we can talk about that later as well. And we will mitigate that. Yeah, so only 1800 points, I should say, votes for answer C and answer C was no, we don’t need an ideation. We know we don’t need existing services on our platform. Depending on how you counted a lot of support for AMD, but then in the end, we will do a pool. So question A was let’s have a pool with existing services. So that’s where we are going for in round two. One of the, I think, KPIs was how many people actually vote on the government’s proportions. Do you have some insights there? Yes. Coming right up. Stupid me. Let me go where can I where the engagement and get. Yeah. So here are the raw numbers. So if you look at all the votes cast based on tokens that would be 19 million. And coming from 84 unique wallets. And in round one, the regular round, we had 158, rollets voting. So it’s it’s more than half more than 50% of the people who voted in the regular, the real round, I could say. Also voted in the governance round, which in itself, I think is a very good result. I was afraid about 10, 20 votes, something like. Yeah, okay, I was reading this. I was thinking I made a mistake here, but I’m reading it wrong. So these 48 unique wallets represented almost 10 million tokens. And if we look at round one, these 180, 58 wallets represented 23 million tokens. So it’s it’s about 50%. Well, let’s let’s put it like that. And here I explain how these raw tokens are related to voting points or voting power by quadratic voting and reputation weights. There were actually five or so from these 84, there were five wallets that were also active on the portal that scored reputation points that were identifiable by their wallets. And therefore got an additional voting weight. And we’ll get to that later, but the voting weight was we added voting weight by a factor one till five. One would be for the one person that did vote that had a log in, but didn’t create any comments or proposals. So it didn’t get any additional reputation. So it stayed the same. And the rest was done on a tier system that we’ll get get there later. And the statistics are that we had 16 proposals. One of them was not quite accurate for the experiment for the purpose of the experiment. So usually I say around 15. The nice number of 269 comments, 198 votes and only three downfroads. So we were all very nice to each other. And 14 based I can based positive reaction reactions like smiley’s, etc. and non-negative. And this is based on the existing functionality of our portal. And this is what we use to calculate reputation. I’ll get back later a little bit on how we exactly calculated reputation. But let’s first have a look at this very nice graph. And I’ll show you the interacting interaction version. This was created by Robert has one of the participants and winners in round one. So very grateful for him to do this. So what this shows you, you can show very quickly how many whatever the voting weight was going to question one a. But you also see that it is very well distributed over all wallets. It’s not that only larger, only the smaller or whatever. So in blue it’s a little bit less equal. But all in all it shows that there is a clear consensus here. And it also shows see that those were really the exceptions to the group. And this is question two question two a. Should we have the pool the ideation pool. And this one is should we not have it. And also here you see that the opinions are randomly distributed across the sizes of the wallets. Which I think is a healthy sign. And a nice touch here is that you see the orange dots dots or maybe should call them gold are the ones that are rewarded with extra voting weight. So if you have already a sizeable wallets and you would get five times the amount of voting weight that will easily bring you in the higher regions. And that’s also the purpose of that additional weight. And five may look like a lot, but remember there are only three people I think that got five people that got a rating of five but three that’s not sure how many voted. I think two or three that voted with a with an additional weight of five. But these were people that put in like 40 comments. So they probably read all proposals. And were really active. So from that perspective, I think it makes sense to have them give them really higher substantially higher weight than any random also very appreciated community member. That is anonymous and that we don’t know of which we don’t know how engaged they really are. So if this is a peak into the future and I hope that in the next round. Yeah, I think there’s no reason not to repeat the same mechanism in the actual deep funding round. If more people will have a web three login and will use and will vote on our portal. And I’m also confident next time you will be able to vote with a GI EXO cardano. Yeah. A few minutes. Remember I should. One moment. So I finished my sentence. And yeah, then then the impact of the people being active on the portal will be even higher. And maybe we can play around a little bit with the weights, etc. But I think the messages that if you’re active and involved and engage you have a higher say in in the results. And I think that makes sense. And if you don’t do that, it’s also fine. You will still have some voting weight, but it’s less than somebody who is really very much engaged. So that’s what what is image shows very clearly any comments on this. Yeah, one thing what quite surprises I think is that let’s say for example the five rewarded wallets. Let’s say we’re meritocratic systems really hit and already they have been in quite strong consensus with the general vote. So say there is not a huge misalignment between their vote and the general votes. That’s a good actually. Yeah. But will be interesting to see how this then actually plays out on proposals itself. Then if you say, OK, that’s implemented to actually real deep fund round. Is it still possible that as five let’s say, par excellence voters are still able to represent the general vote on proposals then as well. Why if we see a bigger friction there then. Well, there are several things here. If we would have done one token, one vote, the outcome would have been different. So this is already changed by having quadratic voting. And these people actually influenced the vote by having a larger way. So in that sense, it’s not that surprising. But you will also see. Let me let me go over them in the video only. So this one. Oh, this one only voted on one question. You see that’s interesting. And this one actually voted for for question one was one of the few that voted against the consensus. Here, this one was against both. But this one probably carried a lot of weight and was with consensus. So it’s not that clear cut. There’s also difference of opinion in the people that are engaged. But their weight is higher than average. And then the end result is that they’re combined engagement and the amount of tokens they have because that also still counts managed to flip the outcome. Yeah. So I really like this image. I’ll put my camera back on. I really like this image because it’s very intuitive and it shows you in one view of what voting behavior was. And by making it interactive, you see that there are only a few wallets there that only voted on one most voted on both questions. So it’s a good spectrum. So yeah, it all makes sense to me. Right, let’s go back to the. Oh, to the. Blockpost. So there was. Yeah, it was quite a bit. Of course, we hope that it will grow in future rounds. But part of this engagement is also that we will never have 100% engagement on our portal. But the most important part is that the people that are engaged are. And the other part is that it’s important that the folks of the people who truly care are represented in the end results, whether it’s for governance or for real deep funding. So we were distributions. I already mentioned the rewards. So it’s we gave 5% of total rewards just for the act of voting. If you have a reputation score also on the portal. So for those people active on the portal that also voted 5% of the rewards is 5000. A GIX was reserved just for that. And 95% was for actual reputation. So yeah, we put it in in in 5 different in 4 different tiers. The reputation I’ll show you that later. And that is represented in these amounts. So that’s why they are. Rather similar to each other. But these are the people that really got a good benefit from their. From their behavior on the portal. I haven’t this burst this yet. I have a good look at the. A GIX on Cardano wallets that they address they provided. And that’s where we will be transferring these amounts to. All right. Oh, yeah, important. These people. They don’t get anything. They don’t get any A GIX. I don’t get any extra A GIX. But I did. Sorry. I’m to write. Let’s give singularity and some talk. Yeah, I think we’re well off already. And. We did assign ourselves reputation. And the reason for that is that I hope that at some point in the future, reputation will also influence visibility of comments and remarks on the portal. And I think it’s a valid that also people inside the singularity that would have a higher. Higher visibility based on their score of reputation. All right. Yeah, one person with a web 3 login that didn’t create comments or proposals. So he didn’t get any additional reputation score. But he did get a vote in reward. So he or she, I don’t know. Also noticed there were people that did get reputation scores, but that didn’t vote. And actually, there’s relatively many people that were engaged on the portal, but didn’t vote. Normally, you would expect everybody who’s that engaged on the portal would also be in motivated to vote. I can understand the reason for this. I don’t vote with a GI X on cadano yet. So for some of the people here in this group, that would be a hurdle, which is completely understandable, which is another reason why I hope to have more connections between portal and voting portal in round two. Whenever you have questions, please interrupt me. The reputation scores gained. So the above the rewards. I put that on top because that might be more interesting to some people, but these rewards are actually based on reputation scores and what we did. We created four tiers for reputation. And the highest tier we got, weight of five and the lowest we gave a tier of two. If you have any reputation gains at all. So here you see the actual numbers. We have a clear winner here, but that’s because I’m a professional in this context. But a very good run rub was Dr. Farham here, a well known guest in this setting as well. Well, here we have T for all and you win a well, won’t name all of you, but you can see that some are really high and some are quite low. But we thought it would be a good idea to flatten that also a bit by putting them in different boxes, different tier. And I’ll get back on the way we did that a little bit later. So scoring rules details. I think this is an easy to understand. Well, almost easy to understand one for everybody. These are the points we came up with for as a reward for all the activity. So creating a proposal is important, but we split that in two things. You would get a reward just for creating a proposal, no matter how lousy it is. There weren’t any lousy ones, but just a matter of speaking. But you’ll get also rewards based on the rating that other people gave you in the portal. And for that, we have kind of a complex algorithm, which I maybe shouldn’t go into in this setting. You can read it in the block. I tried to explain it as good as possible. Basically, we multiplied the amount of people voting with the average vote given so that it’s not just the rate that counts, but also the amount of people that rated you that. So if you have an eight average and there were 10 people voting for you, maybe that’s a better result than having a nine with only one people voting for you. So we tried to take that. Oh, we actually did take that into account. And I’ll come back to the multiplier a little bit later and use both components as a reward for proposals. Now, let’s finish the proposal thing right away. Then I figured that how what would be a good ratio between points given, and reputained points given for actually submitting a proposal versus points received for giving comments and giving feedback and reacting on comments, etc. Any one would like to have a go. What would be a good ratio in percentages? What percentage would you have given to people just for the act of proposing and what percentage of the awards for the act of commenting and one person can earn on both both ends. Any one has an idea and shoot. Then I’ll compare it with what I actually did. I will use a two nomencl approach. So using correct results of, let’s say there are 20 people in the tip-found round and like two in stuff, then perhaps like 25% means that the good amount of proposals would be five. And then if more people join the list of proposals, you actually want in so that you always have like a golden line of proposals and then you have the other hand, you have a hit’s basically the same. So that when the amount of proposals deviates from the golden line, then the does not matter is it up or down, you basically start. So it’s interesting. It’s the percent of how much it should be proposed on work. So you start from the assumption that it should be a kind of ratio between the number of proposals and the number of comments. Okay. I started from the assumption that there should be that there should be a ratio fixed ratio between the number of points earned by submitting proposals versus the number of points earned by submitting comments and assuming that if you have more proposals, then probably also more comments. Or you could say that supposed that would be a huge rush of proposals coming in, then the ratio would shift, then you would have. Then the the weight of the comments would be lower, but actually the importance of these comments would be higher. And I would like to have even more comments if there are more proposals. So by having an hour, I’ll just give it to you by having a fixed ratio between the number of proposals and between the number of reputation points gained by proposals and by comments. I think you will actually incentivize a little bit, motivate people to actually or make more comments, give more comments, or maybe submit a proposal. Right now, the ratio that I defined was 20% of the points go to the submitters of proposals and 80% of the points go to all people giving comments, etc. That if you don’t submit a proposal, and I imagine right now that we will do it in the same way or similar way in round two, you can still score a very high reputation score by commenting and by giving good feedback. On the other hand, if there would only be one proposal, then there are still a lot of points to be earned by creating the second proposal or the third proposal. So that’s another way of looking at it, TFO. So actually by doing it this way, you would motivate a kind of balance there. Feel free to comment if you think this is stupid or brilliant or average. And if you all agree, then I’ll just then let it sink in a little bit and then I’ll continue and we can come back to it later. It will have neither the mark of what for that to kind of represent it. Yeah, it’s not the easiest topic, right? It’s quite complex and technical and I also said to Peter, well, I don’t know how many people actually read this blog from top to bottom. But it’s still important to write it because it’s about transparency and I want to be transparent in how we did those calculations and also be open for feedback and improvement. For me, it sounds solid enough for a bit beginning, but from my perspective, it misses a little bit like some really innovative idea to say, wow, hey, that’s something really fully around. Because I think what will happen also for posts like this, there are two kinds of people who read it. Contributors, not really. They don’t give so much at about they look more on, okay, how can I use on my individual base of the whole platform? People will read it. People want to really commit to the thing and really want to go deep into it and the people will try to cheat the system. And like this late, I think it will break after four or five rounds because people can easily cheat the system actually. That’s what we saw in Qatar, that’s what a lot. And then your highest ranked people actually spend. You’re right. So there’s always it’s very hard to make a system which is completely foolproof and will not be cannot be broken. What you can do is limit the impact of bad actors. That’s the first thing. So maybe not bring it to zero, but make it so small that it’s not that it usually won’t have a lot of big impact on the outcomes. That’s the most important thing. And I think there’s also always the human aspect. I would always, well, maybe not always, but in the beginning, I would keep the option open. That in case we see clearly fraudulent behavior, we will penalize that penalize whatever. So if I would see people creating a script that would submit proposals and then put 100,000 proposals there. So they can have all 20% of the reputation. Yeah, then clearly they will be blacklisted. And I won’t put a special rule in under what conditions exactly they will be blacklisted because there will always be something around that. But at the beginning, we will use our common sense for that, but we will be transparent in reporting why and how we did it and from. And then as we move along, we will have an increasingly good understanding of what kind of fraudulent tactics there are strategies. And then we will be increasingly hopefully be able to counter that to a to an extent and at the moment that maybe at some point we feel safe to have our hands disattached and say we can’t touch it anymore because we have such faith in the system. And if it then goes wrong, then we have one round where it goes wrong and then we have to fix it for the next round. That is how I expect this to be. So it will be a kind of a if we are successful with deep funding and with singularity net and decentralized AI and all that. Then it will be an arms race because it will be popular and there will be people coming here for the wrong reasons. So yeah, that’s also a reason why I’m transparent in all these calculations. So on one hand, it makes us vulnerable because people know exactly what the rules are. So they know exactly how to exploit them. But on the long run, I think it makes it stronger because we will have the intelligence of the of the crowd and all of you. So we have to improve these rules and be completely continuously working on improving those rules. All right, all right, all right, let’s continue. So this was about the proposal score and then so the multiplier is used to bring the total points gained from proposals up to that 20% of the total reputation points. Distributed in around. By the way, Felix one one thing I want that I aim to improve this, but actually two ways. I’m thinking should I let’s just do this organically. That is the risk organically this discussion. One other way we aim to demotivate unwanted behavior is by also having a reputation system for wallets. So we will count in how many rounds of wallet was active. So that somebody who suddenly is new and want to change the whole thing over. Doesn’t have a history. But the latest thing is that we will also be looking at the earliest dates that AGI or AGIX was added to the wallet. So that and the reason for that is that we are doing quadratic voting. And we are very clear minded that quadratic voting gives the opportunity to people to split their wallets in that way have a higher weight. But we mitigate that by giving also weighted amounts to the voting weight based on the history of the wallet. So the older it is, the longer term associated you are supporter of singularity net. The more weight you will have and the main thing is that people will not be able to split it just for this round and then expect to have a huge gain by that. The main thing however to mitigate that is by having more people on the platform gaining reputation by actually productive behavior and giving them extra weight as well. Which also makes it less interesting to split your wallet because those wallets will not gain reputation weights unless you’re really smart and have a script or whatever that will post comments for each of these split wallets on our portal. You understand it’s already getting quite complicated. So you have to be really motivated to do that. But hopefully then we can still read it out by looking at history of people and by other people who will vote down certain comments and maybe at some point we can have some AI there as well doing some some analysis. And the cherry on the cake here would be and that is something not for this or next round but it will take a while because we need proper tooling for that first would be if we make the whole reputation system liquid. So if you get an up vote from somebody with a high reputation that would mean more or a down vote for that matter that would mean a lot more than if you get an up vote or a down vote from a very new person in the system. So that is not a way to account for history and and and accumulated reputation etc. So for all those people here that are listening to this with the purpose of manipulating this system you have to be here quick and you have to be really active and otherwise I think it will be very hard if you start in round two three four and the community and the system around it is maturing to still get a high position based on fraudulent or undesired behavior. So we putting up our defenses here one by one probably there will always be ways to trick it or to influence it but I think with with all these things voting reputation or wallet reputation reputation in the system based on constructive behavior and the liquid aspect of that reputation. I think these three things together to make a very strong defensive system against undesired behavior. Let’s call it undesired not even fraudulent because it’s allowed you can split your wallet, snow, law against it. It’s just not what we would like to see. All right. But there’s there’s ways to go and you know there’s an excel sheet behind this these calculations in the next round. We probably if we are so if we are in the next round if we are looking for awards that I spoke with have another hundred hundred thousand Ajx distributed for engagement we will only look at the newly created reputation but if we look at voting weights we will look at the engagement in deep funding round two but we will also take into account the reputation score gains in this governance round. And maybe it will be something like 2080 or 3070 or don’t know. That is another way so that’s also a complicating factor right so that will make the next excel sheet more complex than this one and then one after more complex so there’s also more room for errors there. We’re all you we can make errors in these excel sheets. But then imagine that you would have this system with a reputation per round based on all these scores here and then you want to add a liquid reputation saying that not every comment is equal but it depends on the person who gave that comment. So I’m not sure if my excel sheet will be able to keep up with that so that is a tool that I really would like to have and I plan to have some discussions with people with sway maybe also with some readily awarded team members and some others to see if we can create such a thing. Well, well, we’re on the topic. Ideally, but I see today maybe I think about different tomorrow after I had a couple of conversations or next week. I can imagine that we would make such a tool for singularity net based on our ratings and based on our systems. But I would what I would really like if we could make such a reputation scoring to generic so that we can also plug it into. Well, catalyst. Of course, then it has to be used and the tooling has to be ready, etc. But also for other programs out there and then it will become even more interesting, suppose that people could get a reputation on catalyst for things that are doing there and people would get a reputation on deep funding for things they’re doing here. And then we could say in deep funding, we will not only count your reputation gained in previous rounds, we will also 10% of the reputation points given or whatever. We will base on your reputation in catalyst. So 10% of all reputation will come from catalysts so people that already have proven that they’re very constructive and helpful, etc. in catalysts will inherit a little bit of that reputation when they come to deep funding. And maybe there are other platforms out there that do the same and then yeah, we can get an ecosystem of people with high reputation and then somehow it will all be connected. And then we can also get a lot of those reputation across ecosystem, cross project, yeah, reputation does would be all goes a little bit also in the way that enough the idea isn’t as I wouldn’t make sense if I have the idea on Ethereum one for a poll cut out one for cadano. And each time I have to create and now this would be all like that. And continue to configure this in higher levels, you can say well, cadano catalyst is really similar. So I will really give a high weight to reputation score gained in catalyst. But suppose that another project that is less related to our ecosystem would have similar system, we could say well, we’ll give them a little bit because at least they have shown that they’re serious people, but it doesn’t say a lot about their merits in this topic. And so, so you can do a topic based, you can do it platform based, you can do a combination of those, you can make it infinitely complex. And infinitely fun. And hopefully also by that way, very nuanced and valuable so that you’re not scored your score. So that’s the important thing of this whole reputation system that it’s not based on one simple parameter, your A or B. But based on a whole string of activities across the line and history, et cetera, and that would make it stronger and more resilient and more valuable, I think. Also more valuable because it would take more effort to gain a higher reputation score based on all those criteria. So therefore we also represent real value. So we’re deviating a little bit from your results of round two here. But yeah, this is an important topic, which is close to my heart. And actually what I tried out to do with this experiment is do really hands on first step, that’s what the experiment was really about this first step in this direction, see if it’s feasible and see if we can continue on that path. So based on that and based on these results, that’s the reason why I’m really happy with the outcome of this. Let’s see if we have anything else important here. Yeah, so the reputation rating tiers. The way I did it right now, the top 20 get away fifth, five, the runners up 60 to 80 get away factor R and tier B get away factor B. And I have a little bit larger box, so to speak of 30% of us and a lot of 30% of that get lower weights. This is really adequate for this round, I think. I can imagine if we have more people engaged and a bigger audience, I can imagine for instance that we will make a separate tier for the top five person, because what you will see often in these kind of context is that there are just a handful of persons people that everybody knows. And that really put all their time and effort in it and are also really accomplished in giving good comments. So I can imagine that these are. At some point in some of an extra these will surface and these will be the people that will be given even more weight relatively than the others. But this is open for discussion. But this is what we did in this round. So we have four different tiers. So whatever your score is, you will get into one of these four boxes. So to speak and based on that, we have given you a weight factor. So it’s not completely relative to the exact score that you gained. All right. The impact of the of the reputation rating. So if we would not have used reputation weights, what would have been the results? That’s what that’s what I’m showing in this table. For question one, there would have been no material change. It would still have been question one that one easily. For question two, however, you see that. Actually, the outcome would have been different without the reputation weights and then question also to would have one. In that case, there would not have been a pool for ideation in the next round. And if you would also take off quadratic voting. So just pure one token, one vote like we did in round one, actually. The outcome would be even different. So for question two, it would still be answer to no separate pool for ideation, but also question one would have changed. If we would have done it one token, one vote, the outcome would be let’s not limit existing surfaces to a single pool, but do a full route just for that. So it does show the enthusiasm for people for this concept of having a pool for existing surfaces. So I think that’s nice. To be clear for everybody here, I don’t have a clear preference for any of these answers. So I have no motivation to steer. I think I think answer to would have been for question one would have been cool as well. And I think answer to also to be honest, I have one preference. I would have been disappointed if it would have been answer three. We couldn’t have done anything with existing surfaces, but I’m in good company there. And also question two, there was even more, I was even more neutral there, both are fine from them. But with the additional reputation weight, it turns out that we will have this ideation pool. And I think it’s good. It’s good to try it out as an experiment. We won’t make it a whole big pool. We’ll keep it a small pool with some conditions there. And we’ll do it for round two. And we’ll see what will come out of it. And then we’ll decide where we will continue that in follow up rounds. Or we may have, we may have, sorry, I was being caught or was distracted. Or we may even have a new vote on the same topic. Listen, we did it once. Should we do it again? Or maybe have three proposals to change it? I don’t know. Maybe after the next round, we’ll have other topics that we think are more important at that point in time to ask the community. But, yeah, this is the background. These are the options. And this is how we treat that. Yeah, so conclusions and what’s next, what’s next is the next round of deep funding. There’s quite a bit of work right now in writing down all new rules and conditions and publishing them on the website. Maybe create a video about it. I don’t know. Before we can start the next round. If everything would have been the same as round one, we could have started it tomorrow. But I think I want to give us some time to properly write everything down so it’s clear to everybody where the rules are. Here it was an experiment. Maybe I will still call it an experiment in the next round. But I think more of the rules will already be predefined because in round two, it’s about actual grants that are more important to people than the awards that you can get as a bonus for your activity. So I think we need to be a little bit more careful there. But like you mentioned Felix, if something really goes awkward, goes strange, then we keep the right to change it if necessary. For the sake of having clear and unbiased representative outcomes. All right. Any questions? No, I just want to say this is very, very cool and very insightful. Thank you for that. Thank you. We appreciate it. And they’re interested in viewing like the exit, not the exit sheet or whatever is public like the calculations and what not if I don’t know where to find them. Yes, I didn’t put them and make them public yet. The main reason is that all the wallet addresses are there and I’d rather not just publish them. Even though there are public addresses, and maybe in some cases they can be inferred already if you put enough effort to it. Still, I don’t just want to throw them out there. If I take them out, then the exit sheet will collapse. Of course, I can create a new sheet with paste everything as values. But then it will be harder for you to see. But what you can see here already are all the variables that are used. So the tiers, the waiting, the points and the scores for where is it, the how we calculated the proposal score. But if you think you really want to see the actual sheet. Well, okay, then I’ll have to make it public and then take out the take out the addresses somehow. I’m all for transparency. So I’m actually surprised that after this long talk and all these details, you’re still motivated to do that. Sorry, I’m a math person. I work with numbers sometimes in my head, so equations and stuff click better. But no, this is completely fine. Whatever. No, no, no, it’s also. It’s also fine. I figured that this information would be already too much for most people actually. So I thought I can put it out there, but who will actually look at it? If there is an interest that’s not my not and I can put it out there. And yeah, if you see an error in the excel sheets or if you see something that you think could be approved one way or another, then please let me know. Felix. When looking back when we started the whole thing, there have been really the main fear was other actually people submitting proposals for the governance experiment and other people actually voting on them. But I think this was a really nice success actually. So from the takeaway. Do you plan to say, okay, once there’s a deep funding round two between deep funding round two and three, what would be a second? Let’s say also improved governments experiment. I think so. Yes, I think that would be a good idea right now. I think it’s valuable. I think I liked it. I think we got good feedback. I think it helps us to help us to learn and to evaluate what happens in around. So the way I’m thinking about it today is that maybe in the first year, the first four or five rounds, I don’t know, it would be good to have a governance round maybe three, I don’t know. After each regular regular round have a governance round. And I can imagine that as we are progressing, we will have fewer governance rounds. So then we will have two rounds and then a governance round. At some point, there will be less things to discuss and things will hopefully get more stable. So let’s see the other way around. It’s more complex than in the second round and there’s more people you have in the world than it’s more important to maybe his further governance round. Well, there are different options. There’s also, so it’s a parallel track, right? So we can in any way, we can continue discuss it continuously on social media, on Discord, etc. At what point we need to organize a vote. Yeah, I can also imagine that at some point, that will be more like an organic thing, that we will base voting, actual votes on engagement on our portal. I don’t think that would work today. We actually tried it. There is a mission. There was a mission from the beginning where you could give input on what you want to have changed. But I think doing it like this in an organized way will bring on more engagement. But maybe if we get there Felix that we have this larger community that is super engaged on all kinds of things, then maybe we will have an organic thing where we say we give some conditions based on which a proposal will be added to a vote. That would be the ultimate thing. So if there are X amount of comments and X amount of positive comments and the rating is more than Y, etc. and maybe some more conditions, then we can say, okay, this is now mature enough to put forward for an actual vote. And then it could also at some point maybe, yeah, I mean outcomes of these things have to be implemented in the next round. So in that sense, there’s always a relation to rounds. But apart from that, it could be a completely parallel thing. And in other way, I can imagine things continuing. So the proposal of starting, of having a pool with existing surfaces. In a different situation, the outcome would have been that this would not be a pool, but a full round. So I will, I’m really interested to see what will actually come out of this if community members for instance will be activated to start looking for surfaces that we can put on the portal or help out actually onboarding the surface on the portal and gather reward for that. Suppose that would be a great success. I really hope it. I think that would be the greatest thing. Then there might be reason to say, well, this is such a success. Let’s put the whole round to that. And then maybe we can have alternate rounds. We can have one round for regular, like we did the first round for regular solutions that still need to be developed. And then one round for existing surfaces and then alternate those. And then in the meantime, do all the governance stuff. So yeah, and then we have the tooling that we need to create to make sure and configure that and improve that. So there’s work to be done. Another question, then as well. Okay, now we know in deep front round two, there will be our own pool for onboarding already existing AI services, which means, okay, now we have a challenge actually because this is something quite different. And if you ask people to come and build their own services, which requires maybe some very specific skill sets, what we had already conversation around. So say if we want, then actually this result becoming in success, what we have to do now in order to make this then the role of that success as well. Because onboarding already existing one requires totally different skillsets, I think, than creating a new one. Definitely, definitely, and I think actually it is, it is more reachable for a lot of community members, so maybe community members have knowledge of existing services they would like to see on a platform. They could reach out to the developer of that service and they could say, hey, listen, that’s team up. They’ll do all the administrative work on the portal and create the proposal, etc. But you give your go code, and maybe there’s a third person that will help with the actual onboarding of the service or whatever. And then they will put a reward to that and that is something that any community member that is not a developer can do. So there’s some overlap here with, well. What we see in Kadam Faxon when it comes into integrations, this is really difficult thing integrations because the thing now is that it’s not a community anymore for the integrations you need to develop our teams from the entity. So in this regard, then there’s much more if coordinated from singlearity, that foundation now to make this integration certain, because the people on board this project, then they will need to come and say, okay, I have this massive nice platform. They want to integrate on the platform. Singularity foundation, we need to develop our teams to make this happen because the community is not able to leverage this integration actually. So what happens now is that the community engagement and commitments shifts a little bit more or much more actually to singularity in the foundation itself and to their requirements of delivering effective integrations. So, well, integration, I would call it publication publishing your service on the platform, but that would be a great problem to have. And in the end, as I have mentioned before, if you have service and you run into issues in onboarding it, publishing it on our platform, we will help you. But the first step that we will take is say, well, here is all our documentation. Please have a look, go through it and let us know where you are running into issues. So there is, we do expect some effort from either the developer or the community member or a third community member was dedicated to publishing. It would be great if two people would team up from the community and one that would be focused on this publishing thing and the other that would be focused on getting developers. There can also be substantial work related. So if you have an AI service that is really interesting but isn’t API ready, then that could mean that you need to refactor actually that service. And that would be substantial work info. And that is the reason why also in the previous discussions, we said that we don’t want to have a fixed award for the services, but it will depend on the circumstances. But to I stay I stick with my former statement that any community member would be able to gather rewards based on creating a proposal for an existing service. You have to go out into the world, you have to find that service. So you need to have some affinity with the topic and maybe know a little bit what’s about. But if you can locate that service and locate the developer of that service, then you have enough to create a proposal. And then you have to motivate the developer to take a little bit, take a few extra steps to publish the service or find somebody else who can do that for you. Or if all that doesn’t work, come to singularity net and we’ll find a way to help you with that. If that proposal is awarded by the community, because that’s that’s necessary, right? And an important thing here as well. The awards will be disbursed for this pool after the proposal is on board it. So that makes it very interesting for our platform, because we will actually disbursed amounts when having the service on our platform. And it also ensures that people have to be serious and not just put names and proposals, but they will actually have to make it happen. I can see community members here as drivers of this program, of this pool. But if that will actually happen, if people have the time for that and take the effort and are motivated, I don’t know. I really don’t know. I hope, I hope that will. And I hope that whatever will come out of that pool will give us enough confidence to do it again next time. Maybe do it as a complete round is a very ambitious would be nice. But I hope it will be at least successful enough to continue having that as a fixed pool in all rounds. So I guess maybe over the coming weeks somewhere, we could also have discussions about that during, for example, this meeting on how to exactly like, let’s say, percentage wise, how we play around with with numbers. So that’s something you want to have on a community level, like a colleague does. Yes, I think we’re discussing now internally what the height should be of the next round, or the total amount is. I have some ideas on how to distribute that amount, whatever it will be among the five pools that we indicate sort of five pools are a regular pool with no maximum. But just for with a with a certain total amount. So we don’t have a pool for 40k and a pool for what was it. 150k and 50k anymore will just be one pool without a maximum because we found out that a lot of people would create proposals for 40k because that was the maximum of that pool. And I’d like for this time to keep that open to see what what will happen there. The second one would be a pool for existing services and then there are three secondary pools as I would call them one would be for marketing proposals. So if you have a very good idea to promote the platform in a quantifiable way, then we can have a small pool for that. We will have a small pool for the proposals in the ideation phase. And we will have I would say medium size pool for tooling. And then I’m also hoping that I can motivate some parties to create a proposal for for instance a reputation dashboard or reputation calculation system that is configurable. Maybe your first step that we can build out at later stages. If the community will award such a proposal with their consensus. By the way, I see there are a lot of in the corner of my eyes. A lot of things happening in the chat. I really haven’t read it. I didn’t have time to read it. And usually when I start reading, then I find it hard to distribute my attention. So please shout out if you have anything or maybe somebody else. If I see some remarks from Nadine, if you’re not able to talk, maybe somebody else can put something forward if it hasn’t been addressed. I can, I just an interesting conversation that Peter and I were having. He was mentioning how can we move data around reputation data from other platforms to ours and vice versa. Like for example, Cardano, like we were saying. And I was just not clear how they would do that. You mentioned through wallet mapping. I was just wondering how they would then know my actions on an off-chain platform like a catalyst because all that doesn’t happen on a wallet. So just because if this is doable, then it’s massive. You know, it’s definitely worth exploring. It’s certainly doable. But you need to have tooling for it. And yeah, there are roughly two ways, I would say, either have it in a centralized database or tokenize it. And if you tokenize it, then yeah, you have proof of reputation from on one chain, then you can even do it cross chain. And yeah, you can use that as related to your wallet, for instance, and then have it transferred to another platform. So that would be one option there. It would be helpful to have some consensus between systems on how they will use metadata on that on such a system. For instance, what kind of it would be helpful to have reputation related to topics. So I have a very high reputation on blockchain, but a very poor reputation on AI. That’s possible or vice versa. So it would be helpful to know from people whether this is somebody with a high reputation on AI or on blockchain or on repairing secondhand cars. But there are actually I did a look up on that a while ago. There are open sourced catalogues for that categorization systems that could give some direction there. But this is quite far ahead. We are not there yet, but this could be potential future. But I think it will be possible to share these reputation rates one way or another. I’d rather start small and we started with smallest way right now with a simple Excel sheet and some handmade rules and calculations and see what happens. The next step would be to make a better tool for that that we can really rely on and that we can configure and then also add additional complexity like liquid reputation. And then the third step I think would be making such a tool generic so that everybody can use it and have some configuration on data that goes in. How you will rate it and what goes out that could be a step. One thing what could help that was warm. We already started generate data there because with the bounty board and the work what we do there. For example, the bounty is for singularity that as was one and we see people which are quite engaged in both parts and we document the whole stuff. And we have around 1000 reverted bounty’s tasks so far. We can see with a direct link to each task where we already created that and say where do the people actually engage and where we see already okay we start to have this kind of resource pool of human of folks which are quite engaged not only in the catalyst but also singularity. Yeah, that’s a start we have that token it’s also there’s also a tokenized part there so we could use that but yeah we have to think that through. In the meantime I’m also thinking about these topics and that metadata probably the better way here would be to AI that basically not to have a static list of topics but to have an AI. With proper NLP capabilities and understanding which is a little bit above the current state of AI that would be able to infer from all the interactions of all people are making all those links and all those following all those interactions that you mentioned Felix and then infer this person is mainly getting good results and these in these in these areas. That would be very interesting. It’s quite important maybe not categories but a certain request for taxonomy or to create a library of tax for example where they I can relate to it’s a figure out okay that this person is more content rate. Yeah, greater this person is more translated this person is more what not. Yeah, it will be more like taxonomy with a hierarchy in it or maybe it will be even more like these source with more like an unstructured connections and hierarchies there are all kinds of ways but yeah if we leave it to an AI to make those. And then we don’t even need to have that structured but it will be more organic and therefore we it could also be targeted better for specific goals because whatever taxonomy or catalog system whatever you put in front of that it will always be limited for in relation to specific use cases. And then we will have future looking stuff and that will be a while. That’s why we are here for the future. Well, maybe it will not be that long. So let’s if we can make a step each quarter or each health year in this direction then a couple of years looks long in crypto space but when looking back to three years. After today we might have covered if we if we do our stuff right we might have covered quite a lot of ground here. And then we have a solution for all social media shits going out there. Let’s say most all is. Well, we have a solution. I’m not saying that it’s implemented. Well, they’re already solutions going around. I hear talks of of charts and ban and so they’re already potential solutions but having them implemented is a hard thing. All right. It’s actually right on time right now. Any other things you would like to comment on or mention. Will you be in Los Angeles because Pete I think he will be there. Sorry, where? In the carano summit in Los Angeles. Okay. I don’t think so. It’s not my agenda right now. Right. I will have a beer. I will have a beer with people who ask me to drive and then I have an excuse. What is the date? What is the date of that? 19 to 21. It’s the official. But yeah, folks are coming already and they’re from the 15 to 22. All right. All right. I might be busy with dig funding around to run and a lot of other stuff as well to be honest. I think I’m going to be there. You see, you are going there Felix. Definitely. Yeah. Well, I would be nice meeting you in person one time. If not this time, then another time. I also plan to use my position and advocate and spam a little bit about singularity that I think is in stuff. We’re always grateful for that and you’re our greatest ambassador. Thanks a lot. There are people on that wonderful note. I would say, let’s end this call for today. Yeah, we started out with bi-weekly meetings. So what would you propose? Would you think there is enough topics for another call next week or shall we skip a week? When does the next round sound to start? I don’t have a date yet. I would say the end of November or beginning of December. That’s what I’ve been saying. It depends a little bit on my workload and how much time I have available to prepare this properly. I don’t have an army of minions yet that can help me with all that stuff. But I hope end of November somewhere. I think we’ll probably be voting realistically. The voting will then be after Christmas in somewhere in January to give people some time to get used to the new rules, etc. And to think about it, propose stuff, create proper proposals, etc. Are there major changes happening in the second round? Do you feel there are things that we can discuss and improve before we launch? Yeah, a lot of things have already been discussed. There’s a blog post out there with changes that I’m proposing for the next round. Of course, there are still some details that need to be filled in. I’m not sure how heavy these details are. They all need discussion. But I mean, I created this blog post with all changes. So for instance, one would be a minimum value for HIX that we discussed last time. Exchange value to protect the treasury from temporary downfalls. Another one would be, well, we have these two new pools, which makes a difference. The difference is that we don’t have those maxima anymore. Difference will be that we will again use weighted folks, weighted reputation waiting. A different will be that we will use quadratic voting. A difference will be that we probably also, if I can create that, use the reputation waiting for wallets based on their age, basically. All that to mitigate the opportunity for wallets splitting. Yeah, I think those are the top of my head the most important ones. But those are quite a few things. And we will have, I hope, also incentives again based on that reputation score, because I’m very curious if people have read this blog or part if they see what kind of incentives we have been distributing in this round. It would be very interesting to see, if then in the next deep funding round, we will have an higher engagement. Then it’s still the question that it’s because of that or because it’s just the next round, but would be interesting. So there’s at least a lot for me to describe in proper detail and to publish on the website, etc. For the proposals out there, the regular proposals, the only thing that they need to take into account is that if the token price would go really low for a month or a couple of months, then we will not be able to disbursed the whole amount. But as a counter measure, what we’re doing for them is that we plan to have to disbursed the amount at the beginning of each milestone and not at the end on delivery of the milestone so that they always know they will get the full payment for the milestone ahead of them. Should there be like last few days with all of us going on a sudden extreme drop, then it is possible that we won’t be able to commit to the full dollar value of their milestone. Then we will be in conversation with them on how to address that and what will happen if the price goes up in two months and we will be able to compensate it afterwards. Those are also things that will require a little bit of thinking and doing actually an experience. I think for now we would do well by keeping it on the weekly. There’s these new rounds which to me feels like we can ideate a little bit on how to best approach it at least. I don’t know, I feel like worst-case scenario after half an hour we realize, yeah, well that’s it for now. We don’t have to go full-time, best click. Do I have these calls? Oh, thank you. I really appreciate that. I do too, by the way. Okay, let’s keep it on a weekly thing until we get bored and then we can do it on a little bit. Actually, I’m glad that you say that, Joanne, because I feel that I am the one talking all the time. Even though I really like to hear my own voice, it would be even better to have more discussion, etc. I think this was a special case, right? I wrote this blog and I had a lot of explanation to do so. Yeah, hopefully in the future. I hope at some point these meetings will also continue and be valuable in situations where for some reason I will not be able to attend one time or something. So that’s what I think it’s also natural, because right now I think UNP have the strongest experience and expertise here. As the people join regular restart to build their own experience and expertise established a role, then might shift also more to, okay, they come with already topics of what they have in mind, what they want to discuss and just spam you and you don’t even have time to speak anymore. Yeah, and then we have to need a real good system to manage all those ideas and suggestions and proposals, looking forward to that. Alright, then thanks a lot and looking forward to speak to you next week.