The meeting begins with an overview of the current structure, inviting questions and clarifications. The system is described as a vertical hierarchy with two DAOs overseeing services to lower tiers. Discussions include determining voting power, protocol and Hamza DAO considerations, and debates over vendor and product rankings.

Chat History:

Michelini — Today at 6:05 PM
https://www.figma.com/file/kVlAmEYLZU6eyBNqOlyiZp/Loadpipe-Tokenomic-Design?type=whiteboard&node-id=0-1&t=XAyML4zLs41xhoyQ-0
FigJam
Loadpipe Tokenomic Design
Created with FigJam
Image
Michelini — Today at 6:13 PM
@john.kosinski for protocol services, you mentioned the mass market relay nodes being included as a Service provider here?
on marketplace side – how about the rewards for the community to find mis-categorized products and other product listing violations?
another service – not sure if on the protocol or mp level – but “reviewers” is a service in web2 – like Amazon vine reviewers https://www.amazon.com/vine/about
Michelini — Today at 6:47 PM
https://www.mturk.com/
Amazon Mechanical Turk
Access a global, on-demand, 24×7 workforce
jlaurentum — Today at 6:49 PM
How are judges judged? Is the only mechanism for judging judges based on how much they agree with the other judges? If so, you could be leading the system into a tyranny of the majority, incentivising group think over individual think
Alex — Today at 6:50 PM
How does current ranking work of public judges?
jlaurentum — Today at 6:51 PM
How about creating automated judges, taking decisions based on an algorithm and weighing of individual “objective” indicators?
Alex — Today at 6:51 PM
So there is no system ?
Jetsetter Jim — Today at 6:52 PM
Also, one thing Beau and I had discussed is the inability to see how others voted before you do yourself, in order to avoid the tyranny of majority (or trying to game the system by trying to vote according to the majority so you can receive a reward)
Alex — Today at 6:52 PM
Can we do the same ? Appointment by foundation?
jlaurentum — Today at 6:56 PM
Appointing judges randomly… did you guys know that’s how ancient greek democracy worked? They even had a random selection device for appointing public servants – called the kleoterion : https://thelivelifecouk.wordpress.com/2021/05/18/kleroterion-what-is-it-how-was-it-used/
The Live Life
Eloise Hands
Kleroterion: What Is It & How Was It Used?
This Kleroterion was a device used by the ancient Athenians to randomly select citizens for juries, offices and state councils as an early example of democracy.
Kleroterion: What Is It & How Was It Used?
Michelini — Today at 6:57 PM
like getting called to be juror in court – jury duty
maybe the community gets requests to be a judge / juror / jury duty

Transcript:

(00:02) so what I was thinking for this meeting is to start with just sort of like an overview of where we’re at and then sort of open the floor to like clarification and question and like any question at this point to to clarify what what these different things mean what you know any sort of like lingering questions people have about the design or like where we are what things mean cuz like as I think about it and so I’ll start with just like a little bit of an overview of how you know which we talked a little bit about last time but an
(00:39) overview of how I see this board looking right now and what it means to you know for all these different things basically it’s like this I see this as like a vertical hierarchy where the two Dows represent like anything that is like horizontal to a dow to me is a decision or a governance thing or something owned by the Dow anything below it are Services provided to to like lower tier Dows and then the same can be said down here right where we have the services we talked about like product liability for vendors possibly warehousing and
(01:22) fulfillment and Marketplace UI which are things that are provided by the marketplace Dows and then these are like the decision genres that they can make and then up at the top here we have this section for like voting power which is like how do we determine the voting power you know the token voting power of this this Dow right so you can see like this is a controlling mechanism for this Dow which provides these Services which is consumed by this Dow which you know and again I I don’t know if we should have a different voting power mechanism
(02:01) probably similar but like that’s something maybe we need to flush out and discuss which provides these Services ultimately to vendors and buyers so and and with that sort of again like vertical hierarchical view drilling down into any one of these like boxes is a system we need to design right so this is like large scale system design how does the whole system look and then individual boxes are like how does this system work right how does the justice system work how does the vendor escrow system work right so so with that
(02:41) I’ll sort of open it up and to to give you some some time to like think about what kind of questions you would want things I would love to hear would be do you think there’s other systems here that need to be fleshed out do you have clarifying questions about the nature of any of these systems or this system is as a whole do you think there are other kinds of governance systems do you think that there are other kinds of services so you know anybody feel free to start but I’ll you know I’ll give it a little
(03:11) bit of time and just shut the up for a moment all right thanks B something that I personally am pretty interested in is the distinction between the load pipe Dow sort of governance you know because I think that’s very relevant for our like vendor Centric Focus conversations that we’ve had before and then sort of talking about what the the Hamada or let’s just say the generic Marketplace out like what sort of actions they would be taking that I think that’s a a very important distinction to make because
(03:46) that well that impacts a lot of other things like the white paper for example but also yeah like the vendors so if I’m hearing you correctly it’s the distinction between the actions that the protocol out and Market D can take correct yeah and maybe the best way to explore that is looking at the actions a Marketplace Dow would take specifically because I think the load pipe out stuff is pretty pretty clear at like broad Strokes level is for me okay distinctions between actions that the protocol Dow and market Dow can take so
(04:22) I would say that that is again that’s this sort of like and and you know let’s let’s start defining some terms right these are protocol Dow actions okay I need maybe to separate these a little bit more BR this up a bit but in in Broad strokes this category I’m creating here these are all actions that can be taken by the protocol Dow right and so like there’s parameterizing the incentive system and then there’s work flow governance and then there’s executive decisions right kicking a
(05:01) Marketplace and kicking a vendor are the only ones that really come to mind for me but I would be open like does anybody have any ideas of what other actions like the protocol Dow should be able to take and then similarly we can dive into that for for the Hamza da yes maybe I’ll and way I understand it is that protocol doubt provides all the necessary parameters for the rating for for each for the hamsa D to construct or or any other Marketplace DS to construct their rating systems and the decision of how to
(05:41) construct the rating systems and weigh each of the parameters is up to the marketplace Dow is that correct for for one for one instance can you can I’m sorry can you say that one more time is one instance of of a distinction between a a protocol Dow and the Marketplace Dow would be that the marketplace Dow determines how to use the each of the indicators or measures to construct its own rating system for the users like a Federal Reserve do we go back to the Federal Reserve concept is actually in a way well not talking monetary policy or
(06:24) or emissions policy I’m just talking about the the user rating system that the marketplace employ each Marketplace could weigh certain parameters differently you know what I mean I see what you’re saying I think that that this is an interesting distinction because I think what that means is like there there actually is I think a rating score right so like ratings and and I think this is maybe broken into several C categories actually like I think that there’s like there’s actually here there’s
(07:06) Marketplace ratings and you know we’ll have to like expand this at some point but there’s Market Marketplace vendor and judge right so these are all three rating systems which are coordinated at the protocol level right so there’s these these ratings but the rating I think you’re talking about is like the rankings of vendors and that I think it has a lot more do to do with like search in internally in like Hamza right like and search rankings for products based on like the rating of venders and so
(07:47) yeah I I do think that there is like like a Marketplace would have an internal ranking system they could just inherit the ranking from the protocol from the protocol yes yeah or they could Implement their own like ranking algorithm but again that would be how things are populated on the front end not necessarily reflective of like writing to protocol stuff so so I think that that’s like we could even call that like a service down here like you call this like internal vendor ranking now these squares are all
(08:17) piggledy piggledy and maybe even uh buyer ratings B yeah like H how how soon does a buyer release the the escrow to the to the the vendor that that might be a decision that that might be an indicator which different marketplaces use differently some some marketplaces could be more vendor oriented others could be more buyer oriented you know what I mean yeah yeah yeah no I see what you’re saying and I think I think you’re right because I think that that that also probably belongs up here and again please please anyone feedback or
(08:56) thoughts on this would be appreciated could we be including like a reputation it’s just worth noting I just want to say for the record you know as much as I’m an internet marketer SEO kind content marketer I I do agree with you know like these search engines now and web to that don’t disclose their algorithm I think that is something that there should be some privacy or proprietary part to I don’t think it should be fully maybe I’m wrong I don’t know I don’t know how a Twitter’s doing
(09:29) it exactly is Twitter transparent about their algorithm I’ve heard they’ve been more open Twitter’s open source but I think again this is this is like this gets to like the distinction between types of ranking right so like a oh man now these are all different sizes this is going to bother me oh well I’m not going to look at it there’s the internal ranking that like Hamza da and the marketplaces can do and that’s like entirely proprietary but like the protocol rankings should be pretty
(09:55) should be open source and predictable and you know again those come after like like you know so we James outline you know like we we work to outline what a Marketplace reputation what what inputs a Marketplace reputation might have that would all fit within this like Marketplace ratings block and then we have to do similar things for for these for These Guys these guys it’s like a little bit simpler right for buyers it’s like how fast after you receive your item do you rece you know do you get do you release the payment how often are
(10:29) you leaving RS you know there’s not not too many things how often does a vendor dispute you you know how how often do you lose disputes vendors similarly you know like there’s there’s only so many inputs whereas marketplaces are a little bit more complex and judges you know yeah there’s like a whole other system which we haven’t even really gotten to for for judge ratings that I don’t know that we want to dive into but does that answer your question Jose yeah pretty much I understand that
(10:59) the protocol would be open in regards to the data collected by all protocol users vendors and buyers right and it would be up up to each Marketplace to determine its own algorithm for the ranking systems yeah yeah yeah I mean like you know Marketplace probably should display protocol level rate rating but again they you know when it comes to like you know Marketplace UI you could even say that this is like a subset of marketplace UI right M because like Marketplace UI is the service that they’re creating and they’re just you
(11:38) know they’re creating a way to interact with the the products and the protocol and one of those aspects of them creating that UI would maybe being them developing their own internal vendor or buyer rankings yeah so it’s possible as as far as like the data that’s visible right as as I I see it at least like vendors for example could see their their Big R score but they can also see the variables that are contributing to that and so if a Marketplace wants to prioritize one of the smaller variables
(12:11) in terms of how they serve up the information in the UI I don’t think there’s really anything stopping right okay we we want someone with zero complaints someone with zero complaints is ranked higher or we want someone who can deliver quickly someone with like quick delivery you know is higher ex excuse me James could you bring briefly remind me again what the Big R means I mean it’s it’s a user reputation but what what do you understand by Big R yeah user reputation there are some variables that we’ve discussed we we
(12:43) haven’t really flushed them out yet but what we have listed is Big R is a sort of sum of all of the little reputations within a category so estro time is a little r unique by a reputation or unique vendor reputation is a little r dispute time is a little r dispute performance order completion rate Real reviews stickiness score I think is something Mike talked about those are all the little bars and okay yeah they’re put together for that big reputations for so what I was saying basically is that protocol provides all the little RS
(13:23) for marketplaces to determine the formula for their Big R I mean we we could do it that way but I think as a protocol service you know especially because we we anticipate vendors are going to participate in more than one Marketplace you know having a universal rating might be beneficial and you know also serve to make load pipe more relevant I suppose I don’t know what well that could also there could also be a big a default Big R from the protocol levels and most marketplaces would sort of inherit that that same formula right
(14:04) yeah yeah so what what would the like use case or Advantage be in your mind for having a why like would a Marketplace want to set their own localized reputation for vendors well in purse I remember that we had different managements administrations and each Administration would place a different value on sellers or earners and sometimes they they would they would Place more importance on the earners or on the sellers Mo mostly on the mostly on the buyers and that created you know different outcomes so I think I think each
(14:49) Marketplace should be free to determine its own the way they rank users and and and you know the the the importance they give to to buyers and and to sellers in this case you know yeah I would I would okay go ahead go ahead Mike I would second that I think there should be like a protocol Global rating but I do feel we’ve had those focus groups with some sellers like Nina and others and they were concerned as a seller if a new seller comes onto this Marketplace and just Imports all their reviews from other
(15:20) marketplaces they they would like dilute the value and the work that that sellers done on that local Marketplace and there should be a way to maybe allow the you know I think it’s about being transparent really there’s maybe maybe the marketplace could display its own reviews and ratings and it also show global global ratings maybe separately I I like the idea of being granular or allowing it to be different data sets that could be used for indicators yeah I mean I Nothing is Stopping someone running a Marketplace
(15:53) from having their own reputation score and and one of the other things this affects I think is like paid ads right so like for instance if a if a Marketplace wants to allow vendors to pay for like better placement that implies that they have their own like ranking system or at least like they have a way to override you know the the ranking that is written on like a vendor profile that’s on chain right so so like they have a you know they they would have they choose how they’re going to display it it could be that a doubt is
(16:26) just like look we don’t care about anything internal We Care care about like what your reputation is globally written on chain and they just display that another Marketplace may you know could could be like a New York City open mic where they like charge you $5 to to go on right and they’re like oh yeah you have to pay the fee if you want to be ranked at all right may I don’t know if that would be a great vendor experience but like yeah marketplaces can do s of whatever they want and you know what you know the
(16:55) protocol could in turn rank or weigh each each of the different Marketplace big RS based on volume so like say you have H hamsa a hamsa B hamsa C they all have their different rating systems and the user score for that system would be based on on the weight for each Marketplace which is which is in turn based on the volume of that Marketplace yeah so we haven’t talked about this in a long time but in the toonomic model playground document we kind of are or at least we were creating an aggregate emission score which takes
(17:33) into account the reputation as well as the well the revenue that a specific Marketplace is making the idea being is you can have a perfect reputation and not be earning any money and you’re not getting any emissions likewise you could be earning a lot of money and have a reputation score of zero and not be earning any emissions because you’re a bad actor and so you know I think that is one of the load pipe governance level sort of decisions within that I think emissions or what do you what do you have it there
(18:04) I was following you within the incentive system and they would sort of be determining how to value that the the load pipe level incentive system you know so like what were giving you marketplaces and yeah so like these all of these are like functionally like like this Marketplace rating is this Marketplace like this emissions here right that arrow that goes from token Vault to marketplaces is determined by this right so you can even say this sort of like governs this but I’m not going to like but it’s also a
(18:44) service that it provides to the to the Dows right ranking vendors buyers judges so so like yeah it would ex the rating governs this up here it also governs all these emissions at the load pipe level in some sort of like backwards way and governance at this level has the ability to change these formulas more or less but that it’s I think that get a little bit more and just not that we have to chart it here just so we understand what he was saying is that it’s not just the ratings necessarily but it’s also the throughput that a
(19:23) specific Marketplace is having so you know if ham is doing a lot of business right that would also you know be part of the formula the incentive system as well as the ratings well well the rating the rating take of Market places in for instance is like takes into account volume well okay maybe it doesn’t maybe maybe I see what you’re saying maybe maybe there’s like okay so so what you’re saying is that there’s like there’s a rating and there’s a waiting for lack of a better term right like there’s a rating which
(19:57) is like how good you are and there’s a waiting of like how much like how much volume you’re doing but it could there could be other things that determine weight yeah okay simply because yeah we probably don’t want that as someone’s like reputation I mean it could be like a sister variable but yeah then you just have the the very very large organizations being able to achieve maybe a similar reputation score even though they’re doing bad stuff compared to smaller who’s a good actor so there’s
(20:29) a few other data points I don’t know we should probably keep it flexible I don’t think we can get them all but like so so we just Amazon just added something on one of our listings for the bar supplies we had a call Wednesday night not returned often and then his name is Christian he actually I think some of you might know him but he works with other other sellers too not just us he’s contractor and he says his other client actually has frequently returned product on his product listing that they’re very upset
(21:00) about so I mean you know I I just would like to I don’t know I just want to make sure we’re flexible and and and kind of you know elastic on this type of stuff because it’s it’s evolving even at the mature I call Amazon level of these different indicators interesting yeah like product rating is another right because that’s that’s not a vendor a reputation issue that’s like how often is this specific like you could have a great vendor sells a shitty product yeah how much there’s
(21:32) actually seller ratings and there’s product ratings are different in Amazon you as a seller you have a rating and as a product you have a rating most people focus on the products but sellers do have ratings separate yeah so like when a user leaves a negative review for a product does that impact the the vendor of the product or so I the there’s a long discussion on this it’s hard to but basically I I most people I work with and I’m mostly called a private label seller so by being a private label
(22:06) seller it’s called is I’m both the brand and the seller but on some so that would hurt me because I’m the only seller on the listing I don’t want other sellers because it’s my brand so but on like Apple iPhone Apple iPhone yeah that would hurt Apple’s listing but of course of course obviously Apple has so many so much volume but of course a negative product would hurt Apple’s iPhone but they could also they try to get the seller to get a re negative review it’s not as easy or
(22:41) as visible but if you the seller also has a it’s usually related to the delivery the customer support it’s not as commonly seen But the seller has a rating in e in eBay there was all only seller buyers or sellers had ratings the products didn’t cuz were expire you know they were they were expiring yeah the idea of gend shouldn’t sell something you know and I put in chat a few three points I don’t know when to discuss those and I I called out John for one of them chat just now this chat here in the
(23:18) chat of this town hall yeah there we go really know being service I don’t know and John feel free to chime in here if you I know if John saw the chat either what I don’t know if he’s recognized seen that chat but yeah for protocol Services I so like I still am not sure where the relay node sits and I think that’ll become clearer next week John I don’t know if you know feel free to again chime in if you if you do know or have an idea but like the idea of like how do we pay the relay Noe I think
(23:58) we have to keep we’ll have to keep like a like a smart contract funded I think but I but I actually I don’t know I mean I think I’m speaking too far out of I like I I don’t know enough about how we’re going to have to keep that funded to be able to integrate that yeah in in terms of like this diagram right like because it would in my mind it sits somewhere down here and it’s like sort of it’s like a service being operated by the marketplace and so I don’t know but but it could would be a service operated by
(24:30) the Dow by the protocol d i i i i just don’t that I don’t know if everybody’s even following you know this just has been evolving I don’t know does James re following this or we could catch up offline but John do you have any comments on this right I am probably okay great I just I know where there’s a lot of changes or you know things happening I’m not sure if John’s with this but that was one point the other point is I think you you actually gave me that idea oh you’re John do you have
(24:58) a comment on relay nodes being I don’t think those are really important distinctions right now like we’ll we’ll pay the relay no somehow but I don’t think we need to it doesn’t really affect the overall architecture I don’t think on the marketplace so the second point on the chat is the the reward I believe I liked your idea of like rewarding people for finding miscategorized or product ER listing errors in in in in listings I so I think that that is like so I posted a little bit ago in the chatting
(25:35) governance Dow like Channel I believe heading governance down let me see see see yeah I believe you saw it yeah yeah the community notes idea I I think that because like when we were talking about like finding miscategorized products that was back when we understood like the filtering of items were for Dows to happen like at this level and they would like decide that they wanted to like include items that had certain categories right and that was when we like understood the filtering as taking place on a per item
(26:19) level rather than on a vendor level right so I think I think we’ve changed now and now it’s like vendors are the people that we care about and like a can kick a vendor if they’re like posting products that they don’t want and I think the the moderation and filtering like ju Justice should operate like like I I want to think about like that in terms of like the the Twitter Community notes and yeah and so I I think it would roll into like the Justice thing and it would be like you know everybody is I mean like I as much
(26:53) as I hate the idea but like everybody’s a cop and everybody is like you know monitoring the products and if they see you know if you see something say something I don’t know why all of this is like such you know National Security speak but like but yeah anybody can basically report a an item that like shouldn’t be in the store and then like if enough people also agree that it shouldn’t be in the store then it like raises a dispute and the Dow can like kick a vendor is how I think that system
(27:24) I mean should should work of course it does have this yeah I mean it does have the negative but you know even back when I was a kid there was Community Neighborhood Community Watch right and it was like I had I went to some meetings with my parents at at the like somebody’s backyard and the suburbs and we talked about different concerns in our community I that’s healthy but of course you’re thinking the negative now is like the the point system in China with like you report bad people and you disassociate with bad people to to not
(27:53) get your points lowered but I I feel like it’s healthy it’s good that we’re all like policing the system and and ensuring the Integrity of the ecosystem compared to you know I I I think something to not forget here too is you know if everyone is having a problem with how we execute that then they can make a vote with a load pip Dow in order to actually change how we we execute things so you know especially it’s some sort of formula yeah okay yeah no I mean and I think go ahead sir well I was going to say yeah so there’s
(28:31) like justice up here and maybe you know I need to block out of space and just say like a product slash vendor cuz I think that they should be similar systems or at least I think that they should be linked right there’s there’s the dispute process right I guess like the way to think about it is like we have a justice system and we have a and we have police right and in our system judges and police officers are the same thing and and it’s just like you know every citizen right is a is a police officer is a judge and and
(29:11) again I you know recommend anyone going through that post it’s I think it’s a post from vitalik actually where vitalik like goes through and like examines what how how the Twitter Community notes system works and how that ends up with like people making judgments about whether or not something is true false right so it’s it’s like a it’s a way of like gaining reputation for making judgments and arriving at some notion of truth right allowing a citizenry to act as oracles for subjective truth and and
(29:42) so yeah I I I think like like when you what happens is like if enough people get to the point where like in Twitter there would be a or like in X there would be a community note added to a tweet then that would initiate the dispute process right so like you have this accusatory arm like the police arresting people and then you have a justice system to determine whether or not like they actually violated the terms of service or something like that and yeah in my mind I think that’s how it should work I open to other
(30:20) suggestions right so there’s costs and there’s like tokenomics inside that action is there a cost to to is there a cost what do you mean if I if I report a product I have to pay a gas fee or a token fee or something if if it if it’s I mean you shouldn’t have to right I think that’s a that’s maybe somewhere between an account abstraction question and a like you Marketplace UI question well what if I’m what if I’m could I be rewarded for doing it at least you know like if I’m if I’m
(31:02) highlighting products that do violate and I’m keeping the thing well I should be getting rewarded in this ecosystem right yeah and I think that that’s like okay so what is what is like our emissions you know what is this emissions formula for judges and what is like our rating system for judges like I think both of those things factor into that but yeah I mean like I agree right like a judge you know and and some of the like you know the rendered verdict here right like if you participate in a judgment like you should be rewarded for
(31:38) making a correct judgment if you have a high rating you know then you should be getting more you know and and similarly to the way that we’re thinking about marketplaces higher rating plus lots of cases should mean more emissions right so like same same thing here right rating and waiting right there’s like we we need to arrive at a way to rate a judge and my initial like pitch for that is I think I I outline it yeah I outline it a little bit here where you submit if you submit a vote on a judgment then you’re also
(32:17) forced to up vote and down vote the a sample of other verdicts and that is how we like get to a rating of you as a judge is how many people agree or disagree with your verdict in general and then similarly here in that Community notes post like you know that system is like the more notes you leave that other people that that end in like an actual action then the higher your rating goes right and so like yeah a combination of like how many times are you right versus how many times in total are you initiating disputes or are you
(32:52) initiating like moderation actions like yeah I think that’s a way of arriving at some you know you know excuse me is the only way to rate a judge taking into account how many votes agree with with that judge or not that the only way to rate judges I is I I I don’t know if it’s the only way it’s the only way that I’ve thought of again I don’t I don’t know I don’t know what else because again it’s like we’re we’re arriving at like subjective truth and so like other people’s opinion to me is the only thing
(33:28) that you know can really like is any sort of like measurable metric for that like but what else what else would there be oh well that depends on the case being judged but I would say there there are some objective indicators out there like for example in in the case of a product dispute you know there are some indicators that would seem to weigh the the case for a certain actor like for example did the guy sent ship the product yes or no did did did the is there evidence of that the product was delivered yes or no and that sort of
(34:09) thing you know yeah but like if the outcome of the case is like you know like the outcome of a case would involve multiple people rendering judgments right and and so like we can’t yeah it it would be people weighing the evidence and then you know deciding what was right or not like I don’t know that like we can have like if we could build an automated system that just said like oh this is true or this is not and weigh that against what people’s like subjective judgment is on it then that would be a way like to to
(34:51) determine a rating score on an objective basis but when but we can’t build a system that can be sure we we can look at indicators and then hopefully based on those indicators the judges arrive at the right decision but but yeah I mean like I I I don’t know how to measure like we can’t measure against objectivity I think in these in cases like this right yeah you you could take you could take into account both for example the judges would seem to to we to vote one way and the objective the so-called objective indicators would
(35:26) seem to vote in another way so if if the if the circumstantial evidence is different from what the judges from from what the majority of judges are saying then yeah I guess that in that case you could you could penalize a judge who’s not in agreement with that you know and reward judges based on how how well they they they agree with each other I don’t know if I if I’m making myself understood that this is a this is a really abstract idea yeah well because it’s just that like you know I don’t think like I’m
(36:06) sorry was it someone else yeah I I’m just I guess the point I’m getting hung up on is how do you objectively determine what an objective indicator is you know without developing a system as B had said because you know at the end of the day if people are determining what is objectively true then it’s kind of a subjective judgment and the only reason it’s objectively true is because enough people think it is objectively true and so you know if the the end goal is still seeing whether or not judges
(36:43) are voting in the direction that is supported by the objective truth or let’s call it the majority truth you know we still are doing that via an up vote down vote system I think there are other ways that we could do it so it doesn’t just have to be up votes and down votes but you can look at the final verdict of the judge and then look at the resolution of the case and see if the judge voted in a way that agrees with the final verdict as well sorry if I’m rambling here no no well the thing is I don’t know I don’t know if I’m too
(37:14) comfortable with the idea that that judges get penalized for disagreeing with everybody else that that’s that’s basically it in some cases it it could be that it could be that way but in other cases not because because maybe that judge is taking into account is the only one taking into account the objective evidence pointing in a certain way yeah I think we we need to penalize them as you know as much maybe if it’s like a repetitive action CU most of the judges I I don’t think they’re going to
(37:46) have any incentive to be a bad actor so really we are rewarding I suppose in the system the people who come to the conclusion that the majority of people support and that’s not always correct but how do you do it better yeah well but and but I I tend to agree I I don’t know that this is necessarily necessary I’m sorry I don’t know this step is necessary like penalizing incorrect judges because you know we’re creating a game to and we hope that the outcome of that game is to arrive at truth right
(38:25) and we imagine that each of the judges is like sort of an independent actor and and so like a process whereby like a judge submits a verdict and then they’re given like a you know a number of other verdicts and and then they upvote and downvote those verdicts right based on you know how much they agree how persuasive how thoughtful right and and then like maybe there’s another question which is like would you after reading these would you switch your vote right and and then we just don’t you know if we’re not well if
(39:10) we have enough judges it shouldn’t I said if we have enough judges the the judge penalty probably isn’t necessary I I think with this system and perhaps I’m optimistic here but you know if you have a large enough sample size I think it’s going to be a statistical anom for you know a bunch of people to just vote incorrectly one one question are are Judges volunteers or are they are they is it is it like the like a job they’re paid for or are they volunteers both I mean it’s like it’s
(39:46) not a full-time job I think I think it’s it’s one of those things that we incentivize and by to through token emissions and through you know again some rewards Uber Uber Uber drivers it’s airbnbs it’s like yeah you know in a cool way yeah we would just like notify a bunch of people like hey you’ve been selected you could you know like we would give them maybe like a a Justice portal where they could go and they could like make judgments and they could earn money right because we have a we we
(40:23) have a budget to supply it right and we would say hey yeah you just like do this and and leave a judgment and then we’ll pay you maybe they can be penalized that that would kind of make it unattractive for them to become judges if if they can be penalized in the judging I agree and and but it’s like the thing is is like you again you want to create the incentive or like I would like to create the incentive for truth rather than for rightness you know like like I don’t want people like I don’t know it’s it’s just like hard
(41:00) right like how do you create go ahead one thing we can do is like the online survey approach I’m not sure if you you guys ever did something like this in college but you can earn some beer money signing up for these survey sites you fill out some user demographic information and then you know periodically you’re informed that there’s a survey you can take you take the survey you get paid for it and they do at the end of the day behind the scenes weigh all of the survey submissions from all of the different
(41:32) users against each other to see who is like clearly just clicking buttons and not caring and they don’t penalize them financially they penalize them by giving them less surveys so you know we we could do something like that well I mean like I think the the something like that again like the community notes is like how often like like we can correct for Randomness through rating right like if if we trust that at least like some percentage of people are honest than ranking other people’s you know like
(42:10) reviews like that that step here where like you lock in a verdict and then you have to like up vote down vote other verdicts yeah yeah similar to Mechanical Turk right you up vote down vote other people’s verdicts that should that that was like the solution I came up with for like correcting for Randomness and like and and people just like you know especially if there’s no like and the disincentive or like the punishment for being wrong was again also sort of a check against like random judgment right
(42:45) where someone just comes in and and says one side of the other but but yeah I think I think if we if we do it like this and you’re r ranking of verdict is what determines your payout of the reward pool then we don’t need to like penalize a incorrect judgment we can just give more reward to the more correct judgment which is theoretically the verdict that is highest ranked of the correct verdicts correct of course being the like Winning Side not necessarily the true side yeah and again depending on what
(43:29) our sample size is for the amount of Judges it’s the same as to you know flip a coin 50 times you know if you flip it coin 50 times it’s roughly going to be like 25 26 though you you can of course have variance at like lower levels of flips and so if it’s a large amount and we’re assuming some people are just going through and always saying no or always saying yes that should balance out it’s only mechanism for judging judges based on they yeah I mean I I would I would love to hear other ideas on how to arrive at a
(44:04) at a ranking for judges it’s the only it’s the only answer I’ve come up with but but yeah would love to hear another another proposal as to how to do that okay my other point I don’t know if you saw it but the Amazon vineer reviewers I do that’s I guess at a Marketplace sty level but if these reviews are left on a protocol level is a service of the protocol like I do think there is a good value of like quality reviews left by verified re approved re professional reviewers in today’s world yeah I think that falls into like
(44:43) the buyer ratings category and probably uh we don’t have an emissions for buyers make aniss for I do think how does current ranking work of public judges you do you mean like I I would ask like where do you mean and I would I would probably say like it’s how many powerful people do you know that will appoint you to like a a judge ship I mean like yeah if you’re talking about the United States then it’s just like do you know the president did you go to Y were you in Skull and Bones there’s no ranking
(45:24) for judges you in the United States you just get OED to like a level of judgeship right and there’s different requirements for being appointed to different levels of judgeship right so you could be like a circuit court judge County court judge that would require like an appointment I think by like a mayor in Texas there’s an election for judges but there’s no like Global ranking no there’s no system there’s no ranking system for judges the system is like the the system you know like the man man you know it’s like the IV League
(45:57) Like You could argue that maybe the ranking system is like placement in law school and the law school rankings but even then that’s just like sort of a backwards metric of who do you know and then how powerful are those people and what fraternities were you in so that was the other proposal Alex and I’ll come back to I haven’t read Jose or or James’s messages yet but that was the other proposal like that was how we were thinking about it at first was like we have a judge registry and load pipe Dow
(46:34) you know then the executive decision would be like we add a judge to the registry and so they’re managing their judge registry and then like the Hamza da is managing its judge registry and but then it’s like you know it functionally be at that point it just becomes like a workflow right whereas like this community note sort of system allows everyone to be a judge it it doesn’t create this like class of Judges so I guess that’s what it is is like when we’re talking about these like rankings and how do we raate a judge and
(47:14) and and everything like that if we were to create it in that way where the foundation like is appointing a judicial branches appointing judges to judicial seats then that would be like the tradition system where it’s a question of like who do you know you have to apply there’s like a top down process like you only have to keep the people in the governance systems happy and you know they’re the only ones with power to like fire you whereas you know this system is a little bit more decentralized and
(47:45) automated gives everyone like there is no like special class of Judges there’s just anybody who wants to participate and the more they participate and the better they participate the more power they have in that system and and I think that’s why at least I shifted in thinking about it this way yeah what’s up I mean I I I guess you follow a bit the Clara decision making process for this yeah a little bit uh so there’s I think it could fit our purpose in the sense that normally like Jose was saying in in in in
(48:28) Ecommerce transactions there is an average things that can normally go wrong either way and that I imagine it could be put into a smart contract in a way that when anybody who is going to be a judge is going to decide this case they already have like the choice you know things that can be done let’s say if you know the the did the pack was the package Shi yes did it arrive in in good condition yes you know did the parties use an Esra account yes or no and based on this the judges you know basically select what
(49:10) what their decision is you know from from the contract itself like that there’s not that much room there shouldn’t be that much room for decision making you know so for others to judge what the judges are you know why they are de deciding what they are deciding it’s more like I guess we can kind of foresee in a smart contract what can go wrong and then you know the judges basing the evidence they just decide and and as you say if a judge decides against everybody else well probably he would guess he gets negative rating you
(49:47) know that that is I guess the simplest way of of at least starting the system and this is basically the way I think that the people in claros are working right yeah so I mean and that’s that that is sort of what we’re thinking of right is that like so so let’s let’s like talk through a test case right like so a vendor sends an item and let’s say that the truth is that it is delivered and the item is fine but a buyer raises a dispute and says you know like it well like that that the item was
(50:27) like never sh you know there’s like harder cases where it’s like oh the item was broken and you know the person could have broken it themselves but let’s just say the truth is that the person says that it was never delivered but it was and so they raise a dispute and it cost them like maybe a little bit of money to raise this dispute which is what the like sort of bounty is on on raising a dispute I you know I could go either way on that the then there’s like this functionally Discovery period where they
(51:00) present they they like attach their evidence to the case and then the vendor has to be notified that a dispute has been raised and then the assistant asks them to attach their evidence right for the case then once that like Discovery period is done you know we need to determine how long that is then it’s just pushed out and you know again to to to Jose’s Point like CL what is it clar terion right a random sample of all eligible judges which would be vendors buyers Governors you know anyone right we we take a large number and we just
(51:36) send a notification to their profiles and we say Hey you know like you you could take up this case and the Justice You Know Tab they tab over to Justice they look at the evidence provided by the vendor they look at the evidence provided by the buyer they determine like oh it was shipped here’s the shipping you know information like here’s the receipts and everything like that they say like no refund you know buyer should for you know and they say like you and and it would be like who do you yeah maybe there’s like a select
(52:05) list of things that can be done right and and generally judges are like oh no we should re we should not refund them that’s fine and then everybody is required to write like just you know just reasoning right like why did you arrive at the decision that you decided right and then when they submit they lock in their like vote and their reasoning and then before they submit they have to like up Vote or down vote other people’s verdicts right their reasoning as to why right and that allows us to like you know who is
(52:41) consistently putting in verdicts whose verdicts are consistently upvoted or downvoted that allows us to arrive at the reputation for the judges and then as people get more you know good reputation in judgments they’re the ones who are primarily being pushed like cases and so yeah it takes a little while to build up like a set of people who are like committed to judging and then those people end up getting getting like that that’s how the system should work in in my mind in theory right does that make sense so I guess the
(53:14) difference is that the the the let’s say the rating is done by the judges themselves instead of by the code right like I I understand like for example in claros don’t judge each other like whoever judges differ they get downrated right yeah so that and that’s why I think maybe that like punish incorrect judges was there is because that’s how they do it but also the way they do it is they require the judges to put a stake forward so like in order to take a case you have to put money up and and you’re basically like betting on
(53:51) yourself right like to to to render a judgment that is correct and inline with the other judges they also have like smaller you know like judgment tools right and they have like longer periods so like the they they have like yeah long like it’s a smart contract it’s what’s called a a state machine and so it goes through like an innumerable like an innumerable list of states before it like arrives at a judgment but yeah again it’s like in their world they have this class of Judges which puts up money
(54:27) in order to render judgments and and I think like yeah maybe let’s try and be a little bit more like Community notes let’s let anybody participate in the Justice process if they want to and then find a way to arrive at a at a like a rating for people who do good work in in judgment I don’t know if you saw my chat Bo but what requiring people to sometimes participate or even like using as an indicator of their like reputation in the ecosystem potentially even Sal themselves could yeah we could I mean I
(55:00) don’t know I I I think that people should just have the freedom to do what they want to do right like in in a doubt right we should if someone doesn’t want to be a judge I don’t think they should have to be I think they can be select like part of the reason why when I say like randomly selected I mean like say we have like 20,000 people in the system we should randomly select like 5,000 of them and then it’s like when we hit like a 100 people that’s when the the casee is considered close so like we just make
(55:31) a ton of people eligible to to be a judge at for any given case and then we close it when a certain critical number of people render a judgment it it is is sort of my idea and so like in that world we wouldn’t require people to to make judgments but we we would allow people to to make as many judgments as they wanted or they could just like turn it off they could say I don’t want to engage in justice but it would be an Avenue to like earn money in the system okay yeah we’re at the hour Mark James had the Hop I think this has been
(56:05) pretty I think today’s call has been pretty structured how do you feel B yeah I mean I feel good about it I mean like like I said at the beginning you know like we can I mean like we do end up just going into to different and and like into different areas all of these like little boxes need to be like filled out and fleshed out and I think they’re coming together a little bit more you know and maybe we need to start like deciding beforehand like okay this is the Box we’re going to talk about today and you know again any any work
(56:40) anybody wants to do in in these boxes in between please feel free and I know that we’re going to be pretty busy next week but I think we’ll do a little bit more work on it then but yeah I mean you know I don’t know how I mean like yeah I think we figured some things out but yeah we have we still have a lot more work to do yeah I’m just chatting I think Jose maybe could also had some value in between the calls I mean he’s always got great insights here but we can talk about that later any other I should we here or keep
(57:11) going or by the way sorry I just wanted to say like Jose when you link this like Clarion you you link this like a long time ago as well and that was what served as like the basis of my like theorizing about that whole like like uh Community note Justice idea so so first I just want to recognize thank you for that yeah well I think I think Randomness is a great mechanism for ensuring fairness in judgments you know it it’s what levels the playing Ground yeah prevents prevents high level collusion exactly mhm and again I think
(57:54) that that’s part of the drawback of having a class of judges right similar you know to what to what we I was talking about with you Alex like you know that that’s what exists now right the current justice system is a class of Judges there’s no there’s no you know like accountability there’s no there’s no like public accountability except in the case of like Texas judges who are you know who have to have elections right and so if someone doesn’t like what a Texas judge is doing then they go
(58:22) and they lose the election right and that’s happened in the recent past they’ve been campaign because like one I remember in particular was a Texas judge who was like giving measurably harsher sentences to black people in her District I believe and she was voted there was a huge campaign to get her voted out and they did right but generally accountability doesn’t exist within the judicial branch in the United States you can look to the Supreme Court and like you know the Clarence Thomas getting like houses and and jets paid
(58:52) for by like a a billionaire like uh sugar daddy and nothing happens to him there is no accountability in the Supreme Court because you’re just you’re like you’re just appointed and then that’s that’s it and it it mirrors some of the problems I think that we have with with Amazon and so how do we do it differently hopefully like this correct okay all right I yeah unless there’s any other comments questions I think we’re good for today yeah thank you this is great all right thank you
(59:23) guys till next week yeah have great weekend everybody cheers [Music] [Music]