DAPPCON 2025: A Technical Deep Dive Into Circles - Paul Boes

Transcript

Hi guys, everybody. Can you hear me? All right. Yes. Very good. My name is Paul. I'm with the circles team and it is my absolute pleasure to be uh doing this workshop together with you on uh a deep dive into the circles protocol. So um when I told Vanica, our Devril member about my plan, she was like, "Well, you might not want to call this a technical deep dive because people would expect to be building handson stuff using SDKs and everything." So I'm sorry to uh disappoint. This is not going to be a sort of a builder's workshop. Instead, circles is um a great protocol that we've been cooking up for a couple of years, but it's not the most intuitive and simple protocol to wrap your head

around. So this hour is mostly about making sense of the protocol um in terms of you know why why do we think the currency we've designed is sort of a reasonable currency and uh what what what it works towards and I'm just going to keep talking a bit until the technicians have solved the issues with with my slides. Um so maybe in terms of sort of setup um as soon as the slides go up they're going to see a bunch of QR codes. So um the first of them is linking you to uh an an app called menty meter where you can add questions that you have and then everybody else can

vote on these questions. So the idea is so you know I've been told that this is live streamed and having lots of Q&A in between with the mic being handed around can be a bit sort of cumbersome and and then for the people watching online you know they might miss out on on questions plus you know I think it's important there's lots of questions and we want to address those questions first or prioritize those that you know most people find find interesting so I would suggest that questions of sort of immediate understanding in terms of just not getting something I said or something that is on the slides that doesn't make sense to you. I think raise your hands, wait for the mic uh and and then we can address that immediately.

But if there's of bigger questions you have um or things you wanted to always kind of like uh you know ask ask people from the circles team about then um use the QR code uh add them there and then I'll have a look. So uh on top of that I already added two questions that you can vote. One is I think called more detail and the other is called less detail. And this is basically something that serves me to give me a kind of feedback whether I'm moving too fast or whether I'm moving too slow. Right? So feel free to also um you know participate and sort of upvote um if if if either one of these two if

you feel like I'm I'm you know moving in too much detail or not. So this is the QR code in question. But I've you also see this code here at the top, right? 1373 0226. And you can go on menty.com and use this code. Um so the QR code is going to disappear in a sec, but the the the code isn't. All right. So removing this code thing. Also the first time I'm using this, right? um to get hands-on experience it really helps having a circles metric profile. So just to be clear there's circles the protocol and v1 of the circles protocol has been launched in 2020 already and last October we

deployed the version two contracts of circles. So the circles protocol v2 that we're representing today has been live uh already for for half a year or so. But what is only very recently launched is Metri and Metri is a wallet that is sort of the go-to um wallet to interact with the circles protocol. We're going to see that it's not the only way of doing so. There's other tools that we've built mostly, you know, first for internal use and now we're sort of publishing them um to interact with circles. But it's kind of important conceptually to understand that there's this difference between metry the wallet and and circles the protocol.

And yeah, if you don't yet have a metric profile, I mean, who here, not shaming it, who has a who has a radio circles or a metric profile? So, lots of people. Um, that's great. That's great to hear. If you don't have yet have one, um, you know, downstairs we have the booth. Um, after this talk, we can we can invite you and then you can kind of get going with with networking with with people here. Scope of this this this this hour we now have together. Um, I'm one of the co-authors of the circles white paper. So that's going to be something I want to talk about. Um, plus I also want to present some of the tool tools we use to kind of inspect the state of of the circles network and kind of details you don't see when you just use the metric

app. What I won't be talking about is sort of the high level motivation in terms of um, you know why we think why we think the world needs circles. I'm going to kind of touch on that but not really make it the focus. So for the f for the for the purpose of this hour we're just going to assume that circles is a great addition to the world and we're just going to be interested in how it works not not so much why why we need it. And what I'm also not going to be talking so much about is dev tooling in the sense that you know if you wanted to build a DAP on top of of circles how would you go about doing this? We have great docs um at docs.abouts cycles.com and I think one of our tech leads is

going to have a workshop on this tomorrow. Um I don't know at what time. Uh yeah. All right. So to this morning at 11, Martin who is the CEO of Gnosis and also the kind of originator of the idea of circles was going to take do a talk on circles. Um, now that got postponed. So for the sake of getting everybody up to speed, I'm first going to start with a simple recap or sort of high level intro to circles. So it begins by understanding that money is a technology. Uh, it's the world's most ubiquitous protocol. So we use it every day and we think of it the rules as sort

of something that maybe sort of just set in stone or given to us. But that's not the case. It's a social technology uh and uh it's a technology that isn't neutral. If you're in control of deciding who gets to create money under what conditions and and kind of hand that technology to whom, then you can have a lot of impact on terms of real economic development. So we can think about given the the tech stack of the blockchain and all the other great things that we now have uh at the at our fingertips, how should money work um in a way that we think it would be would be fair. So I think

Martin sort of sat down at least in this sort of expost constructed thought experiment. kind of sat down one morning and he thought about how can we build money that works better in a multipolar world where we don't want to necessarily rely on you know uh the US dollar as the main reserve currency but instead we want to have a currency that is underline that is universally accessible in the sense that everybody can go and create their account regardless of you know their passport or anything that splits or that democratizes access to the technology which is money creation and lets everybody benefit from their technology and where also we don't have sort of the first mover's advantage in the sense

that you know the first of two generations of people doing this are going to run away with the bulk of the value and then you know sort of everybody else misses out. So we want to somehow have fair access to the money across time meaning across generations of people and circles is sort of the one answer to this to this design challenge. Um so it's unaligned in the sense that it's it's a completely unpacked fiat token. Um it's not a stable coin and it's not representing the interests of any particular entity. I mean that is to be taken with a caveat. Of course right now is a big sort of stakeholder in the in the in the currency from the point of view of you know kind of funding the development etc. But Konosis isn't hardcoded into the protocol. The the the

contracts are open source out there and they could be deployed on any other chain right now. Right. Um, and we don't have as a team, we don't have any say in what people do with this currency. Um, yeah, it's universally accessible in the sense that everybody can create an account. Um, we don't have a KYC layer or anything like that on top. It's democratized uh in the sense that every user can create their own personal CSC token at a fixed rate of one circle per hour. And fair access across time is ensured by making by having this funny counterintuitive property of demorage which is a kind of daily burn. So all circles circles balances evaporate at a

rate of 7% per year on a daily basis and we can look at how you know how how this leads to this this fair access across time. Um there's an additional layer of complexity to circles. Um and I quickly want to talk about why why it is there. So circle's intention is to enable every person to create one circle per hour unconditionally. Not more not less. And we have this problem where since we does this have a pointer oh but I'm not anyway. Uh oh uh sorry. Yeah. Uh you know we want to allow anybody to create an account. We

want every account to start creating their own money. And we also want sound tokconomics, which is sort of a bit of a mouthful. By this I just mean we want there to be one circle per person per hour. Now, of course, how are we going to deal with civil attacks? How we how are we going to keep people from creating several accounts uh you know, and if if Guinea here created secretly 10 accounts, he would make 10 times as much money every hour as I do. So he could buy 10 times as as much uh coffee uh as as I can. and and so if I know that he's able to do that, then why would I use this currency? So lack of confidence uh in in in the ability of the protocol to produce sound tokconomics in the sense I define here uh would of course be detrimental. So to

safeguard against uh civil attacks or people kind of exploiting this protocol uh or the openness of the protocol, we have a web of trust approach. Um what this means is people um on circles who have an account can trust each other and this is a binary relation that is simply sort of registered on chain and then there's the rule of trust and the rule of trust basically says this if Ellis trusts Bob then anyone in the network holding circles created by Bob circles can swap them at any time for any circles held by Ellis at a rate one to one. So in this example, we have um Ellis here holding her own circles and

Bob and Ellis trusts Bob. And now Carol, who happens to hold um Bob Circles, can go in and without Ellis approving this um kind of sign a transaction that lets her take grab the five Ellis's circles from Alice and in return for five Bob circles resulting in this final state where now Carol has the five Ellis circles and Ellis is left with the five bob circles. And this actually is independent of these being Ellis circles. Like if Ellis held Dave circles or Paul circles, she could still be doing that, right? And so this rule of trust um is interesting in the sense that it lets um it lets uh disconnected circles users um

transact while ensuring that everyone only ever receives circles they trust. So in this example, we have Dave who wants to send Bob five circles, but Bob doesn't trust Dave. So Dave cannot go and take five of his circles and hand them to Bob in exchange for a cup of coffee because Bob's going to say, "No, no, no, no. I don't know you. Who are you? I'm not going to give you anything for your circles." However, Bob knows Ellis and trusts her and Ellis happens to know Carol and trusts her and then Carol knows Dave. So there's this path of trust and this exist the existence of this path of trust together with this rule this swapping rule lets now Dave initiate a sequence of traits where

he'll first get five Carol circles for his Dave circles. He'll then use those five Carol circles to get her his hands on five Ellis circles and then he'll send those five Ellis circles to Bob who now is going to say well Ellis circles that's different. I'm going to give you coffee for for Ellis circles. Right? So this is how the rule of trust enables transitive paths of exchange and the we're going to talk about how this now is is is meant to sort of safeguard against civil attacks in a second or in a couple of minutes actually. Um but just to kind of close this brief overview, there's an additional feature that um we probably want to talk about as well which is

groups. So groups was weren't weren't a thing in the V1 protocol. And groups are essentially accounts that don't create their own circles at a fixed rate of one circle per hour, but they have members. So they can make other circles accounts members. And then there's a mechanism for how these members or people who hold those members circles can exchange them can kind of turn them into group circles. And we think groups are a really interesting mechanism and vehicle. Um, but I'll talk about them in depth uh later.

Right, eight questions already. I'm going to check the you know the Can somebody tell me whether there's a lot of votes for move faster, move slower so I don't have to mess with the Okay, take your time. I mean, if anybody maybe Sean, you can tell me if if people are screaming, I should be speeding up or something. All right. So, this is the beginning of the deep dive. Right now, we just s of we're just sort of swimming at the top learning learning how to swim. Now, we're kind of going deep. So, um and to do so, I kind of want to structure this in in in in two sort of stages. And by the way, I plan to talk in total like 35 minutes. So, I want to leave 25 minutes or so for Q&A and discussion. Right? I'm not going to be

talking like this for an hour. by by six votes. Okay. Right. Let me just create some extra slides. I was I was expecting the opposite, but yeah. Um okay, then uh I'll take more time and um yeah, I I don't really have an actionable way to I'm just going to talk more talk a little bit more. Uh but so yeah, here here's more detail. There's the macro level and there's the micro level. really helps me to think about this this apparent complexity of of of circles and break it down.

So the macro level is the level at which we assume that all people that have a circles account are honest. There's no sibils, right? And we're interested in the kind of macroeconomics of circles. Circles has kind of counterintuitive rules, right? There's demorage, there's constant issuance. looks very different from money as we know it. So how would a world look like in which there was this circles currency and we don't want to deal with the web of trust. We don't want to deal with symbols. We just ignore all of that. So that's the macro level and here we kind of just see users who have different circles balances and we assume that they're sort of fully

liquid with each other. And then the micro level is the one where where we kind of drop that assumption and we now start differentiating between the different kind of kinds of circles they hold and who created them. And we also drop the assumption that all accounts are kind of honest or you know kind of use the protocol as intended and this is going to be sort of the micro level where we deal with the web of trust and uh pathfinding and so on. So right let's start with the macro level. So assumption no malicious parties are fake accounts. And the first thing we can think about is well how would the balance of a single account look like over time right? Because

there's two forces in circles. There's the force of me creating one circle every hour and then there's the force of 7% of these circles evaporating over time. So the the way this looks like is that if I started creating circles uh at on day week day zero and I just did this every day every hour I kind of you know used my metry account or in fact we didn't talk about metry fees but if there were no metry fees uh I would use metry to create circles and then over time initially my one circle per hour rate dominates these 7%. So I kind of quickly increase uh my my circles. Um and then over time the those 7% kick in because I have more. So you know I lose

more. So because those 7% of more is more I guess. And so uh there's this magical number of 120,84 where these two forces exactly cancel out. And so if after 80 years I was to die and to stop pushing that button, this is how the total supply of circles I created would look like, right? And well, why is that why is that interesting?

Um well first of all circles has this super simplistic base monetary policy which is that the total supply of CC in a group always converges to that magic number times the number of people in the economy right so we kind of naturally scale the supply of um the currency with a number of people we don't have an interest rate we don't have you know a lot of the sophistication of of the Fed But um we have this um another way to look at this is that oh I guess the vote now deals with this but there of an effective tax rate. So if you have exactly 120k 804 circles

then you know your balance just remains unchanged whereas if you held more then your circles balance would reduce like in the absence of transaction and economic activity right and so this is if you're paying that tax although I mean I don't really like calling it a tax because it's not nobody receives that tax it's not like metriognosis gets the circles you burn they're just literally burnt whereas is if you just start out then you have this negative tax because well you get a lot more um than than you started out with.

And um here's why um we think that the combination of this steady issuance rate together with um the 7% demirage leads to fair access to money supply. So there are two types of fair access, right? as the fair access at any given time which is just to say that there's no nobody has a monopoly over money creation like the creation of the new supply of money is completely um distributed uniformly across the whole user base right so it's sort of fair in that sense I guess depending on your notion of fairness but I think that sounds pretty pretty fair to me and then there's this more

interesting fairness over time and to appreciate that consider a user that joins uh an existing group of N people like you know we we now have a circles economy and we all start creating our circles like crazy and then in two years um somebody comes and they consider joining this group just like lots of people might today consider buying their first Bitcoin and we want to give them a good reason to join, right? Um, and with Bitcoin, I mean, you know, I'm not going to hate on Bitcoin, whatever, but like the idea is it's not very attractive to to join. Now, if you look at the distribution of the existing

supply when there's so few people that hold so much of the the the the supply of Bitcoin that's ever going to be around, then they are in a position to go into any shop and pay whatever they want for, you know, this cup of coffee. So, do I want to be part of a society where my ability to buy coffee is is basically dependent on other people's desires? Well, I guess no is the answer. Of course, I don't I won't let them do that to me. Instead, I'm going to use circles because the combination of demorrage of this burn and constant issuance says that if somebody joins later, then the

share of the circles that they have created that they contributed as a function like as the share in the total supply is going to look like this where capital n here is the size of the group. So maybe like 50 in in the the example of this room. And this gamma constant is a constant that kind of goes to zero exponentially quickly in in the number of years that they've participated. And so one overn is obviously their share in the in the group, right? So this means if I start whenever I start, if circles was to come become the next global currency, if it's just a tiny

experiment, it doesn't matter at all. If I start joining, I have a guarantee that I will start building up a significant share of the total supply like a significant share of the total supply of circles is going to be under my name. It's going to be ones I have created and they of course go into my pocket and I can invest them. I can spend them but nobody else can decide what the purchasing power of me is going to be in that society. So that's something that's what we mean by fair access to money supply over time. Oh yeah, I have a slide on this.

All right. And there's of course in the white paper we talk about time varying populations. And of course that doesn't change the gist of of the statement. Another thing I want to talk about at the I think I need to speed up at the micro level is the value of senorage. Um and what does that mean? So, Circus 31 was kind of marketed as as a UBI and lots of people think of of Circus as a kind of UBI, which of course makes sense because we we kind of airdrop uh money into everybody's pockets.

But, um the sad truth is that the value of the circles that you're going to be creating over a month is probably not going to pay what you'd want the UBI to cover in terms of your monthly expenses. it's going to be a much smaller share of this. So senorage here is basically defined as I mean it's typically defined as the revenue that the government makes from the issuance of currency. It's this this sort of value creation out of thin air example where if you if you print a dollar note but you only paid 1 cent for it then you have sort of 99 99 99% profits right 99 cents profit um and that's the senurage and of course you can have senurage also when you hand out credits so this isn't really about uh

cash only and in circuits of course the senurage is with everybody and we democratize access to money creation technology so we also uh democratize and distribute the benefits of that. So the question is well what do we expect the value of the c create circles created by an individual to be there are two approaches to answering this question. Um, one has been conducted by by Martin on in a blog post where he basically uh looked at you know different economies um and he looked at the current total money supply the M2 supply he looked at the population and the GDP and he basically said well if we have so

and so many million people and we have take this times 1284 right this magic number Um, and then we look at the total money supply. Then we can kind of like take the ratio and then depending on stuff like velocity, we can cook up a ho hypothetical value of one circle in equilibrium. It's like if circles was to take over the world um and completely replace uh the USD um what would kind of the circle like what would one circle have to be valued at to kind of cover the same um GDP? And so this is like you know these these numbers here range wildly. So they're like 30 cents like 50

cents 50 cents. So this is of gives you roughly an intuition right where where you would expect uh circles to land at if it was to become really successful. And this is not like 100k right per circle. Uh this is clearly very low but much higher than it is now. I think today circles trade at like one and a half to two cents. And there's another approach um which basically well looks at looks at um the the fraction of the value of circles created at at a time t divided by the demand. Uh and so it kind of just looks at you know kind of what if this uh currency was at equilibrium. What if

there was enough circles to cover the demand? Um blah blah blah. You can look at up at the dep appendix and then you get to this number here where this is the velocity of money which we can set to one. Um um and if we look at like the euro for example and so what basically this says uh if I if I translate it is that we would expect the total value of the circles you create per month to constitute roughly 7 to 10% of your demand for money every every month.

Right? So if you have monthly expenses of€2,000 euros that that on average, you know, it takes 7% of that and that should be roughly the the the the the value. So these these two approaches, they kind of roughly align. They they're not completely don't give you completely different numbers. So maybe the value the like in today's prices, you'd be looking at two$200 to $400 um of purchasing power. um in o over the course of a month. So circles is not a fullyfledged UBI, right? Because for UBI, we'd want maybe 50% 40% of of your expenses to be covered or even more. Uh

I'm not an expert on UBI actually, but um they should constitute a significant source of income. And those circles that you create, you could of course spend them on stuff, right? But you can also start investing them because there's going to be financial products built on circles. And so the the hope is that you know this this this democratization of money creation technology could lead to um sort of allow mass access to interest bearing products with the hope of kind of distributing the benefits of participating in a kind of a growing economy um among sort of more uh more of the population. And of course, it could also kill sort of extreme poverty,

right? Where if if people that were living on the street or something, um I mean $200 a month of things they could buy, that's a very significant uh amount. All right, there's a bunch more things that we cover in the in the macro level, but I guess you all you're all here for the graphs, so you don't really care. So, let's move let's move on to the micro level. Um, so yeah, now we kind of turn into sort of switch on paranoia mode and we don't trust anybody here unless they sell me an ice cream or something. I don't know. Um, and uh yeah, let's maybe look at this uh

15 questions. Okay. Um yeah so first things first um the possible flow of circles in this graph depends on both the trust graph and the current balances. So when we usually introduce circles we always make it seem as if there's just this path of trust and then everybody can pay each other along the path. But that's that's simplified depiction um of reality. um it really what matters is who holds which which circles and only then only if you give me the trust graph and the current balances um can I calculate you know what's the possible movement the flow of circles uh in the economy and that is simply because of this word any here right so I

told you that it doesn't matter whether Alice holds Ellis circles or Dave circles or Paul circles the rule of trust lets Carol take any circle she might hold so we we need to know exactly which circle she holds to calculate the implications of this rule of trust. Um yeah so usually we say um you know the state consists of kind of the trust graph and the the the matrix of balances if you will and then we can think about the st state transitions that are possible and these state transitions you always have to kind of think of relative to the ser right so clearly I can initiate or trigger state transitions that others cannot because I can send my my circles to Shawn but Shawn cannot

send my circles to him right so state transitions are always relative to to the party that initiates them. But um you know I'm not going to go into a lot of the of the technical setup and the formalization. What is important is that transacting in circles boils down to the following problem. What is the maximum amount of circles that the sender can pull together by applying this rule of trust right that are trusted by the receiver. That's what you that's the sort of transferable trusted balance. So I might hold 10,000 circles. If I can only use this rule of trust uh to get 10 circles that Sundep trusts, then I can

only pay him 10 circles. Doesn't matter how many circles I hold. And we can actually map this problem to uh a well-known combinatorial optimization problem which is called the max flow problem. Uh and so um you know H Google is the our internal uh you know maestro of uh max flow problems uh and and when whenever you pull out your metry account and you send circles around Hugo's algorithms are crunching the the numbers for you. The max flow problem is actually it's a complicated problem but it's not an NP hard problem.

So I think uh it's it's kind of famously um polomial in the sense that it scales the complexity scales quadratically in the size of the instance. And I think two years ago Google published something where they actually have sort of a quasi linear time algorithm for max flow. So max flow is not I mean if it was an NP hard problem we should just throw circles out of the window because then you know we couldn't scale just from the combinatorial nature of it. So luckily it isn't and and uh we have that and let's talk about more of this if you if you're interested.

Yeah. Okay. Um so now we're at the point where we can um talk about how this web of trust leads to civil resistance. And so now um using this mapping of the so sending circles around to a max flow problem and then using the famous max flow min cut theorem which you can look up on Wikipedia we can basically uh make the following kind of statement which we call relative stable resistance.

So let there be some account of some set of accounts that are controlled by a malicious party and let f be a set of non like accounts that are fooled into trusting these malicious accounts. So they're fooled to trust people that they shouldn't have trusted and that we shouldn't let into the economy and then let R be the remainder of the network. Then if these malicious accounts initially hold only their own funds, right, which is always going to be the case initially because they can only create their own funds, then the transferable trusted balance, like the amount of circles that the malicious accounts can transfer into the remainder is limited by the amount

of circles that are trusted by accounts in R that are held by accounts in F, whatever. Like uh if you look at the white paper, I'm not going to introduce notation here, but basically what we're saying is that this is the thing we want to have very small. This is the amount of circles that malicious people can pull together and spend in the honest part of the economy. So ideally we want this to be zero and we can bound this by this thing which is independent. I mean all you have to know is that m isn't in this formula. This is now uh the amount of circles that the fooled people hold that are trusted by R by the rest. So

what that means graphically is that here we have the malicious people. Here we have the fooled people and we here we have the remainder and the remainder is exposed to the malicious accounts only to the extent that these accounts who sort of have been fooled hold circles that they trust. And so this is of the the the the damage is independent of how many circles the malicious people hold. It's only limited by the size of the leak. And importantly this leak will kind of close itself because if there is if these people hold significant resources like circles that they trust then they

can kind of use that but this that that will reduce like the damage is sort of bound by what they initially hold and it s of leaks close themselves. So that's our relative simple resistance property. Um I guess if we had a perfect proof of personhood and we wouldn't need the web of trust then you know we we might be even safer but of course it's also difficult to build. I mean I mean we know that there's a lot of other protocols that are also uh trying to build sort of a proof of personhood that are that are struggling and and because this is this is s of the achilles heio of the protocol is if people fall prey for malicious accounts then we have a problem at circles and you know that's why we we trust in the wisdom of the

crowd to not not fall prey that's sort of an important point okay before I finish um emergence of global fungeability So fundamentally we not only want to exclude civils, we also want the honest part of the um network to be liquid with each other, right? So I mean so if we're going to want two things, we want the honest people not to have exposure to the bad people and we want the good people to be able to pay each other, right? If I have a super simple resistant network, but every time I want to buy coffee, I cannot because they don't I don't there's no path. um then I also haven't built a very successful currency protocol. So we can measure this by the average spendable fraction

like if if if we have this honest part of the network uh you know how much are people going to be able to pay each other relative to their current circles balances. Um and so we can look at different graphs and the kind of average like average spendable fraction that we get here. And so if we had for example um a complete graph where everybody trusted everybody else then it's it's easy to see and kind of intuitive that independent of the balances of these people they can all pay each other right if if any one of them is offering services for anybody else they can kind of interact like they do with euros

right like always 100% of the circles they hold they can spend with others because there's always going to be paths but importantly like crucially for circles the same holds true, albeit less robustly so in a setup where there's where everybody's connected with everybody else and there's always a path. Um, and and in this case, you can also have complete average spendable average spendability, but it might sort of die down like so you you might be able to do this once, but you might not be able to keep it up. Um, and so yeah, I'm kind of I'm speeding up a bit because I want to leave time for

questions, but uh, you know, the the the end goal or like one goal is of course for circles to become sort of a big global currency where there'll just be one price for circles. Um, and we can all forget about this microlevel detail of who created what because the we we trust in the web of trust doing its thing uh with respect to keeping the sibles out and we just sort of enjoy the ice cream we can buy and we don't really have to deal with the the complexity and that will depend on people's behavior uh and the connectivity properties of the trust graph and so whether that will happen will depend on user behavior but the hope is that kind of the trust network

inherits the small world properties of social networks, right, which kind of famously connect have high connectivity despite being sort of pretty sparse and um you know that is to be seen that that depends and I don't want to sort of spend too much time here. There is one slide I have here on exchange rates and arbitrage that I people find arbitrage for some reason always very exciting. I don't really know why. Um but uh um of course all these circles everybody creates their own circles. Um and you know you can create a market for them.

Nobody can stop you from doing that. So there's an obvious question you know all of us have different circles. You know what do we know about the um the equilibrium like the the fair prices of of these different um tokens that we now have on the dexes. And I think there's a simple um statement here to kind of develop an intuition for this which is that for any state in which every account holds some of their own currency, right? And that's pretty likely because well you you create one every hour um or you can then exchange rate in the absence of arbitrage reflect the trust connections between users. So if n trusts n prime then the price of um

n prime the one that's being trusted has to be at least as high as the one of the one that you trust right so um trusting everybody like I mean this is a bit of a sad statement but trusting other people can never increase your exchange your s of your equilibrium exchange rate but being trusted can also never decrease it right Um, if it was different, then we'd be also talking about sort of a rubbish currency because I could build a bot that trusts everybody and then just lift my price, right? That obviously can't can't work. So, we at uh in the circle theme, we build arbitrage bots that are constantly looking out for for these opportunities and try to keep all the

all the prices in check and if we have time, we can we can look into that. Oh, man. I don't have yeah I'm just uh so we have this really exciting feature called groups and this is a depiction of the trust mechanism but let me just finish here and then if you really want to talk about groups we can talk about them but I kind of really want to uh uh start looking at the tools um I already talk way longer than I wanted. So there's of course metry which is the app we are building that I think many of you are using to interface with uh the protocol but there are a bunch of other tools and I want to um kind of highlight um some of these here. Um so the circles

explorer is kind of the explorer kind of like you know it from fromnosis scan or e ether scan. Um the circles core app is an app that we built for um admin uh admins of groups and organizations to deal with u members. It lets you create new groups. Um and the circles tools is a bunch of informal um tools that we use most internally but um that all of you are also you know kind of invited to check out. This is a QR code for the circles tools. And yeah, I mean my my plan was to kind of like accompany a lot of my talking with real exploration, but I think now is a good time to actually just start looking at

the questions. Um, more detail. Why would you back your circles? Yeah, very good. Um, can I order them by see all questions? Okay, I think these are already ordered. So take a deep breath and uh go through them. So um do all of you know what backing your circles means? Okay. Who doesn't know what backing your circles means? Okay. Okay. Good. Um, so backing a circle is not part of the core protocol, but it's something that we offer people to do on on metric and it

basically means investing $100 and uh um Oh yeah. Uh and and uh okay uh and now so workshop dummy Ellis is not is not a backer yet, but uh I for example am. So, it lets you it it it lets you back your individual circles with 100 USD. And what we do is we create a pool uh on on balancer um with 48 of your circles against $100.

Um well, I mean, I think the the main reason for um backing your circles is, I guess, ideological, right? I mean s of it's it's it's a way of signaling that you support the protocol and and uh you um buy into the idea of circles. Um when you back you get a um a nice backers badge. Um and there's a group of backers that um is exclusively available to you. Also on the Metry app, you uh only when you back will you be able to swap your circles. So for examples here um this uh circles don't show here. But if I if I sort of log out

and I log in with um my my main code. So all of these circles accounts, they're clearly marked as test accounts or dummy accounts if they're not from uh from ABC. I have this I have this this back badge here. Um so you now because I'm I'm a backer here uh when I go on the wallet I can swap circles it shows them here and um that's I mean I there might be sort of economical reasons for doing it but primarily right now we we think that just like a central bank has a reserve

um we now kind of democratize the central bank when we split it up and some of the people sort of constitute the the the reserve And by backing you contribute to you know the the reserve of circles. Um and of course the more people contribute the better the more more of an argument we have when we go out to people. Um so if if I was to uh go go somewhere and say please accept my circles um then being able to tell them that they can now if they wanted to sell those circles and turn them into cents or euros um on the market. That's a great that's a great you know argument to give somebody. Um right. So yeah.

Um what's to stop flight of capital when people get nearer to that amount where someone might prefer to have a currency which is or even more funible but decays at a slower rate like regular fiat. Uh I don't I don't understand the question. Uh can you swappy? Um no no I mean um you mean another circles currency or USD?

Yeah. Mhm. Mhm. But how would that how would that keep you from suffering from demarrage? Yeah. Ah, sorry. Yeah, the microphone. I forgot about the microphones. Um, well, I mean, so all that demarrage is is that you know, all these circles nominally disappear. And so, of course, the total supply of these circles affects the price. But I don't think by I mean so of the the the prices of circuits on the market reflect that supply. So I don't see how I just don't see a specific

strategy for how you would be able to deal with with um demorrage or like trick it. Um but maybe I'm missing something. I think it would be reasonable. Sorry. It would be reasonable once you reach the plateau to keep your circle level below. That's the idea behind this question. Yeah. So that you never lose more circles than your daily amount you gain. I don't we are 80 years in in the future now but but once we we get there I think many people will do that. Sure. So you're saying that we in this world everybody has 12 804 circles

and now they they they sell death circles change for something else. Yes. And then they get them back and then you sort of and then they they uh buy them again and then I mean if if you have a sort of very specific I I I don't see how this would let you gain an advantage. I I don't see this arbitrage opportunity in doing that because of course you know I mean the demarrage the prices factor in the existence of demarrage. So, I would be curious to see a specific strategy to do that. Um, I don't see it off the top of my head. There's a question down here.

And is is the balance uh in the liquidity pool of circles also demoraged or just the in the wallet? Yes and no. So, what happens I mean the circles you create, they're all ESC 1155 tokens. and the the contracts apply the demirage in the sense that if you check somebody's balance they factor in you know the number of days that have passed and then they subtract that um on the market we have 20 tokens so you first of wrap your 1155 into 20 tokens and then those are the ones that you buy and sell and um that exchange like when you wrap them that's not a onetoone uh exchange right And so and of course this 20's

token the balances they are static right so the in the pool your tokens don't suffer from demorrage but the way it works is that the um kind of if if I if I have I guess if I have a 100 circles and I wrap them into ESC20 and then a year later I would unwrap them then they would only turn out to be 93 ES 1155 circles. Um, so I I quickly I mean I'm going to I'm going to look at the questions, but sorry for talking so much. I wanted to to quickly also make sure we I at least show you these these these um these tools um because we only have 10

minutes left. So these this is the first of the three links I showed you and these are these of sort of informal circles tools. So you can for example have a group checker where if you if you type in um you know you can look for any group you can put it in here and that look let's look at for example at the at the Dvcon group so maybe let's kind of get rid of the other complexity and so this we every group has an address and owner a fee collection um this group yeah probably a price should pop up in a second but I can um yeah I can kind of check the price of this uh token right now and I can see the address um and I see a bunch of helper organizations that

that are also helpful and in particular this group actually has a has a membership condition so there's sort of a smart module that um checks that people who are added to this group are actually human avatars only right so no group or organization could be added to the group and we enforce this with this with this membership condition and now I can kind of swap um I don't know Like I mean the the this is not a very liquid group unfortunately. Oh yeah. But uh if I was to for example um go back and we have this metric core group for example here. Where is it? The metric core group. Then um this is more liquid and I can kind of like see what I could get. So this is the um group

checker. There's a group creator. There's a tool for group management. Um there's a profile checker. Um, some things that are really quite quite quite nice. Uh, for example, this um a a Pathfinder Pro app here, which is a way to um look at these paths. So, let's see if um sorry. So, this is Martin's account. I'm going to look for my my main circles account here. So if Martin wanted to send me, you know, a million circles or something, then um oh well, we can look at the graph here,

right? So this this is why I consider Google a magician because he kind of like while standing there uh immediately tells me what the what the graph what the path should be like um between us and then I can actually say look I actually want to be paid only in my in in my tokens and this is going to reduce the graph and so this is a very helpful tool to let you deal with these these paths and actually you can then um execute them directly. So this is this app is going to give you also the copy the call data and I could kind of like go and and execute this transaction directly and Hugo's edit a bunch of very very fancy stuff um in terms of path statistics. So this is really really

interesting um app as well. Um, we have an equilibration analysis which is kind of like a tool that looks at um how well the arbitrage bots that we're running um are at equilibrating um the prices. Um and so uh this is not a not a good sign. Uh I mean we we don't use red for good. We use red we use red for bad. And so but uh I always can just kind of like uh make a bunch of things green. uh but not that green. Um so um right there's a bunch of a bunch of interesting tools here to help people. Also if you have a group and you want to start putting

liquidity against it, this is one thing you could do. Um then of course there's the explorer um which uh you know I I can kind of like uh give it my circles account and then it's it lets me um inspect my trust relations and um you know what I've been doing. and my trust graph and um app.about circles is interesting because if if I if I have groups um or organizations, I can yeah manage them from here. So for example, we have the Dcon group, I happen to be one of the admins for the group. So I could now kind of add people to um the DPCON group. If any of you is

keen on being added, you know, let me know. Uh we can talk about it. Uh no, you all qualify. Uh and uh yeah, so sorry I just went quickly wanted to tell you about these things because I wanted to make an interactive workshop where we constantly use these tools to like explore the beauty of circles and I ended up just talking a lot. Um let's move back to the questions. Um all right, we've seen some cases where circles get stuck along the trust graph. Why does this happen and how can you avoid it? So this is a real problem.

Um we have had I mean I guess um we've suffered from this as well. I think the mental model I have of this process is if you have a circles network where there's very where there's sort of points of high demand uh then they kind of suck in circles into that direction. Um and this might lead to a situation where you genuinely don't you know people have your circles and you can't get them back. Um, and so people have been shocked to find out, you know, where have my circles gone? I mean, they all of a sudden they're very very um they care a lot about their circles. Uh, when two days ago they didn't care about, you know, creating a metric account for months, but all of a sudden it's very important to them to have

their circles back. Um, and this is to be expected. I mean, this is the implication of the rule of trust, right? We go around and if we trust everybody, then lots of people can take our circles from us. And if that is not a bilateral decision, then you know, we won't get our circles back. So, think twice before you um uh uh uh trust somebody. Um only trust people who you'd be happy to kind of like have a lot of their circles in your in your wallet. But that said, um it's not just an implication of the protocol because we've only had um a bunch of small groups that we sell offers for in circles and only very few people in the network have been members

of this group, right? We have like a couple of thousand members but only a couple of hundred of them had been in this group. The people the members of this group suffered disproportionately from of loss of their own circles because their circles were in a lot of demand. So we have a job as the team to kind of be mindful about how we distribute demand um for for different kind of circles across the network. Um but this is the effect that you've you've seen um two minutes. I mean uh I would probably suggest that those of you who because we don't have right I think with I I feel like going me going through this is not going to be the right right thing to do. So, let me just kind of

basically stop here. Um, and those of you who still want to discuss questions with me, let's find a spot where there's oxygen and um, you know, discuss them. I'm I'm really happy to. Uh, I don't want to sort of shy away from from any any tough questions in there. Um, but I think it's it's important that we finish in time and give everybody uh, their chance to check out all the other cool stuff that's happening. So, uh, please um, create a metric account if you haven't yet. Please run around, look for look for trustworthy people to trust. Um, get in touch with me or anybody from the circles team if you have questions still. Um, Marty's going to give a talk tomorrow. Uh, Daniel, uh, raise your hand maybe. Daniel, he's our tech lead.

He's going to give a workshop at the at the on the sponsor like at the you know where the all the booths are in the corner I think tomorrow on all of the things I didn't touch on like SDK building stuff. There's going to be sort of a civil hunt. So that's very exciting.