New Ethereum talks, every Monday. The week's conference uploads by event, in your inbox.

Circles 2.0

DevConflictSat, Nov 9, 2024, 07:03 AM · 28:00

No description

Transcript

Cool. Yeah, let's continue. So the day today, so the event is all about credible neutrality. And of course, the first discussion was already about Gnosis Chain as a credible neutral chain or platform. Now I will talk about, personally one of my, maybe my favorite topic, circles.

And framing circles as credible neutral money. So first starting with that reminder about the question of what money actually is. And money here, under Professor Bernard Lieta, Belgian central banker, is first and foremost an agreement within humans, within a community, to use something as a medium of exchange and as that it lies in the same space as let's say marriage or a lease agreement so it is something humans choose to do and of course we absolutely have choices so we can design and arrange our money systems and it's up to us to define rules. So let's just together go through this exercise and just imagine we're sitting on an island, somehow we stranded there, 100 people of us, we somehow ended up on this island and there's nothing. And we think, oh, having money would be a good idea.

And now we need to kind of somehow design that or come up with that. So how would we do it? And how would we convince everyone that that's a thing to join? And of course it only makes sense if all the 100 people agree. If only three or if one guy says, okay, I'm creating here 21,000 or 21 million units and then says to the other 99, that's the money.

The other 99 will unlikely agree with that setup. So the next setup could be okay let's start equal everyone gets everyone equally gets 100 units and that's it. We do this one time transaction but then you could come up in a situation where where new people join this island. Then the question is, okay, do they also get those hundred units? What if they kind of join but they leave already the next day?

Can they still get the full amount? So a good idea might be to say, okay, everyone gets some amount of money equally, but you get it over time. So, first, the rule you might come up with is to say, okay, everyone who participates in this new form of money gets one unit per hour. So, the longer they are in the system, the more they get. But now if you think about just that rule, that essentially means over time the amount of money that is already in the system will of course go towards infinity essentially.

So then those who are later in the system, the amount of money they will get in their one unit per hour will be compared to the rest, will go towards zero essentially, that ratio. So to counter that, the second rule you could come up with is to say, yes, there's constantly a generation of new money at a rate of 1% of one unit per hour per person, but at the same time, you have decay, so old money disappears essentially over time. And here we are suggesting a rate of 7% a year. And you can boil this down. So essentially at the end of every day, you multiply all balances by this number and you get this 7%.

So those two rules already create this kind of curve where that would be the money supply generated by a single person. If you assume they become 80 years old, so in the beginning it's almost linear, the increase, but then the 7% kick in and essentially the supply then reaches pretty much exactly those 120,000. And let's say at 80 years they die, so they stop generating new money and then it kind of disappears at the 7% rate. So with just those two rules, one unit per hour and 7% decay, you will end up in a situation where the total money supply will over time converge or will always converge to exactly 120,804 times the number of people in that money system. Another consequence of those simple rules is that you can think of it as some form of implicit tax rate that is generated by that, where if you hold those 120,000 as magic number, then the 7% you are paying on that money will exactly equal out the new issuance.

So, therefore, your quote-unquote tax rate will exactly be zero. If you have less than that, you have essentially negative tax. You get more than than you pay and if you have more than that as Your balance would go towards infinity the tax rate will approach This seven percent So we have those two rules now To design that money system. We would need to ask okay who can get who can join? What are the rules to join and here we also?

system, we would need to ask, okay, who can get, who can join? What are the rules to join? And here we also give in circles the radical answer, everyone. And there's no one gatekeeping. There's no entity that decides who can get an account, who can start issuing that money.

And that first might sound like a problem because, well, then someone can just create many accounts. But we will solve that, and we are solving that by saying the decision, by saying everyone issues a unique form of their own money. So everyone creates a personal version of their own money and then we will see that people can accept, can choose decentralized whom's money to accept. So concrete person A will issue A money and B can decide to what we call trust person A, meaning they are willing to accept their money, their circles. And a transfer can look like that.

If A wants to transact with C, then C does not necessarily need to trust A. But as long as there is a transitive trust connection, then A can send A circles to B. And because B is trusting those circles, they can send B circles to C. And C is happy with those because they are trusting B. But ultimately, it just tries to solve the problem of who can join this money system and trying to give it a completely decentralized web of trust answer.

So, again, we had our exercise. How would we design money if we could kind of just start from scratch. And we ended up with those three simple rules. So one circle per hour can be issued. There's a 7% demurrage.

So kind of that means, yeah, again, balances are reduced at that rate. And then, yeah, we have this concept of trust connections where everyone can choose whom circles to trust, so who they recognize as real humans or real other participants in this network, which indirectly solves the Sybil attack problem without introducing any centralized gatekeeper. So Circles is a system that actually launched already in October 2020. And here's how this early web of trust then developed. So started just with a handful of people and an hour later it already looked like that.

And a day later it looked already like that. And a couple of weeks later it looked already like that. So we had huge interest at the time in circles. Social media was for a while, for at least for a few days, full of posts kind of about circles and people asking to get trusted and find connections in circles and that was all great. But on the other hand, then after that initial hype, we then also had a long time now where circles kind of circles then in much smaller rounds was kind of where people tried to bring this new form of money already to the world and make it use.

So here a picture of an effort in Berlin, the Kaffee Grundeinkommen. Grundeinkommen is kind of the German word for UBI. And, yeah, farmer's market where you could buy those vegetables was, yeah, for 450 circles. This is, by the way, how a concrete circles transaction looks like. So kind of what I showed was this person A, person B, person C.

In reality, it's then much more complex. So this is here, Martin and me, sending a transitive transfer to this Bauer direct, so kind of this vendor from the last slide selling the vegetables. So they would trust those accounts, and those accounts would trust those accounts and those accounts would trust those. And essentially kind of this is the trust pass to essentially convert my circles into circles that this vendor would accept. So that all happened.

And then there was another larger experiment with roughly 1,000 people using Circles in Bali just a year ago. You can, if you want, read more about that here. So Circles 1 was running for a few years and the question is what did we learn at that time for our upcoming Circles 2.0. So first that the concept of creating fair money definitely resonated around the world in very different settings, very different people all kind of, yeah, were appealed by this concept of this radically of money but also that money requires trust and the 2020 blockchain technology I would say was still too early for that so money that just works 90% of the time is definitely not good money and yeah in practice our I would say this early circle attempt four years ago was hindered by simply technology not being at the stage where it was really ready to use And that is here an important point that Circles in its core design relied only on peer-to-peer connections.

But in reality, there is another form of connections, what we call groups or communities, where it's not just peer-to-peer, not just connections between people, but connections across people, or one-to-n if you want. And that had not really placed in Circles, in its core protocol. And that we will, or we have added to 2.0, and I will talk about that more in a minute. And finally, that if you want to do such a crazy and ambitious project like creating new form of money, ideally you can use it within applications where you already spend or use your money today.

And of course, the hint is here towards many of you might know or might have heard about Gnosis Pay. So again, here already the hint is in our new attempt that I'm introducing, we will integrate Circles tightly to current forms of payment. So your Visa, Gnosis Paycard and stable coins and that will all be usable from the same wallet. So, yeah, this is just giving a brief or short outlook to Circles 2.0 that we are releasing shortly.

And the one thing where I already hinted at was a slide about communities that we are adding is Circle Sets, or previously we also called them groups, where in the Cir core concept again you have per person a different type of money of circles and that is made quasi fungible through those trust connections. If you have the trust connection then you can use the circles interchangeable but they are still different tokens and that makes it very hard to, let's say, put them on an exchange or even on a DEX because you would have, well, in Circles 1, we had over 100,000 accounts, so we had 100,000 different Circles. So you would have 100,000 different marketplaces or markets. Now we are introducing circles sets where you have the ability to create a set of circles and across them create one fungible version so everyone can commit their individual circles to that and create a fungible version. But I just want, again, to give super brief outlook.

So again, Circles launched in October 2020. We had the Berlin experiment. We had experiment in Bali. And now we are very close to, we already did the developer release. So Circles 2.

0, the contracts are already live. If you want to be involved, definitely reach out to us already. Then there are plenty of things you can already do. But the public launch is coming extremely soon. So that means there will be a proper wallet, Metri wallet.

It will work nicely with yeah, with Gnosis Pay and we are trying to give this crazy idea to bring new money to the world and challenge the dominance of the dollar. Another shot. So... Oh, not working? No!

Okay. I see. Okay, can you maybe... Can you..., can you, can you, okay, okay, I mean, just go to aboutcircles.

com. And, I mean, the QR code was linking to some waitlist directly, but on aboutcircles, you will also find that waitlist. All right. So aboutcircles.com is the domain.

So again, well, thanks for listening. And hopefully many will join us in our Circles 2.0 rollout and experiment. Thank you. Thank you, Martin.

We have time for a few questions from the audience before we take a lunch break right here in the yellow shirt. Yes, hello. I have a quick question. Why base circles around the concept of UBI? And the kind of underlying thinking is that if the group of people get money continuously, it doesn't represent any interaction between them.

Whereas, for example, you could base it on debt, meaning I transact with you and I issue my tokens to you when I buy something from you, right? And this would be different from just continuously accruing tokens. Yeah, so I would say it's in the previous in Circles 1.0 we put kind of some emphasis on this UBI in Circles 2.0 we decided against that and said it's basically just the core idea is just fair money and the core idea is basically just to say how would money need to be designed that every participant would want to agree to that.

if you look at things like Bitcoin that also tries to become money, there's certainly, to some extent, value generated by together deciding or there's value kind of projected into the thing that you agree to become money. And basically the answer of Circles is that that value that is generated just by the fact of people together agreeing to use something as money, that that value should be equally distributed across those that agree to use that as money. So I wouldn't even say that UBI is kind of the goal or the start. It's more a consequence of successfully together agreeing to use something as money. Back.

Thank you. I have two questions. One was do you have some insights about how effective was kind of like this civil system mechanism based on the trust that one person had with another and then that other person with another. And the second question is regarding, I forgot what was my second question. Well, let's start with the first one.

JAN-FELIX SCHWARTZMANN- So very effective. So the important thing to note is the goal of, in the circles trust graph, this big graph that I showed there with 100,000 nodes or people, there are definitely many, there are definitely Sybil's accounts, we know that. But the good thing is that the damage those can do is is extreme or is essentially limited to those that are actually trusting them. So let's say I want to create simple accounts and I start to trust five accounts and those five accounts trust another five accounts and a hundred accounts, then I can create this huge amount of accounts here that all trust each other. But as long as they have no trust connection to the actual network, which is over here, or let's say I'm the only connection because I'm trusted by the real network and I trust all the Sybils, then I didn't get any advantage from that because as soon as the Sybils can in the very beginning send Sybil money to me and then Martin coins go into the real network.

But as soon as I run out of Martin coins, then I basically hold only Sybil accounts that are not accepted here. And I didn't get any advantage of creating all those simple accounts. So, again, the summary is this web of trust approach is not a solution. If you just look at circles accounts and say, okay, it's circles accounts, therefore it must be human, then that give the hard guarantee if you only trust, well, non-civil accounts, then you are not in any way affected by civil accounts. Okay, have you dived into how important is also to have some kind of like specific demands to have more velocity in money?

Because, I don't know, if we are 100 developers that are looking to use this as new money, but we don't really have much to trade because we are all developers that do the same thing. Yes, yes, yes. No, for sure. So that's also why in these experiments that we did, both in Berlin and Bali, that was not at all mainly developers. It was really kind of food production, kind of all kind of things.

But clearly that is a challenge and the way we want to address this challenge in our 2.0 approach is by saying you're exactly right, so the question is you are now issuing your own money, but the question is, and now in a, it's your responsibility to back it in some form. And yes, there could be a backing by saying, I am a developer and I do some work here. And I might accept my own circles for that work. So, it is possible for people to back their own circles by offering something.

There is another form that is much easier or that might be much easier for some people is just to put some other assets behind it. So in the simplest case, use some Ether, some SDAI, some whatever, some Bitcoin and create a small automated market maker between that reserve asset and your very own personal circles. And that's actually how we want to start our Circles 2.0 rollout, that we first want to gather a group of people that are willing to back their very own circles by at least $100 worth of some assets they choose, be it Bitcoin, be it Ether, be it, you know, whatever. And then we have first a group of, ideally, we are targeting hopefully 10,000 people and we can create a set circle for those 10,000.

And that, because it simply has liquidity, kind of in the beginning overcomes this problem of is there enough stuff to do to close an economic loop? Because, well, if you can trade it against other currency, you can indirectly trade it against anything. And you will even be able to use it, Circles, with no pay. Because under the hood, then, you have liquidity. But, of course, the actual goal is to have things natively denominated.

And, yeah, use Circles as a medium of exchange and unit of account. Okay, I think we have time for just one more question. Up to you, Martin, who would you like to call? No, sure. Is there a market right now?

Well, right now just for some individual Circles. I think like for a little bit like diet or something. Like Circles and other Circles. Yeah. Well, right now just for some individual circles.

I was going to ask, what can I buy with your coins? Yeah, well, I have already put on various places liquidity for my very own circles. So yeah, you can get indirectly probably any asset you want with my circles. All right, everyone, give it up for Martin. Thank you so much.

Thank you.

Automatic transcript — names and jargon may be misspelled.