Carbon.xyz

Transcript

[Music] thank you all right we continue with our program and now we are gonna dive into token design optimization strategies after we've learned a lot about token launches and the first Speaker actually that I asked to come to this token engineering track day was Mark Richardson from carbon and I told him you know what I would love to discuss carbon as a new primitive in the amm sector and basically a new approach but I also want to learn more about how you designed it

how the design process looked like how you organize the collaboration between researchers smart contract development and the banker Dao as well and yeah I'm very much looking forward to your talk thank you for coming bye bro thank you for the introduction this talk is actually going to dovetail remarkably well with the last one I think that that was by Design so Angela thank you for putting together the speakers in a in a chronological order like that my name is Mark Richardson um I've been with Banco now since 2021 and I've taken over the CEO position for the project since October of last year

with the development of our new product um and so part of this presentation is going to be very highly technical if you don't understand the symbols on the screen just listen to what I'm saying okay I will be interpreting it for you it's important that we start at the beginning okay and I hope this isn't controversial but bancor is one of the most plagiarized protocols in the industry right we invented the amm and we have uh what's called the the original bancor formula which is actually a a big component of our intellectual property and our patent family and the way that it works is essentially that it's a derivative of something called the Douglas Cobb constant production formula which if anyone is familiar with microeconomics

you may have seen that before but here it's applied to token pricing rather than economic productivity the reserve invariant here is just a simple expression that says that the total number of tokens inside a liquidity pool is always a function of the other tokens inside the same liquidity pool the most basic version of this is the constant product formula which was popularized in uni swap V1 and uni swap B2 and countless other Forks of those protocols the important thing here is that in bangkok's original implementation we didn't have any restrictions on what the weights of each one of those tokens could be so you could actually bias the overall portfolio value in the direction of one

token should you want it which was critical when you when we developed the protocol during sort of the Ico era so there were a lot of projects that didn't want to do an Ico via traditional methods and would prefer to do it all on chain and so having a reserve ratio constant that would allow Market participants to interact with the token without crashing the price or without causing it to you know pump to some ridiculous or unbelievable level was critical the this is the graphical representation of it which I hope we've all seen before this is a conic section this is actually a perfectly symmetrical hyperbola and um the important thing here for the purpose of this discussion is that you need a token balance on both axes to

understand where you are in this coordinate grid okay so the um so that brings us to the end of this component the most significant part of it though is that this allows for what's called price Discovery no matter what is happening in the Market at any one time you can query the pool and it will just tell you this is the price of that token right we call this a marginal price this is essentially the price that you will pay if you trade nothing right if you trade zero tokens because as soon as you trade a non-zero amount the price slips right and moves away from you and this is generally you can think of it as a um sort of a a pragmatic understanding of Supply demand curves if anyone has a

strong economics background now that marginal price quote it's still a function of all of the token balances in the pool okay so this is um uh the specific way to calculate it in in bangkok's ecosystem but also in balancer if anyone is familiar with the balancer ecosystem and again the press it has two tokens in it in this case it's got the token weights but when we're also referring to another token balance so here all you're looking at is that this is basically the price of let's say the J token and as the relative proportion of some other token let's call that token I as it starts to dominate the pull the value of the J token goes up red if you've got a lot of buy tokens you not really want to absorb

any more from the market so the the supply demand curve is kind of responding in an intuitive way the special thing here is that that Reserve function that I showed you the invariant is asymptopic at zero in N Dimensions so if you have three tokens you have three asymptotes zero it means that the pool will never run out of liquidity ever it will just keep marking the price up and up and up and up right you can take it down to the last way of a token but the price there will be close to infinite right there will be no more tokens to to buy because no one will pay that price so that gives us price enumeration between 0 and Infinity but then with our V2 formula so this is going back to

um June of 2020 we started to explore the concept of using virtual token reserves maybe you don't want to enumerate the price between 0 and infinity maybe you only want to enumerate prices between near where the market is and a little bit into the future right this is a much better use of capital because why provide liquidity for Bitcoin at a trillion dollars if we're not expecting Bitcoin to get to a trillion dollars anytime soon so this is the um the uh the the proper way to describe um that Arrangement and the way to depict it graphically is essentially to shrink the boundaries um of your typical Reserve right we're literally just pretending the liquidity pool has more tokens than it does and

then just trading with what it has um we coined this Amplified liquidity a notable Market competitor uh ripped off the idea a year later and called it concentrated liquidity but it's exactly the same idea we still have two token balances and this is key right the even though you're trading with virtual token reserves those virtual token reserves predominantly don't exist right there's only a small proportion that actually represent actual tokens that you can trade with so the pricing of this thing is still dependent on those token reserves we haven't really got out of the woods in terms of being able to price something independent of something else so the Amplified marginal price quote looks like this and you can apply

that same sort of heuristic in order to understand how its price changes over time just by shrinking um the the boundaries of your um of your window so here we still have the weighted ratio of now virtual token balances determining the price versus some of the other tokens it doesn't really matter which one that you choose but now instead of enumerating the price between 0 and infinity you can see that the price is actually hitting that boundary right there is a maximum a minimum price here and so we now have this min max sort of price range that we can decide to provide liquidity within so this was Bank of B2 like I said this is 2020 this is a very very long time ago

there is a problem with this when you are assuming that the price of a token should be dependent on the state of another token in your liquidity system it necessarily makes that system dependent on all of the things inside it I know that sounds like a redundant statement but I'm going to show you exactly what I mean in just a second so here we have the price of J whatever that token is versus some balance of I like why should that be true right if you are a market maker or just someone who's placing an order in a system why do we necessarily require that you provide both of those tokens simultaneously because we need one to

press the other it's actually a design Choice coming back from the early 2016-2017 Ico era when Bank Corps was developed that assumed that look that projects would literally provide their own liquidity right if they raise a certain amount of capital they have a a moral imperative to make sure that the market participants that they're interacting with just what we heard in the previous presentation actually have a way to interact with a fair market an Open Marketplace so it would come from them but then when uni swap came out of the gates and they said no no the projects won't do it the users will do it their community members will do it now we're in a slightly different situation because those people that are speculating on a token they're not market makers right ensure they're

making markets but they wouldn't categorize themselves as a market maker right what they're actually doing is speculating I would rather buy a token at a certain price and sell it at a different price but the financial primitive that you get out of this is not bad at all the financial primitive that you get out of this is I will ensure the market against any price movement down or up would you sign that contract in crypto I bet you that the price of eth will not move that is the insurance contract you buy when you provide liquidity on an amm it's fine for a project to do it because like we said they've got these Oodles of tokens pouring out of cupboards that they can't do anything with you may as

well make markets with it but for users those are the tokens that they're speculating on why would you Insight why would you sign that insurance contract so the other thing that it does is it creates exploits does anyone know what a sandwich attack is or you've heard of Sandwich attacks because it's actually derived from this floor the problem is that as the balance of eye changes in One Direction sure it affects the price of J but a sandwich attacker can easily reverse that trade put Jay back into the pool put I back into the pool and move the price back to where it was the problem there is that if I see your trade coming through I'm just going to change the balance make you pay a trillion dollars for that Bitcoin and then background the whole transaction so what we really have in the amm

context and I've got an animation that's going to I think make this a little bit easier to understand is that you have prices of tokens always quoted in the balance of some other token okay and again that's not a design plug given the purpose that we originally envisioned amfs to serve but from a Trader perspective this is insane this is this makes no intuitive sense with the way that they actually want to trade markets so the solution is just let the token price be denominated in its own balance if I have a lot of tokens left I will sell them maybe a little bit cheaper but when I'm running out of tokens let me Mark the price up a little bit and I don't need to refer to another token balance I can do this with one token if

I want to why not the other thing is that it makes that let me go back it makes that trade irreversible you cannot just put tokens back into that liquidity curve if you buy those tokens from me I'm sorry no refunds right if you've made that trade that was the risk that you took I have absolutely no obligation to you the person who accepted my offer to wear additional risk that if you change your mind you should be able to trade it back with me which is exactly the Primitive that amms give the market but again if you are a token project that makes sense because you want your Traders the people that are speculating on that market to feel very safe but if they're trading

against each other it's PVP right it's literally someone's risk versus someone else's risk and if they enter into that transaction with each other there's nothing wrong with that there is no refund requirement so this is what we've developed recently I'm not going to go through all of the math if you want to see it the invention disclosure is on our page there are 300 equations in it but this is the most important one this is the marginal price formula and what you'll notice is that it refers to only one token balance right there it is agnostic it is immune it knows nothing about any of the other tokens associated with this position just going to quickly go through what some of these other terms are a which is the the parameter that we use in the

contracts it's called something else in the white paper but this is basically telling you the relative width um that a market maker is wanting to create a range with right so if you're offering a price between one dollar and two dollars that's a certain width right if it's between one dollar and four dollars that's a different width so that's what that parameter is this B value is just some point of reference this is telling you either the maximum or minimum price where to begin that width at so if someone says my width is four dollars but I want to start at 10 then we know that that means they're making a market between 10 and 14 as opposed to if I start at 50 cents then it's between 450 and um sorry between 50 and 450. so this is just a point of reference that tells you where

that width should be measured from and then this Z value is uh maybe a little bit too advanced to get into in a 45 minute presentation but this is essentially the maximum number of tokens that that position can hold before we need to start adjusting some of these other parameters and so I've noted here that it can be dynamic we actually do implement it dynamically on carbon um but it's um yeah we need to sort of get into the differential geometry a little bit more to really understand what this thing means the other design principle behind carbon is that there is no requirement that two tokens priced versus each other have to have the same price in fact that was unique to amms right if you go to binance today or coinbase right or kucoin or wherever you're trading the

price to buy eth is not the same as the price to sell Leaf right we call that the bid ask spread if you are asking to buy something from someone they are incentivized to charge a higher price than what they could get that if they just sold it to someone else right so if there is a bid for eth at 1800 you might find that the first offer is at 1900 and so even though you know even though someone might say the price of beef is 1800 because that's what they can sell it for it's like yeah but I want to buy it it's like oh well in that case it's 1900 which kind of comes back to this idea that there is no such thing as the price right and this is uh very very closely related to the presentation from before which is that sure you can sell

like 10 of your tokens at a certain price um but once you've exhausted all of that material then um the rest of your tokens might become worthless right it does not make any sense to calculate a market cap by taking some marginal price and multiplying it by the total number of tokens never has there's actually a really cool story about someone in Britain who became the richest person in the world by creating 5 trillion shares of stock valuing it at one dollar and selling one of them to his friend and became a trillionaire overnight and was then arrested but this is important so the way that I know that the notation here is confusing because you think that you should be able to split this thing upside down but remember these are two different functions in two different directions

so what we've invented is what I'm calling asymmetric liquidity and it is something that is much more Trader focused it's something that is innate I think that is more intuitive for real people who aren't categorizing themselves as market makers who aren't beholden to the objectives of the project who are actually just speculating on a token and want to decide the price the specific price that they buy at the specific price that they sell at and don't want to be sandwich attacked or front run or anything else so let's look into um a quick example here so this is the eighth Bitcoin chart a lot of constant product amm again that primitive that you're buying is I will

agree to buy and sell at any price again weird right like who would do that who would say I will buy you nominate the price I will buy it from you right or you tell me what to sell at I will sell it to you that's an amm even uni swap B3 is very close except instead of saying I will buy and sell everywhere instead we'll say I will buy and sell until I run out of liquidity but at any point in this range again you choose the price and I will just agree to that price carbon is different carbon says I will only buy in this part and I will only sell in this part okay so it's in a way it's the opposite Financial product to

UNI swap B3 um I have this comes back to a little bit on the development side of things in order to explain this idea to my smart contract developers it became very necessary to create something that they could sort of interact with something that I could present and show even to our users and investors and things it's not enough to just write a white paper with 300 equations in it and say go and do your homework and come back to me when you understand it you need to actually give them something that they can sink their teeth into and really get that visceral you know like you know it in your bones the same way that you understand driving in your bones but you might not be able to actually explain to someone else how to drive so we have a market back test though at

the moment it's closed Source but if you ask me for it I'll just give it to you and what we're going to do here is we're going to look at the eth BTC chart from this range and we're just choosing some arbitrary Earth portfolio evaluation and we're going to simulate all of uni swap B2 uni swap B3 and carbon across this range so here we're we're choosing a a bitcoin price of 12.99 to 15 eth per BTC and again it's going to trade non-stop throughout that range and for carbon we're going to split it right this has actually been geometrically split so the um on a logarithmic scale the actual width of these boundaries are I think pretty close

so carbon is only going to buy in this part of the market and it's only going to sell on this pocket part of the market the way that the simulator works is you can choose specifically which protocols you want to simulate which animations you want and whether or not you want the the final summary plots so let's look at how a constant product functions for people in a Trad fire Universe they tend to find this representation in the bottom left as being the most intuitive because it resembles an order book and it actually took a few years for people to realize that you could present an amm as a legitimate order book and so you can see that it's got the the the depth chart and as the price kind of wriggles around it doesn't look like it but the

the liquidity is actually moving so because the um we're covering the price from wherever it currently is down to zero and up to Infinity it's basically impossible to um to animate this in any way that would that would make sense so it looks like it's always down the middle now what we're looking at here is as the price moves around we're simulating a 100 efficient Arbitrage this call this um this uh price data is taken I think from the coin gecko or CMC API and then we're monitoring the portfolio versus a huddle position in the top right so when this is in the negative numbers this is generally referred to as impermanent loss and you can think of positive numbers as being impermanent gains but if you are again from the Trad

fire world you might just see this is p l same idea on uni swap V2 it's important to note that how the fee earnings work so when someone trades against you they just kind of leave some residue behind in the pool which means that that pool gets slightly bigger over time your pro router ownership of the pool is the same therefore you're entitled to withdrawal funds so in the bottom right those two those two bars are equal Heights almost all the time because we don't actually know what the fees were paid in all we know is how much bigger the pool was or how much bigger the pool is now compared to how big it used to be and so you can see unison's not doing too bad right if you had actually provided on the fbtc pair over the last year or so you actually might be up one

percent compared to compared to your starting horrible position so not too bad I'm gonna fly through some of these other animations because there's quite a few of them um but you can see that I also animate this in a defy Centric context which is people that aren't used to seeing a depth chart but actually prefer to look at bonding curves and this is the unisaw B2 bonding curve now remember when I was talking about asymmetry and Symmetry and the fact that you're always quoting one token it's press versus the balance of another token I think that that is really well understood from this animation you can see that these things move in tandem right when the when this one's moving up this direction this one's moving the other direction this is because the balance of one token is literally so in the y-axis here is the

x-axis on the other chart this is that symmetry it means that the price is always tied to both token balances and you can never Escape it okay this is the summary so at the end of that whole simulation and if anyone wants to see the whole thing I'm happy to share the the presentation or like I said the simulated code so you can generate it yourself we see that the the blue line here which is the portfolio value whenever it's above the Gold Line you're making money whenever it's below the Gold Line you're you're losing money so this is not too bad I think uni swap did finish with about a um a one percent gain here okay so let's talk about standard concentrated liquidity Dynamics for a second

in this one what you're looking at is again exactly the same price chart but now we're only providing liquidity between those ranges that we nominated at the a couple of slides ago now you can see just how efficient uni swap B3 is right this is concentrated liquidity really looking like it's it's worth something because we're not even a couple of weeks in and we're already at um five you know 0.5 percent higher gains than we were at Uni swap B2 after a year but now look what's happening as the price continues to move out of our range our p l goes crazy right we're now down to negative 15 whereas on uni swap B2 we never even got close to that level of loss and this is why the best way to understand Unity is that it is a

leveraged uni swap E2 your upside is higher but your downside is also higher okay and the depth chart here looks nuts because as you're depleting liquidity from one side of the pool there is no infinite liquidity after it to replenish it okay and this is in a way the exact problem that uni swap B3 has created for token projects that are looking for de facto liquidity is that you can run out of liquidity very quickly when you are for example needing to uh to liquidate some tokens to pay the bills or whatever it happens to be so token price is a lot less reliable in a system like this because it's a little bit more volatile in uni swap V3 we collect the fee tokens

outside of the curve so it's important to realize that the um the uh the heights of these bars on the right don't necessarily have to be the same height whereas they did on V2 okay again quickly on the invariant side of things they still move in tandem they're still paired together they still are symmetric but now you can see it's massively Amplified we no longer have that gentle curve up you know to the upside and to the right Okay so uh how did it fare not too bad so it looks like the hoddle position was worth about 110 and the um the final position was worth about 130 but we can we'll return to that in just a moment for asymmetric liquidity and this is

what carbon is offering you'll notice first that the depth chart doesn't respond the same way that uni swap v3s did as the liquidity here is being depleted the price point that the user has set is not coming down to match it right we are continuing to observe the price bands that the user set so if they said that um I don't want to buy I don't want to sell my Bitcoin for any less than 14 eighth then we won't sell any Bitcoin for 40 until it gets to 14e this is exactly how the system works on the smart contract side this means that it's also very difficult to route and we were actually it seems uh again maybe something of a trendsetter because we gave up on on chain routing very early

on in the development of carbon and said look the only thing that matters is that the user's intentions are on chain and that the execution is on chain everything else can be abstracted into an SDK and even if you don't like our SDK you feel free to build your own right all of the information is there to get started and all of the theory is very well explained in our white papers and now if you're uh here at the conference yesterday uni swap X has just announced that they're also giving up on on-chain routing and doing everything off chain for that side of things so it seems that like at the time I was worried that this is something that the industry was going to reject but it turns out that that was actually totally fine and now the rest of the industry is actually headed that direction anyway

something to note the performance of this position is actually pretty good right and the reason is is that it holds on to tokens longer right the hoddle position is the thing that's hard to beat so if you're always buying and selling at any price point it kind of comes back to this idiom of trade well not often and carbon is exactly that right it is the natural expression of the of the token holder's desire to sell um and buy within the ranges that they choose on the invariant function side you can see with the asymmetry here is also demonstrated I think rather elegantly as the as the liquidity is running out of this curve you can see this curve is growing but it's not

moving okay so the there is no nascent connection between the price of one token and the other here what I'm simulating is what we call connected liquidity or rotating liquidity orders where you can actually say if I'm selling my BTC in this range use the eth that I've sold it for to buy BTC again when it comes back down okay so this is very similar to grid trading if you're in Forex or now it's been in crypto now for about a a year maybe a bit longer but carbon really is you can think of it as an on-chain grid trading bot but a gasless one because all it does is report the prices and the market interacts with it as it wishes so at the end of that simulation you can

see carbon's actually done quite well um it looks like with the Huddle position again started about 110 but we've ended up pretty close to 140 something so the simulator generates these kinds of readouts okay and again this is what the the sort of thing that I was um talking about with my Engineers is this is the data right like when we go to tenderly or when we go to a test net when we're running arbitrary sort of transactions through the system make sure that it replicates exactly this because this is like the de facto truth of our system and sure there might be some accuracy of precision loss or something with respect to um to floating Point versus fixed Point arithmetic but that's something that we can predict right so this is the system

and when you've got something that's so new right and so complex that hasn't really been seen before I think that this level of verification um not on the um you know on the the smart contract side with respect to bugs but the fact that it's even been implemented in a way that matches the economic model is Paramount so if you're doing a DEX or a lending protocol or something from scratch I highly recommend building a simulation like this one because it has massively reduced the develop element time on this protocol it is by far the most difficult the most complex protocol we've ever built and we built it in I think four months shorter time than any of our other protocols and I think that this kind of development process is

definitely a part of that the other thing is that it allowed us to very early on start interacting with people that we thought would use the system especially market makers when we speak to them and we've been speaking with them for years about do you use dexes to use uni swap do you use bancor like what are you doing and it's like no we don't use them and the reason is is that it's prohibitively expensive to manage a position on a decks even uni swap E3 where you can choose your ranges if the market is moving in a way that is let's say against your wishes to constantly withdraw tokens burn that nft put tokens back in the system get a new nft this is a very gas intensive operation and so in this system we made sure that smart

contracts were like you want to change your position short just change the number on your nft that's it there's no need to withdraw there's no need to put tokens back in fact this system is so simple that it doesn't even have liquidity pools right and this means that it has a huge gas efficiency that for the first time means we're really having um let's say productive conversations with market makers who are interested in doing stuff on chain and I would say that it's good timing with all the stuff that's happening with binance with the FTX collapse last year and things I don't know about other people in the room but anecdotally I can tell you that there is now a lot more interest from legitimate players right the the healthy responsible market makers in the space that are actually trying to get off of centralized exchanges and move to a

fully on train system and this is designed with them in mind the um the punch line here is that that ability to make markets asymmetrically and this is the same price chart using a similar range for carbon and uni swap E3 you can see the performance is much much better and it's you know it shouldn't come as any surprise to make markets well it's not automatic right they apply very very sophisticated algorithms about moving their prices around and they use a three hundred thousand dollar a year subscription to the Bloomberg terminal to work out exactly where they where they should put stuff average users don't have that capability and they have day jobs right they're making markets automatically was always meant for

projects was never meant for end users this is the opposite this is built for Traders and coming back to what is your token worth well it's worth what people are willing to pay for it this enumerates all of the people that are speculating on your token and exactly their price points it will tell you this is where the bids are this is where the uh the asks are how much liquidity is there and you can simulate transactions if I interact with this market and take out those orders what will the price be then right it's a much more it's a much more uh detailed way I think of examining a market and ironically it's a much simpler one to to

interrogate at the same time so that's carbon um we have just to show that this is you know out in the wild we recently ran a trading competition this is someone who managed to um to clear seven percent ROI in only 30 days and this is organic right this is no token emissions this is you know we're not paying them to do this this is just hey where did you want to buy your link and where did you want to sell your link and you can see here they set up two discrete limit orders and when the price came down here they bought it and when the price went up there they sold it and they didn't have to pay gas for any transaction other than the one associated with starting that position and if they left that

there it will trade perpetually between those things so I think this is a radically new paradigm and I'm very thankful to have been invited here to share it with everyone and thank you Angela and I'll take your questions and everything else awesome thank you Mark we have some time for questions any questions great presentation Mark um so um quick question what kind of attack vectors does this equation open up to um are they can you repeat that what kind of attack vectors does it open uh your equation out to so you've looked at the sandwich attack based on you know

the previous bonding gaps we've spoken about it what are the vulnerabilities of this existing equation oh well yeah so the the the equation itself is more refactorization so the the fact that we're ignoring one of those token balances mean that you just need to find another reference point so geometrically and algebraically it's actually the same as the bank or V2 formula which is the same as the uni sort B3 formula which is the same as the Trader Joe concentrated liquidity formula so yeah algebraically it's exactly the same so there are no um like vulnerabilities with respect with respect to Smart contract risk or something that haven't already been thoroughly explored elsewhere in the ecosystem um the thing that it might have interest

reduced is obviously we need to choose a reasonable realm of precision and so I'm you know I tried to keep this talk as concise as possible but there is a version of this presentation where we get into once you've done this and you want to compare the token prices of something like ship which is practically worthless and something like wrapped Bitcoin which is you know five figures and then you have to multiply that by the fact that rap Bitcoin is a six decimal token and Shiba Inu is an 18 decimal token you realize very quickly that one of these tokens in terms of the other is worth 30 orders of magnitude more right and so trading it's like asking a Ferrari dealership how many Ferraris can you give me for this

sandwich right like a Ferrari is so much more valuable than a sandwich that it doesn't even make sense to try and buy a fraction of a Ferrari but the opposite is fine how many sandwiches can I get for this Ferrari yeah that I can give you that number with amazing precision um so the way that this formula is implementing you I hope that you saw that it's actually relatively short right there's a single division there's no exponentiation other than the square um we don't need a Taylor series or anything else to solve it there's no numeric approximations it is symbolically provable um and for that reason our Precision loss is very very very low in fact the the all of the data that we need to

um to implement the system is stored in three blocks in memory um so yeah there's there's it's so out in the open and there's so little to interrogate that the the vulnerability and attack surface area is is much much smaller compared to more complex protocols right any more question things somewhere I have a question mark um you I know that Stefan Lush was at your team can you say a little bit more how you set up the team and and who was involved in collecting the requirements coming up with this idea um yeah and of course yeah yeah okay so

um for those who So Bangkok has a chief Economist we always have um the the first Chief Economist of Bangkok is a gentleman named Bernard Lita who was actually the co-creator of the Euro right so we actually we come from fighting stock um uh after uh Bernard later passed away um we had a search around for uh basically someone with a similar background someone who we wanted someone who was a an economics and financial academic right maybe someone with a quantitative Finance background but has applied it to things like the regulatory landscape which is becoming so much more valuable now with the um the chaos that's happened in the United States and

so Stefan Lush is a is exactly that person he was a quad Trader I think he ran the um the options desk on Wall Street for like 30 years or something he's published a huge amount a huge number of books on financial regulation and you know different economic devices and things and so when when I presented this model to him it took a little while it's what I found with carbon early on is that after I had conceived of it it wasn't a short conversation with anyone because I didn't have the simulator I only had half the equations it wasn't something that people got immediately right it was even to to the point where I eventually just started saying forget about liquidity pools forget

about amms it's not an amm this is something else we're doing something else when I finally got Stefan on the same page as me it was really a matter of then going through the let's say the financial definitions of what we're actually building um and uh yeah it's I would say that that was that that was his role then I guess um as a mathematician having someone to check everything that I'm doing but then also he's an extremely experienced programmer so solving the Precision loss issue was something that he was very very key to developing and even now as we approach other Financial professionals he and I are the people that lead those conversations and you

need someone that has that sort of let's say um gift of the gab right when it comes to speaking Financial language with other Financial professionals um so that was a very big component of it so it's not yeah so top to bottom it was uh proof of the math um helping me validate the concept and it even made economic sense and then translating it into let's say correct Financial language and now contextualizing and discussing it in sort of a contemporary way with regard to where the rest of the industry is headed and some final words on the then there was a doubt decision right yeah yeah so everything that we do on chain is uh determined by the bank Hotel so in a way

all I can do is think of things and present them to the Dao and I think we I take this personally very seriously there are um I have noted in other projects where a proposal might be you know a few hundred words and maybe a figure some of the people in my dial might resent me for this but the things that I put up can be 100 to 1000 pages and have like full mathematic proofs and and economic sort of walkthroughs with examples and example code um with carbon I tried to not be as overbearing as that because I did want to get this to Market very very quickly and so instead what the process was is essentially we waited for the

provisional patent to be filed so that we could start speaking about stuff in public um then it was a matter of warming the Dow up to the idea having lots of one-on-one calls public calls Twitter spaces Discord meetings making sure that everyone that's involved in the project who held the token understood it and that they liked it right we can't do anything without their support after that it was the white paper development and then presenting that to them and giving them time to process it and that was a couple of months and then the proposal right so it was really a um trying to keep all Dow members or everyone who wants to be involved

everyone who wants to be informed um getting them access to everything that I'm doing at every stage of the process before there's even a proposal right when the proposal shows up I wanted the Dow to already be familiar with all of the the things that I'm discussing there um and then that also meant that the proposal could be much shorter because we could then say you know link to the white paper at this particular chapter or whatever is required for a certain component of it and it also helped to organize it right there were certain decisions that we needed the Dow to propose you can't just say here's the project and here's the product please approve this there are you know this is going to be the implementation of this

bonding curve yes or no this is going to be the implementation of this monetization aspect of the protocol yes or no this so on right there are about you know I think there were about half a dozen decisions relating to every aspect of the protocol that we needed the data to understand and bringing something this complex on them straight out of the gates and saying look we're going to go to snapshot in a week please read you know it's just not a practical way to do that so it was really about making sure that they understood what was happening from its Inception you know minus the the time it took to get the provisional filed up until the proposal this was a I think a part of the process that we actually do quite well and something that I encourage other other projects to think about

awesome thank you so much for sharing Mark thanks again for being on stage today [Applause] just taking a look iscabe already around