 And that's one of the conditions under which we do this because we are completely open source consortium and we operate in a completely open manner. The other two things we have to say, one is of course the court of conduct, which says that we are bound by the court of conduct, which says that you can disagree with people but do not be disagreeable when doing so and do not hog the floor and do not, you know, denigrate people. The details can be found in the wiki page. The other is of course the antitrust provisions such as they are right now, which basically prevents price collusion between companies participating in such fora. And that's about it and obviously bankor is well known to everybody and Mark is going to talk about as he announced the institutionalization of mm, which is on everybody's lips both from a regulatory standpoint, as well as from a disruptor standpoint in the sense that the when will the disruptors become disrupted themselves or when will the disruptors join the mainstream and start taking advantage of all of us. So with that I yield the floor to Mark. Thank you. Thank you very much. I have a presentation. So I'm going to share my screen right now if that's okay. Can everybody see the title slide. Yes. Wonderful. Yes. Okay, so, yeah, thank you for the introduction. Yeah, my name is Mark Richardson I'm the head of research at bank or a little bit about myself I was actually a research scientist for 14 years my my field was actually organic synthesis. So I didn't come from a traditional economics or financial background. Obviously in sort of synthetic pharmaceutical compounds nano structures and other nanomaterials total synthesis that sort of thing. It was only during the, the pandemic in 2020 that my career was was pretty thoroughly disrupted I was I was supposed to visit Berlin and complete a about a year of work at the Max Planck Institute for colloids and surfaces in Potsdam. And so when that didn't happen it wiped out, you know, in some about about two years of work and cause one of my research grants to become rescinded. And so I immediately started finding other ways to spend my time. I became extremely interested in in blockchain technologies and in particular defy which I thought had the best product fit for for blockchain in general. And this was during the 2020 defy summer period. I was researching becoming an active community member in a bunch of different defy protocols, including uni swap. I was there for the sushi swap vampire attack which is extremely interesting time. I was really interested in Kyber network, but eventually I kind of settled down on on bank or because of the types of conversations that were happening in that community were very different from what was happening everywhere else. It was a lot more fundamentals based it was. I thought much more collaborative and much less. You know, much less, let's say toxic, I think is one of the ways that I would describe some of the stuff that's happening in cryptocurrency. And so as a community member there I started to help guide the, the establishment of the bank or doubt. And then, you know, within a few months I was collaborating with some of the bank or founders on on new financial products, including an interest free loan system that we call the bank of what and shortly after that, the bank or foundation reached out to me with an offer for full time employment and seeing as the pandemic in Australia is still ongoing and I'm that the borders have not yet opened up. So even if I had my, my funding from the German government still I still wouldn't be able to go. So I decided that I should really just do a hard pivot into into defy. And so what I'm going to be talking to you guys about today is where I sort of see bank or as competitive advantage with when it comes to institutional level product offerings, but also sort of the lie of the land and also a general introduction to not just bank or business in general. But I'm happy to let Kerti sort of moderate this. If you feel like you want to ask a question as we go, please feel free to just interrupt me that this presentation is kind of an amalgamation of a couple of different presentations that I had. And so it is a little on the long side and I apologize for that in advance. So if you need to interrupt me please feel free to just interrupt me. So when I asked Kerti what the, the assumed knowledge should be for this audience. He said, somewhere on a range of one to 10 I should assume a three or four in terms of expertise in in a man technology. So with that in mind, I have prepared a couple of introductory slides featuring familiar faces characters from the Simpsons and that sort of thing. And this is Martin my favorite character from the Simpsons and we're going to assume that he is a defy native. Okay, there are two cryptocurrency tokens that he wants to participate in defy with. And in general, there are sort of only a couple of different primitives that are available other the spaces advancing quickly and there is becoming more and more sophisticated ways of participating and earning money. Providing liquidity and making markets is still one of the most profitable and certainly one of the most popular. And so in order to make a market with an automatic market maker. It is fair to assume that the status quo is that you need to provide at least two tokens, although protocols such as balancer allows for multi token bonding curves and other things. So when you are providing liquidity it's important to realize that you need to combine your tokens in an appropriate ratio, and that ratio is determined by their price that you know you might be able to to fetch for these tokens on on secondary markets like Coinbase. So if Martin was to provide liquidity with both of these tokens. You would need to create a set of them such that however many of token aid that he is providing liquidity with is equal in value to the number of be tokens that he's providing liquidity with. So in this case if a $64 you could provide 125 of those and then 80 of token be that's worth $100 so that both piles are worth 8000. Okay, so that first row again is the single token price that second row is the total number of tokens that he's providing liquidity with, and then that last row was just the token values. So if this was Martin's liquidity position you could say that it's worth $16,000. So just to think about what what Martin would actually be doing these things if he wants to participate in an automatic market making system is here if you're providing that to a smart contract or a series of smart contracts that we call a liquidity pool. Okay, so you may have heard that that term before it's actually been around since our ICO in 2017 when we introduced the term. The pro rata share represent representation that Martin receives has changed his name a couple of times it used to be called a relay token and then it was called a smart token and now we just call it a full token. So, what's important to realize about liquidity pools is that while Martin and the rest of us will know the price of these tokens. So the liquidity pool has no idea it doesn't even know what price is really. It only knows how many tokens there are in the pool and that will tell it how it should balance them when it's performing in an exchange. So prices is an abstract concept, you know, in general that's where liquidity pool it's completely obscure. There actually is no price quoting in a liquidity pool. So what we have is what's called a constant function bonding curve, and the the most basic kind is what we call the constant product model, where the number of tokens contained inside the pool, multiplied together is always going to be equal to some unchanging number so in this case I chose the numbers 125 and 80 so that that constant is some easy number to remember like 10,000. So, if Martin provides liquidity with these tokens, he can then start making markets with them. And this is completely passive it's not something that he has to actively manage it's not something that's overseen. It's just automatically executed by the by the blockchain and in this case the Ethereum virtual machine. So if we have a trader who is interested in exchanging between the tokens that Martin has provided the trading with, for example, selling v tokens in this case and purchasing a tokens, the liquidity pools bonding curve will determine exactly what rate of exchange the trader is going to enjoy. So we can imagine a situation where our trader wants to purchase 10 a tokens. And so the question is, how many v tokens is she parting with. It's a relatively simple algebra to solve. We know that by the end of this process the pool will have depleted by 10 of the eight opens because this is what the trader is taking away. And then the unknown quantity is just how many be tokens is she providing. We know that at the end of this process it still has to be equal to 10,000. And so it's a relatively simple rearrangement to calculate what this number is. And compare that to the starting condition. So if it's going to be 86.9 at the end of the transaction, and it was 80, then we can tell that it was 6.9 tokens that were sold. So that's the end of that calculation and that's exactly what the smart contracts are actually doing. And so the, the trader in this case would be providing those 6.9 tokens and then receiving 10 a tokens from the pool. And so this would be what we consider to be a fair trade. Okay, this means that the, the constant product formula is being perfectly obeyed that it also means that Martin is not receiving any, any value from this transaction right he's basically losing, losing a tokens and gaining B tokens. And this is especially important if the a token is gaining value and the B tokens losing value on secondary markets, then this might not necessarily be a transaction that Martin is always comfortable making. And so we need to make sure that there, Martin has some incentive to participate in these kinds of smart contracts. We need to make sure that it's not a perfectly fair trade that on top of whatever is required to satisfy that constant product that the trader also pays a commission or what's become known in the industries as a pool fee. It's generally somewhere in that 0.1 to 1% range for that transaction. And so if there are a very large number of transactions happening on that pool, then Martin's passive revenue can actually be quite high. The numbers in the sort of 25 to 40% ranges is not unheard of, and that's partially because of the lack of intermediaries right because you're dealing exclusively with the trader who's trading with you directly. There isn't these, you know, middlemen that are siphoning value outside of that system. And I think that that's what blockchain in general are very good at is connecting the value provided directly to the customer. So that's kind of the general two token constant product AMM. The bankor is slightly different in that you cannot use any two tokens that you want. One of the tokens has to be the BNT token. Now at first that might seem a little jarring like that could be an inconvenience that liquidity providers need to purchase the BNT token but I'm going to show you in just a couple of slides that this one, it provides a huge benefit to the network overall in the sense that there is always a common exchange base token, but two, with the release of bank or version 2.1, it's actually no longer required that all liquidity providers personally have to own BNT, just that all pools have to learn it. Okay, so why have BNT or why have the requirement to be in all pools. In the case that we saw where we had a trader just swapping between A and B, that's fine. If we know that that's the only market that's going to exist for, for example, the A token. Then imagine now that we're in a more complicated system where there might be a bunch of different tokens. Imagine then that the trader wants to swap the A token now instead of for B tokens but for C tokens. The same as BNT is in every single pool, then you can always perform a trade between any two assets via that BNT is as a conduit right as an intermediate transaction. So the swap from A to C would occur first from A to BNT, and then BNT to the C token. And this is true no matter how many pools are a part of the system. The bank or protocol currently has something like 200 pools in it. And you can swap then between any of these tokens using BNT as the intermediate exchange. And so BNT kind of comes a utility token right it's the it's the thing that actually powers the bank or network, and its tokenomics was developed under the, you know, with explicit guidance from FINMA. So that we can continue to classify BNT as a non security. And this has a huge amount of advantages as we move into, you know, what, I think a lot of protocols and defy for them is kind of uncharted waters. But you know bank or as a slightly older protocol, and we've been very smart about how how the protocol and its tokens were developed, such that we're not sort of encroaching on anything that would be the most regulatory stress. So that I just wanted to point that out that in terms of institutional adoption BNT is actually already in a really good position. And within the within Switzerland we already have a collection of private banks that allow BNT to be bought and traded directly from their customers internet banking user interface, and you can actually take BNT in the bank or protocol from these internet banking interfaces. Right so it was kind of built with with institutional adoption in mind. Okay, so to understand sort of, you know what some of the more significant changes in the bank or ecosystem have been. The first thing that we have to do is go back to April of 2020 with the launch of bank or version two, well AMMs are really great. And you know I do think that they are disrupting a lot of the traditional, you know, sophisticated professional market maker space. They're, they're far from perfect. And we kind of identified back when we first announced versions who was being released, but the four major pain points are with AMM design currently. And two of them, the most prominent two are certainly the, what we call impermanence loss and I'll explain a little bit about what impermanence loss is shortly. And the, the requirement to maintain exposure to multiple assets. So as I said before, it is inconvenient to ask liquidity providers to purchase some amount of BNT in order to support their liquidity. Especially when we're dealing with other token teams who as for now let's call them a stand in for institutions. They often either have some limitation acting on them where they actually can't purchase BNT for some reason or they can't sell you know half their token treasury to buy BNT. They would rather just use the tokens that they have. It's also true with, you know, some existing defy, well, sort of see defy applications that are built on top of bank or such as Celsius. They use the bank or pools to provide the credit and earn yields for their, for their customers but also crypto.com uses the bank or protocol in order to earn it to you. In both of those cases for Celsius and for crypto.com. They can only use the, the tokens that their customers have provided them with right and they can't sort of behind the scenes start swapping out customers funds for for other tokens that their customers didn't want to be exposed to. And so the ability to to provide liquidity with just one token without having to exchange for something else and provide liquidity with both is something that we thought was a huge bottleneck and something that absolutely had to be resolved before institutions would start taking a seriously. The last two I'm going to talk about a little bit later in the presentation this is capital and efficiency, and this has become something of the meme I would say over the last six months, especially with the launcher of things like curve v2 and unisport v3. But again, we'll talk about that shortly, and then the opportunity cost providing liquidity as well. So let's talk about the bank or version two, which is a now retired version of the protocol. It was developed exclusively to deal with impermanent loss and capital efficiency using the same mechanism. And so if you consider this hyperbolic curve, this is the constant product bonding curve and the same thing that bank always using back in 2017 to perform. To perform exchanges between any two tokens. The problem is is that when you're trading on any part of it. All of the other parts of the curve aren't being used. So if you're providing for example a million dollars in liquidity to one of these things, it could be that no trade is actually using any more than $10,000 worth of it. And so that means the other $990,000 of liquidity is effectively idle right and it could be put to better use somewhere else. So that's the capital and efficiency problem. What we decided to do was to essentially artificially, mathematically, synthetically blow up that that hyperbola such that it no longer was contained within the X and the Y axis, and then allow this curve to sort of move around under the influence of a chain link price protocol. So we call this liquidity amplification we're using 20x, but this is the same idea that has gone into curvy to an unisort b3. So how this worked was essentially as traders would stop moving one of the assets away from peg. This would essentially cause that the pool to become imbalanced or you can treat it as becoming imbalanced. The chain link price oracle would then sort of detect that imbalance and then move this fulcrum if you like, in order to bring the liquidity back into range. And this was a dynamic process that happened in every block. So as traders continued to move the liquidity in and out of the curve, the chain link price oracle would continue to keep the liquidity in range. 100% automatic, every block this was being updated. And so what this was good for was one, it was extremely capital efficient. And two, it also protected the liquidity providers against a permanent loss, because the pool is always trading at exactly the market rate, right or exactly the rate that you're getting on Coinbase or Binance or anything else. So why did we have to retire this model? Okay, it's an important point. And I think that it's something that a lot of our competitors have neglected to study because the same flaw that we had is also contained inside some of their product offerings at present. If you think about how these kinds of strategies would be implemented, right, ours was completely automatic, but there are several sort of manual strategies that are being developed now with yield aggregators and other sophisticated intermediaries that are trying to use products like Uniswap V3 for precisely what we were trying to do with Bankor version 2. You're basically always looking at the present block, right? There's the mem pool and a certain number of transactions are going to be included in whatever block is about to be mined. And you are using basically the past data to influence that decision. Okay. So if the, for example, if the liquidity is moving out of range, it's only moving out of range with reference to the past. The problem is that arbitrageous and miners, they don't just, they can't just look at the future. They can determine it. So if the liquidity is moving out of range, you might think that they have an incentive, for example, to rebalance it or perform an arbitrage trade in order to extract a small amount of value from the pool. But what we found was that you actually provide them with an alternative way to extract even more value. And that's to throw the pool even further off kilt in order to force the oracle to rebalance the liquidity in such a way that you can then perform a second transaction at the end after the rebalancing that causes even more of an arbitrage incentive at the end. So really, it's the adversarial nature of Ethereum that meant that this model was was pretty unsustainable and I'm going to show you some data later on in the, in the presentation that strongly suggests that this is happening on on some of on some of our competitors products right now. So we had to retire version two and we immediately followed it up with version 2.1. So during that version two, let's call it an experiment. We had a lot of opportunity to discuss with, you know, with both our community and also, you know, these kinds of these braver institutional influences. So one of the, you know, the market makers on on Coinbase, for example, we were in close touch with them. We were talking to people that were running sophisticated arbitrage butts and things. And, you know, DeFi was really starting to find its legs. And so when we asked, what was the most important thing about version two to you right what was the what's the one feature that you wouldn't want to part with, if we had to move to a different protocol. It was the impermanent loss. A lot of the, a lot of the people that are have become assimilated into the bank or community are the ones that have been, you know, taken advantage of or at least they feel they've been taking advantage of on some of the more AMM protocols like like Uniswap version two and sushi swap. And also, you know, bank or version one, right, we invented the AMM and so we kind of invented IL at the same time. And so this was really the, the feature that we felt that we had to maintain, especially when we're talking to institutions, because they kind of want the assurance that if they're providing money to the protocol that they can, that they have no risk of leaving with less money provided. It also struck a chord with the chain link community. Okay. And I think that, you know, the way that they have flocked to to to bank or and the value that they see in what bank or is doing. It also is, it has direct parallels with the kinds of institutions that we're also speaking to. So for example, it's kind of a meme that we're going to make it means that the chain like community members in general have absolute conviction that the token that they have chosen will be the thing that outperforms the market. Okay, so they don't want to entertain the idea that they should sell part of their chain link to buy something else in order to provide liquidity with it. So the idea of that is, is, you know, disgusting to them. And then the other thing is that they, you know, they had a strong desire to have as much of that token as possible. Right, that it's, it's not just a trading game for them. It's much more about sort of that wealth accrual that long term accumulation of a certain asset. So in traditional liquidity protocols, you can't really do either of those things right you actually are asking people that they have to either hold less of the asset that they prefer or sell some of it in order to buy something else to provide liquidity with that. And on bank or we say that actually you can just provide the one token. I'll explain how that works right now. So remember I said that on bank or we have bnt is being the base token and all of the pools. That's still true. But it no longer has to be provided by the same liquidity provider. So you can have a second person who is extremely enthusiastic about one particular token, and they can enter effectively into a partnership with the bnt token holder. This is advantageous because that means that Tony in this case, he gets to maintain 100% exposure to the asset that he wanted just like the chain that community wants to. Whereas the bnt holder, they get to speculate on what assets are going to have the highest velocity, which things are going to have the biggest trading volumes. And it's usually the case, you know, volatility is itself volatile right. And, you know, AMMs are kind of anti fragile in the sense that you want to have your tokens in the pool that has the highest volatility, because that means it's processing the largest trade volumes and collecting the largest commissions. So it's not necessarily the case that you want to be in a pool that with something that just has, you know, a highly valued asset or something. You want to be in a pool that has a lot of, you know, a lot of trading volume associated with it because that means that your profits are highest. So this relationship means that as a bnt holder you now get the flexibility to move your bnt around in order to extract the most value from anywhere in the protocol where where you find it. But it also means that the other side of the pool that the token provider has the freedom from having to sell some of the tokens or or hold less tokens than they would otherwise like to. So I'd say that this is a commensal relationship or a, you know, a collaboration and that's one of the things that I really appreciate about that core. So it's not always the case though that there are enough bnt community members in order to support all of the Tony's of the world. You know, I we've got something like a third of a billion dollars just in chain link liquidity right now about the same amount in Ethereum, there just isn't enough bnt in the in the world and support that. And so we needed to make sure that there was a way around that. What we developed was a system where the bnt supply becomes elastic, and where the protocol can supply its own bnt. And so this is bnt that is minted out of nothingness so it's an inflationary aspect of the protocol. And then that bnt is then contributed alongside Tony's tokens. What's interesting about this is that it means that the protocol is earning its own revenue. And of course the minting of new tokens. It does bring an inflationary element to tokenomics and so there have to be deflationary mechanisms in place to counteract it. And in this case, one of the most pronounced, the most pronounced deflationary effect comes into a comes into play when Tony removes his tokens from the protocol. When Tony is removing the token, the protocol then removes its bnt and destroys it right so the bnt was created just for this liquidity provision event, and then it's destroyed at the end. However, even though the minted bnt is destroyed, all of the bnt revenue that was generated which is bnt that is kind of sucked in from external markets over the course of the trading. The protocol gets to keep that. And so it's an interesting question then. If the protocol is earning all of this money, and it's earning a lot of money now right if the protocol is now the largest bnt holder in the world. What is it going to spend its money on right this is valued that it is actually, you know, a crude for itself. And this doesn't go into a community treasury. This isn't something you know that the bank or protocol is completely funded out of its ICO. We've established a Swiss nonprofit that, you know, keeps the lights on pays the legal fees and pays for development. So the protocol isn't doesn't actually need money in order to support itself. And so the question of what it's going to spend its money on is extremely important and I will be coming back to it very shortly. To understand what it spends its money on first we need to think about what happens when the people that aren't a part of the bank or protocol are providing liquidity elsewhere in the ecosystem. So for example that the meme is that chain link is usually outperforming everything else, which means that you are kind of in an option straddle between the two tokens that you're providing liquidity with. So as your chain link in this case is is moving up versus the other token you provided liquidity with, you're actually selling that chain link off automatically but through the an arbitrage process and accumulating the other thing. And so this is why we say, you know, impermanent loss results in broken hearts, because in general when people are providing liquidity, it's with the intention to accumulate more of that particular token. And so as that token starts to outperform the market, they end up with less of it. And so as a token project, for example, you are kind of asking your most devout followers to sort of short their own token in the sense that they are going to end up with less of it or, you know, they are going to want to in order to be the most profitable pair with another token that is likely to perform as well as the thing that they're providing liquidity with. And so this results in these kinds of charts, we can see the collected fees blue line, but the actual potential losses of these are red and green lines. And so often, you know, especially in the retail space, people don't actually, I don't think they realize precisely what kind of losses they're making, but it's a lot of effort to go to to make less money than doing nothing. Right, that's what these charts are being compared to. Okay, so on bank or because we offer the single single asset exposure you actually get to keep 100% exposure to just that asset. And so on bank or the relative profits of this orange line. And so this, this difference right this chasm is it's not insignificant it's like a 10 to 15% difference between providing liquidity on on bank or versus anyone else. And that single data point. And so this is real data by the way this is, I think I scraped this from some sushi swap but it could come from any of the, any of the major events out there. This one feature is why one of the largest cryptocurrency market makers that in the world actually provides liquidity on bank or that arbitrage is uni swap. Right, so they don't actually trade where they provide liquidity, because they don't want to be exposed to this type of effect. Okay, so let's understand where that comes from right why is it possible to lose money because it doesn't seem like that should be true. Remember that we set up the situation with Martin where he's providing 125 of the eight token 80 of the be token. And that was because they were both worth $8,000. So the question is, what happens if one of those tokens changes its price immediately after Martin can commits it to the liquidity pool. So let's say that, you know, and this is an outlandish at all. Let's say that the eight, the eight token doubles in price, right, immediately after Martin provides liquidity with it that this happens, you know, that's just a regular Tuesday for cryptocurrency it seems. Now we're at $128 per a token, but the number of tokens that Martin provided hasn't changed still at 125. So now, to Martin's mind, he's provided $16,000 of a and $8,000 of B. And so that means a total of $24,000 of value he's contributed to the pool. So remember, the pool doesn't know what prices are. All it knows is how many of each token there are and that their, the numbers multiplied together has to be equal to some constant, right, which in the example that we, we looked at before was 10,000. So my only conclusion is that Martin is going to start selling the token that is moving up in price at a pretty profound discount until the position on the bonding curve starts to match what the secondary market, secondary markets are quoting for a. So I'm going to skip over some of the math because it's a lot less trivial than some of the other things we've been looking at here but if you want to ask me about it later please feel free to contact me I can show you how to do it. There's going to be a different kind of market participant now, who is going to try and extract more value from Martin than he meant to part with. And so the a token is going up in value, this is the one you're going to be buying, and you're going to be selling the thing that either isn't moving or is moving down. So Martin in this case is going to be the person who's performing this process, and he's going to sell I think 33, yeah 33.17 e tokens, and he's going to buy 36 a tokens. And so if we apply those to Martin's apparent balances, you can see that Martin is ending up with $11,000 on each side. That's a total of about 22,600, which is less than the $24,000 that he thought that he had. And so this difference of $1,000 is what we call impermanent loss. It's called impermanent, because if the price ever comes back the other way, then Nelson will actually sell those tokens that he took back into the pool to bring it back into balance again. So it's called impermanent because he might not have lost this money forever, but more often than not it's lost forever. He also made a little bit of fees though as Nelson performed this trade Nelson still had to pay, still had to pay the commission or the pool fee, but this $33 is pretty pale in comparison to the $1,300 that the Martin's losing. So this is an arbitrage cost and it's completely inescapable. It's easy to think that Nelson is doing something wrong here, but just like arbitrage in traditional finance, there's nothing fundamentally wrong about it. In a sense it actually keeps the market healthy. It means that Martin's liquidity pool is always buying and selling assets at a rate that is commensurate with what you would expect to get on Coinbase or Binance or something like that. So remember this question of what is the protocol going to spend its money on and it's making a lot of money. Is it uses it after it's calculated what Martin's impermanent loss is, it then uses that money to refund Martin. So Martin is not exposed to impermanent loss when he's providing liquidity on backcourt, and it means that at the time that he withdraws the protocol basically sections of funds and gives it back to the liquidity provider so that they get the full exposure that they were expecting. And yeah, this is the this is the thing that we have found has really struck a chord with with institutional players. So is it some sort of insurance product. Yeah, so yeah, being a bank or is an insurance product. It is both it's currently both a dex and an insurance company. And we're hoping to it to grow that kind of that kind of synergy even further. So this is version two and two point one. And we have just recently announced that we're developing version three. And so this is really not too far away now, our community is going completely nuts, waiting for us to reveal more about it. But this is really not too far off. So we're, we're, we're, I have said previously that I will be very sad if it doesn't hit by the end of this year, but we still haven't actually released a an official date yet. So we've had to thinking about what were our mission statements with with version two right that the super, the super optimistic, highly automated, academically, you know, beautiful thing that it was but was, you know, unfortunately in highly exploitable. We still managed to strike off those first two important points right the exposure to a permanent loss is now no longer something that we have to contend with, and exposure to multiple assets is something we no longer have to contend with. So it's really these ideas of capital efficiency and the opportunity cost of providing liquidity on a dex that are sort of the remaining known problems of an AMM. I want to quickly explore this idea of capital efficiency, because I hope, you know, some of the listeners are at least familiar with the idea that what Uniswap v3 brought to the table was, you know, a capital efficient AMM product right. And even though I it is slightly derivative of what we had with version two, it is sufficiently different and it's very interesting in the way that it is. It is implemented and it's extremely popular, something like 85% of the trade volume on Ethereum right now is being routed through Uniswap v3. So in terms of attracting huge volumes, it has certainly done that, but that doesn't mean that it's a successful dex. In order to demonstrate this, you know, you can take a screenshot right now if you want to prove this for yourself on an Excel or something like that. But when you are providing this kind of liquidity amplification or what Uniswap is calling concentrated liquidity, you can think of it as basically moving this amplification factor up. So when you've got an amplification factor of one, you've got the standard, the standard hyperbola. And then as you move it up through two or three or whatever. This is the amount that that hyperbola is growing. You can think of it as like a magnification. And so that means that traders who are trading on this code now, the price impact of their trade is diminished. So they get slightly better prices compared to the standard hyperbola. But that is also the proportion of the slippage or the effective slippage paid by the liquidity provider. So there is no free lunch, right? If there is a lesser price impact, it means that the liquidity provider is giving up more tokens for a lesser cost. And I think that this is something that has gone a little bit by the wayside, mostly due to some very clever marketing by the Uniswap team and all the more power to them. But it is something that has kept, I think, institutions from actually becoming genuinely or sincerely interested in using it for anything except trading. So the question is, when they talk about capital efficiency, efficient for her. And I have an example that I want to show you, it has been cherry picked because it is an extreme case, but I want to demonstrate just how severe this problem is. And why I want you to be confident when I say that the institutions that I'm speaking with are really not interested in using products like Uniswap. So you can verify this if you like on chain. This is a AXS, a wrapped ETH NFT. So this is a liquidity position that was created on Uniswap B3. And this is the transaction hash. So again, you want a screenshot right now and verify this for yourself you're welcome to, or if not just contact me after the presentation I can share this details with you. So this position. It was $200,000 around about, and it currently has about $64,000 worth of unclaimed fees and this position is 123 days old. So very quick back with the envelope calculation shows that this is about a 32% gain in only 123 days so that's about what almost 90% over a year which seems really really great. You can see the information that Uniswap actually provides you with. However, if you actually investigate on chain at how much money was provided. You can see that it's almost $700,000 worth of tokens. So, it's not a 30% gain and 123 days it's actually a 60% loss over that time period. So this kind of leakage from from that position is catastrophic like I don't think that this is trivial. And I don't think institutions do either. And so if we're developing an institutional grade product something that is appropriate for pension funds something that is going to be you know that people are going to put their savings in. I don't think this level of risk is acceptable. And I also, you know, we've got our research paper coming out about this in the in the coming weeks so please stay tuned. But it looks like even the professionals are also really bad at making money on Uniswap. So there are people that are providing liquidity and removing liquidity in a single block. And those tend to be, you know, for extremely specific purposes. And I think that those types of sophisticated players will continue to operate that way that it's highly exploitative. In order to perform these kinds of activities, it actually requires that someone is providing liquidity beneath you. And you're basically coming in just as a profitable traders occurring and then collecting all of the fees for yourself. So it's not a retail game and employing someone in order to perform these strategies for you is not the kind of it's kind of the antithesis of what DeFi was invented for. It's meant to be very self driving, highly efficient by itself without having to without having to have an extremely well educated, highly sophisticated individual responsible for managing it. Right, the whole idea was to get rid of of incumbency and middleman. So, yeah, the capital efficiency argument I'm just going to say is the jury is still out. I don't think that what's presented on this slide suggests that amplifying the liquidity on the curve is necessarily a good a good approach to becoming more capital efficient. In short, institutions want assurances and so do I right this is something that is applicable not just to the, you know, to private banks and, you know, family funds and things but also the retail space. There are people with nine to five jobs that don't have the time or the patience of the knowledge to manage liquidity in the same way of these sophisticated individuals are on B3. And so I think that it's a much bigger market to provide a product that lazy liquidity providers the uneducated liquidity providers can participate in. So yeah capital efficiency let's say that this is still an unresolved issue, and I think that there are better ways to handle it. So Dan Alitza published a blog post in in February 2019 talking about what he called super fluid collateral. And I actually agree with almost every single thing in this blog post I highly recommend that you go in and read it. But I'll just skip to sort of the end. He's basically saying that this idea that you can use tokens for more than one thing simultaneously is a much better way to think about the, the type of breakthrough that defy is about to make. So not the idea that necessarily that you can provide liquidity for trading with a token or stake it for something else or perform, you know, lending and borrowing with it. But the idea that you could have a single protocol a single program that can do all of these things simultaneously is a much better way to think about it. And this kind of it flies a little bit in the face of the defy Legos paradigm that we're currently living in, but you can see with announcements like like sushi sorts trident and what Trader Joe is doing now an avalanche that I think most of the defy teams. They're all kind of thinking the same thing and they all want to offer the full the full suite of services that modern banks offer their customers. Mark, conscious of time do you think we can take a few questions from the audience and. Yeah, of course, let's um I can end it there actually. If you like. Yeah, let's send it there. Okay. I'll just skip to the end. Yeah, let's go ahead and I'll go ahead and take. Take questions there is a lot of my, my contact information. We will be sharing those slides with us. Yeah, of course. Yes, please yeah. So, Manny has come up with a question asking about liquidity staking. Can you quickly explain or describe liquidity staking. Yeah, for sure. So, there's when we talk about staking is really two different. It can mean two different things depending on the context. So when you say for example, Ethereum in a liquidity pool, what you're doing is basically allowing that liquidity ball to perform that constant function market making process with them. But when you provide those Ethereum to the pool, it will usually issue you some receipt, right, and we call those pool tokens, and those pool tokens basically represent a pro rata share of everything that the pool contains. And so staking in this case, it doesn't necessarily mean it's not for proof of consensus or something like, like Adder is used for on Kedana, or what the beacon chain is used for on on Ethereum. In this case, all we mean by staking is committing, committing tokens to a smart contract. You can think of staking as being a synonym for deposit or something like that. And the other type of staking is what I alluded to just then. And this is a part of the proof of consensus or proof of stake consensus mechanism that's currently being investigated and may offer a alternative to proof of work, but not a part of this presentation. So yeah, when I say staking, what I mean is give broken give tokens to smart contract with the ambition that that smart contract will generate yields for the user. Brilliant. A couple of questions around. I've been trying to understand. So, all of this is run gone. Thank you. Thank you. And Mark, I just want to say a fantastic presentation really, really grateful and I really appreciate it. We'd love to take conversation with with you and the others on this team as well in my old world back in the institutional days. A lot of what you were discussing and slippage it falls under a category that we used to talk about around implementation shortfall. One question I do have also is it to the extent you can discuss it, those institutional conversations. The other component that they we in my past I've often seen that they're concerned about is moving at size. Can you talk a little bit about how institutions perceive that in the next world. What what bank or is doing to allow them potentially maybe in v3 or beyond to move in size or is it simply matter to them spreading across the decks. Yeah, so the, the scalability is certainly is certainly an issue. And yes, we do have an answer for it in version three. I can't. I think I'm still under gag orders. I can't talk about that feature specifically, but just, I can tell you that you're absolutely right. If there's, for example, a potential client that says that they want to contribute, you know, a billion dollars or something to the pools that could potentially, you know, double our TV L overnight. They might only want to provide it in a single pool. And that would mean that the protocol is forced to sort of make markets now with this enormous amount of capital. And in general you get diminishing returns right as the pools get deeper and deeper. The ROI actually comes down. And so it's not always the it's not always the case that a deeper pool means a more performable. But as the ROI goes down the exposure to that divergence loss that the impermanent loss that continues to scale linearly. So yeah that there is a scalability issue, and that I encourage you to have a look at that blog entry by Dan Alitza, I don't think that just using that kind of capital to make markets is going to be the, the winning strategy for much longer with with or without liquidity amplification. I think there's going to be and this is the sort of thing that I'm researching in depth, you need to offer a sort of a diverse range of different yield generating primitives that support that scalability right it can't just be one thing. You don't want to turn people away when they're bringing that kind of capital to a protocol and tell them, oh you know you should probably put some of this in in some of our competitors products or you should probably, you know, do this that or the other thing. You really want to make sure that you've got everything that you need in order to accommodate that level of capital in house, and that's something that we're trying to bring bring to the table with version three. Thanks Mark really really great presentation really well done thank you. Thank you very much. Yeah, go ahead. So, let's, let's step back a little bit. Yeah, basically, what you're talking about is the basic premise which is the question product, which implies a certain price conversion between one to the other. Remember, I don't know whether you familiar with the US history of the US economy. There is something called by metalism, which did the same thing with gold and silver. And, in fact, it was shown to be having most of the things that we have observed here, which is basically one asset moving moving in price with respect to another. So, don't you think that that is a, you know, it's a basic problem with this, this philosophy in the sense that all of these methods that you have done to mitigate those effects are in effect because the price between these two assets are. Yeah, yeah, so let's come back. Yeah, so I think one of the, yeah, you're absolutely right. 100%. But the problem that we're trying to alleviate is the requirement for a human market maker. Right, it's the, the, the one of the oldest kinds of white collar crime is to, is to run do what kinds of stuff. Yeah, not just front running but deliberately misquoting something. Right, and defy blockchains in general are about trustlessness. And so what the AMM is doing is it guarantees that the price that you receive for something is agreed upon by everyone and is not under the influence of anyone. And so that's really the problem that we're that is trying to be addressed, but the, the specific, the specific thing that you've drawn attention to which is the fact that these, you can have two different things that aren't necessarily that are in no way anchored to each other or that, you know, their prices aren't derived from each other. It's dynamic. That's all I'm saying. It's dynamic, right. Yeah, I'm not saying that you cannot cover a price at a point, but that doesn't remain constant. Yeah, correct. But what's interesting is when you start stringing a very large number of these things together. And so with with the balance of pools, for example, rather than having bonding curves they have bonding surfaces, and those surfaces are highly customizable. So rather than have, you know, what we call a 5050 pool which, again, bankable will always have that because it gives traders the best possible slippage. But at the sacrifice of slippage, you can actually create a custom curve that does something else. Now again that it's, it's not addressing the problem that you're raising, but it does diminish it a little bit or at least alleviate it a little bit. But in Bangkok's case, when you've got a very large number of these pools. You end up with such a flexible kind of price discovery vehicle that the price of something even though it's expressed in terms of BNT on the protocol, the price of BNT and the price of Ethereum and Bitcoin and everything that we have are always going to be moving independently of each other. And as long as the protocol is large enough as long as there's enough BNT liquidity there to support it. It doesn't really run into, into too many problems. One of the things that I didn't discuss by the way is exactly how our impermanent loss refund mechanism works in depth. And there's basically by allowing for BNT to to inflate and deflate. You kind of have this, this amazing like value cushion underneath everything. Even if something is going to, you know, move in price parabolically next to BNT and things have right we've had we've already had thematic pool for example moved 100X during a time where BNT only moved, I think 20X. And so you know that that relative difference. It should be a problem in the sense that if there was only one liquidity pool for that asset pair, that you know it would basically throw the whole thing off kilton that the market would become dysfunctional. Because the, you know, the BNT bleed off is being able to is being absorbed both from, you know, from centralized exchanges, and also from other pools in the network. There's really, it's boundless right how far these, these prices can move relative to each other. And I think if you would, if I was to give this presentation a year ago, you would say that's highly speculative there's no way that you can you can say that with confidence. You've now been through, you know, such a, such a turbulent year of price discovery, not just for BNT but for a whole bunch of assets across the network, that there's really good reason to think that this this thing this idea is very sustainable. Well, I mean, beautiful. The, the, the point that I'm trying to make is that when you have these pairs, I mean some kind of, you know, the pricing exchange rate between two things. And in the regular markets, in regular times that price doesn't move that much, but in, in the cryptocurrency market that price seems to be highly volatile even during regular times, including, including even some of the worst downturns we have seen. But as we know, sometimes when the market moves heavily, like for example, the mortgage crisis of 2008, where the price of mortgages was undiscoverable. In other words, one price goes to zero, then you know, all bets are off. So in the regular marketplace, there are what are called providers of liquidity of last resort. Yeah, which, which in our case happened to be let's say central banks would buy stepping to buy stuff when they plunge. Sorry to interrupt you, I think conscious of time I think people from the other groups are coming in so maybe we should wrap it up and probably invite Mark again for another session. So maybe we can just ask him the question which is basically, you know, what happens to something goes to zero. What happened is there an equivalent of a liquidity of last resort provider. Can we make one in AMM? Yes. Yes, we can. Yeah. In fact, you know, we're already at the bank was already at the stage now because the BNT token appreciated in price. So quickly, during 2021 that if every single human liquidity provider was withdraw from the protocol today, the protocol would still have liquidity left over. And so there's like the protocol will now it will just maintain some basal level of liquidity across essentially every every single pool. A couple of exceptions. But yes, the that liquidity will remain, you know, essentially forever and let's not forget and this is going to sound a little macabre but you know take with it what you will. During 2020 and 2021, there are probably a whole bunch of liquidity providers who have died. Right. And if you know because of the pandemic or other things. And if their family or lawyer or something didn't have access to their private keys, then that liquidity will remain on the bonding curve forever. And there are also people that you know that lose their private keys or get their wallets hacked or something like that. Or who, you know, turn their back on cryptocurrency or for whatever reason, decide to leave their liquidity in the protocol and never return for it. And so in a sense there is this residual right some people call it dust, right sometimes when you want to withdraw from a protocol. You can withdraw like 99.999% of it or something because of rounding errors. And so the rest of it kind of remains behind as as residual and the bank or protocol along with every other decks has that property about it. The other property is gas fees. Generally, when these huge market crashes are occurring. The price of doing something with your liquidity is, it goes through the roof. And so a lot of people end up, you know, and this is both good and bad, but they end up kind of shackled to whatever DeFi application that they've committed to, because the price of removing it is just astronomical compared to the capital that they've provided. You know, for better or for worse this has a stabilizing effects not just on the protocol but on the assets themselves, because it means that these panic sellers or would be panic sellers kind of have this minimum level of friction between the drawing those assets and then and then selling them back into into that death spiral. I would say that DEX is in general have actually been a highly stabilizing influence where we would once have to rely entirely on on human market makers who let's face it are probably getting pretty nervous when they're taking on a large amount of exposure to crypto assets as well as selling them off. We've seen you know 99% drops in cryptocurrency pretty regularly that I think that those market makers they start to price things a lot lower than they otherwise would because they themselves are panicking. So the DEX space is actually started to even the keel a little bit. I think that overall, we're headed towards a much more, a much more stable ecosystem, and you know, cryptocurrency is one of the hardest things to tame you say that, you know, the crashes of when mortgages, the price of mortgages become undiscoverable. I promise you, you know the kinds of downtones we've seen a cryptocurrency dwarf that, and the fact that DEX has have managed to weather that and come out stronger on the other side I think it's testament to their fortitude. So here's where I want to call it a wrap guys. Thanks, thanks Mark. So much for for taking the time to explain these things to us. So please reach out to Mark for all your questions. He's a great guy.