 So, I've started the recording and we'll wait for one more minute before we start. This was a last minute sort of call to action to have this meeting and to talk about the things that we discussed in our email. So, I think even though there are only five participants, we have to start now. And like Mani said, this is a pretty slow weekend here in the United States, I mean week in the United States because of the Memorial Day weekend. And of course, people are probably making up for their lost time. So, let's go to it, a couple of things, one, you know, we need to talk about two or three things before we start. One is the fact that we are part of the Linux Foundation and we have to obey the antitrust policy of the next foundation. That's the first thing. The second is the fact that we treat each other with respect, even when we are disagreeing with each other, even if we are disagreeing with each other, and that is a second aspect. So welcome Alfonso. Since we did not announce this a long time, you know, we gave only a short notice. I have a feeling that we are going to have less than ideal attendance, but we will nevertheless go with this presentation, which I'm going to start. But let me introduce Kirti. Kirti is the vice chair of this group and he's also a guy who's been very active, especially in the insurance sphere. And he's working on a lot of things, most importantly liquidity for certain types of securitized insurance-linked securities. So that's important to consider. So I lead off with a little look at the concept of recordian contract, even though people do not know about this, it was started in 1996 and Iran is here, so we have all the important people. I don't know about that, but thank you, good to see you. So we are going to discuss, you know, the main thing we're going to do is take a very broad view and then link it to specific ideas in insurance-linked securities, securitization, and then go from there to liquidity as a concept and why that is important and what are the proxies of liquidity and how do you monitor them and how do you link it back to the initial issuers contract and how do we also automatically manage liquidity, which is of course the failures that we have seen in the crypto market linked to that because liquidity disappearing is what causes tremendous price movements. So I'll start off with a presentation where I'm going to share the screen. Let me go back to this and try to share the screen. Can you see my screen? Yes. Can you see the presentation or is it not visible? Presentation. Is the screen changing? Yes. Okay, so first we go with recording contracts, like I said, you know, this is the basic agenda and there should be a fair bit of participation, otherwise this is quite uninteresting. The first thing is what exactly is a recording contract? It has been variously described as a software pattern and other things, but basically it is a single document that is both human readable and machine possible. Proposed in 1996, maybe 13 years before Bitcoin itself came online. So it has some history behind it. And the other aspect to it is that it is an immutable commitment. That means the document can be hashed and then it can be signed. And later on it was discovered that it can be implemented using a smart contract, but in the beginning there was no blockchain, but it is meant to be an internet-based issuer contract, which is how it started off. And the dispute resolution uses the original contract. So the two aspects of it, that it is human readable and it's machine possible at the same time, makes for a linkage between the world of us humans, the analog world and the digital world. Which of course we have seen a tremendous explosion of ideas that are underpinned by something like this, even though we may not even call it a recording contract, in fact, we still would recognize the basic elements of this in almost all of the activities in the crypto sphere. But we have seen that not being aware of this gestalt, this story of the whole thing, makes people not realize certain elements of it and hence causes problems in the market when stress hits. This is the main story here, that is that any digital contract has its feet in both the human world and the digital world, even though people do not recognize that fact. And they only recognize it when things start going bad. They are very happy when the prices are going up constantly, but when the prices take a hit, they are not happy. Anyway, so what are the features? The hash creates a contract that cannot be changed. The additional part of this is that there is a Recording PKI that delivers clarity on the, so first thing is hash creates a contract that cannot be changed. We are familiar with this concept in ideas like CDM. CDM is actually a recording contract, strangely enough because you can actually generate the contract or at least generate some form of contract from the original CDM, the common domain model of ISDA. The other thing that is not commonly known is that the Recording PKI is a self-contained piece of data that is the public key of the issuer and of the signing key are in the initial contract. People may say, oh, you know, that doesn't make sense. You've got to have a trusted third party. You've got to have something else. But here the concept is that the key is known well known through relationships, which are the basic fact of human interaction makes the key a reality. And the last point, which is the presumption of possession, the user has the contract. Otherwise just knowing the hash doesn't make sense. This is not some kind of a blinded commitment. It is more of the fact that the user can say, okay, this is the contract and if I hash it using the proper method, I will get the same hash as that in the contract. And there is a presumption that the user possesses the original contract or at least has access to it, which is not the case in many, many ICOs, many other things because there is no contract. There is the ERC-20 smart contract, but the issuer doesn't stand behind it in the same or similar way. So this is the seven layers that started off the Ricardian contract. So it is recognized that you have cryptography at the basic at the base level, then software engineering, which is, you know, how do you send messages? How do you transport the actual goods, actual work, a product from one participant to another? And then the rights, in this case, the rights are nothing but a linkage to the identity. Then there is accounting, which if you look, if you know CDM, the accounting is things like how do you, you know, what are the promises the issuer has made? And the events that impel the issuer to pay up according to certain events. And then of course the governance, which is non-technical governance and which is basically what some people may call a type of crypto economics. And then you have exchange of value and then from there on the final layer you have finance itself, which is nothing but the topic of our discussion here, capital markets. So it is a slowly emerging concept that finally ends up in finance. If any of these things are missing in the original contract, then you have difficulties, especially enforceability is a big, big problem. And, you know, how the contract takes care of that is important. So now we go to the concept of securitization, which is very important because it is nothing but the issuer saying that, you know, this is a financial instrument that I have created from a pool of assets. Normally the asset has generates cash flows and that's how the, the securitized instrument pays out. And of course, I'm not going to go into the details of the tranching mechanisms where, so some people have asked questions on the chat. I think accounting does accounting have events. Accounting is basically caused by events that are promises to pay inside the, inside the contract that is sent out by the issuer. So what does it do with the securitization? I mean, this is only a specific form of digital asset, which is a securitization. And that's what we're going to discuss, especially Kirti is going to discuss. It creates fungibility from non fungibility, which is a tremendous concept. People do not usually think about this like a home loan is not a fungible asset. But by creating a pool of home loans and securitizing them, the mortgage backed securities that it creates are fungible. In other words, each unit of the securitized assets is the same as others. And in fact, I have proposed that as a way to securitize even the so-called NFTs, which are artworks and so on. But the problem with those assets is that they do not, they do not generate cash flow as a matter of fact. The only way that they generate value is through the rise of value. What happens if the value falls? All of these have to be handled in the issuer's contract, which they, you know, which takes us back to the Ricardian concept. And because it creates this sort of fungibility from non fungibility, it creates liquidity. That means people can buy little pieces. Institutions can buy knowing that the prospectus of the issuer, which is another recording contract is enforceable. Currently, there is a lack of transparency, but again, going back to the recording contract that can increase due to the fact that user possesses the contract and then can link the events, the accounting events to whether they should get paid or not. So, next slide is going to be where Kirti takes over and leads us through the ILS, Insurance Link Securities, and he's been investigating liquidity also. And you can tell me whether you want me to advance the screen. Okay. Well, thanks. Thanks, Wippin. So, Wippin, you know, covered quite in detail about Insurance Link Securities, I wouldn't say Insurance Link Securities, sorry about Ricardian contracts. The reason why I think Insurance Link Securities are really interesting is because it creates that whole element of fungibility from non fungibility. Now, why is it lucrative for both the parties, like in this whole ecosystem? There are two parties. One is the issuer or the sponsor of that specific risk. It is nothing but an insurance company. On the other side, you're looking at. Oh, sorry. Okay. And on the other side, you're looking at the investor. Now, what is ILS doing? ILS is just helping, you know, investors come in, take a part of the risk in return for a reward. And what does, how does that help the issuer or the sponsor is the sponsor has passed on a part of his book. He was diversified the risk further into capital markets using special transformer vehicles or structures. And this creates better diversification of risk over a period of time. So it creates a win-win for both the parties. Of course, they're very different complex structures, which work in specific jurisdictions as well as in various regulatory environments. We'll get into that a bit later. But one of the most lucrative things for the investors, especially for the institutional investors is it's got very little correlation to, you know, your traditional capital markets. Most of the events are triggered either through some, some majority today, majority of ILS backed instruments are into something like cap bonds, site guards, ILWs, et cetera. We'll take a little more into the detail of that. The pin, move to the next slide. So this is what we spoke about. And while it looks very simple, the way the risk is transferred from the sponsor or the insurance company to a seeded or a protection buyer. We have a first element of risk transfer from the sponsor to reinsurer, reinsurer to what is known as an insurance special purpose vehicle. And of course, through the insurance special purpose vehicle is how the capital market investors interact with this whole mechanism. Now, where do Ricardian contracts come into the picture is we look at the part where there is an ISPV involved. What we've been traditionally trying to look at is to model the whole cat bond or an ILS instrument as a digital twin. Now, if anyone understands complex systems, a digital twin has two parts to it. One is the analog signals as well as the other part is the digital signals that come into it. And how can we do that using the basic principles of Ricardian contracts is something that I will talk about right now. So what's of interest to us is to look at the ISPV structure as to how the structure in itself acts as a vehicle to transform risk. And how does that happen? So every insurance or reinsurance contract is wrapped, it's got a fund wrapper around it. It could be either traded as a cat bond or it could be traded as a sidecar. So in the sidecar, you have like a specific fund which is created. And when this is done, what happens is you have investors who are issued a prospectus. So generally this is institutional investors who are issued a prospectus. The prospectus has the terms and conditions under which it clearly describes what are the returns, what is the liquidity mechanism, what is the proposed earnings and things like that. So it's very, very comprehensive. And the aspect of Ricardian contracts comes into the picture again because we want to look at it and decide for ourselves to say, hey, what part of this can be computable? And what part has to take like an analog signal through human intervention, perhaps governance? And how do we address that? Like, you know, when we create that digital twin and how do we take it forward from there? So that is the basic principle of how we look at a digital twin and why the importance of digital twin is having a digital twin can create certain liquidity mechanisms, and which can also be very, very compliant, the way regulator wants to see it. Say, for example, the regulator says that the reserve capital, which is required by the insurer to be able to write insurance, which sits today in this ISPV, has to be pre-funded. And the obligation or the intercontrovertibility of the contract is that the reinsurer has the rights of that reserve capital rather than the capital investor. So this is the trade-off where the capital market investor has taken on some risk onto his portfolio and returned for a reward, which is nothing but a part of the premium, which is passed on to the capital markets investor, either as a coupon or an interest payment. So that is a simplified structure and moving on. So if we go back to this diagram, each one of those contracts can be modeled using something that is both human-readable and linked to something that is machine-possible, usually using a markup language, which has been done through things like, you know, many, many people are active in the space. There have been a lot of, like, lead brains activities there that, you know, people like... So one of the things people like, you know, the Corda people have made this a important part of their whole selling proposition, because probably because Ian Grigg was, who's the originator of the recording contract was one of the original participants in Corda when he started, when he designed it, and I had talked to him extensively about this stuff. The other part, which is kind of not shown here directly, is that the collateral trust interacting with the capital market investor is also predicated or dependent on a contract, because the coupons and interests are paid according to a certain schedule, looking at certain events. So anything you want to say there, Kirti? Like, how does the capital investor know, okay, I have to, you know, I am protected so I can invest in this? So there is a lot of transparency within the prospectus, which talks about how the collateral trust invests most of the reserve capital, which is in there, into, like, high-rated short-term investments. So that is clearly articulated. And that income which comes from the investment, it also acts as sort of a buffer in case there is some sort of, I would say, trigger event or some sort of, I would say, claim which is coming through into the system. And of course, there's a lot of clarity on the promise to return both the collateral. If there is no trigger event, the prospects clearly articulates that there is the collateral as well as the coupon will be paid out as a part of the closure of the ILS prospect. This is the happy path. So let's go back to the other slide, which talks about the benefits. Now, let's concentrate on a little bit on what are the downsides, because that's something that people miss, especially when they are selling the idea. And how does something like a recording contract protect you during those times of, you know, a downdraft? That's important. Do you have anything to say about that? Well, when a risk event or a trigger event occurs, the fund prospectus covers most of the risk which is associated with these investments, because most like 80% of, I would say 60 to 80% of the market today is cap bonds, which is climate triggered events, right? Now, these climate triggered events, like Hurricane Katrina and all of these, when if they trigger by any chance, which is very, very, you can see that most of the stuff is parametric in nature, and which is a straight signal into the SPB where it has to pay out immediately. When such things happen, the investor is clearly aware of the risk that he is taking on to his portfolio. Or she. He or she, sorry. There. And of course, there is always like every investment prospect. There's always a probability of losing both your capital as well as returns in the prospect. So that is something which is no different from any other financial instrument that exists in the market. Yeah, so do you want me to go to the next one? Sorry. Yeah, of course, there are certain buffer mechanisms which are built in, like for example, when I spoke about the collateral trust, the income from the collateral trust acts as a first level of buffer to take in any hit which comes into, you know, the ISPV. And of course, it's a contract of a contract. So at the first level, you have the sponsor treaty contract treaty or a. So this triggers from the sponsor side, which hits the reinsurer and the reinsurer goes into from the reinsurer goes into ISPV from where the claim is paid out. So just want to articulate that the the income from the collateral trust not only pays for the whole ISPV, but it also pays sometimes for the initial hit or a threshold within which it can manage specific trigger elements beyond which then the the highly rated short term investments are then liquidated to pay out. If it's a big, big trigger event, the highly rated highly liquid, you know, investments are then liquidated to pay out the rest of the trigger. So just wanted to highlight that as well. Yeah, so here we are dealing with a very specific form of securitization. We just wanted to, we were going from, let's say a 10,000 square 10,000 foot 30,000 foot view to, you know, something very close. But all through we see that the contracts are the animating feature. It is contracts all the way down. Meaning the contract between the collateral trust and the capital market investor is backed by is is dependent on the counterparty contract between the reinsurer and the collateral trust and the reinsurer and the sponsor have a risk transfer contract. Of course, even the customers who actually come to the sponsor have already signed a contract for paying premiums in order to get that insurance. So contracts all the way down and how they link to each other. Absolutely weapon and think about it as a complex system right so if you think about it in the terms of a Ricardian contract. There are commonality of elements which transfer the risk from, you know, the actual customer to the actual capital investor. So they're elements of the contract which act as digital signals, which then become very, very computable over a period of time. And if they if the right type of data and the right type of structure is created. It creates almost seamless payout structure like for example the certain parametric insurance covers exist today. They pay out as soon as an event has occurred and because they have certain sensors and they can always do, they always are there from the models of risk and when a trigger event happens, they're very, very clearly at the heart of all the parametric information they get through and the whole objective is not to make it a lot more complex is the objective is to kind of see how we can pick and choose elements, which can be, I would say, automated, which can be computed clearly easily and focused on to create that specific liquidity and flow from customers to capital investors across the ecosystem. So that's the basic idea. Of course it will have a lot of elements of governance, which we may not be able to implement at this point in time but probably in the future state we should. model the aspect of the asset in a digital twin format. But governance, in that sense, and also the whole crypto economics is an emergent quality, which proceeds because of the base, the bases are covered in the fact that, you know, you have the direct correlation between the human and the credible contract and the machine possible contract and it all starts with us humans. So that's that's very important. And, and the fact that you cannot change the contract the power asymmetry. That means the investors have less, lesser power than the collateral trust, you know, managers and the reinsurers. Even though they have lesser power, they are protected because of the fact that there is an immutable contract, which can operate and plus, you know, smart contracts have known to have bugs. So we can always go back to the to the paper contract. So the everything is codified in the paper contract. So if there are bugs, then they have to be fixed. And, you know, it has to follow the contract, not, you know, the paper contract, not the smart contract. Next slide is about market liquidity and how does this relate to the contract, because as long as there's someone assuring that there is a market maker, there's someone who says that I will be your counterparty in most circumstances for a reasonable price, market liquidity refers to the ease in which an asset can be sold in a market at a reasonable price. So reasonable price is the key point here, which means that the price stability is important. If you can sell something at a fire sale price, it doesn't mean that the market is liquid. You can always give it away, or you can even, you know, this happened during the during the whole financial crisis 2008 2009 and we have seen similar similar things happen even last couple of weeks in the crypto market, where buyers disappeared. So if there are no buyers, and all their remains are sellers, there are no transactions and price discovery is impossible. And there were times even when bigger big exchanges like Coinbase, you couldn't transact or finance or whatever. So when the price is going down, especially this becomes a big problem, unless the counterparty emerges that can do a support price in regular capital markets, especially for the highly liquid instruments like treasuries, it happens to be the central banks who have done it these days. And that's what supports the price of those bonds, even of corporates, even of junk bonds, even of, you know, all the other stuff and people complain. But they don't realize that if that price, those prices go down like crazy. Then there are problems with the with the market. Anyway, Dan, you're going to ask something. No, it's just the point I was going to make is that we have flash crashes, even in treasures. Yes. I think you're, you're establishing kind of two features of liquidity, the ease of transaction and the ease of discovery of interest, which automation provides. And the interest to provide liquidity either side, which is related to the former but isn't the same. And I would argue that, you know, whatever Bitcoin sliding, you know, by thousands of dollars in price was was still liquid from an instrument perspective, it just, there was it was a one sided market. Price discovery was, you know, maybe it was exactly, I don't know, but we'll have time will tell but the prices were discovered for transaction without impedance by wait, what is this, what is this asset that I'm buying, which is I think, you know, that was your former point is that there isn't a question. Because you're cutting in or other smart contracts to express these kind of transactions, there isn't a question about what it is, which, which is a huge barrier to finding liquidity. Well, I mean, you are saying that prices were discoverable, but I would argue that point because if there are no transactions, there are no buyers. So, I guess that you're discovering is a one off transactions that are either wash transactions or somebody attempting to prop it up with some, I mean, you know, not real liquidity, meaning the amount of transactions and the block, you know, the size of the transaction are also important because, you know, you can have a straight transaction, like you said, the flash crash, if you're going to sell the tertiary is automatically, and if it causes negative feedback loops that the prices crash. Then, you know, all those transactions which people engaged in and ends up losing money for them, especially when the flash crash passes. The point about the recording contract is, I mean recording contract is, you know, too much said, but the whole idea is that there is a human and digital aspect to any transaction. And most of the time you can let, let the automation run. But we have to be also very, very clear about, you know, certain other proxies of liquidity that is, you know, the size of the transactions, the number of those transactions and who's actually making them, you know, all those things are important. And of course, it's very difficult to find out who's actually making some transactions in the crypto market. Yeah, I mean, look, at the end of the day, I suspect that we're all closer in understanding an agreement that we are farther apart. And that's a semantic or pedantic argument with respect to Bitcoin, if it goes from gaps down from 40,000 to 30,000 price was discovered at 30,000, just not necessarily happy for those who were trying to sell volume decreased also which isn't interesting but these are all market dynamic. Those my point was really those are, there's a part of liquidity that's market dynamics and the part of liquidity that's comes from clear transparency about what it is that's being transacted. And the automation smart contracts recording contracts as I understand them, which is limited. Those are addressing the second point, the clarity about what it is that's being transacted right if it's, and I'll give the example if you have a catastrophe bond link to hurricanes. And all of a sudden, we, you know, the market perceives that hurricanes are going to be more, they're going to be more hurricanes and they're going to be stronger than those bonds are going to decrease in price because risk is higher. That's appropriate. Another news is that people who are evaluating buying or selling know what it is they're transacting and because the contract has been described systematically. That's right. I'm happy that you think that people have. People are sophisticated and are. But they always cry when they lose. That's the whole. But again, coming back to recording contract, these practices should shore up price stability. That's that's that's all the I mean, without looking at the whole picture and taking care of it in your issuance contract. You are more subject to this while swings than if you're not that that's all there is. I mean, it's not, it's not a question of that they won't crash either. They, you know, depending on the on what is happening in the real world. Of course, there's no protection. And that is the whole argument but having these protections should make the bonds, you know, whatever it is cheaper, the asset cheaper and people more willing to transact in them. So going back to this liquidity proxies. Some of the observe observed ideas are to look at the bidders spread, the price volatility number of number and size of transactions per period. And these are some of the proxies of liquidity that can be observed in the market. And this also kind of brings us to an interesting point about automated market making, right, the AMM aspect of things for ILS. So, looking at it traditionally. The secondary market for ILS is limited complex. And of course liquidity generally is provided as as a mechanism in the prospectors where you know the issuer generally says that every quarter we give you a choice to buy back some of whatever, you know, the other instruments that we have issued from you as an institutional customer. But the experiment where I'm trying to explore at this point in time is looking at the possibility of a regulated capital pool where the interacting entities are known, like for example the institutional customers obviously known entities are looking at what is known as a bonding curve. A bonding curve is nothing but an algorithm which helps you assess liquidity for a specific asset based on a value, a stable value that it's picked. So this, this is the area that they are absolutely exploring. And the liquidity proxies that you've highlighted as a part of this conversation. That is, the bid ask spread is addressed by the algorithm. The pricing volatility again is addressed by the algorithm. The number of transactions and period. Of course the number of transactions of course is a very important element because as the transactions increase. If there is a pair within an amm pool, and it's linked to a specific. What is called as a constant value curve, then there it creates, you know, the whole aspect of price discovery in an automated way, and gives you that the liquidity as well as ability to look at arbitrage opportunities within the automated market market pool. So this is, you know, something that I just wanted to add to the conversation and what we are trying to explore today in the lab. You are exploring this in your interactions with balancer protocol or something else right. Yes, it's, it's a part of the exploration with balancer protocol. And if it was easy, it would have been done already but because there's a whole not always true but there's a whole element of, you know, regulation, which comes into play as to how we can put certain circuit breaker mechanisms, also the ability to trade between certain part of the bonding curve, not the entire spectrum of the bonding curve. And of course, also certain risk mitigation mechanisms that we need to put in place are all of the things that we are trying to discover again it goes back to the whole aspect of the guardian contracts to be the objective of looking at everything is a complex system which has got many inputs, assessing these inputs either as analog, that is where we step in to provide some sort of an input and digital which is possibly generated automatically as a part of the process either through an oracle, or, or perhaps through the algorithm in itself. So money has raised his hand. If you want to let him ask the question. Thank you. This is a, how is this, how is the balance of protocol different from, let's say the Uniswap version three, where he allows you to put liquidity and certain parts of the yield of the other curve. Maybe you can Balancer has got now balancer is in version two, and they are currently looking at implementing some sort of a mechanism called as an asset manager, and the asset manager kind of helps you to look at certain limits within the bonding curve as such so this is still new to me, I'm exploring as much as I can possibly at this stage so hopefully when I have a better answer for you, I will definitely share that output with you. Thank you. So the last slide, which is going to be about something which is actually called a liquidity contract, which is an accepted market practice. It is also done by the issuer with a financial intermediary, which has a, they do have a bunch of, you know, a certain sort of a liquidity pool, you can call it that. And of course it is an intermediary in the, you know, that that is actually a centralized party but they are regulated entities, and they can carry out the purchase and sale operations to create or not create liquidity. But this relates to, so Mark has a question. What is Mark's question. Yeah, thank you for that actually. I mean, I'm also in the same balance of research groups. So, yeah, I just wanted to also second the point on on how balancer is not just looking at these asset managers but also then what dynamic weights and fee functionality might better support those liquidity pools but actually my question was going to be more around. I was just listening in on the Are you, are you still there. I don't seem to be able to hear you. What. So, Mark, perhaps you can talk about the legally smart contracts and what auditing. We seem to have breaks with you Mark, your question doesn't come through. If you can type it in. It may be better because then we can actually see the question and answer them. But I believe what you're asking about is how closely do the smart contracts follow the actual human readable contract. And if they're auditing mechanisms. Is that what you asked, or are you there still there. Anyway, I think we have lost Mark, or we have lost me either. Yeah, sorry. Yeah. So sorry, I just put it, I put in the chat but I was just saying that it was it was about yes, the ability to better mirror the smart contracts to the actual legal understanding in the, in the physical contracts, as well as then having and this is what it was discussing this Frankfurt crypto asset discussion was state state mechanisms or at least seen as trusted auditing and other mechanisms that would also give better security to those participants in knowing that it was sort of given that certification and auditing that it does meet those certain standards. And then also just in the design, just looking at what happened with the with live or was sort of a lawyer behind because you had, you had the discontinuation of live or many contracts, before and actually in thinking around how taking that lessons learned and designing us legally smart contracts, so that they could then be more easily amended between the parties, and having that flexible that sort of built in flexibility, sort of a little bit lining with the dispute resolution but sort of in the amendment provisions and clauses, and how that could be well supported actually if properly designed so those are my points thank you. Certainly, I think, I think the main point is that the more we cover in the initial contract, especially based on history, the more secure we are. Because there is always, you know that dispute resolution is baked into the contract. But of course, the emergent effects will never be able to be fully aware of. And we have to sort of be forward thinking about that. By saying, you know, under what conditions is, should people redesign the contract, the smart contract, I mean, even if even in the case of people that have not been observed before that that's that's a tough, tough one. Yeah. And may I just add like one last comment before. Yes, yes. I think we are amidst an opportunity where it could be the best of both the worlds where enforceability for contract legally, as well as execution of certain parameters on the specific contract through competition has to be a hybrid at a specific stage at this point in time, because while we explore that rabbit hole, we find that, you know, optimal balance, and inside the legal purview, that's the reason why we are trying to take the project of a less liquidity to FCA as well to get their, you know, their view and experiment this as well within their sandbox framework to understand, you know, some of the broader things that Mark also highlighted as a part of the conversation so that's that's exactly what we're trying to do. We say that we're not trying to boil the ocean I think there's exactly a great opportunity to kind of explore hybrid models, which could possibly work as a, as probably a stepping stone. Yeah, I mean, the point of this whole discussion was that these ideas around it just concentrating on one level of the stack is not good enough. You know, the full thing and it's not that. Yes, it is boiling the ocean in a certain sense, but if you follow small accepted practices that are, you know, very limited in execution and principle, then you're safer than if you had not. Of course, the end of the world is nigh, and, you know, you, you, you cannot escape it, but at least we are safer. That's all for today. And hopefully this sort of session was interesting to you guys. And if so, we are going to write a paper on, or at least some thoughts on AMM, following on from this. And I hope to see you on another meeting soon. Thank you. And this is of course the references. Dan, are you still around the East End? Going back and forth right now in Brooklyn. Oh, okay. And you? East End. That's pretty. Wisteria behind you. Well, that's Central Park in May 2020. I'm still thinking we have a garden filled with Wisteria. Well, that's nice. I love Wisteria. All right, guys. It has been great. Please interact on the mailing list. And we are also of course looking for ideas for new talks. And we'll continue the AMM discussion. Thanks. Thank you. Thank you. Thank you.