OnlyMP3.to - Governance and Risk Ep. 191- sSKWsFlNBz4-192k-1... Fri, 5/20 1:57PM 1:13:39 SUMMARY KEYWORDS d3, dai, protocol, risk, collateral, maker, governance, market, oracle, proposals, polls, compound, passing, vault, votes, delegate, module, ave, case, emergency shutdown SPEAKERS Prose11, Robert, Brian McMichael, Kirk, Artem Gordon, Pablo, Primoz Kordez, David Utrobin, Niklas Kunkel Prose11 00:04 All right. Hi, everyone, and welcome to the 100 and 91st government's nursing. My name is Payton, I know my prayers the webinar online and I'm one of the government's facilitators here at Maker. Now, I'm joined by a live by a bunch of awesome people who both work for are generally interested in the Maker protocol. And this is our weekly call where we discuss so I'm drawn to the governance and kind of risk states of the protocol. few ground rules before we get into things, obviously, this meeting is being recorded, although if you're watching us on the recording, so let's try not to speak over one another. There are some great tools on zoom you can use to help the modern AI if there's something you'd like to add, to speak next to the conversation, you can consider raising your hand. That should be under the reactions section of your zoom dashboard there at the bottom. Also, feel free to drop something in the chat, I'm more than happy to read out loud for you. If you're able or otherwise unwilling to hop on the mic. With those warnings out there, this is an open meeting, we do encourage your questions, your comments, your feedback. So please don't be shy and consider getting involved in in our conversations. One, I guess, heads up on wanting to give before we move on to things is next week, we are trying out a new meeting time. For the last meeting of the month. We're going to try during the meetings a little bit later. So next week, this call will be taking place at 2100 UTC, rather than its normal time and 1700 UTC. More on that later and on the forum. But I figured well at least say once before we get too distracted here. All right, David, I think that's enough blabbering. Let's go to the agenda and let everyone know what we're going to discuss. All right, thank you. So as usual, we'll start out with our governance roundup that includes the votes, what's going on with our Maker improvement proposals. And the latest from four. There are no initiative updates today. So we do have a lot kind of cool discussion topic, centering around d3 items that are risk and kind of what they mean for the protocol. d3 m for those that are unfamiliar stands for Dai direct deposit module. I'm getting the orders right there. And that's basically our way of injecting and Dai to different lending protocols or similar, similar so losing a word, or two similar protocols or other teams that are using something where we can have Dai and be able to inject it directly for one. Cool, let's go on to votes and David so I can look at my lunch. Not bad. We had a fairly busy week on devote site with two weekly polls, and P they both pass those were raising the emergency shutdown module minimum threshold and adjusting the lid parameter on the rate limited prepper favour. So on the ESM threshold, that is the amount that needs to be deposited for our emergency shutdown to take place, and the Euclid controls the amount of Dai that's able to the auctioned off at one time, for MKR. Cool, and then we had to greenlight polls that concluded on Monday, first beam, a TD combustors. That passed with a little over 47,000 MKR to 40,000 against. And the other Greenlight poll that concluded was the India power green bonds, that one did fail with 40, little over 49,000 MKR against compared to 30,790 supporting for active green lights, there is one that voting ends on the 30th. So please express your opinion there if you haven't already saw on failure, speak. And then we of course have five ratification polls that are going on that we covered in our MIPS segment next. All right, so that takes us to the executive proposals. Shortly before this meeting started, we had yesterday's pass. So that's still pending execution. But that took care of some vendors in the world today. We're authorized a new Flashman module and some MKR missing the executive that will be going up next Wednesday and we have three things planned for it. That is the racing thing. ESN threshold as I mentioned earlier, up to 150,000 MKR. And we are adjusting the lid parameter down to 30,000 Dai which would be essentially one, one MKR function at a time. And finally we have MKR transfers for ducks and sidestream action services. Awesome. I think that does it online. So I'm gonna pass it over to Pablo to give an update on the MIPS Pablo 05:22 Thank you, Payton. And hello everyone public here with a MIPS update. Next slide, please. So it's only four days until ratification polls close. We've had five formal submissions of four days governance cycle. We'll take a look at the votes right now. I'm not actually right now. This is how they looked about two hours ago. But it is probably the same. So all ratification polls are passing. For me IP 65 Project Clydesdale and MAPE 68 mentalis Lusitano both are passing with radian? 65% Yes. Then both are MAPE 69 l to 201. Fast withdrawals and the coordinate budget process amendment are passing with 100% yes votes. And finally the budget for the strategic happiness Core Unit led by Andrew bourbon is passing with 75% of approval. Again, these ratification polls will close on Monday. Next slide. Yes. So we'll now take a look at the proposal scenario. See many of these we've been seeing for a while now. But we have some new additions. So we've had no new top level MIPS that is WB MIPS supposed to serve proposals. Since last week. We have our MIPS 72 delegated collateral attachment real world asset arranger application and six is capital, which I don't think is going to be formally submitted in the near future, MIPS only for permissionless open market operations and MIPS 75 tax task forces sorry, for which we are seeing its first sub proposal which introduces the growth task force and this requesting funds for 1 million Dai. Next slide. We have one Core Unit onboarding for the lending oversight Core Unit whose codename is love. This one we've already seen. Next slide. So we have a facilitator onboarding proposal for Psychonauts for the mean five security Core Unit second, that is currently a deputy facilitator for this coordinate. We also have a voluntary facilitator recording from data insights by Thomas who hopes to be succeeded by Taddeo. Next slide. Amendments Yes, I'm sorry. So we have three amendments. The coordinator for the process amendment whose will request I think Tim, who's running point on it has recently updated. We also have an amendment to make 16, the weekly governance cycle, which adjusts the map so that the governance the new governance cadence is reflected in it. And we have a new amendment since last week, May 14 to SB 21, which amends MIPS zero to introduce retrospection dates for MIPS. We have coordinate budgets for the Strategic Finance Core Unit, which is asking for a one year budget that totals 1.4 million Dai and one for the growth Core Unit also for one year, that totals 5 million. I know that both of these budgets P use the new year to be ratified coordinate budget process, the one that's currently being voted on so I guess it's a good way to assess how that works. Next slide. We have one special purpose fund for the baker Shire Hathaway. This special purpose fund intends to experiment with various yield generation strategies. This one we've already seen for a while. Next slide please. And finally, we have two Oracle related proposals for modifying the data models for BTC USD link USD nine USD T Wi Fi and for updating the article expiration time for all Oracle's these proposals can be found under the Oracle category in the forum and will go on chain as polls on Monday, the 30th. And finally, now for real important dates. So, ratification polls close Monday, the last April modifications for proposals eligible to enter June's governance cycle is Wednesday the first and the formal submission window for sale cycle opens on Monday, the sixth. And that's it for me. Thank you very much. Prose11 09:58 Thanks, Pablo. I really appreciate that. All right. That should bring us to our forum and opponents are now ready. Take it away. Artem Gordon 10:08 Hey, yep. Yeah, thanks. Happy Thursday to everyone. Welcome to a quick format a glance, recap. Quickly going over the past week's Top announcements, discussions and signal requests, and this is covering may 12, to the 18th. And we're starting up with the announcements. And first up we have mamoun, putting together an extensive website traffic analysis of the Maker dao.com website. It's a report comparing the website to some of our biggest competitors such as compound curve, and Ave and the report includes in depth analysis on various traffic metrics, as well as an audience analysis, growth opportunities and much more in there. This is pretty helpful source of information for anybody interested in checking out our websites annual performance, and yeah, I liked it. And next up, we have three delegates this week, who have submitted their delegate platform. We have the governance house delegate platform, the blockchain blockchain at Columbia delegate platform and the London Business School lbs delegate platform. We have a mature delegate number 12 Coming up may 25, at 6:30pm UTC, which will be featuring London Business School delegate platform and blockchain Columbia delegate platform. Each delegate will have the usual 30 minutes to present an answer questions. And moving on to our top discussions. GovAlpha produced a really comprehensive, total 61 Page DAO governance report, which is crazy, which provides a deep analysis of governance structures among Ethereum Dows and also to help develop an understanding of how to apply this research to help improve our own governance. And the report covers participation, resistance to capture resistance to attack openness and incentive alignment. And the authors who wrote it, examine how different organizations work to achieve these goals, and also discuss potential implications for Maker. Next up, we have an acquaintance to the votre committee meetings hosted by rune I am yo has created a thread to promote MKR stakeholder feedback and discussion and help include MKR stakeholders and to some of our internal discussions going on. Additionally, I Emil has taken the liberty of summarizing the DAO singularity meetings that have been organized by ruin throughout the Prose11 12:40 last month. P A P last month. Artem Gordon 12:44 And now moving on to our signal requests over the last week. First up, wink stick inadvertently sent 3.2 MKR to the MKR governance address and submitted a signal request in hopes of a way to return that MKR to the sender's address, several members and recognized delegates stated that the unfortunate reality is that it's almost impossible to return the MKR because nobody has keys or control over the MKR governance address. And next up, the signal request was prepared by the command initiatives and its stakeholders and they are proposing to offboard a uni a uni v2 Dai EAFE, a uni v two w BTC if a uni v two uni if a and uni v two w BTC Dai A. Now these volts lack in demand, which is why they're being proposed to be off boarded. And the proposed parameters are also set to minimize any damages to existing bolt owners that have them. Also ensuring that sufficient amounts of debt are repaid or the worst case scenario liquidated at minimal costs for vault owners. And the details on the parameters that are proposed are available in the signal request. And next up, we have GovAlpha, who submitted a signal request to gauge support for increasing the minimum threshold to trigger the ESM from 100,000 to 150,000 MKR. This adjustment will help minimize the potential threat of any hostile emergency shutdown to occur. And the reason that the signal is happening now is due to the recent fall in MKR price making it cheaper to trigger an emergency shutdown. There is risk to this which may create a challenge to perform a beneficial emergency shutdown if needed. But in conclusion, GovAlpha is asking the community whether we support in raising the ESM threshold to 150,000 MKR. And this vote ends tomorrow and so far, it's leading with the US. And the next couple of signaling requests are either signal requests that have just concluded or recently concluding or right now concluding and the first one not Up is by Sebastian to deploy the balance sheet. And if and as of right now, it's, it's has actually concluded Oh, yes. So it has passed successfully. And yeah, we'll just go on to the next one. The next one that has ended as by ACE to launch the Maker teleport one basis point fee, which has also successfully passed with yes votes. And next, we have Abdullah, who was requesting kind of D three M for the bacon protocol. And it looks like this is not passing at the moment. So yeah, that's unfortunate. And next up, we have an alpha institutional vote, which has passed with Yes, votes at 66%. And yeah, that's it. So yeah, that's a signal requests just concluded, either today or yesterday. But yeah, for more information, check out the forum at a glance on the forums that has a three point summary, more announcements and discussions as well as ongoing initiatives and Help Wanted. Prose11 16:20 Awesome. Appreciate it. Yeah, lots of signals means lots of polls coming up. So stay tuned to the Learning Portal. On Monday, I believe we should have several boats going up for an official. All right. Moving on, as I mentioned earlier, I don't believe we have any initiative updates tonight. So that should take us straight into our discussions. I don't know if Thomas was looking to lead this learners or not. 16:50 Yes, thank you, Payton. I appreciate it. All right. And thank you, everyone. I'm Thomas flitter engagement lead for the GovComms Core Unit. And today, we have a great topic, revisiting d3 M. Next slide. The direct deposit module interfaces with third party lending protocols to enforce A P M. Next slide. The direct deposit module interfaces with third party lending protocols to enforce a maximum variable bar rate for selected assets. Maker governance is able to pick a target variable bar or weight brought borrow rate, excuse me. And the module will automatically deposit or remove liquidity to ensure that target rate is hit. And we've got a lot of questions that we're putting out here for think both risk and PE are here to talk about. And I'd like to hand that off to them so that we can go through these and have a great open discussion with the folks out here on the call. First question, we just said what is the three out kind of gave a very high level? Describe the explanation that but second question is, what are some of the use cases? What use cases apply? And I'll open this up for both risk and PE to kind of take it from here. Primoz Kordez 18:11 Yeah, I can, I can talk to this. So use cases. So I mean, why the tram is really cool is because it's really easy to scale by supply. So you know, if you imagine now, if you want to scale the supply through bolts, you first need to identify collateral, the collateral needs to be liquid, you need to identify if there's any demand behind borrowing that are Oracle costs. You need to determine interest rate. I mean, it's it's not simple task to be honest. And you know, we felt at many points, because as I said, it's not simple. We did run that's really straightforward, right? You just select the lending market or whatever market that has pulled the Dai and you set some parameters, the debt ceiling count, in August case or in August Ethereum or compounds, you select the target rate, so it's straightforward each, it's faster. And some degree, it's also more controllable, which is like the the the second I would say, advantage is that, you know, you can really control the supply, which becomes really important when you might perhaps need to contract it samples. So you can imagine that, you know, let's say PSM, unwinds, and we need to contract the supply. In order to defend the back. The DT rams are probably the best tool to do that really fast, right? You just pull the pull the Dai out. So that's that's really useful. Another thing is when you do the three amps with Atlantic markets, you start offering something that Maker doesn't offer and that's cross collateralization. So we get we get benefits of that right some, some people are Most illiquid Maker because they cannot pull collateral and do cross collateralization. And, you know, they're just normal users of Ireland compound therefore. And you can have benefits, you know, through the trend to that. Yeah, and there's a bunch of stuff, there's also the competitive angle that you can somehow control, right? Because you can control the rates to some degree, currently, of Dai data. You can even address this. And yeah, there's, there's bunch of stuff. But yeah, we'll probably get to that later. 20:40 All right, thank you. Any other comments or thoughts? For this? Prose11 20:47 I can, from a, from the sort of technical perspective, I'll just add maybe an analogy, which is that the sort of vaults that we have in the system, which this is, like, kind of pseudo vault, like these things are like pseudo vaults under the hood, I guess. But effectively, the D three M is, is in a way, it's its own sort of Vault engine, right. So the vault engine is our way for generating die. And, and, you know, you've got to entice vault users in order to put stuff in and, and we have carefully sort of scope the risks around our existing sort of Vault engine. And it's sort of P P direct to consumer, right like that. That's the existing sort of mechanism for generating Dai. We have shimmed, in a PSM that it's a sort of another sort of, generation mechanism. And under the hood, it's kind of using the same bolt mechanism. And this is like yet another kind of like fault mechanism, if that gives you like, a good, like, mental place to sort of hang on this to understand like, what the d3 M is so. So you could almost imagine, you know, if we had like, removed all of our sort of permissionless direct to consumer vaults, that the d3 M would still function to generate Dai supply the same way the PSM would still function to generate Dai supply. So I don't know if that helps, like, give any additional concepts or like, gives everybody like, sort of a good mental model of, of what it is. Yeah, and then and, and then as far as like, I think I'm like, What are its use cases, I pretty much nailed a bunch. But right now, it's, it's primarily, I would say, a Dai generation tool to satiate insane amounts of Dai demand. Right. So we were looking, let's say Maker was looking for ways to try and add, like a lot more dye onto the market. And the PSM, obviously, like helped us with that sort of overpaid pressure to to generate Dai but it did it at the expense of ingesting a bunch of, you know, centralized stable coins. And so the d3 M is was like an answer to generating Dai supply when there was a ton of Dai demand outside of the normal vaults engine. So that that's being very pedantic about about exactly what the d3 M is and the problems that it's solving. Brian McMichael 23:31 Can I get them like check your Okay, so yeah, what is a d3? M Ethereum is I would think of it maybe more like a push model for Dai than a pull model, where right now we have our traditional Baltz, we require someone to come to us and affirmatively generate Dai, where the the d3 M is something where we reach out to other protocols and generate Dai to, as Chris said, satisfy demand. So we know there's demand at in the most obvious use case, then is Ave we know there's demand there, because there's a high interest rate. So what we do with the existing v3, Emma's will set a target interest rate, which is for the last, you know, since we've been using it has been lower than the actual interest rate there. And so when the d3 M sees that the interest at Ave is higher than our target, it can generate Dai to drive down that rate at Ave which which there's Dai supply on the market and it earns us a little bit of of yield. This is risky at a smart contract level and kind of a community level because we're exposing ourselves then to these external protocols, smart contracts issues multisig issues. Chain you know, the So obviously, for example, is exposed to chain links, but now we've got external Oracle risk. We don't know all the time what, what the composition or ability to control these protocols of multisig is. So it's, it's quite a bit riskier. Additionally, you know, the PE team has been building these things doesn't have expertise on these these external protocols. So it's their potentials that there are going to be some gotchas in the way we implement these. We're trying to mitigate those by having external protocol developers, you know, from from these other protocols take a look at this. But for example, there was a we didn't even notice the like channeling exposure in in Ave until just recently. The surplus buffer theoretically can limit us by, you know, if if, for example, think of a good example here, if if one of these tokens like ust is involved in a protocol, and it ends up taking down that protocol, the surplus profits should theoretically limit our losses up to to that amount. To answer the how should Maker voters think about this, you need to think that that it is a highly effective tool that we have, but it is quite a bit riskier from a technical and social perspective. And I've mentioned some of the risk scenarios. So if there's any questions or additions from the team, that would be helpful. Prose11 26:42 B P I'm actually very curious, how did how did our vase d3 behave when USC blew up? Niklas Kunkel 26:53 So they don't allow using USD as collateral, they only let people borrow USD. 27:00 Okay. I've got a couple of notes in the chat. But that Robert, you guys can raise? Robert 27:11 Yeah, I do. And I really appreciate the discussion with the features of the d3 M, where I'm struggling as is that being the recipient of collateral applications, where what tends to happen is is that we want to onboard this one onboard x as a d3 M. And unfortunately, that doesn't mean a lot. Because I need to go back and have to understand what's the business case for this particular application as an example. And so, you know, I've heard one example as it relates to rate stabilization. Okay, great. That sounds like it's one very specific use. And yes, this is generating Dai. So I get that. But then when I take a look at, okay, that's I look at like Ave, or compound. And that's one utilization of the d3 M or the one that's currently being developed. Then we have on the other side of that we have other projects that are proposed, for example, with true fi or maple finance, which is a little bit different. So I think the objectives of the utilization of the of the technology is, is it provides some similar benefits, but the use is different or not. So it would be great if we could talk from from that level versus the features on what does this actually do for the protocol. Now, if you're just telling me it generates Dai, that's great. But what I'm not hearing that I'm hearing rate stabilization, or potentially, like working with things that are hybrid real world assets. Brian McMichael 28:58 So, Robert, yeah, so under the hood, the way they have a d3 M in particular works is we are actually collateralizing the dye that we generate with a dye. So instead of being you know, over collateralized, like your traditional vaults, these are, we would say fully collateralized by the equivalent amount of a Dai under the hood. It's going to be a little bit different with some other protocols, like ideally, we're going to have some sort of token that we can put into the system and retain this. This full collateralization. But that might not always be the case. And we might have to do receipts or something like that, to make some of these situations work. Prose11 29:44 Yeah, maybe I'm going to sort of add to Robert's point because I have a really good idea of where he's coming from. So with normal collateral engineering, so the collateral engineering services group has taken over this sort of collateral prioritization spreadsheets. And there's, in those sheets, there's a number of these sort of like scored weighted attributes and risks of each collateral that we would add. And it seems to me that what we need to do is enumerate all of N R B P the particular benefits that a d3 could give us, whether it be Dai supply rate stabilization, maybe it gives us access to a collateral that we can't onboard, because it doesn't have a high enough debt ceiling or like, you know, it would be too expensive for Oracle's to to add it or something. So we enumerate all of the sort of like positives of the d3 M in its own sort of collateral onboarding sheet. I feel like we could then also enumerate all of the, of the risks of of those d3 items as well, to get an idea of their rankings. And also, then ultimately, to like help kind of review and steer governance towards whether or not we would want a d3. I don't I don't know if that. Like, if that's sort of the answer you were looking for Robert, but I feel like that is something that CEOs can really sink their teeth into. Robert 31:05 Yeah, because what happens is that we get however the collateral application might come through, it's specifying things on a level that we don't even even understand the business opportunity. Like, okay, what are we trying to accomplish? Or what is the actual collateral in question? Right, the utilization of d3 M might be applicable in some of those cases, and it might not be so so what I'm really suggesting is that for applicants that are that are coming in to the system, you know, to, you know, help us educate them, right, in order to do that, you know, one of the one of the key tools here would be is Krista to work with PE to, to get, you know, a summary of, here's a d3 M, here are the use cases, here's some potential areas where you will utilize this, here's where you wouldn't, here's where some risks are. And I realized we're still in development of of the d3 or at least the refactor of the d3 M. So I think that would be really helpful. One, Primoz Kordez 32:12 one other proposal I'd have coming from the risk angle is that so the tramps are really, you know, they have totally different risk profile than wolves. There's, there's financial risk that we have with waltz as well, or you might call it collateralization, the risk but then you have a ton of integration risk, right? So you're basically relying on external protocol on, you know, all that Brian man mentioned. So political risk, smart contract risk, all of that even their risk management basically. So and this part is actually really hard to quantify. It's not, you know, for us to look hard to calculate the risk premium of depositing dye to other like, we can do that for it a vault, but it's really hard for that, because for our because we need to evaluate, you know, what's the risk premium of channeling failure at RF for Dai market? Like that's, that's really hard. So we have all these tail risks that are unquantifiable. But you know, they're very devastating, right. And, you know, when it comes to surplus buffer debate, how, how high should exposure be versus surplus buffer, you know, normally should be, you know, it shouldn't be hard and surplus buffer because you still have this low probability, high severity event of losing everything created if there is an exploit, and adapt right now before the v3 is officially. So Ethereum market is in v3, you know, one token can actually record the entire market, you know, Dai markets, if if they want, so what what you would normally what, what I would propose to do is to have as many Dieter Rams as possible with some limited capacity, right? Because it's very unlikely, all of the protocols that the trams would be integrated into would fail at once, right? It's, you know, it could be correlated events, you know, it could be something which they did and everything fails, or chain link, you know, you of course, could have this, but, you know, even if channeling, you know, if you go to compounds, they have some fallback mechanisms. So, Oracle is, is probably safer it compounds. So you get some kind of is isolated R P risks. And that's why I do prefer more three more d3 apps with some limited exposure, it would be probably more catched with this than, you know, just one large exposure towards protocol, even if it sounds the safest, you know, it's not necessarily the case because you have this huge tail risks. Brian McMichael 34:46 We probably also need people monitoring every external protocol that we integrate with, just to keep track of their governance to see that you know, something's not going to come down the line that's going to run us well There's a lot of considerations around. Primoz Kordez 35:05 Yeah, I totally agree with this. We actually are, you know, we have an internal dashboard for us right now. And we're looking at, you know, every single market what's happening. So, you know, I tweeted last week about steak, tater, because I was afraid, you know, there's a bunch of Dai being borrowed from steak did, it appears that is much more than it was in the past, but we are still okay. From that perspective. But, yeah, basically, you need to, you know, know, every single decision they make. And, you know, the, the worst part is, even if you monitor good, and you can, you know, you know, you should react, it's, it's this, maybe one other bad part of this module is that you cannot react really fast, right? You know, if you want to wind down the whole thing, it takes time, you need to have Maker votes for it. And you know, it's all public, and it gets discussed, and you might be lost out of the market. And that's another reason why you don't want to be too exposed to to particular market, right, in terms of, you know, Shadow the poll, but also in terms of the of the depths ceiling. Prose11 36:17 It occurred to me that whatever we integrate with, with a d3 M, you also need a sort of incentive alignment between their sort of token holders and MKR token holders, right. So there, there's a huge incentive for another protocol to want us to be injecting dye into their market, if it means that MKR holders are footing the bill for any losses in their implementation. Right, they need to, they need to feel the same, like pain and suffering from a loss, or their token holders do that MKR holders would only then are we sort of, you know, separate but working in the same direction. As far as like risks go. Yeah, I'm wondering if it would be instructive to get into sort of specific risks. I mean, because we've we've kind of been beating around the bush a little bit with like some of the like Ave risks that have come up in the last week or so? Does everybody feel like maybe we're at a point where we can talk about that? Or do we still want to sort of talk about things in the abstract? 37:30 About the emergency shutdown support that we have. So just wanted to mention that currently, there are like two kinds of things, we try to like code, specific scenarios that would allow permission that's shut down. For so for example, if AVI or compound changes, some kind of parameters, like the rate model, contract or stuff like that, then we support like, immediate B P P thing immediately and winning over makers position. So that's like one thing that we have currently. And the other thing is just having like, a support for shutdown without the governance delay, that Maker usually has. So that's like the other thing. Prose11 38:31 Yeah, in in our new version, actually, I don't know if that doesn't matter, it's gonna say there's a, there's a bit of like a diagram that kind of can give you an idea of the model. But in the new version, there's sort of a hub that manages all of the d3 items. And then each d3 is sort of has its own kind of plan and pool. And I and on the plan in the pool, there's, there's a function that we can call from the hub, that that gives us the ability to potentially reach into these other protocols and look if certain conditions have been hit and prevent them from sort of winding up or unwinding. And it may also make sense for us to have, you know, maybe a call that does sort of an emergency on line. So it sort of, in a way, it's like an instant access module that minimally sort of either circuit breaks the function of it, or maybe could unwind it. So that's one sort of risk mitigation tool in general that that we may have across these D three arms, but it requires us to identify the technical need and these other protocols, which is, I mean, just a phenomenal amount of work if you consider that PE or CS needs to be digging deep or Oracle's in this case needs to be digging deep into these like entirely different protocols, and then understanding all of their sort of technical rules. It's, it's a big ask. So Niklas Kunkel 40:04 that's one of the reasons why I'm not very bullish on this thesis that pretty much kind of gave earlier when he was like, oh, we should just have like a bunch of d3 M's, right. And that way, okay, if one of them blows up, right, you know, at least we've been making enough returns from the other ones right to and you kind of get like a good sampling, right? But yeah, the the overhead for monitoring many, many, many D three M's is just is just so high, right. And if it's not something where you're really diligent, and by diligent, you know, means like really putting resources behind it, right, you're almost guaranteed to miss something I'd like so I'm definitely more of a fan of like, very small number of the three M's, right, that's just a concentrated bet on a specific partner, who has a specific stature in the market. And we think having a d3 M there serves us strategically, rather than every, you know, DeFi protocol coming at us and going like, yes, We'd like some Dai liquidity, thank you very much. Prose11 41:14 Maybe if they wrote their own d3 module, in a sort of sandbox way. There might be a middle ground there. But it's, yeah, like, in order for us to get any traction out of dye supply, or to solve those sort of critical business problems that Robert was talking about, it's going to require like an enormous amount of probably effort and ongoing monitoring, in order for us to have like a d3. And that's sizable enough, that it justifies all that expense. So and then even then, I think there's just risks that we can't mitigate, that we just have to accept in that model. And that's very unnerving to think about, especially like, you know, last week, when the markets were really tearing down because of Terra Luna, and that, and then USD T looked like it was D pegging. And then, then this whole, like F staked up, it's like, all of a sudden, like, multiple P N P assumptions you make across the ecosystem start to fall apart. And, and on my mind, all I can think is like, is are they going to be okay, here, you know, so it gets very difficult when we have our own or its own sort of risk factors that need to mitigate slash accept. Niklas Kunkel 42:29 And a lot of these things are going to fail in unison, right? Like, like, say, there's a some kind of chain link hack, or they go rogue and malicious, right? All of DeFi starts falling apart, because they all use chain link Oracle. So if you have a d3, MTA, you know, a bunch of protocols, right? You're gonna be you're gonna be hit on all of them, right? It's not like you have one blow up, and you're, you know, you're okay. Right? So, depending on how these things grow, right, you know, we can say, Oh, well, Maker, you know, we have our own Oracle's, we're not susceptible to chainlink yet, but through the d3 M, you are, right. And so like, like, what is the inflection point? Right? Where, okay, for a little bit of Dai, you know, we're like, okay, it's not really worth the resources. We're sinking into making this d3 M, right. We're not making enough money. For a medium amount of dye, we might say, okay, the risk reward is good. We're making a good chunk of money. But we're also not overexposed if it blows up. Right, versus, oh, we're making a ton of money off this. But now we are adding this existential risk. Right. And at that point, there's, there's there's this inflection point. And I'm not necessarily like sure where it is, but but it does exist, and we have to be kind of conscientious of of that. Sorry, yeah, that's all right. Go ahead. Prose11 44:05 Yeah, I was. I mean, unless there's another I was gonna move into like, maybe one of the specific threads because it seemed like a good segue from what Nick was just talking about. Yeah. But yeah, so. So with Ave in this case, first, let's think about like the vault engine. So our typical vault engine, we put just a enormous amount of effort into making sure that we have very safe Oracle's for the vault engine and in a significant expense into publishing those prices. We've put a lot of work and a lot of thinking a lot of risk modeling into that. And that has led us I mean, we could have easily switched over to using chain link Oracle's years ago, but we decided back during a bear market back when you know, we were trying to be safe and secure. That this is definitely is the sort of safer, more prudent approach to take to control our own Oracle's now, with when we accept a sort of B three M that has its own sort of Dai generation mechanics, like an Ave, for example. It has, it has chainlink, Oracle's and, and so over the last couple of weeks, like two sort of major, like, Oracle risks have presented themselves from the audit perspective. The first one was published by Chris Bielek. And it was basically the the chain link, Oracle's themselves have a contract that I'm trying to remember the name of the begin with an A. Anyway, they're aggregator, I think is what it was. So they're aggregators basically protected by a three of 20, multi SIG, and that three of 20 multisig, you can propose a new aggregator and then promote the aggregator and then do a price update. And there's no time lock in that in that process. So in theory, you could do that entire update, all in sort of one transaction. And instantly change the prices across all of these different DeFi protocols, which you can imagine. So that means there's three people out of a group of 20 people across the world that basically you can like, turn DeFi to ash over, over like one transaction. And prior to the d3 out and have a, we would be the king of the ash pile, right. But with our current sort of debt ceiling exposure, we are now part of the ash pile, right? So or at least MKR holders have to back literally hundreds of millions of Dai in this case, right? So or have them back die. And so N P that's the sort of first risk. And I have a like light, I went out to their community. And I'm trying to like get one of the devs to sort of talk to me about possibly adding a time lock or a way to like signal that this is happening so that we can mitigate it in our new sort of modules. So that we can see that, you know, that that this aggregator of something has been changed and the price, you know, the entire like price aggregators about to flip over. And we can just say, okay, like, like, stop these modules or unwind them immediately. But I haven't I haven't spoken to anybody there yet. And then so the second one emerged from the sort of Terra Luna problem, which was on I forget which chain and doe