Episode 171: December 9th, 2021

--- tags: G&R --- # Episode 171: December 9th, 2021 ## Agenda - [00:00](https://youtu.be/PjfAEZHdLEw): Introduction - [02:28](https://youtu.be/PjfAEZHdLEw?t=148): Votes and Polls - [05:54](https://youtu.be/PjfAEZHdLEw?t=354): MIPs Update - [10:14](https://youtu.be/PjfAEZHdLEw?t=614): Forum at a Glance - [14:24](https://youtu.be/PjfAEZHdLEw?t=864): Discussion - Mandated Actors - [1:06:06](https://youtu.be/PjfAEZHdLEw?t=3966): Cpen Discussion ## Video <https://youtu.be/PjfAEZHdLEw> ## Introduction ### Agenda and Preamble #### Payton Rose [00:00](https://youtu.be/PjfAEZHdLEw) - Hello to everyone. Welcome to Scientific Governance and Risk episode #171 at MakerDAO. My name is Payton, and I am one of the governance facilitators. I will be guiding our way through the meeting. I appreciate everyone who joined. We are with a bunch of awesome Maker people, both those who work and contribute to the protocol and those who are just interested generally in Maker governance. Thank you all for being here. Also, those who might be watching this later on YouTube, we appreciate you. I will go over a couple of ground rules before getting things kicked off. As you probably notice, this is being recorded. Be mindful of that. Try not to speak over one another. If you have questions, we encourage them, and if there is a lull in the conversation, please feel free to hop on the mic. If you are not comfortable, or for whatever reason, you would prefer not to do so, feel free to drop your comments or questions in the sidebar. We will try to work them in when conversationally appropriate. - We have got a bit of an agenda to get through. I am happy to present that today. We will start with our governance roundup, going over the votes and what is happening in the forums, and our Maker improvement proposals. We have got a couple of neat discussion topics today from our community discussion concerning the authority and powers of mandated actors and defining that and talking about what it is now, what it could be in the future, and what we think it should be. Then, handling Dai recovery situations. This topic that anyone who spent a long time on our Discord might know is a common issue; people accidentally send Dai to the contract address when they need to be sending it to a friend. It can also happen on an institutional scale, which we might get into later. Finally, we will go into the discussion section if there is any time left over. If you drop any questions or comments that may not relate to what we are talking about now, if we have time, we will circle back and get to them later. ## General Updates ### Votes #### Payton Rose [02:28](https://youtu.be/PjfAEZHdLEw?t=148) *Executive:* - Last Week’s executive - PASSED & EXECUTED - Delegate Compensation Payments - OMC and `dust` Parameter Changes - GUNIUSDCDAI-2 Onboarding - COM-001 Budget Transfer - Executive proposal up for vote tomorrow. Will include: - Adjusting wstETH-A and MATIC-A Parameters - Changing MKR Vesting Source - (Unconfirmed) MegaPoker Updates *Polls:* - 2 Weekly Polls - PASSED - [Adjust wstETH-A System Parameters](https://vote.makerdao.com/polling/QmYuK441?network=mainnet#poll-detail) - [Adjust MATIC-A System Parameters](https://vote.makerdao.com/polling/QmdzwZyS?network=mainnet#poll-detail) - 2 Active Greenlight Polls - [aUST (Anchor TerraUSD)](https://vote.makerdao.com/polling/QmPBRpCW?network=mainnet#poll-detail) - [G-UNIv3-DAIUSDP (Gelato UniswapV3 DAI-USDP LP Token)](https://vote.makerdao.com/polling/QmVTiToU?network=mainnet#poll-detail) ### MIPs #### Pablo [05:54](https://youtu.be/PjfAEZHdLEw?t=354) [Weekly MIPs Update #65](https://forum.makerdao.com/t/weekly-mips-update-65/12033) ![MIPs Updates](https://i.imgur.com/yOBnOzL.png) ![Proposals galore](https://i.imgur.com/YwUFcud.png) ![CORE UNIT PROPOSAL SETS](https://i.imgur.com/jBW0gll.png) ![MIPs](https://i.imgur.com/RPv12rA.png) ![BUDGETS](https://i.imgur.com/eIjh4t4.png) ![OTHERS](https://i.imgur.com/uLM2f1U.png) ![That's all for MIPs](https://i.imgur.com/3IX9ORo.png) ### Forum at a Glance #### Artem Gordon [10:14](https://youtu.be/PjfAEZHdLEw?t=614) Post: [Forum at a Glance: December 2 - 9](https://forum.makerdao.com/t/forum-at-a-glance-december-2-9-2021/12119) Video: [Forum at a Glance](https://www.youtube.com/watch?v=e9E7NDumWQE) ## Team-led Discussions ### **Mandated Actors** [14:24](https://youtu.be/PjfAEZHdLEw?t=864) ![MANDATED ACTORS](https://i.imgur.com/tzd59ve.png) - David Utrobin: Generally speaking, to decide which topics we will cover during this G&R call, we have an extensive list of current events and current topics, and between the mandated actors, we choose what we think makes sense. We are open to hearing suggestions from anybody on what topics they would like to discuss or hear on these calls. Please feel free to talk to us on our public chat channels and give us good suggestions. - This week, we settled on talking about mandated actors, specifically, sharing this idea that mandated actors have also delegated authority: special vested powers depending on the type of mandated actor they are, whether they are the facilitator of one core unit or another. Because different core units have different functions, their facilitators have different particular types of authority in some cases. - I created a few prompt slides. We will talk about the mandated actor role in deal negotiation, the importance of authority in partner relations, and what can help various mandated actors be more effective at producing impact and driving business growth at Maker. [16:24 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=984) - One area this came up with and talking about the mandate actors was our institutional vaults and the ability to make deals as it is hard to with the DAO. The DAO has to approve the deal. When you are in negotiations, you lose much leverage from not saying; this is what I am offering, and having to say; this is what I hope to offer if the community supports it. That is some context on the origin of how this topic came to be. [17:05 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=1025) ![Prompt 1](https://i.imgur.com/w2wWuBd.png) ![Prompt 2](https://i.imgur.com/PTPVOnb.png) - Prompt 1 was digging into the types of delegated authority, examining this whole topic of what direct authority should be given to mandated actors and what needs to follow our traditional governance process. For example, how much authority within a deal should a mandated actor have, compared to what has to be signed-off on it at the end, by governance. Also, what sort of checks are in place to prevent abuse of special power privileges, and how do we support transparency around these permissions. Because when you give people permission within a business, it gives them extraordinary power and a leverage point. We, as a DAO, are very mindful about who holds power and how we distribute it in a way where there is a healthy amount of balance throughout. That ultimately leads us to make wise and good decisions and not blunder and have people abusing their powers. I am happy to stop ranting and ask Nadia to come and speak to us. Since you are the facilitator of the Growth CU and you are on the frontlines with many partners. #### Types of Delegate Authority [18:37 - Nadia Alvarez](https://youtu.be/PjfAEZHdLEw?t=1117) - At Maker, we need to redefine how we will manage the onboarding and all the process of Institutional Vaults and regarding the relationship with the partners who agree on having a Vault with us. Institutional Vaults is a nifty tool and product for Maker because, thanks to that concept, we have attracted a lot of crypto collateral into the protocol. As Séb stated in his latest post, Nexo passed from minting 200 million to 1.5 billion thanks to that idea. We talked about this with Celsius, and they got excited about that and started using Maker Vaults again. As a product, as a concept is amazing. However, things are messy when you start creating something permissionless, like a product or a service that you offer to individuals. It is not easy because, in Growth, we are the face of Maker. We represent Maker at a point with our partners. We talk to them, tell them about the Institutional Vaults, and explain what it is all about. Right now, we have many institutions interested in using this product. It is a long process. We have to explain many things to these institutions. The concept of being a DAO is challenging to explain to a legal and compliance team. Still, they are very interested in trying out this institutional product because Maker works. Then we have to negotiate the terms of the proposal with the institutions, and we are at a disadvantage because we are a DAO. We have many people who want to participate in deciding about the terms of the agreement against one entity who can decide fast, and they are expecting answers as soon as possible. I took one approach because I did not have all the knowledge to decide the best parameters for this, to create a working group with Risk, Protocol Engineering, and Ace from Real World Finance. We work on these agreements, and we propose risk parameters. Until we posted in the forum, the institution had many discussions to agree on the terms. Sometimes it would be great if the working groups could decide and not wait until what the community thinks or what the Maker holders want. I do not want to have all the power of decision. Working in a group is the right approach to do that. We need to understand how the protocol is affected from different perspectives. That is why people are from different CUs. That is an example on the Institutional Vaults side. We did not want to discuss this particular case in the mandated actors' call because we see this in different cases. We are having a liquid problem with Dai on other networks, and we need a market Maker and a lot of Dai to deploy on those networks. What should we do? Going through the MIP path will take a long time, and we need a quick solution. It is an open question. I brought these two cases because those are the ones that are on the top of my head. Every CU faces a lot of these situations where they have to act fast, make decisions, but at the same time, we want to talk with the community and ask for their feedback. If we do that and go through a MIP, that takes time. The first approach or solution will be to have these working groups. That is something that we have been organizing in the last month. We want to know what the community thinks about this. It will be great if other CU wants to jump in and talk about their particular experience. [Juan - 25:06](https://youtu.be/PjfAEZHdLEw?t=1506) - I wanted to add that Joey Santoro from Fei Protocol published today, a good article about, I forgot the exact words, I think it is optimistic governance. Something where DAO gives more power to some actors, and then you have all the checks and balances in place. That is also something that I believe that Rune was bringing up in the governance channel. It is something worth checking out if anyone is interested in this topic. It is very hard to agree as a whole. Again, I believe Rune promised a good document in weeks. - Looking forward to that and seeing what he brings. - David Utrobin: What are the main takeaways that you had from that article? What were the most significant points? - I do not want to focus on the article per se. What Nadia is saying is true. Sometimes you need the speed and the authority to have judgment calls in different situations. At the same time, because it is a DAO and the MKR holders need to keep the power, you need to add the check and balances to ensure that this power is not abused and that there are different mechanisms for different situations. It is something that Maker needs to work on. We are still very green in that sense. In the immature sense, not in the clean sense, unfortunately. #### Deal Negotiations and Partner Relations [27:03 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=1623) - For those that are unfamiliar, I am thinking of the way our governance processes now. The protocol stays the same unless we pass a proposal through our governance that indicates that we want to change. On the other side, Optimistic assumes that a proposal will pass and go through unless a veto occurs. In theory, you could get more done with less governance overhead while still having checks and balances in place. There are benefits and drawbacks to doing both ways. We do not have to have all or nothing either. We could set up certain opportunistic ways of approaching specific situations and keeping the standard model for others. Kind of a rich governance topic that gets me excited. I hope it is not too boring for others. [27:57 - Nikolai](https://youtu.be/PjfAEZHdLEw?t=1677) - With these sorts of private negotiations and how companies do it, it is private for a reason. It gives a huge advantage in negotiating power if you work out what you are willing to offer, what you are willing to take, and all that kind of stuff behind the scenes. The problem with the DAO is that we have to present this in public. It is required to get the authority to proceed with things, but it is a disadvantage because people we are negotiating with can read this. They can understand what we are offering, what we are willing to take, and all these kinds of things. It is something that is not well suited to being out in public. We need to think about this. It will come up again and again with these Counterparty negotiations. [28:51 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=1731) - My question is specifically for Counterparty negotiations, Institutional Vaults, and deals with partners that need to be negotiated behind closed doors. If a mandated actor were given some special authoritative permission to make deals and engage in those talks, what kind of checks would be designed around that to prevent a bad deal or bad terms from going through? Is there any real-world example of something like this? - Nikolai: At the very least, we can evaluate what the end proposal is. Ultimately, something needs to go to governance. But maybe the in-between is something that needs to have a little bit more autonomy. - Ultimately, the solution sign-off still has to happen through governance. That way, governance always has a final veto on any deal whereas the whole bulk of the deal. Is that what we have now? - Nikolai: Not quite. I do not know exactly what it looks like. I am still working this out. What the correct balance is in my head. - Juan: Governance needs to approve each deal compared to Nadia, or any given actor could bring this deal to the table, and it is a done deal. We could reject it or even revoke this power from them. Until something happens, they have the power to negotiate that. Right now, that is not at all what we have. Governance is even approving every single MIP6 for RWA that is popping up. There is no way a delegate can have the bandwidth to analyze those, let alone a voter properly. It is something that we definitely should be looking into. Unfortunately, I do not have the solution, but we will think about something. [31:22 - Nadia Alvarez](https://youtu.be/PjfAEZHdLEw?t=1882) - Yes. It is exactly what Juan said. For example, the problem we had with the Nexo deal was when we agreed with Nexo after discussing it internally with other CUs. Then with Nexo, we had to pass through the forum, ask for everyone's opinion, and create a signal request and follow all the steps. It was good. We got much excellent feedback on things that we did not think of in the first place, and now we are incorporating them in the next Institutional Vaults. However, it took time. It was great that the community was super receptive. You all understood that we should give it a try. If the community rejected that agreement, it was going to be complicated. Still, the possibility of being rejected and voting no for the signal request existed. That is the first problem. The second problem is the time we spend on that validation from the community. We spend many days trying to get an agreement with Nexo. Then, we have to, at least, have two weeks with the community to get feedback from them and wait until it is approved. Those are the problems, and we are trying to figure out how to solve those challenges. [33:29 - Robert](https://youtu.be/PjfAEZHdLEw?t=2009) - My perspective on this and for consideration to the community. We have an extensive vetting process to approve any CU facilitator as a mandated actor. When we say yes to a CU or yes to a facilitator, I hear the community say that we want you to do your job based upon the mandate you have outlined for your CU. Inside that, there is this: help me get my job done. You have set me up to do something very specific inside the DAO, and for me to do my job, I need a little bit of leeway. I am not asking for carte blanche. I am simply saying that there are some parameters that I know that I can work within. If it exceeds those parameters and goes outside, we go back to governance to ensure that we agree. For example, to build on what Nikolai was saying, one of the ways in the Collateral Engineering Services CU is that we have an enablement strategy. Very shortly, we want to reduce the friction for onboarding collateral. Suppose we have to run through the governance process every single time for a piece of collateral. In that case, that could be quite laborious, and it might impede getting collateral onboarding into the system. If we have some parameters that we can agree to say: hey, up to this debt ceiling, or up to whatever might look like from a risk perspective as well, that we can work within and then go check back in with governance when we need to exceed those parameters. I am asking the community, and maybe we need to be more precise as mandated actors. I am asking the community to place more authority on those individuals that have been designated mandate or actors and allow them to do their jobs. If we are not, we already have a process as we are seeing now to remove facilitators. I just wanted to share some perspective there. As it relates, and again for your consideration. [35:48 - Ultra Schuppi](https://youtu.be/PjfAEZHdLEw?t=2148) - I envision that we stop doing this on repeatable stuff. Institutional Vaults are something that we envision that will happen many times in the future. What we need to do, and this is something where we need to be a bit slower probably than what we would like to have in the actual deal-making process, is that we clarify and we need to be flexible on that, what the dimensions are that we need to cover debt ceiling spread to the average stability fee. Those working groups need to find that out. Then those working groups can propose boundaries, and within those boundaries, they can have total carte blanche. If we agree on some boundary for debt ceiling and stability fee spread, and Growth wants to close a deal, this can go directly into an on-chain executive. Assuming we follow the procedure before going through the forum, going through an on-chain poll. If we have this, we can even have a carte blanche. We do not need to be super slow on the whole process. It needs to be super clear what the dimensions and variables are and the boundaries you want to set for, essentially, the autonomy of the mandated actors. It is probably best to start with something happening many times, like Institutional Vaults. We will learn along with this and how to proceed with other things where we need more autonomy for the mandated actors. [37:49 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=2269) - There are several mechanisms and different ways of implementation. Through executives, we can delegate certain on-chain authorities to multisigs. That is a tool we have in our pocket, and others should be alluded to with going direct to an executive. That is a power we already give our mandated actors to do in certain situations, to a certain extent, depending on the urgency or emergency of the situation. It is a conversation about how the authority and the privileges are not clearly defined. Mandated actors have, in some cases, broad authority to recommend or reject proposals that they deem unsafe, and they put their job on the line when they do so. They could be outvoted for doing it earnestly or with bias. The system as a whole, we have yet to define clear guard rails as Tim and Robert have said that would make it easier and much more apparent to the mandated actors what they are allowed to do and what authority the governance has seated. That is probably enough talking for me. I want to be mindful of our time. We have another topic to get to, but this one is juicy. [40:00 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=2400) - There needs to be a single place where people can look for special authority to audit, like who has what authority across various situations. To deal with negotiation and who has authority over Oracle feed relationships or whitelisting privileges. It could be anything. As @monkey.irish alluded to earlier, our current expectations are set by the MIP41 and 39, and it is the actual mandate. However, the mandate and the facilitator proposals are the two things that inform us of what those authorities are. I am getting from this conversation that future MIP authors or revisions should aim to be a lot more explicit when explaining the authority they are requesting. I did not mean that the MIP set itself, but within specific CUs and MIP sets. I know that certain CUs and their mandates outline specific authority. [41:44 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=2504) - It is interesting because, with a lot of the MIPs and our bylaws, they are essentially soft authority. There are no on-chain powers unless explicitly asked for. That is a part of where the confusion and gray areas come in. - We have talked about this quite a bit. Feel free to keep the conversation going in the forums and on the chat. Very exciting to engage there. ### **Handling Dai Recovery** [42:17](https://youtu.be/PjfAEZHdLEw?t=2537) ![HANDLING DAI RECOVERY](https://i.imgur.com/AYrxvQm.png) [42:29 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=2549) - I want PE to come in and weigh in. I want two fronts here. In handling Dai recovery situations, and in general handling user crises or third party crises, I am curious about how we as a DAO approach this topic. There is a huge benefit to helping our users in this way, but there is also a trade-off of bandwidth—number one. If you can help the user, whether they are big or small, so far, the culture has been helpful. If we can, we do. I am curious about the volume of help requests we, PE or TechOps, or whoever has gotten over the last year. How do we handle these requests? Are there cases where we should not handle requests? That is the basis for the discussion. [43:44 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=2624) - I will kick it off by describing. The Dai token is decentralized and censorship-resistant. Governance has no powers on that token. Governance has an interesting amount of power over creating new Dai. For now, the most common cause is that people are taking Dai. Due to several UI or UX problems, they tend to get used to looking at a normal Ethereum transaction seeing the to and from, and then they usually pick the from address to send something to, but the from address, in this case, is the Dai token. Then they think they are sending Dai to their friend when they have sent Dai directly to the Dai token itself. We debated whether or not we would have that kind of protection on Dai when we were originally launching it. - Because of that socialized gas costs, we decided not to prevent people from sending Dai to the contract. Since then, we have come up with other schemes that would work. But it is too late because Dai is already deployed. That is the general problem. Now there is all of this Dai accumulating in the Dai contract. In theory, this Dai can be sent to other contracts, but these are completely incapable of sending it back. I think people have sent Dai to other tokens sometimes. We can probably show that there are many Dai locked in several contracts. We can use a trick to remit this style, though it would take greater protocol changes. The trick is similar to upgrading from single collateral Dai to Dai or MCD. We would instantiate some dummy token type that probably was like dead Dai or something like that. So it is signified that it was a representative of this dead Dai. Let us wave hands and say that we would figure out how we would determine whether or not someone has a claim on dead Dai. Maybe there are on-chain mechanisms that we do permissionless. In other cases, governance may need to intervene and periodically anoint this. Once we have that dead Dai, we can mint Dai against it and give it to the original person that had lost their Dai. The downside of this is that this is the term I constantly have to say; things get a little tricky. With this collateral type, you would not want to hand out dead Dai components to everybody because there would be no way to redeem that dead Dai. You would need to scrape this special collateral type out of the accounting not to get rolled up like total Dai in an ES. We reach the end contract, then ES would work with dead Dai, and in theory, we could have a way to give Dai back to all the people that end up losing their Dai. Then we can discuss if we periodically do it around the holidays, and, once a year, as a sort of holiday gift to everybody. We end up giving them their locked Dai back. It would be too intractable for the DAO to be dealing with on a case-by-case basis. There probably needs to be some greater aggregator of a bunch of this. Unless we can do it permissionless, we may do it with the Dai token itself. That is the general case. There is an institutional case that is driving us right now. I do not think it is subject to the same problems. Maybe Sam or someone else would want to talk about that. Maybe we should discuss the general recovery of Dai that gets sent to locked contracts first, and then we can take the second more specific case after. [48:56 - Brian McMichael](https://youtu.be/PjfAEZHdLEw?t=2936) - There is also the issue of how do we determine who sent these. We would need a customer service CU or something like that to process and verify that the people are the people who sent them. Following back, the original sender that had sent to the contract may mean that we would end up sending these tokens to a different contract that had sent these, and it would be locked there because it could not process tokens itself. There are a lot of tricky things about it. It sounds like we can send this back to people we do not always know, and it is going to be a bear to verify who these people were. [49:49 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=2989) - Good points. Maybe because I am an admin, I imagine a number of the PE and other people get it too, but I am constantly in flux with DMS about: I sent my Dai here, what can you do, and that sort of thing. Even though we are not addressing it, it still requires much cognitive load. Do we want to talk about the institutional case, Sam? - David Utrobin: It has not been processed in the forum yet. [50:25 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=3025) - This almost happened. I got pulled into one of these war rooms on a hacked protocol a while back. I cannot remember which protocol. Millions and millions of Dai were potentially locked in this thing. It looked like we may end up having to have dealt with this back then. They found an owner key or something that had access to it and could move the funds. - Payton Rose: From the governance end, it is much easier for us to negotiate with protocols and big amounts of Dai and see the problem there. But we have to think about the precedent that sets if we are only returning Dai for some people and not others. There is an interesting governance question there. [51:22 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=3082) - For a 10 or 20 million Dai block, it makes sense to devote some special resources to help them get it back. What about the 380,000 Dai sent to the contract by 50 or 150 people who did it? 380k is a slight drop in the bucket compared to the amount of time, resources, and investment that has to happen to create a permanent solution. If sending to contract is a temporary problem, we expect wallets to block it off and things to be designed so you cannot do that. Does it make sense to handle it case by case? Perhaps do a yearly assessment and reimbursement of the smaller locks, even out of the surplus buffer. Would that be a cheaper way of solving the problem? - Christopher Mooney: One time a year, the small locks, we get a miracle distributor and allow this whole process to happen. It would be a good gift back to the community. - Someone: It would be good. They can spend $300 in gas to get their 200 bucks back. - In an ideal world, we do it through a push mechanism, where it just shows up, and we pay the gas. [53:15 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=3195) - What other user blunders and issues besides sending to contract addresses and accidentally sending to the wrong place have we experienced? Are there other cases where PE has stepped in to help a user? - Christopher Mooney: The Arbitrarum Oasis thing. It was not Dai. It was collateral. It was a very happy accident that we were able to fix. We controlled the address that deployed the original ds proxy factory. We could reproduce that same chain of hashes that led to the same contract where the funds ended. We were able to recover that user's funds. That is one other big example. We put the working on that. It is a lot of work. We want to do this as efficiently as possible. - Sam: For L2 DAI, we can manually pull that out. L1 DAI is sitting in an escrow contract in governance's full control. If somebody burns their L2 DAI, we can potentially recover that without any special mechanism. - Christopher Mooney: Right, if it is the burn mechanism. Sam is talking about a hypothetical case worth paying attention to. What happens, and this is for anybody on PE if you want to conjecture, if it is not the burn function called but rather Dai that gets sent in locked in a contract on L2? Do we have a recovery path for Dai like that? - Sam: On L2 DAI, we blocked sending to the contract itself and addressed zero. It depends if they are accidentally sending it to some other contract. It has to be provably lost. - Brian McMichael: He meant that if L2 freezes up and Dai is irrecoverable because everything on the chain is lost, can we recover that to settle the balance on the mainnet? - Someone: That should be super easy, right? We could recover it in theory because we have got all the Dai in the bridge. - Someone: It cannot get locked because the whole point of L2s is that you can operate even when the L2 goes down. People could exit themselves. - Brian McMichael: What about certain sidechains like Matic or xDai? - L2 is not sidechains. [56:27 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=3387) - If there is an incorrect state transition and no fraud-proof is submitted within the seven-day window, you cannot make any assessments about the state if that does not happen. The roll-up is dead in that case. Everything is lost. - Brian McMichael: We have to assume that for any secondary solution, any wormhole solution that can be completely nuked? [57:04 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=3424) - I was talking about the case of somebody aping into a buggy protocol on an L2 and getting their Dai stuck in that protocol, where we can probably say it is stuck in that protocol on L2. The downside is that the Dai token on L2 would have a total token balance. It should match the bridge balance. If we move Dai out of the bridge to give it back to users on L1, that is no longer an invariant that would end upholding. There is some interesting accounting there. [58:06 - Paper](https://youtu.be/PjfAEZHdLEw?t=3486) - Does that mean that law enforcement agencies can force us to act upon Dai on Layer2s? - Christopher Mooney: Paper's question is: Can law enforcement force governance to act in any way possible? Can they compel governance to do anything? - Sam: That is what the ESM is for. If an action went through where the law enforcement is just going to start stealing people's stuff, that would be hard to do. They would have to get a majority vote and then get it through, and then ESM would not be triggered. It seems unlikely. There are lots of worse things you can do. [58:56 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=3536) - Someone could attempt to drag Maker holders into that dispute. A government could try and get Maker holders to act. Since Maker holders are a swarm of anonymous token holders, anybody trying to force compulsory action would probably want to start with honey rather than vinegar. In the end, there are lots of protections against censorship and compulsion of MKR holders or in the protocol. That is why it is a DAO. It goes back to the first sort of discussion topic. It is way more efficient to run an organization and have business relationships as a major company. There is a reason corporations have evolved to have this sort of hierarchy structure. DAOs are our response in a way against centralization. They respond to certain regulatory climates and prevent actions in certain areas. That is why you have doubts. [1:00:24 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=3624) - That loops back to the first conversation about optimistic governance actions. In an optimistic situation, you could compel an actor. Then it would go through unless Maker holders took action. - Christopher Mooney: This is why in the optimistic case, you still need the DAO to have veto power over actions to prevent the optimistic case. It is maybe a talking point we did not discuss. Think of it like the DAO is a school of fish. Except that the school of fish can take the shape of whatever wants to be. When a predator attacks one of its appendages or even its core, the school of fish has some invasive predator evasion technique. As long as we always revert to this amorphous blob, a school of fish, we will be resistant to predators. Predators, in this case, maybe centralization. It may be censorship. I am not trying to make it sound like the DAO is a criminal organization. Just so everyone knows, the DAO is a criminal organization in North Korea, and the DAO is a criminal organization in China. There are regulatory regimes where the DAO would not be allowed, or let us say, a single company acting the way the DAO does would not be allowed to act. [1:01:53 - Someone](https://youtu.be/PjfAEZHdLEw?t=3713) - We also do not have the overwhelming majority of our customer base in those countries. It was hypothetical because it was new information to me. That is all. We do not have to take up all day going in circles about it. - Brian McMichael: We do not choose where our customers are either. - Christopher Mooney: That is the permissionless aspect. That is why it is important to be permissionless. [1:02:47 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=3767) - What would the veto power look like if Maker were to move towards a more optimistic governance model? Who would have it? Is it something like you need like 1000 MKR to veto? Because now you need to block the vote via your actual voting weight. It is not an efficient way to veto. What are the downsides of Maker were to do optimistic governance? - Christopher Mooney: Any decisions or actions that require an executive vote mean that we are not framed to do optimistic governance. MKR holders will always have to instantiate some change in the protocol in that particular case. If it is an arrangement outside of that or a parameter that we do not need governance for, maybe a pole with a low bar threshold or something for people to veto, that would probably be sufficient. [1:04:14 - Brian McMichael](https://youtu.be/PjfAEZHdLEw?t=3854) - We have to consider that Dai recovery can be good for the protocol. Not just the users. Having liquid Dai on the market is good for the peg. If somebody loses 10 million, 100 million, it might be in the DAO's interest to recover that just so that we do not get back into a situation where we are pushing the peg up because there is not enough Dai. [1:05:04 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=3904) Nikolai said an unintended consequence of having this recovery flow for Dai. In the traditional financial system, people can usually claim their credit card and get their money back if they go and spend their credit card on some shady website or something. We do not want to create this weird second-order effect where people are more likely to ape or not scrutinize where they are sending their Dai. As Marianna used to say, we do not want to incentivize sticking your Dai in crazy. #### Comms when Affecting Stakeholders ### Open Discussion [1:06:18 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=3978) - When we talk about gas funding, do we mean specifically for Oracles, or are we also talking about the gas rebate program for voters? - Payton Rose: It could also just be for CUs in general. Many of us have to use Eth for things, and we buy them. Long and I, for instance, when we submit polls, we have to use gas, and we refund ourselves in Dai, which we then have to swap. We spend a decent percentage of that chunk just for the transaction fee to swap it. It feels like there would be a more efficient way to do that. I do not know what aspect of the chat was called for. If someone wants to redirect your flow here, please. [1:07:05 - Someone](https://youtu.be/PjfAEZHdLEw?t=4025) - Some of us had been interested, maybe trying to separate the Oracle gas proposal into two. One for the actual Oracle gas that is looking forward and one for figuring out this whole amount that would be repaid to the foundation, considering that it covers a period that has not even completely elapsed yet. So anyway, I just wanted to make maybe that I can get the ball rolling. I do not know if MakerMan is listening. - Payton Rose: From the governance perspective, the person proposing would have to divide it up and do two separate ones. [1:07:57 - Someone](https://youtu.be/PjfAEZHdLEw?t=4077) - Since MakerMan does not have a mic. Some of the concerns I had, were number one: getting some documentation, not just about costs, because that is a pain in the butt, and that does need to be done probably for 3 million. However, who is the Counterparty? What were the terms? We probably should have been thinking a little earlier about where the gas money is coming from. I would like to see a little more detail about all the specifics. [1:08:39 - David Utrobin](https://youtu.be/PjfAEZHdLEw?t=4119) - I want to note that Nik Kunkel is on vacation and is not on the call. Mark Andre from the Oracle Team is here. Maybe somebody else from the team. [1:09:09 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=4149) - I see a comment on the sidebar. He is going to ask, like Paper, to break the costs up from retroactive and forward. Also, he is trying to see who is being paid when this refund goes through. Nik is not here. He is the proposal author, and the way our system works is that who is proposing has the right to, if it follows the rules, for the proposal to go on-chain. People can make competing proposals, which would be a fun governance situation to find themselves in before moving on multiple sets of the same object. [1:09:50 - Someone](https://youtu.be/PjfAEZHdLEw?t=4190) - The question of who the DAO would be paying is probably a good one. We should know to whom we will send 3 million. You wonder why is a nonprofit foundation exists to bootstrap Maker DAO, asking MakerDAO to pay $3 million. It is not to say that it is inappropriate, but I would like to know a little more, especially since, presumably, it is dissolving. And whatever surplus assets are, they are likely being moved into the old Maker ecosystem growth Holdings Company, a Cayman company, not like a foundation known as UDHC today to everybody. We would like to have a little more clarity about why the DAO needs to pay them $3 million. [1:10:58 - Someone](https://youtu.be/PjfAEZHdLEw?t=4258) - Was it a loan? What does it matter to whom it goes back? You get money, and you pay it back. [1:11:05 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=4265) - I have some more details about this, Paper. During the handoff stage, from the foundation to the community, it was unclear that Oracles would have the operating budget to continue to operate the Oracles. The foundation had made a gentlemen's arrangement with the Oracles' team and gave them enough funding for gas to pay for running the Oracles during that transitionary period. The gas calculations for Oracles were very difficult at that time. They did not have the actual amount that they were spending offhand. PE was easier to look at our deployer addresses and see how much gas it would cost. It was non-deterministic, and there are more addresses to look at for Oracles. They do have that calculation now. It is just a loan from the foundation and probably harkens back to the original topic, which is how much autonomy mandated actors or teams have here. Regardless of what the foundation does, the DAO should honor its debts. If the DAO was given a loan to run the Oracles, we should pay the loan back just so that people know that they can do business with us. However, I do think Paper brings up some interesting points. It goes back to that original discussion about exactly how much autonomy teams should have; taking on loan agreements that the DAO is responsible for might be a limit of that, or maybe there is a monetary limit. It may be a one-off anomaly because it was part of the transition. [1:13:05 - Someone](https://youtu.be/PjfAEZHdLEw?t=4385) - If gas was spent, then something probably should be owed. If a for-profit venture fund owns parts of lots of affiliated entities that the DAO pays money to, if they are part of this ecosystem, maybe there should be some kind of cost split. I would like to see a wider discussion. I am not saying we absolutely should not pay them 3 million. Still, it strikes me as a little weird that you take a foundation that existed to get Maker DAO up and running and then say, Maker DAO, pay us some money to help fund this private for-profit entity. [1:14:06 - Brian McMichael](https://youtu.be/PjfAEZHdLEw?t=4446) - I would suspect that that loan helped Oracles' CU leave the foundation sooner than they might have otherwise been able to without it. I am not defending it; it is probably an executive decision on the facilitator's part. With Mooney and others, it is a debt that we would have had to have paid either way. [1:14:35 - Someone](https://youtu.be/PjfAEZHdLEw?t=4475) - Oracles has a lot of excess budgets right now. Are we still using borrowed gas? - Someone: Gas prices are expensive, really expensive. [1:14:55 - Andrew Burban](https://youtu.be/PjfAEZHdLEw?t=4495) - Oracles need a huge buffer for the guests. The other night, I wanted to do one transaction that was $10,000 to do a single click, and I was like: I am going to pass on this right now. The safety factor of three makes sense to me. Any unused funds roll over. [1:15:19 - Someone](https://youtu.be/PjfAEZHdLEw?t=4519) - I would like to clarify one specific element. The Oracle gas costs are still being paid by the Maker Foundation today, and it will remain the case until the gas funding MIP gets passed. That is separate from the Oracle CU funding. [1:15:54 - Someone](https://youtu.be/PjfAEZHdLEw?t=4554) - I see 1.2 million in undrawn streamed Dai, and there is whatever the emergency fund was. - The CU cannot hire as many people as we would hope. There is a budget surplus. Eventually, we will be drawing some funds from the DssVest contract, but until we have spendings to do. The idea is that, at the end of the year, any surplus is returned or accredited, or I am not sure what the exact formula will be. The difference between our actual spent and what sits into the DssVest is that we did not hire as many contributors as we hoped. - Could that be used to pay for gas until our January vote and disbursement instead of continuing to borrow? [1:17:14 - Payton Rose](https://youtu.be/PjfAEZHdLEw?t=4634) - The other issue is requesting funds for specific purposes. You generally do not want to turn around and use it for different ones. [1:17:24 - Someone](https://youtu.be/PjfAEZHdLEw?t=4644) - The loan is interest-free, so it does not matter. - Paper: It seems that it is to be held over here. It is being bundled into a must-pass budget item that is coming up. [1:17:45 - Andrew Burban](https://youtu.be/PjfAEZHdLEw?t=4665) - I feel the real issue is that there was no transparency around the original establishment of this loan. It is not even clear who the counterparty was. We are saying it is the Maker foundation. But there is no agreement on the form of the DAO and the Maker Foundation, as specified by some legal entity agreeing, and there was no executive vote to approve the taking of this loan. It can be a one-off thing, but it would be good to have some documentation produced as to what the terms and agreements were in the first place. - Someone: Maybe it better to chat about this when Nik is here. [1:18:38 - Tim](https://youtu.be/PjfAEZHdLEw?t=4718) - If we take this discussion to break things out, it would be good to have the person bringing the proposal on that discussion. The second thing is a call for transparency that is valid and probably easily resolved. If we can get those terms established, it seems like it was an essential decision to make, and they just made it, and it was for the good of the protocol. It is best to assume that there is nothing malicious. Some things need to be more clear. Daylight clearing all hills and such. [1:19:31 - Christopher Mooney](https://youtu.be/PjfAEZHdLEw?t=4771) - I have one quick one on a different topic: there is a hard fork for Ethereum in the next couple of hours. [1:20:03 - Prose11](https://youtu.be/PjfAEZHdLEw?t=4803) - Prose11: Thank you to everyone for joining us today. We appreciate the discussion and all the participation. Let us keep it going in the forums and our Discord. There will be one more GNR call next week for 2021. I look forward to seeing you there, and we can wrap up the year in style. Thanks, everyone. [Suggestion Box](https://app.suggestionox.com/r/GovCallQs) ## Common Abbreviated Terms `CR`: Collateralization Ratio `DC`: Debt Ceiling `ES`: Emergency Shutdown `SF`: Stability Fee `DSR`: Dai Savings Rate `MIP`: Maker Improvement Proposal `OSM`: Oracle Security Module `LR`: Liquidation Ratio `RWA`: Real-World Asset `RWF`: Real-World Finance `SC`: Smart Contracts `Liq`: Liquidations `CU`: Core Unit ## Credits - Artem Gordon produced this summary. - Kunfu-po produced this summary. - Everyone who spoke and presented on the call, listed in the headers.