Other Presentations and Updates
Nik Kunkel
Liquidations and Uniswap LP Tokens
37:22
- I was just gonna talk about it, but now I’m kind of feeling like I should have made a presentation. Now that we’re getting deeper into the weeds with the onboarding process for the Uniswap LP tokens, one of the things that came up is the liquidations process. Currently, there are no really active markets for LP tokens where they’re being traded. If there are, they’re definitely not doing that much volume. This leads us to this thing where keepers, especially in the next version of liquidations, can kind of win in a single transaction. However, even to an extent right now, I think what they’re just doing is arbitraging. They’re just going to win this auction here and then immediately dump it somewhere to capture that spread. With Uniswap LP tokens, you don’t really have that. I think there needs to be a discussion in the community on how we want to handle this. If you’re doing it manually, you can just buy the LP tokens. You can take them apart into the components, and then you can sell the components. Still, keepers also need to know how to price LP tokens. While that is doable, it’s not entirely trivial to get the price of an LP token. Therefore, if we want to onboard Uniswap LP tokens, the keepers need to be aware of the challenges of participating in liquidations for those tokens. I want to encourage everyone to join the keeper’s channel on RocketChat. Anyone interested in the liquidation side of things, especially anyone running a keeper, can discuss in there as a community about how we want to handle this and how we want to go about it. I have a couple of different things in mind, probably a little bit too long to go into on this call, but it’s a topic that has to be covered.
- Chris: Yeah, I was just gonna say that we should create a library for both Python and Javascript that allows one to price an LP token. That way, people could just consume that library. Whatever keeper they would write, I think those are the two most popular languages for keepers. If I were to envision how they would handle it, this would be one of the ways. Also, in liquidations 2.0, there’s going to have to be some whole process for unwinding these. Those keepers who implement this will have to have a more sophisticated unwinding process if they’re going to arbitrage.
- Nik: Yeah, that’s the idea. Either on the keeper side or on the Maker protocol side, you want to break the LP tokens and the components to make them easier to handle. I’m leaning towards that the keeper side as the best way to do that. I’m also attempting to see if we can build up some kind of LP market in that way where we don’t actually end up having to do any weird breaking of LP tokens in the liquidation contracts. That would just be weird to make a bunch of exceptions in there and potentially riskier.
- LongForWisdom: Lev, I think you were trying to talk earlier. Did you have anything to add?
- Lev: Yeah, I just wanted to say that MIP 30 is potentially applicable in the future to Uniswap and SushiSwap LP token collateral types. It’s designed to be possible to adapt it to capture rewards on those two. I mean, Uniswap rewards have just ended. However, if those were to restart or start, or Sushiswap rewards are still a thing, I think that would really turbocharge those collateral types. Hence, it’s worth considering those two things. For example, most of the work for onboarding the LP tokens like this stuff just discussed liquidations and keepers. The price feeds would apply in the same way if they were being farmed.
- LongForWisdom: Yeah, Lev, I’m not entirely sure I caught all of that Lev because of the microphone, but perhaps you can comment on the thread.
- Chris: Yeah, Lev, if you could just leave your public key address, and we’ll all send you a little bit of DAI, get you a sweet new microphone.
- LongForWisdom: Cool.
- Chris: Yeah, with Zoom on Linux, it’s game over.
- LongForWisdom: Cool, does anyone else have thoughts about liquidations for Uniswap or LP tokens? It feels like the pricing issues are the primary concern because keepers should just redeem the underlying tokens from Uniswap or SushiSwap or whatever once the auction’s finished. They just need to know how much they’re worth.
- Chris: Yeah, our reference implementation of keepers don’t do any sort of market cycling. We put that on whoever the keeper or user is of that software to cycle their collateral back out. I think that would be fair to say for LP tokens, but if we could provide a reference implementation of pricing, that would be really helpful.
- LongForWisdom: They said something about Uniswap themselves or any other LP token providers have worked on, right? Because it seems something that’s more generally applicable rather than just…
- Chris: That’s a really good point. There may already be a library that prices these things, so we should look, and if there isn’t, maybe, we create a bitcoin grant or something to have someone make one.
- Juan: That’s a good idea. I was thinking if there’s a secondary market for the LP tokens that’s big enough, then you wouldn’t need to redeem them, just sell them.
- LongForWisdom: So, Uniswap has an implementation of the pricing. The keeper channel in RocketChat is probably the best place to discuss this further. If anyone has thoughts or is a keeper and has concerns or questions, please leave them in that channel. Anything else you want to see on this Nik before we move on?
- Nik: Yeah, I guess just that the action item here is to go to the keeper channel so we can discuss this more. This is not a situation where we talk about it once on the governance call as a one and done.
Charles St. Louis
Collateral Onboarding Handbook
48:13
Collateral Onboarding Handbook
- I’m going to share the form announcement in the chat. With the recent launch of the collateral onboarding handbook at collateral.makerdao.com, I’d love to just give an overview of the purpose, motivation, and action items that this handbook entails.
- The handbook contains all this great content. If you are new to the collateral onboarding process, begin with the getting started pages. The first page is an overview, which provides a high-level description of the entire collateral onboarding process over nine steps. It has an absolutely wonderful diagram from the DX team at the Foundation, giving all the little details of the process, including when things are posted to RocketChat and when things are on the forums, when votes happen, and so on.
- The next page provides an overview of the stakeholders involved in collateral onboarding. The main stakeholders are the Maker community, Maker integrators, the Domain Teams, MKR holders (AKA governors), and the collateral application proposers.
- The next major section details the community involvement guides. There are four main guides:
- The first one provides an overview of the coordination and communication process between the involved stakeholders, adding collateral types to the protocol, their interaction, communication, where they communicate, and so on.
- The next document talks about the collateral onboarding calls that Juan has been hosting over the past couple of months. This document provides the information of the calls, such as the details for when to join, how to join, and links to the previous calls if you are curious to see which collateral proposers have come on to do AMA’s and vouch for their collateral type.
- The next document talks about the collateral prioritization spreadsheet. It gives general information of why it was created, how it’s used, and also a link to the actual public spreadsheet for anyone to see.
- The last document details the funding process or grants process for this. It mainly highlights the goals of the initiative and how it incentivizes and rewards contributors for participating.
- The next four major sections are the governance process, the oracle’s Domain Team guide for all their work, the smart contracts guide, and the risk guide.
- The governance process goes into a little more detail than the overview page. The goal is to help understand the governance process from start to finish and learn how to coordinate and follow the necessary steps for successful collateral onboarding.
- As for the Oracle Domain guide, it’s a comprehensive guide teaching you about how the Oracle Domain works and what they do continuously. It also goes into details about how to onboard an Oracle using the MIP10 framework.
- As for the Smart Contracts Domain Team guide, it highlights what the Domain Team does. They do a lot of things outside of collateral onboarding, but this gives a thorough explanation of all the steps involved, such as how to create a token assessment, what resources to use, how to add collateral types to the Maker protocol code base, as well as all the requirements for testing and so on.
- The Risk Assessment helps understand the purpose of Risk Assessments when onboarding collateral and provides the data and methodology behind building these assessments.
- There’s a resource section for having a glossary of all the terms used throughout these guides and documents to have a go-to place to figure out all the key terms. Lastly is the resource list, which also collects all the links and resources throughout the guides in one go-to place.
- Frank Cruz: This is super awesome, Charles. Thank you to you and the rest of the team. My question pertains to RWA. I would imagine you’re going to expand this when we get a clearer picture of how to onboard a Real World Asset. I guess what I’m saying is if this is a little too early to push this to say, and I’m going to exaggerate a bit here, to tokenize the New York Jets? It may be a little too early, right?
- Charles: So, with the Real-World Asset progress, there are currently two for those who aren’t familiar with Matthew Rabinowitz and Centrifuge. As we go through those processes, I’ve been documenting them. Once we get to a state where we’re comfortable with the RWA onboarding, we’ll have a pretty extensive guide detailing the difference between the ERC-20 collateral onboarding or more crypto-related versus RWA. So, yeah, it’s definitely in progress, and it will be more formalized as we experience the current RWA onboard.
- Frank: All right, thank you.
- Charles: I would be keen to know about the NYJ you want to tokenize, though.
- LongForWisdom: Thank you for that Charles, I guess I would just like to quickly race through Charles’s point. This documentation feels like a huge step forwards in terms of decentralizing the collateral onboarding process. Saying that this is the first time we’ve actually had transparent processes laid down for how to actually go about doing Domain Assessments for each of the domains and what needs to happen has preached to me to onboard something. Therefore, it’s a huge step forward, in my opinion. Thank you, Charles, for helping put it together or for coordinating to putting it together.
- Charles: Yeah, I definitely want to give a shout-out to everyone involved. There were quite a bit of people working on this. By no means was it just myself. The integrations team at the Foundation, the DX team, and the Marcoms team, and some community contributors were involved. Thanks, everyone.
- LongForWisdom: Yeah, I know the Domain Teams themselves, so I obviously helped source all the content.
Christopher Mooney
Timeline for DsChief 1.2 Revealed
1:00:36
- This is a little bit more detail on the timeline. We’re currently in that code review first step. I know the timeline says the 13th of November and the week of the 16th, but we’re finishing up the code review right now. We just made some changes to the tests and whatnot, and we’ll probably merge that today or tomorrow. Then I’m hoping that tomorrow afternoon after we do the executive or some of us in the morning begin working on an executive to deploy DsChief 1.2 to Kovan. That’ll probably sit in review for the weekend. I’m hoping that we can deploy it to Kovan on Monday so that anybody who needs to integrate with DsChief 1.2, which means any UI and governance portal stuff, will be ready next week. Of course, next week is also a holiday in the US; it’s the week of Thanksgiving, and the executive is on Black Friday. On the 27th, we’ll put the executive up for mainnet, and people can begin voting on it. This is where we’re going to want things to hopefully be pretty aggressive on the voting. I’m skeptical that we’ll get involvement on that week. I’m guessing it won’t be until the following Monday, the 30th, or the first of December where we’re finally garnering enough votes to pass the executive. Once that passes, we’re in a 72-hour delay. This period of time is pretty critical. We want to keep as much MKR on the hat as possible, just defending the old chief before the executive’s actually executed and the old chief loses its permissions. Once the executive is executed, the old chief loses its authority to manipulate the system. The new chief is granted authority to manipulate the system. However, it’s not active yet. At that point, we can pull all of our MKR out of the chief 1.2, the old chief, and we can put it onto. I think we’re gonna use address zero or proposal zero to launch the new chief. Currently, there’s a hundred thousand MKR threshold hard-coded to launch that new chief. Once we get to the hundred thousand MKR mark, we’ll be able to activate it. Then the new chief will work, except it’ll be flash loan resistant or actually maybe even proof. Although I don’t wanna say proof because someone noted that recently and then they got flash loaned, so we’ll be flash loan resistant after that point.
- I wanted to mention that we should have a signal poll for how much MKR we should hard code into the chief to activate it. I believe 100,000 MKR is a reasonable number that we can typically hit. However, we have to keep in mind that many custody providers Coinbase Custody Anchorage have or are all tooled up to vote on the old chief and may need to adapt their code to allow their people to vote on the new one. We may not easily get the 100,000, so we might want to consider something that’s still resistant. I guess the new chief will resist the flash loans anyway, so maybe we really want to think about like 80 or 75 to activate the new chief. That’s it. Sorry that it was longer than I was expecting. You got to stop me when I start yammering.
- LongForWisdom: No, no, I think that was fine, Chris. that was a good explanation. We’ll probably just discuss the hard-coded amount and figure out a number rather than polling. It doesn’t feel like a thing that would really benefit from polling in this case.
- Chris: Yeah, if anyone has a strong opinion on that, then please let Long or myself know.
- LongForWisdom: I don’t think it’s gonna be drastically different from the 80 to 100 range; it’ll be either somewhere in there or slightly around.
- Chris: Pretty much yes, we will also re-enable the osm mom and the flipper mom, so that’s the ability to freeze oracles and enable and disable liquidations.
- LongForWisdom: I guess the question is, would we also consider changing the GSM as well?
- Chris: Yes, actually, maybe we should consider changing the GSM as well. I don’t see why we couldn’t go back to 12 at that point, but…
- LongForWisdom: Yeah, I think it might be something we poll on again because it’s potentially worth discussing. I believe it’s true that 12 hours is not a realistic time to get together a sufficient amount of Maker to cancel something if needed.
- Primoz: I was mostly asking to find out how gas-intensive this all is because we’re going to make so many other changes and inclusions mid-December.
- Chris: How gas-intensive the new chief is?
- Primoz: No, sorry. If we do all the collateral onboarding, everything that’s following in December, and then include these two votes as well, is it possible we’re gonna fill the whole block?
- Chris: Well, the chief is gonna go out on its own. We’re not bundling anything with it aside from pushing us back into the state we were in before disabling the moms. All of the collateral onboarding stuff which is going to come in the following weeks will be separate. Although we’re going to have to see how it pans out. We have such an aggressive schedule that we may hit deployment limits, so the contracts themselves may overrun the size of contracts that we can put into a block. Then we’ll have to split them. Of course, there’s only 12 million gas available in a block. We’ll have to see if we run it to either that limit or if we also possibly run into execution limits. Those are the two things that we’ll end up having to look at in the executives that come after this.
- LongForWisdom: Cool, yeah, excellent Chris, yeah, something’s becoming more relevant now that we have more collateral types.
Open Discussion
Descriptive Header
1:08:39
- LongForWisdom: If anybody has any sort of questions or comments about any of the topics that have come up in this call, now would be a good time to voice them and spend five to ten minutes to discuss. I guess Derek is here. I’ll ask Derek if there’s been any movement on the polls. I’m figuring out the results.
- Derek: Good question. Sam’s pulled the data for me. I’m just checking within the Foundation to make sure that we get all the data we need. From our first take, it looks like the new portal wasn’t correctly counting negative transfer votes. When I look at the old portal again, there’s a discrepancy with who the winner is. At the moment, seems like one percent but because of that discrepancy, let us just check the on-chain data first before coming back to you with a final answer. We’re working through it, and I’ll post on the forum or in the chat once we get some final numbers.
- LongForWisdom: Thanks, Derek. I’ll update the same thing on the forum once the results come out.
- joshlevine: Yeah, there’s one thing I would mention real quick if you don’t mind. Regarding the polling, we should verify it against on-chain data. Still, as far as I can tell, the old portal is correct with that specific poll’s tally. It has to do with how it was running the instant runoff voting algorithm. I’m talking about the lower stablecoin stability fee poll. However, one thing to note is that even though the portal says that there is a winner, it’s kind of erroneous because no individual option has more than 50 percent, which technically, if you’re following the strict instant runoff voting rules, means that nothing is won. So, there’s no clear mandate going forward if that’s the result, just as an FYI.
- SamM: Yeah, I’m just looking at the initial data. It seems like a lot of people have voted for only one thing. I would recommend doing multiple votes in ranked-choice voting.
- LongForWisdom: I’ll just comment on that briefly as well. In general, for the ranked-choice polls, I do try to specifically say in the ‘next steps’ section what will happen if we don’t get a majority. For this poll, I did include a statement saying if the result gains a majority of votes, then the stability fees will be changed. Therefore, the absence of any of the options having a majority results in a default, which is for us to not change the stability fees. I was briefly talking to Primoz before the call about what we would do in this situation. Primoz can maybe talk about it if you want.
- Primoz: Yeah, sure. I see things that if stability fees stay as they are, that essentially means tomorrow we will already start to have
flap auctions. Since we can’t really make any executive votes for about two weeks or more, we’re definitely going to spend fees from stablecoins, which is about 50k per day. In about 20 days until the next executive, you’re going to spend 1 million on stablecoin fees on flop auctions. I don’t necessarily see this as a problem if we already know that the PSM will be implemented. We’re going to implement a low fee because doing this will collect the stablecoin accrued fees, which will measure about 4 million. However, suppose we don’t do that around the 15th or 20th of December. In that case, the total accrued fees on stablecoins will measure to four million, which is the same amount as the surplus buffer. It really comes to the question of what do we think about PSM? We clearly didn’t have a vote from Maker holders. In case this doesn’t get implemented with a low fee, those fees are actually not necessarily going to be collected in the future. In that sense, we would probably need to immediately increase the surplus buffer in the next possible executive vote. The other option is proposing a two percent fee on stablecoins. It is kind of a vote in the middle in some sense if I look correctly. I’m not really sure. However, if it doesn’t pass, we have a problem. We have other votes this week, and we want this executive to pass. I’m not sure if it’s wise to propose two percent right now.
- SamM: Yeah, it’s unfortunate. Only 15 percent of the vote went for ‘no change.’ Many people want to lower to some degree, but it seems people just didn’t check off enough votes; that’s a problem.
- Derek: Looking at the old portal, which is correct like joshlevine said, the winning vote was one percent, and that’s from first choice 3000 MKR and transferred votes 23.000 MKR, which is a 37 percent amount.
- LongForWisdom: Yeah, Greg, to answer your question; We think the new UI is definitely wrong at this point. There is no clear winner for the poll. The top three options have are 37, 35, and 15 percent.
- Gregory Di Prisco: Will that result in a change or no?
- LongForWisdom: No, which is what I was saying. This is due to no single option being greater than 50 percent.
- SamM: Yeah, but from a technical perspective, that’s true. I mean, this is kind of a weird case because it’s pretty clear that a lot of people want it lower to some degree, and they’re fighting over which degree it’s lowered.
- LongForWisdom: Yes, I agree. This is why we’re discussing it. Two percent is potentially a compromise option because, as you say, it’s correct that most people seem to want to lower it to some degree.
- MakerMan: Aren’t we just dealing with this one vault, though? I don’t understand why we didn’t only attack this one vault with a low SF to buy us time just in one shot, one number.
- LongForWisdom: I mean, you can’t start this if you’re on a single vault right now because I’m not sure what you mean.
- Chris: It’s all CR 101 collateral types. They all went out around the same time. Therefore, they’re all hitting their expiration dates.
- SamM: A conservative, safe option to me appears to be lowering to three percent because 70 percent of the vote is lower than that.
- LongForWisdom: Yeah, I guess. I don’t hate three percent about how that actually affects the fees and the implementation timeline. I guess what I’m saying is how much better than four percent is three percent in this case?
- SamM: Gives us another week or so?
- Matthew Rabinowitz: Yeah, something is better than nothing.
- LongForWisdom: My question is, does a weak extra help a lot?
- SamM: This is problematic because now we’re locked out of the executives for two weeks, and we don’t know the result. Do people even want the PSM? We’re very undetermined right now. One thing I did want to say, which Primoz mentioned, is increasing the surplus buffer. Option three of the stablecoin action plan is currently the main plan we are proceeding with until there’s a vote otherwise to override it.
- LongForWisdom: Previously, I guess, was to drop them to zero percent and then replace them with newer Vaults, wasn’t it? Potentially, it is more evidence that people want to lower them to some extent. All right, I think Primoz and I will discuss this after the call. This isn’t the best place to decide what we’re doing and consider whether we want to include a change to reduce it by one, two, or percent in the executive.
- SamM: That seems like a reasonable compromise.
- LongForWisdom: All right, so it’s 25 past the hour. I think we’ll wrap up the call. A couple of final reminders: next week is Thanksgiving. The call falls on the 26th. Therefore, we’re gonna go ahead and skip that call next week. There’ll be no governance call on the 26th of November. I don’t think there’s a huge benefit in having it when half of the participants will be busy enjoying themselves. I’ll post this in the forum as well. Hopefully, everyone will get the memo. I’d like to remind everyone once again to vote in that poll I mentioned at the beginning of the call about whether we want to move this time for the call an hour later because, in this case, the people in this call are the main stakeholders for that vote. It’s not gonna go on-chain or anything, but it’s for the people here to determine if that’s gonna be a thing or not.
Links from Chat
Common Abbreviated Terms
MCD: The Multi-Collateral Dai system
CR: Collateralization Ratio
DC: Debt Ceiling
ES: Emergency Shutdown
EV: Executive Vote
GF: Governance Facilitator
GP: Governance Poll
SF: Stability Fee
DSR: Dai Savings Rate
MIP: Maker Improvement Proposal
OSM: Oracle Security Module
LR: Liquidation Ratio
RP: Risk Premium
RWA: Real-World Asset
Credits
- Anna Alexa K Produced this summary.
- Anatta produced this summary.
- Artem Gordon produced this summary.
- David Utrobin produced this summary.
- Denis Mitchell produced this summary.
- Gala Guillen produced this summary.
- Jose Ferrari produced this summary.
- Juan Guillen produced this summary.
- Tim Black produced this summary.
- Everyone who spoke and presented on the call, listed in the headers.
