Governance and Risk Meeting: Ep. 48 (August 15 - 2019)¶
Video | Audio | Discussion
| References | Person | Text |
|---|---|---|
| # / 00:00:00 | Matthew Rabinowitz | ... known of [crosstalk 00:00:03]. Go for it. |
| # / 00:00:05 | Richard Brown | No, you're right. This is actually interesting, so let's continue this discussion for a bit. |
| # / 00:00:11 | Matthew Rabinowitz | Yeah, the core question being whenever we ... let's just imagine right now we were voting on turning on multi-collateral DAI on Friday of next week, just as a gross example. And I know we've talked about how we're putting on different collateral types; which ones are ... each as we vote them in, but we really, as a general understanding, I don't really think we've got a general consensus about how we're going to get where exactly we're going. We know where we're going to go in five years. We know which in theory collaterals are coming on because we have a process on how to onboard them, but do we have the exact plan on when we launch the DSR comparative to the collateral? |
| # / 00:00:51 | Matthew Rabinowitz | My concern is that we just kind of hodgepodge and smush it all together, and we'll end up with a scenario where we have trouble distinguishing signals of, what do we do with the DSR? Is that more important than doing extra collateral types? Or are collateral types more important than doing the DSR? |
| # / 00:01:10 | Richard Brown | That's a timely question to ask, and so obviously it ties back into this thing where there's obviously this sense in the community that not a lot has been going on and we've been wrestling with trying to make sure a lot is going on, but also falling down and letting the community know that a lot is going on. So, that's on us. Happily, though, all of your questions are going to be answered this week. At least, we're going to add 50,000 more questions, because that's what I want to get into in this call. And I'll start droning on about that in a couple of minutes, but the calendar, every ... I shouldn't say every single step. Maybe 30, 40 steps that need to occur between now and multi-collateral DAI launch have been mapped out, described, and loosely placed, assigned a date and that's going to be released today, sometime after the call. |
| # / 00:02:12 | Richard Brown | So, the community is going to have this opportunity to take a look at exactly how much time is allocated to each one of these steps, how much time we're allocating to the DSR, how long we're going to talk about Oracles, we're going to talk about the risk teams, we're going to talk about the general model when we begin the onboarding process, how long, what that onboarding process looks like. So, the answer is it's all going out today. And the community is going to have the opportunity to start pulling that apart and vetting it for sanity and playing with numbers. |
| # / 00:02:45 | Richard Brown | I feel like sometimes you ask the question and then about halfway through my answer, I realize I'm probably not answering the question you asked. So the answer is that you have concerns about us mapping out MCD and SCD, handling our priorities between those two things, and making sure that we check all the boxes between now and multi-collateralling. Those, from the community perspective, is a very confusing set of questions to answer because you don't have any data to work with and today is basically the day when data begins to arrive in earnest and all of these questions start to get answered. |
| # / 00:03:24 | Matthew Rabinowitz | Fair enough. I mean, in general the point was just the bifurcation until like a software upgrade. Until we can retire, per se, the production environment of what single collateral DAI is and it gets the complete settlement shut down and it's done and fast forward six, eight months from today, whatever the date is. Until that day, logically, I would that we would want to have at least part of the call allocated to the market as it is today and we already somewhat doing that but then the actual transition. And that's the piece, maybe that's in the material and I'll have to await to see what's in the material. |
| # / 00:03:59 | Richard Brown | Yeah, well that's ... I'm not a financial analyst and so many from my uneducated or perhaps I didn't deeply explore this thing, but in my mind the existence of single collateral DAI and multi-collateral DAI at the same time isn't really a bifurcation of the ecosystem. And maybe Cyrus you can explain why this is a naïve perspective. But the larger system is the same where we're not expecting to see two different rates for single collateral and multi-collateral versions of the DAI are we? Is there going to be a divergence here? Do we need to manage these two systems independently? Cyrus do you think? |
| # / 00:04:36 | Cyrus Younessi | In the transition period, before global settlement is called on single collateral DAI, potentially. |
| # / 00:04:45 | Richard Brown | Really? Okay yeah, well, all right so ... |
| # / 00:04:48 | Cyrus Younessi | Especially if there's DSR for multi-collateral, which I'm sure people are going to want to use, it would lead to different rates for MCD, but in the meantime we'd still have to manage stability of SCD. But I mean don't think it should be that difficult, we seem to have the process down pretty well right now. Things are generally steady right now. |
| # / 00:05:17 | LongForWisdom | It's also about migration from one to another though. Isn't that going to cause [crosstalk 00:05:22] |
| # / 00:05:22 | Cyrus Younessi | Yeah, I mean the migration is something that there's a plan for. I haven't personally sent and enormous amount of time looking into the mechanics of the migration. But yeah, I don't think, from what I understand, it's not going to be some chaotic issue; it should transition seamlessly. |
| # / 00:05:44 | Richard Brown | All right, well, sorry to be a moderator here, but we have a lot to go over today. I think that that's a great question though Matthew and it's something that we should probably start addressing. Happily the person that would potentially answer some of these questions is this call today. So, Soren, I won't ask you to make a presentation right now about what the migration from single to the multi looks like, but perhaps next week or the week after we can get a review from you of what that process is all about. Because I have to admit that I'm slightly hazy on that as well. |
| # / 00:06:15 | Soren Peter | Sure, or we can do it on the commute abroad or whatever. So whatever you think is best. |
| # / 00:06:21 | Richard Brown | All right, cool, I'll give you head's up so you can be prepared for that. But we're 10 minutes in for what could potentially be a long call, or maybe nobody cares and it'll be a short call, but here we go. I'm going to do a preamble that I always do in these calls, this is the 47th time possibly that I've said this. We are extremely interested in feedback from the community and this is the opportunity to, what is that word I hate, ideate and talk about things. So if people have questions, they have opinions, they have perspectives they want to share, please do not feel shy about doing so. Speak up, jump on the mic, interrupt people, it's all cool. |
| # / 00:07:00 | Richard Brown | If you have questions and you don't have access to a microphone, please paste your question in the chat in the side. I'm pasting a link into our chat as well, that link takes you to our forum, where this is the forum where governance happens in earnest. So if you have questions that didn't get addressed in this call, you want to dig into something more deeply, hit that forum thread. If you want to figure out what's happening in governance in general, and risk potentially in general, hit up that forum and browse, there's a lot of really smart people saying a lot of really smart things. It's important to be involved if you want to know what's going in governance, that's the place to go. |
| # / 00:07:40 | Richard Brown | Speaking of that being the place to go, let's talk about that agenda. The agenda today is interesting we have a special guest. I'm going to talk about the governance facilitator. At least I find it interesting. This is the first time ... can I say that? No. Well, potentially. This is the first time that the Maker Foundation has formally proposed a brand new role for a median rep or debate from the community and an eventual ratification from the community. So we're using this role to bootstrap governance and we're using governance to bootstrap this role so it's going to be a bit murky in the short term, but we'll talk about that for a bit. |
| # / 00:08:23 | Richard Brown | We have special guest LongForWisdom is in the call today, he's come up from the shadows to join us in the light of governance. He's going to talk about what's been happening in the forum, which is good because we were just completely plagiarizing his threads up until this point, so now he can come and do it himself. Primoz is going to give us a review of an article that he published on the Santiment blog, which is timely because I've had questions myself about this, and I know other people have, about this moving of collateral into secondary lending platforms, bad or good. A way from MakerDAO to secondary lending platforms, so that will be interesting to hear. And Vishesh is going to explain crypto to us again today, and tell us about what the state of the peg looks like. |
| # / 00:09:17 | Richard Brown | Okay, so let me launch in immediately to the governance facilitator discussion. I'm not entirely sure what the level of interest is from the community about this, so I don't know how much time we should spend ... interest should be high. Should be very, very high, but I'm not going to tell you guys what to do. I'm going to explain what the role means, what we would like it to be, and where people can go to dig into it. But I don't want to go over the entire docs, I think that there's a lot in there and I don't want to overwhelm people. So this is one of those things where I'm desperately interested for people to ask questions about this and interrupt me, throw curve balls, whatever you like, but I really want to get a conversation going about this thing because I don't. |
| # / 00:10:07 | Richard Brown | I posted a wall of text in the forums last night. Obviously the expectation that everybody has spent their free time pouring of this and doing homework in advance for this call is unrealistic, but please take a look at it and let us know what you think. |
| # / 00:10:22 | David Utrobin | So Rich, I have a framing question for you. What's the main problem that this role solves? |
| # / 00:10:28 | Richard Brown | Thank you for asking, David Utrobin, because that's exactly what I'm going to start talking about. So here we go, we have this, we have sort of this organically developed role which is whatever I've been doing in these calls up until this point which is light moderation duties, encouraging people to do this and that, and trying to facilitate other discussions. That has worked out well, at least I believe so far, things haven't been overwhelming so I'm assuming it's all right. |
| # / 00:10:59 | Richard Brown | Things are going to change very, very rapidly in the MakerDAO ecosystem. I've already sort of teased, semi-announced at the top of the hour, we have a road map that's coming out. It's like an official-ish road map that's the first time ... well, I shouldn't say that because we've had a road map that people dig up links to ... we can't seem to find wherever it's posted. But there's an official road map now and it outlines all the things that are going to happen. There's like 20 or 30 steps that need to happen and there's detail for each one of those things and we have a limited amount time to do these things. |
| # / 00:11:35 | Richard Brown | And it's important for ... well, this is a bit off topic, but one of the things that I want to keep on reiterating is that in order for us to get to MCD launch we have to align with a series of different stakeholders and I can put them into these big buckets. One of the major stakeholders obviously is the governance ecosystem, governance in risk in the community needs to do a pile of work in order to set us up. The foundation itself needs to complete a bunch of development work, it needs to complete a bunch of security and auditing work, other miscellaneous administrative tasks need to completed. Our partner, third party ecosystem and our partners need to complete some work as well. And so that's why it's so difficult to come up with an authoritative road map because these three large groups of stakeholders need to get their dependency chains in order and sort of execute around the same time frames. So that's why we don't publish hard dates at MakerDAO, we're trying to get as close as we possibly can, but once these three things align we go live with MCD. |
| # / 00:12:47 | Richard Brown | The area of concern for this group though is the first group that I talked about and that's governance, risk and community. We need to optimize what we have going on so we're not the blocker when the third party ecosystem says, "We're done," and the foundations says, "We're done." I don't want the governance community to say, "Oh, we're not done," because we're going to be talking about this for a year. |
| # / 00:13:07 | Richard Brown | So we need to facilitate this thing as quickly as possible and so the interim governance facility at a role. Here's the segue. I'm going to start reading liberally from the forum post. But in a nutshell the interim governance facilitators are tasked with ensuring a smooth operation of the governance process. This involves a wide range of activities anything from general administration to signals gathering to governance and scheduling them. Essentially this role is about facilitation. It's not a decision making role, it's not an authoritative role, it doesn't set ... well, I shouldn't say that. It does set policy and procedure in conjunction with working with the community, but it doesn't make governance decisions. This role isn't empowered to decide that everybody should vote for X or that time for voting for this new thing, Y, needs to start happening arbitrarily. It's about facilitating and working with the community so it's an evolution of what's already been happening. |
| # / 00:14:10 | Richard Brown | The governance facilitator works with stakeholders in the governance threads, in the chats, in these calls and identifies emergent process, helps to refine workflows, helps to encourage the community to work together in an open sourcy type manner to generate usable pieces or work that we can all agree upon and then transitions that work from the community into the governance infrastructure proper. One of the primary responsibilities for this role is making sure that there is this clear path from a forum thread where someone says, "Hey I have a good idea," to it eventually showing up in the governance portal. |
| # / 00:14:53 | Richard Brown | And that's where we've had friction in the past and a lot of meta questions have arised, primarily around signal generation. We've had lots of really interesting posts in the subreddit that have gone unanswered or that haven't been pursued. We have lots of interesting post in the forum now. It's very important that we figure out how to bootstrap the process from the signal has arised, there's intention in the community to achieve some kind of goal, to actually achieving that goal. |
| # / 00:15:30 | Richard Brown | So the question like I said is, a signal is here, how do we determine a consensus has been achieved on that signal, how do we make sure that signal turns into usable work, and how does that usable work get presented back to the community, and how does the community vote on that. And that's essentially what the governance facilitator role is in nutshell. Is ensuring that from idea to governance decision, that that pipeline has been fully explored, defined and adhered to. The goal here though is that it's not governance facilitator that makes up all that stuff, it's the community that does all that and the governance facilitator there just makes sure that everybody's following the communities rules and they've been formalized, and understood and published. |
| # / 00:16:19 | Richard Brown | So those are the very broad strokes. No questions yet so I am going to go over the part of the forum thread that just outlines the rules and the responsibilities. Because ultimately this is a lot like a job description and whether the community accepts this proposal is technically in line with whether they accept the concept of this being a role that the organization, and the foundation, and the community at large needs. So primary responsibilities for a governance facilitator, number one, ownership and management of the official communication channels. This is an important one, our forums, our chats, our Subreddits, these calls are the official communication channels. |
| # / 00:17:05 | Richard Brown | The health of these channels, the integrity of these channels, is extremely important. They need to be maintained, tone needs to be set, trolls, talk to behavior needs to be mitigated, signals need to be surfaced and ultimately in the case of some kind of an emergency in the future, people need to know that these communication channels have not been subverted and that they can still be trusted. So that's one of the main responsibilities for the role. |
| # / 00:17:31 | Richard Brown | The second, this is my favorite one actually, the second responsibility here outlined in the forum thread is defining the governance problem space and this aligns heavily with some philosophies that I hold near and dear. Governance is a hard problem, it's going to be an increasingly hard problem in predicting the edge cases and how hard and in what form that hard problem is going to take in the future is very, very difficult. So difficult is, as you go farther out into the future, it becomes meaningless to speculate. So part of this governance role is to help the community define exactly the role of the governance facilitator should be, defining what governance should look like, defining what a consensus actually means, defining what a process looks like, defining what work looks like. |
| # / 00:18:19 | Richard Brown | So, it's basically a meta role whose job is to figure out what the processes are that manages the roles that people with meta roles, so it's a bit circular, but the governance facilitator is about working with the community to figure out exactly what it is the community wants from governance and making sure that that happens. That's one of the most exciting parts in my mind. |
| # / 00:18:43 | Richard Brown | We already seen some really cool emergent behavior and one of the people responsible for that. I promised not to put them on the spot, but I will anyways. LongForWisdom is a perfect example of that kind of emergent behavior. So we've seen the community members rise up, aggregate, collate, refine thoughts and feelings and processes from the community and formalizing them. And the job of the governance facilitator would be to work with people like LongForWisdom and ensure that that work doesn't go unrewarded and unremarked and un-codified, that's the word I was looking for. |
| # / 00:19:20 | David Utrobin | Rich, I have a quick question. |
| # / 00:19:21 | Richard Brown | Sure. |
| # / 00:19:23 | David Utrobin | Since posting that, have you had any really clear criticisms of the role? Or any foreseeable problems with the role that the community should like discuss or think about? |
| # / 00:19:37 | Richard Brown | Yes. Why are you asking hard questions? This is going on your permanent record. But yeah, that's a great question. So there is some concerns from various groups, and I'm not going to speak for people and or name them, but the concern is that there is a lot in here. There's a mandate, there's a calendar, there's a process and voting for all of that or none of that might be bit weird. So I'm going to trust to hear from the community whether they want split this thing up into pieces and we need to balance that against efficiency and granularity. |
| # / 00:20:16 | Richard Brown | I think is one of the questions we need to address like do we split this up into 27 pieces and then we vote on which of each 27 pieces of this mandate do we like or do we just say, "Yes, that's good enough," and figure it out as we go. I don't know, but that's up for the community to decide. The other ... yeah, the only other feedback I received so far was in the chat and that there's some lack of clarity about what the term ratified means and the implication that we ratified other teams in the past and that's something I'm going to get to in a second because I'm about to talk about bootstrapping processes. So I need to cognizant of time and I'm still waiting for decent, that always freaks me out, I don't know whether people think this is great or not so you know what we should do, we should start leveraging the poll mechanism. Dave, do you have polling powers in this one? |
| # / 00:21:11 | David Utrobin | I don't think so. |
| # / 00:21:16 | Richard Brown | Well, maybe I'll do it after I'm done talking. I'll read through all of this and then I'll set up a poll whether people think this a good idea or not question. |
| # / 00:21:24 | Cyrus Younessi | One of the things that might be helpful, and maybe you touched upon this, maybe talking about some concrete examples of how this role even came about. What led to the perception that this was something that needed to be put out there? Like what was wrong with the old way and what are we expecting in the future out of this? |
| # / 00:21:45 | Richard Brown | Yes, that's a great point. Can you remember that and then like ask me in like three minutes? I want to go through the rest of what the roles and responsibilities are and then we'll try and figure out, how do these roles and responsibilities ... |
| # / 00:21:58 | David Utrobin | Rich, we can instead of doing the polls, I think everybody if they go into their participants tab, they could either signal yeses, nos, raising hands, things like that, we could use those as signals. |
| # / 00:22:12 | Richard Brown | Yeah, looks like we already got one. LongForWisdom wants me to speed it up. All right. |
| # / 00:22:18 | LongForWisdom | I'm just pressing buttons. |
| # / 00:22:20 | Richard Brown | So the other big thing, I'm a big fan of this whole concept and that is the concept of advocacy. So it's important that concerns, questions from the community actually make it back to the foundation and somebody from the foundation is held ... I shouldn't say held accountable, that sounds ominous, but it is obligated to respond and to provide some feedback. And we don't always see that happening when people ask questions. Everybody agrees that that's a great question, but people at the foundation is fantastically busy and these things don't get followed up on. So the concept of advocacy is big for me. Somebody in the community, potentially is the facilitator, would be responsible for saying, okay that's a great question, I'm going to find out and I'm now obligated to find that answer for you and let you know what it is. |
| # / 00:23:16 | Richard Brown | To David's earlier question, this role primarily is about bootstrapping. We need a governance facilitator ... this also ties into what Cyrus was asking. We need a governance facilitator to start formally collecting and refining and publishing and facilitating and encouraging processes that have sort have been percolating on their own in the past, but have no schedule associated with it, nobody officially driving it, nobody mandated nor power is obligated to make sure that things happen. And that's where we've lost some traction in the governance ecosystem up until this point. People want to do things, but how do we determine whether a poll should happen? How do we move something from this place to that place? Where are going to put these docs? How are we going to work on these docs? All of these are open questions. |
| # / 00:24:10 | Richard Brown | If we rely on the mechanisms that we've had working up until this point, where somebody post in Reddit, six months later somebody remembers that Reddit post, and then they do a post in the forum and then somebody thinks it might be a good idea and then two weeks later somebody picks up that post and says, "Yeah, how about if we do this instead?" That's fine actually, because that's the way that the entire open source world works and it's super low stress but, it's not going to maintain the velocity that we need to see in our ecosystem in order to launch into MCD when we want to launch it so we need to formalize this stuff. |
| # / 00:24:45 | David Utrobin | Rich is it also a point that in the case of governance, you know eventually taking over on its own and not having it facilitated by the foundation? Being able to re-create this role and just having a very well defined position for anybody to fill, that the community can basically vote in or vote out right? |
| # / 00:25:05 | Richard Brown | Yeah, precisely and that's part of this bootstrapping process. Believe it or not MakerDAO is fanatically devoted to the concept of decentralization. It's very, very, very important to us that we remove ourselves from the front lines of a lot of this decision making process. We need to hand it off to the community and we cannot hand off a role off to the community if that role doesn't exist and the community doesn't understand what that role is so. I've mentioned this in the chat as well, the long term, the short to mid, mid to long ... yeah, let's call it mid to long term goal for this, is that the first interim governance facilitators' tasked with finding training and bringing other governance facilitators out of the community. |
| # / 00:25:56 | Richard Brown | This has been a long standing complaint of mine, and I reiterate this in these calls frequently, is that the reason why governance decisions don't happen here, the reason why we don't set the course of MakerDAO in these calls, is because this is a one and a half hour time window on Friday's in a North American time zone where we're all speaking English. That's not very global of us. So the goal for the governance facilitator is that in each one of the major regions we have additional governance facilitators. One in the APAC, LATAM, EMEA, all over the place. We have these governance facilitators repeating these same concerns and processes and conversations and paradigms to the rest of the world and those facilitators can all work together to coordinate and then eventually in some glorious future, the foundation just steps aways and lets the community governance facilitators just do it themselves. And I'm highly motivated to do that, because I got enough to do already. |
| # / 00:26:54 | Richard Brown | So let's finish the bootstrapping process because this is already going on way too long. In this bootstrapping process the governance facilitators are tasked with working to the community to figure out the answers to question and not to dictate the answers to these questions. So it's a collaborative effort but we need to figure out what the frameworks look like, so how do we create canonical reference materials for the community? Where do they live? We need to figure out how general communications work, how do we effectively broadcast governance issues to the wider community? How do we increase engagements? |
| # / 00:27:28 | Richard Brown | In the polls, votes and signals area we need to figure out what the best practices are for collecting, adopting, filtering, identifying new issues as they come up. We need to manage debates, figure out where it lives, figure out how to set the tone, figure out when it ends. These are all really important philosophical questions that we haven't quite addressed yet. So at what point does talking start, debates occur, talking end, polling start, polling end, broadcasting of information from that poll start and end? And then when does it go into the governance ecosystem start and end? These are all really important scheduling questions that are dead boring to implement but we need to figure them out in order to get enough velocity to get MCD out the door. |
| # / 00:28:15 | Richard Brown | So, where does work happen? What kind of pipeline workflows do we have? It's my opinion that what we're doing here is open source governance and it's not just a catch phrase. It's literally, if you take out the word open source development and you swap it in with governance, very little changes mechanically. We need to collaborate in online forums, we need to reach consensus about ideas, we need to commit documents into a repo for version control, we need to iterate on those documents and the issues against them, and then we need to put them somewhere people can see it so in pretty presentation there. So all of these mechanisms already exist but we need to formalize which mechanisms we're using, where things are going to live; stuff like that. |
| # / 00:29:01 | Richard Brown | I think I touched ... yeah so, I am going to condense this down. You guys don't ask me any questions, I'm going to stop talking, how's that? I'm threatening you with ending this rant. How do you like it? So framing the debate, establishing the consensus, communicating scheduling, coordinating teams, et cetera, et cetera, et cetera, We have the governance facilitator role because we need to optimize and put a fire under things that we have been doing. So a process that might have taken us two or three months in the past needs to take us two or three weeks in the future and that's what this role is all about. Before I hand it over to LongForWisdom, this is the last chance for anybody to offer up a question, a comment. |
| # / 00:29:46 | Cyrus Younessi | Is this a permanent position? Not yours, but in general. Do you expect governance facilitators to play a permanent role in the ecosystem? Or is it maybe part of this like five year bootstrapping system? |
| # / 00:30:01 | Richard Brown | I can't ... yeah, this is where I'm afraid of futurism, I can't predict that. If I had to guess it it's a yes probably. But it could be in two years a team of governance facilitators work with the community to determine that that governance facilitators don't make sense and they should do Y instead so. I'm super hesitant to predict what the future looks like. I think that for the next year or two, it's very likely that governance facilitators will be the ones that are ensuring that the risk teams, the Oracle teams and whatever other teams we have arise are all working together effectively, that an idea from the forum actually walks all the way through the pipeline, ends up in the portal. That the community is able to more effectively and intelligently enact it's will and make decisions. |
| # / 00:30:50 | Cyrus Younessi | Is there going to be some sort of operational plan where governance facilitators are not around? What if you're out sick? Does governance push forward or? |
| # / 00:31:05 | Richard Brown | That's a good question and I think that that ... and this is where I have this out where we're defining the problem space. So maybe one of the first question the governance facilitator should be answering is how do we ensure coverage. It is a good observation because this is a role with roles and responsibilities and can be treated exactly like a job. And so ensuring coverage, ensuring communication, ensuring transparency, ensuring accountability; all of these things are going to apply to the governance facilitator. |
| # / 00:31:38 | Richard Brown | But like I said we're extremely hesitant to provide the community with an operator's manual of like 200 plus pages of what this role looks like. It's best to frame the skeleton, the scaffolding or the problem space here and then rabidly reiterate with the community and determine what they think it should be. It's my expectation that we create this mandate ... well, we've created the mandate. We release it to the community, we talk about it for a couple weeks. People say yes or they say no. We iterate if they say yes, we go with it and then in two or three months some omission or problem arises and we change it. And then we ratify that role again and now we've improved things. That's the way that governance has been happening up until this point. I've said this before, but that's the best thing about the way that MakerDAO does governance, is that it's always better than it was last month. And as long as we keep on doing that, that iterative improvement means that a year from now, we're going to be amazing. So I'm not that concerned. |
| # / 00:32:47 | Cyrus Younessi | I actually think it's important to stress that are a bunch of open questions in this talk. Right? And we're asking community to ratify. I mean there's certainly a lot of information in there, but also a set of questions. And if the community doesn't pursue answers these questions, they don't push for iterating the doc and more or less filling in the blanks. I mean, it's just not going to operate as we envision it. |
| # / 00:33:16 | Richard Brown | No, that's great too and I want to address that because this is something that happens more and more and something that as a community we need to understand. That the foundation wants the community to pick up these roles and assume these responsibilities and define them and we're going to do our best to help that to happen but if the community doesn't, then the foundation will. That's just the way things have to be. In the absence of the community interaction, feedback, the foundation will keep on iterating on this role. We'll keep everybody up to date on what's happening and we'll publish our changes and we'll refine it ourselves and keep everybody up to date. |
| # / 00:33:57 | Richard Brown | In a perfect world that this decentralized, self-sovereign, crypto anarchist thing is actually real and people want to seize control of their destinies and they will step up and help us to do the stuff. Or we might find that people just wanted the opportunity to watch what the foundation is up to and as long as we're open enough about what we're doing with our role and how things are being managed, they're happy to let us work out the finer points of the situation. I have no idea how it's going to shake out. |
| # / 00:34:27 | LongForWisdom | I wonder if there needs to be a push for like recruiting for the community generally because, like ... well, it'll be great to get all of these things, feedback for these things from that many people to look at it at the moment. |
| # / 00:34:38 | Richard Brown | I agree with you completely. I address that in this thing too. Broadcasting of information, external communications and campaigning are things that I'd really like the community to begin thinking about it. Because we have a situation where somebody comes in and says, "Okay, well, I vote X on this poll," and then two weeks later find out that only six other people voted or not everybody agreed, and sort of being vaguely confused about that. It's very important for us a group and the people that are deeply into governance to get together and say, "Well, I think the community should do X." And therefore I'm going to tweet about it or I'm going to talk about it in the chats, or I'll do a blog post, or I'll get the word out. That's the part that we're kind of missing right now. |
| # / 00:35:28 | David Utrobin | There's an additional question in the chat, from Frank Cruise. He's asking if the role is going to be compensated or if it's purely voluntary. |
| # / 00:35:38 | Richard Brown | Yeah, that's another great question and we are in a transition phase at MakerDAO and this is going to be a question that arises for every one of these mandated roles. We need groups of people to be ratified, mandated, ratified by the community in order to do work on behalf of the community. And the examples we have so far in that world are the governance facilitator and the risk teams. Until we have more sophisticated mechanisms to disburse proceeds from the ecosystem itself, so once we figure out how to allocate MKR, the stability fees, to cover expenses for the ecosystem, everything is going to be covered by the foundation itself. |
| # / 00:36:24 | Richard Brown | So that's where we are with the governance facilitator role, that's where we are with the risk team. Presumably at some time post MCD launch, the mechanisms will be in place to begin talking about those questions because this is where, it's one of the primary distinctions, if the foundation is creating and paying for that role, it's not a community role essentially. It can be philosophically, and we can make our best efforts to make it pure and behave as if it was, but until these roles are created by the community and being paid for by the community, there is not pure autonomy. So that has to happen post MCD once those mechanisms exist. So there's questions. Does anybody want to read these or actually say these questions? |
| # / 00:37:18 | David Utrobin | I could summarize the conversation. |
| # / 00:37:22 | Richard Brown | Hold on second. I would like to encourage people to actually speak. |
| # / 00:37:26 | David Utrobin | Yeah, definitely [crosstalk 00:37:28]. |
| # / 00:37:27 | Richard Brown | So who's next? Charlie, do you have a microphone? Do you want to ask that question? He does not. So David, do you want to read that? |
| # / 00:37:41 | David Utrobin | Yeah, I could summarize it subversively. Charlie's asking if there's a preference between MKR inflation versus like carving out the stability fees for role compensation. And I think the answer to that is generally multi-collateral DAI, whenever stability fees like accrue to a certain point where they trigger a surplus action, where they're like sold off for MKR that's then burned, it's at that point that you're able to steer them in whatever direction you want to steer them to. That's my understanding, but the whole thing this yeah. [crosstalk 00:38:18] |
| # / 00:38:18 | Cyrus Younessi | Charlie I don't think they're the exact same thing right? Because if you carve it out in the stability fee you're just adding an extra cost to the borrower. Where if it's from MKR inflation, MKR holders are self diluting. [crosstalk 00:38:33] |
| # / 00:38:33 | Richard Brown | Yeah, should somebody ... we need a system architect in here, actually, because I'm reasonably sure ... I've heard references to names of buffers or something. I can't remember what it is. |
| # / 00:38:43 | LongForWisdom | I think we need some assistance. Sorry [crosstalk 00:38:46] |
| # / 00:38:46 | Richard Brown | No, go ahead. Do you have answers for us, LongForWisdom? |
| # / 00:38:48 | LongForWisdom | I was going to say I think when he says carving out, he means out of the portion that we'll be burning Maker right? So it's like it's either burn Maker or you can use that money to pay someone or ultimately you can print Maker and then use that for someone. |
| # / 00:38:59 | Cyrus Younessi | Sure. Sure. Right. |
| # / 00:39:03 | Richard Brown | Yeah, this is probably a question we should follow up with in the chat, forum thread, or maybe I can ping somebody after I'm done talking to find out the answer. But I know that there's mechanisms in play. Embarrassingly I haven't been super interested in that aspect of things because that's too far in the future and I'm focused on the next three months usually so we can follow up and find out. Were there any other questions here that went unanswered? If not, yeah, this has already gone way over time. |
| # / 00:39:38 | Richard Brown | All right, so we have a special guest today. LongForWisdom is here. He's been doing a lot of really great work in the forums. If you are in the forums, you stumble across his avatar. LongForWisdom, we have to create a nickname for you, I just don't know if I'm going to be able to say that over and over again. |
| # / 00:39:57 | David Utrobin | LFW. |
| # / 00:39:57 | LongForWisdom | LFW is fine. |
| # / 00:40:04 | Richard Brown | LF Dubs. Is going to talk about governance at a glance and give us a short five, 10 minute review of what's been happening in governance and then I'll hand it off to Primoz after that. |
| # / 00:40:13 | LongForWisdom | Yeah. Keep it short, yeah. Yeah, well, there's been a whole lot of activity in the forum. So we've seen a couple new threads in current affairs. One is the mandate which Rich has already spent a bunch of time talking about, so I'm not going to go into that too much. We should all know what that is by now. The second is a thread by Iammeeoh, who listed out some of his concerns for the apparent lack of progress from the foundation. A couple people chimed in there with their opinions. So we have strong feelings on that either way. I would encourage you to go and share your view. |
| # / 00:40:52 | LongForWisdom | In terms of seeking consensus, I don't think we've had any new signal requests go out. The exponential rate for the stability of fee polls has progressed slightly. I've added a new poll to try and determine which of the four options discussed people favor. So if you have a opinion there then, please [inaudible 00:41:19]. So other than that, much the same. I made a post the other day about consensus decision making versus plurality voting, which I've added to On the Horizon, that section. So now I just detail what consensus decision making is and what are [inaudible 00:41:40] plurality voting. So if you'd like to read that, yeah, and what I think is important. One final thing, just a whole bunch of threads up going on dashboards, documents are already set up for the community portal and the styling for forum and portal et cetera. [inaudible 00:42:05] |
| # / 00:42:06 | Richard Brown | All right. Cool, thanks. I drifted for a second. That forum post that you did about plurality voting, there's a lot of interesting stuff in there and I actually want to dig into that. I don't think we have time to dig into it today. So maybe next week it's be interesting to hear some of your thoughts about what that was all about. |
| # / 00:42:27 | LongForWisdom | Sure. |
| # / 00:42:28 | Richard Brown | Give us a bit of a tour. All right cool so in the interest of expediency we should probably move on to Primoz. Actually, Cyrus, I'll let you set this up because I'm going to hand it over to risk at this point, so you want to take it over for us? You're muted right now. |
| # / 00:42:50 | Cyrus Younessi | Sorry. Yeah, today we're going to have Primoz talk about some research he did on a question that was floating around on Twitter earlier this week about if falling collateral amounts from the DeFi space is a bearish sign for the space and measured even through, for example, nominal ETH amounts locked up in DeFi or dollar value collateral locked up. So we basically spent a little time discussing it last week and came up with up just a little bit of discussion. Noting crazy, but there's some interesting insights for sure that I will let him talk about. |
| # / 00:43:37 | Primoz Kordez | Okay. Thanks, Cyrus. I'm going to share my screen because it's going to be easier for you to follow. Can you see this? |
| # / 00:43:48 | Cyrus Younessi | Yes. |
| # / 00:43:48 | Primoz Kordez | Okay. So basically here's the tweet Cyrus talked about, and here's the article which is also posted in the forum thread. Yeah, I wanted to show that actually. I wanted to show that DeFi's actually at all time high. If you can see, there's some other metrics. The collateral metrics are represented in terms of risk but probably not really in terms in actual usage if DeFi service, if you want to call it that way. But yeah, let me just show you what caused the collateral to drop and the circumstances around it. |
| # / 00:44:30 | Primoz Kordez | First of all, yeah, the collateral dropped in total DeFi space for about 10% in Ether terms or about 30% in dollar terms. Collateral in Maker fell mostly because of refinancings of larger CDPs to Compound where much lower collateralization ratio is required, but also because a lot of large CDPs that have moved collateral from the system as they either deleverage after bull run it, or either because their collateralization ratio was already high right? Or as you can see here, some of the largest DBT's collateralization ratio reached 600%. Of course they have most collateral because they were well above liquidation ratio. |
| # / 00:45:20 | Primoz Kordez | Also someone wants to actually focus on collateral metrics, I think it's much better to look at the median amount of Ether logged in CDPs, because this tells you about the borrowing production amongst retail and not so much about what rates are doing. And you can see here clearly that the red line shows that the median amount locked hasn't been really falling as heavily tells you that there's still ongoing production to borrow amongst retail or amongst the masses which is in my opinion crucial. |
| # / 00:45:54 | Primoz Kordez | Of course we are seeing less collateral in the system because everybody's monetizing on their profits. The leveraging from the Ether bull run. What they want to say is that the nature of CDP usage and borrowing itself will always lead to high volatility of collateral locked. Because it's shown in one of my presentations in last calls, huge amount of Ether that's actually locked comes directly from leverage and directly from trading the issued DAI and Ether unlocking it again into collateral and estimation here is about 20%. And you can see here clearly that the appetite for leverage decreased in last month of course because Ether price dropped and that was kind of expected. And we could see that probably CDPs are much more inclined to deleverage, monetize profits, remove collateral, repay them and so on. |
| # / 00:46:57 | Primoz Kordez | To conclude, what I wanted to show is that collateral metric is really more of a risk metric of the system. I relate it to general appetite for leverage. Of course this fell last month because Ether price fell but this doesn't necessarily mean that DAI usage is falling or DeFi usage is falling. It's not entirely clear what DAI usage entails, but if you ask me, still much of the DAI transactions are due to borrowing it and selling it short for crypto mostly Ether. And if you look at outstanding DAI borrowed, the different platforms, mostly at Maker, Compound and through dYdX, the bigger spot you can actually see that we're at all time high. So there's about 110 million of DAI borrowed on those platforms. |
| # / 00:47:51 | Primoz Kordez | And if you add USDC trade which is also becoming an important token that's borrowed you end up with 113 million of stable coins borrowed and that's again a record and you can see this here, right? The trends is still picking up, there's a lot of borrowing, a lot of origination of loans and if you want to measure activity in DeFi space, these are the figures you need to look, not the collateral itself. Collateral is a risk metric. |
| # / 00:48:25 | Primoz Kordez | Now if you want to measure DAI activity and the usage of DAI, that's again totally different story and in such case, you want to focus on actual onchain activity of DAI. So focus on velocity of DAI, which I've already showed once, is this very high and trend is picking up is actually presented by this graph but perhaps even more by this graph that we just introduced, but it's not color uploaded yet. Some of you might be familiar with so called holder waves that Bitcoin analyst use when trying to size the patterns of Bitcoin hodlers and how is this related to Bitcoin price. Now, for DAI you can do similar presentation and this is what we did. But you really want to focus more on higher velocity because this is expected from a coin that wants to medium of exchange type of currency. |
| # / 00:49:26 | Primoz Kordez | And you can see here more than half of DAI coins in existence ... so, there's this purple line. More than half of them were actually transacted in the last two months. You could actually see in some points in July that one third of whole supply was actually transacted onchain in one week. Or in one day you can see 10 to up to 15, 16% of coins were transacted in a day. This levels are really high and I think no other crypto actually can compare with this and that's actually greatness for DAI adoption. So I hope this helps to better understand what metrics need to be considered when estimating either risk of DeFi or when estimating the lending activity or when estimate onchain transaction activity such as this because those are all totally different things. I think that's all what I had to share with you. |
| # / 00:50:34 | Richard Brown | That's really cool, do we have active coin comparison charts? Like can we see what USDC is up to by comparison? |
| # / 00:50:44 | Primoz Kordez | Yeah, we can simply just include that. That's really just in BETA that staging that link. But, yeah, Santiment actually have for every coin. It has, this metric is calculated and it's quite complicated. It's not so easy to, it's not as easy as you would think to calculate these things. But [crosstalk 00:51:03]- |
| # / 00:51:04 | Richard Brown | I imagine it's super hard. So this is tracking exchange, volume, all the rest of it. It's not trying to address user wallets or anything like that is it? |
| # / 00:51:13 | Primoz Kordez | No, that's totally onchain volume. [crosstalk 00:51:18] |
| # / 00:51:23 | Cyrus Younessi | For me, the takeaway here is that if you want to measure the growth of the DeFi space, you want to have, you want to try to isolate metrics, you want to use metrics that try to isolate something that represents usage or growth of some leverage origination. For example, collateral value by itself is subject to third party factors such as the asset price itself, which is by itself not an accurate metric for DeFi. Right? |
| # / 00:52:06 | Primoz Kordez | Yeah, totally right. |
| # / 00:52:09 | Cyrus Younessi | Now that true metrics that we would like to see are not necessarily the easiest to quantify or track. Their often subject being gamed or what not. Collateral value lock-up is typically not something that is easy to gain, but then it has these trade offs. I think maybe down the line we can create some cool index or something, but for now I think just keeping the overall picture in mind is the best. [crosstalk 00:52:44] |
| # / 00:52:43 | Vishesh Choudhry | Sorry just a quick point on that real quick. So this was actually something that we talked about quite a while ago. About how we consider through filtering of transaction values used to get to better or less gameable approximation of how much is actual usage. And I think that one of the things that really improved was looking at sort of a net transaction value after like a delay period right? So it's harder to gain value by saying, hey, I'm going to exchange his DAI back and forth and if that delay period is one hour right? So if it's a washed trade or something over the course of one hour, that's more easily gameable. If you net it out over, net transaction value over say a week or a month, it's must harder to game that because them somebody has to pay the cost of holding those assets for that time period. |
| # / 00:53:38 | Cyrus Younessi | Yeah, good point. So what does that converge to though? That's just convergence to what the DAI supply is as a metric? |
| # / 00:53:52 | Vishesh Choudhry | No, essentially you're looking at how much is being transferred from which address, let's say consider three addresses A, B and C. Consider the amount being transferred from A to B, B to C, C to A, and all the different edges between those nodes and then what you do is effectively net it out. So if there's a transfer from B to C and a transfer from C back to B, if they're perfectly equivalent, then it's net zero. But if over the course of say a one week observation period, somebody transfers 100 and somebody else transfers 200 back, it's net 100 transfer. It's a sort of a less gameable version of transaction volume as a like [inaudible 00:54:33]. |
| # / 00:54:35 | Cyrus Younessi | Got it. Cool. Very cool. All right, well, thanks for that presentation Primoz. If there are no questions then let's move on to Vishesh and state of the peg. |
| # / 00:54:52 | Vishesh Choudhry | Sure and I will try to keep it brief as well, because I know there's been a lot packed into this call too. So state of the peg, broadly good. There's been a huge transaction, huge trading volume over the last couple of days. I mean very strong for like the past month or so, definitely not quite record setting. Most trades were either just above or just below a dollar. The volume weighted average price was slightly above a dollar. These, broadly DAI is trading pretty well at the moment so a big portion of this was unfortunately selling ETH for DAI which tends to help stabilize the DAI price because there's a fairly large quantity of liquidations recently. You can also see that here, there's about a little over a million dollars worth of liquidations in the past 24 hours which is obvious with the ETH price drop. |
| # / 00:56:03 | Vishesh Choudhry | So just one other thing I wanted to touch on was it wasn't the actual topic of Primoz's tweet, but it is worth mentioning, sort of that relative role in the market share of what's going on with DAI versus USDC versus secondary lending platforms. I think there's been a fair amount of like FUD recently around DAI recently, at least on Twitter, and I think people like to glum onto sensationalist stories so I think it's good to take a good, hard, critical, data driven look at that and understand what's going on. So obviously we've addressed the whole collateralization ratio component, but the other component is USDC market share, secondary lending market share versus DAI supply. |
| # / 00:56:54 | Vishesh Choudhry | So the DAI supply is down to about 78 million, but that's been fairly consistent over, as you can see here. There's not been much in draws and wipes in August or at least in the beginning of August. And the total value of those transactions is fairly net, so the supply has been relatively steady lately at the 76 to 79 million level, but that is definitely significantly lower than what it had been two months ago. So the question is, is DAI losing market share? Are people less excited about going leverage on ETH or what's the explanation? |
| # / 00:57:48 | Vishesh Choudhry | So, just to touch on the secondary lending platforms, what we'd seen is actually the borrow volume has come up in the past couple weeks on these platforms, but so has the excess. So we saw this sort of this cratering of the excess in the end of July, beginning of August, especially with what was going on with InstaDapp and things like that. And what we've seen is that utilization rate has come down with the DAI supply. So there is a little bit less demand for leverage, but is that being caused by Compound? Is that being caused by you know people switching over to USDC? Not necessarily, it's primarily a function of what's going on with ETH price right now. And that should sort of continue to improve, A) as ETH price gets cheaper, we will see an increase in leverage on ETH when the price is right to the market. I think we may be starting to a little bit of that actually now. |
| # / 00:58:55 | Vishesh Choudhry | But as that demand for leverage picks up the transaction volume for DAI, which Primoz showed, is fairly consistently high lately. So DAI is getting used, it's just there was significantly more DAI that had been minted prior than was used because there's primarily just being sold for leverage and then supplied to secondary lending platforms for a yield by the purchasers on those decentralised exchanges. So essentially I think a little bit of that FUD is unwarranted as the conclusion. Obviously, you know, things are down a little bit when ETH is down significantly, but if ETH sort of sustains these price levels and that DAI usage continues to grow, I think you're likely to see somewhat of a recovery. |
| # / 00:59:50 | Vishesh Choudhry | But the really interesting thing will be what Maker holders sort of start to do with regards to the rates. So we see these borrow rates on secondary lending platforms have come back down, so in tandem with that utilization, so now there is a bit more excess that exists on the secondary lending platforms. So it's sort of a tenuous situation, the moment where you could decide to lower the stability fee in a effort to make leverage a little bit more attractive now, it's likely that DAI, the way it's trading, and the relative stability it's enjoyed over the last couple of weeks, could sustain a slightly lower stability fee. I mean that would definitely I think help push up some of the leverage, but I think we need to be very careful about reducing that stability fee when there's a significant excess supply here that's effectively this should track pretty well with through this invisible metric that we've always had of what is the over-supply, which is basically an implied metric of what is demand. |
| # / 01:01:05 | Vishesh Choudhry | So as this spread kind of continues to widen, what you're effectively seeing is a little bit of over-supply, not necessarily increase, but at the very least an increase of the amount of that over-supply that is captured by the secondary lending platforms. So I would say it's a tenuous situation at the moment, maybe a bit of wait and see is advisable or maybe reducing the stability fee a little bit is advisable; I think that's a very interesting discussion that the governance needs to have. |
| # / 01:01:41 | Matthew Rabinowitz | Is that somewhat like and undercurrent of demand? |
| # / 01:01:50 | Vishesh Choudhry | Yeah, I would certainly say that that is basically an implied metic of demand. |
| # / 01:01:57 | LongForWisdom | I had a question as well. |
| # / 01:02:08 | Vishesh Choudhry | Yeah. |
| # / 01:02:09 | LongForWisdom | Is it possible to see the seven day volume by its average price? Seeing that the period that we need scores is like seven days. I think you've shown it before. |
| # / 01:02:17 | Vishesh Choudhry | I have, yeah. Give me a moment. |
| # / 01:02:20 | LongForWisdom | Yeah, sure. |
| # / 01:02:21 | Vishesh Choudhry | In the meantime, answer another question if you would have. |
| # / 01:02:23 | LongForWisdom | I didn't have anything else, but someone else might. |
| # / 01:02:36 | Matthew Rabinowitz | One of the questions we've addressed for the course of the last six months has always been when will we find a moment when core underlying demand, or core supply, wherever it finds itself really are in harmony, when we see ourselves with excess supply dropping off in secondary markets and the price of DAI breeching above one and holding there. Not very much, but slightly, I guess we're pretty close to the point where we're at that harmony. |
| # / 01:03:06 | Vishesh Choudhry | Well, that's an important question I think is, there's been this continual assumption that in order to keep DAI stable, you have to have a high stability fee. It's kind of been the mentality that a lot of people have been operating with. I don't think that's necessarily the case. And there was a lot of discussion on some of the first posts on the governance forum where basically people were raising this idea that the high stability fee, low stability fee would have very little impact on the price in either direction, and that the price was going to do what the price was going to do regardless of what the stability fee was. |
| # / 01:03:51 | Vishesh Choudhry | Now, obviously that's extremely generous and I don't think accurate, but it's not necessarily true that the opposite is the case where if you just jack up the stability fee, you're not necessarily going to make DAI stable and if you lower the stability fee, DAI is going to become unstable. It's like moving a single piece on a ship and the wind is a very, very important directive. Moving that piece obviously has an impact, but I think moving it at the right times is the most important thing. And so, I think right now the question is like, if you were to reduce the stability fee by say half a percent and you observed the impact on both the price and the excess supply on secondary lending platforms, I think you would learn a lot about the extent of over-supply versus demand levels. |
| # / 01:04:53 | Vishesh Choudhry | And I think the increase in cash or that's what Compound calls it, the excess supply on Compound recently is not necessarily a function of increased over-supply. In other words, it's not necessarily a function of decreased demand. Because overall, supply has also come down. So it's not very likely that supply came down, demand came down, and demand came down significantly further than supply. I think it was more likely that supply came down, demand came down in tandem, and then a greater proportion of that excess supply was captured by secondary lending platforms, because they were simply the best opportunity for yield. |
| # / 01:05:38 | Vishesh Choudhry | So, to touch on maybe what you were getting at, can the stability fee be lower at this point? And I think there is certainly a case to be made for a slight, small decrease. But I think that's an important discussion topic. Maybe Rich would say a better topic for the forums than these conversations, but am sort of seeking signal there to get agreement, disagreement, what are people thinking? |
| # / 01:06:12 | Richard Brown | One of the questions I have is, what is the primary driver for this? Is it this sort of a social impulse to, people don't like a high stability fee, so we should lower the stability fee so people will like us better? Or is it because there's enough room that if we lower the stability fee, it won't really be all that bad for the peg? Or is it because [crosstalk 01:06:32]. |
| # / 01:06:36 | Vishesh Choudhry | Well, it's because demand for leverage is neither of those. I mean, I think those wouldn't necessarily be the best reasons to make changes. What I'm saying is, the demand for leverage has obviously taken a slight hit. The question would be, has it taken more of a hit than the supply took a hit? Which I find to be rather unlikely, given what a dramatic drop we experienced in supply. And we experienced that drop even after all of these refinancings to secondary lending platforms. So, that drop was very likely attributable to the decreased demand for leverage, essentially people giving up on some of their leveraged long positions. |
| # / 01:07:21 | Vishesh Choudhry | So, again, it's sort of like once is happenstance, twice is coincidence, three times is enemy action. I don't necessarily believe in coincidence there, so I don't think it's likely that there were two separate, unrelated effects occurring at the same time. I think it's more likely that the drop in demand for leverage what caused that proportional drop in supply, and so the excess supply in the secondary lending platforms was simply just an effect of some of those leveraged long positions being capitulated effectively. And so, some excess DAI that was being held or being used for derivative structures, et cetera, making its way onto the market and now being locked up in secondary lending platforms because people want to get that yield and the well is dry a little bit, and there are not a lot of great opportunities for them to get yield at the moment. |
| # / 01:08:24 | Vishesh Choudhry | And I think there's still a little bit of uncertainty in the market out there around what's going to happen with ETH over the next two to three weeks. And so, maybe they're not quite ready to jump back into a leveraged position, but if the stability fee were slightly lower, maybe they would be slightly more inclined to do that. So that's where that- |
| # / 01:08:47 | Matthew Rabinowitz | [crosstalk 01:08:47] just this general comment, though. I think it's more of almost a macro question, a macro governance question. I know kind of a long go; eight months, nine months, 10 months, it was the vote that we were using the scientific method. Again, unless I'm mistaken. I don't know if the community needs to, or I guess it's somewhat implied what the fundamental mandate of these polls would be. And in additional to obviously voting on collateral and general governance, that the core, fundamental piece of DAI sticking to the peg of one, and then what is number two and what is number three? And if one of those peaces hits right now ... I mean, while I think it's an absolute great endeavor to identify to do our best to figure out what caused it in a scientific manner, at the end of the day, if DAI is above a dollar or DAI is below a dollar, that largely determines our governance action accordingly. |
| # / 01:09:34 | Matthew Rabinowitz | Today, we can somewhat say, because of ETH, the market as a whole in leverage, which I don't discount any of that. I think as we get to adding two types of collateral, 30 types of collateral, when we start adding stuff that's off-chain collateral, our ability to identify that signal of it was this versus that will become very gray. And the general mandate of ... what is our general mandate? That's a question as much as it is a statement. I think it's important to identify- |
| # / 01:10:00 | Vishesh Choudhry | Well, Matthew, [crosstalk 01:10:01]- |
| # / 01:10:00 | Matthew Rabinowitz | ... and then just follow that. |
| # / 01:10:04 | Vishesh Choudhry | Necessary in your description there is defining scientific. Scientific with respect to which variables is also important. Because I'm also saying an important part of a scientific management process would be an element of experimentation. And I don't mean, let's just try whatever to see what happens. What I mean is, given the current demand levels, given the way the DAI price has been, by the way, for ... I think it was LongForWisdom who asked, the seven day price graph is up here. |
| # / 01:10:39 | LongForWisdom | I see it. |
| # / 01:10:42 | Vishesh Choudhry | The question is, the stability fee is currently significantly higher. It is higher than I think it has ever been, and the question is, when the stability fee was lower in the past, was there simply a much higher amount of over-supply? And so, were demand levels actually lower than the market or most MakerDAO community participants were estimating? Was that demand level lower than what we thought? Or is it potentially the case that it's not quite as simple as a higher stability fee, better DAI price, lower stability fee, worse DAI price? And I think the latter is more likely, is what I'm saying. And so, there is the potential that- |
| # / 01:11:39 | Matthew Rabinowitz | But I don't dispute that. No, I think you're exactly right. I think it's very simplistic to say when we do X, we have certainty that for example, higher DAI ... like you said, the stability fee goes up, things become more stable. I don't think that's accurate. With a single collateral structure of one type of metric that we're using, just the stability fee, and we've used it in the past both to encourage demand as much as we've had it almost use ... destroy supply. It's riding the unit cycle. Very difficult to determine if you're falling or accelerating. I guess the ... go ahead. |
| # / 01:12:17 | Vishesh Choudhry | Yeah, so I was pretty much done. I mean, essentially the last point was just ... so, the question would be at 20% or 19.5% again, would the DAI price necessarily drop? The only way it would necessarily drop is if what would happen is people would immediate mint more DAI and immediately sell it for ETH and there is low liquidity. There's not a lot of demand for purchasing that DAI. |
| # / 01:12:47 | Vishesh Choudhry | If all of those things are true, then yes. You would lower the DAI price by lowering the stability fee. But I think that the last leg of that process chain is what's maybe quite a bit questionable, is if more DAI were to make its way onto the market, would there be a demand for it? And I think as a group, as a community, there's actually relatively little visibility into the liquidity for DAI, and there's not a lot of work that's been done around tracking the order books and the depth of those order books. |
| # / 01:13:26 | Vishesh Choudhry | So, I don't think we necessarily have great insight into the answer on that question at the moment. We know that at a ... Compound had hit roughly 19% or potentially even higher on their borrow rate just two weeks ago, so the question being is, perhaps that was also too expensive for people who wish to go leveraged and borrow. They don't want to pay 20.5%. Maybe they don't want to pay 19% either, right? So, maybe that one and a half percent spread is not enough to shield the demand curve from the stability fee. |
| # / 01:14:09 | Vishesh Choudhry | And I would be less likely to say that at 17% versus 19%, and if there's only one to two million in excess cash on Compound or dYdX, I would be less likely to say it than if there was in excess 10, 12, 15 million. I think in the past, we'd seen much higher buffers of excess supply on the secondary lending platforms, which in the future will be a proxy for what happens with the DSR. So, if you were to think about the same problem in DSR terms, it's kind of like saying, "How do you think about the stability fee when the DSR is seeing high utilization at relatively low rates? And how do you think about the stability fee when the DSR is seeing very low utilization at slightly higher rates?" |
| # / 01:15:08 | Vishesh Choudhry | I think these are the kinds of questions that we need to start thinking about and developing answers for, and I think it would be informative to understand what would happen to DAI price and secondary lending platforms if the stability fee were now slightly lowered. I don't think anybody ever envisioned that the stability fee needs to perpetually stay at 20.5%. |
| # / 01:15:34 | Matthew Rabinowitz | It needs to stay wherever it needs to stay for the peg to be one. At the end of the day, [crosstalk 01:15:38] |
| # / 01:15:37 | Vishesh Choudhry | Yeah, but [crosstalk 01:15:38] know what that level is? |
| # / 01:15:39 | Matthew Rabinowitz | No. I mean, the answer is, "I don't know." It's a blended proxy of whatever the average ... every person who's ever taken a CDP is a selfish investor. Their objective is not to do anything except for the benefit of themselves, and if their blended supported return can generate and withstand 20 and a half percent, that's where the peg will need to stay. And if that number needs to be 23, 29, or 2, it'll be whatever it is on a blended, selfish basis. It'll get to the point where investors who can't generate a 20 and a half percent return will either find something cheaper or they will exit the market. |
| # / 01:16:16 | Matthew Rabinowitz | And that's the thing. As we bring on different types of collateral, that blended overall number will be whatever it has to be, and that's the challenge with how you bring on different types of collateral that have different risk metrics accordingly. And that's the whole challenge of the multi-collateral aspect, is the blended rate of as we bring on new participants. Which is why it's so essential that we bring on risky or exceptionally risky assets in parallel to bringing on less risky assets. |
| # / 01:16:51 | Vishesh Choudhry | Yeah. I feel we got a little off book there, but yeah. That's essentially all I wanted to bring up around what's going on right now, and starting to raise some of these questions of, how will we continue to iterate how we're thinking about these rates and interpreting some of these supply and demand levels? |
| # / 01:17:19 | Matthew Rabinowitz | Yeah, for a 20 second deviation, it kind of goes back to the point at the very beginning, what I mentioned about bifurcating. You know, going from a single collateral world where we only have the stability fee, which a.k.a. is at a risk premium just for ETH. And as we go to multi-collateral DAI, and this is a question for I guess Cyrus, risk team, et cetera. Whenever we go to what we intend to launch with the multi-collateral DAI is kind of my question on the general sequence, and if it's going to be in its documentation that's coming, so be it. |
| # / 01:17:47 | Matthew Rabinowitz | I don't know if we have a estimated and approximated risk premium that we're going to associate, or if we're going to launch with ... I assume we're launching with ETH at the very minimum. And I guess that's an open question of, what is that? What are the parameters? If it's going to be the same 150% over collateralization structure, then the real question is, what would be the estimated? It doesn't matter if it's finalized. Just a guess, because these two things kind of dovetail, because if the risk premium on multi-collateral DAI ETH is going to be 20 and a half, fine. Then we know that basically the equivalent of the DSR will start off very, very low. Then the real question is, is that risk premium of 20 and a half in multi-collateral DAI actually reflective of the risk premium? Or is it reflective of an interim risk premium until we launch within a collateral of ... a risk premium of ETH that's reflective of it, actual risk given how liquid it is compared to the market. I won't pick a number. I'll let Cyrus do that, but ... I would make the [crosstalk 01:18:52]- |
| # / 01:18:52 | Cyrus Younessi | Well, actually- |
| # / 01:18:53 | Matthew Rabinowitz | ... that the risk premium would be much, much, much lower. And the question is then, how do you handle that ... that was the genesis of my point; I just didn't articulate it, was how do we handle that transition from single collateral DAI to multi-collateral DAI when we will knowingfully have movements that will cause the price swing from single collateral DAI to deviate off of one while we're doing the transition? |
| # / 01:19:13 | Cyrus Younessi | Well, actually, that's a good topic of conversation, because we are gearing up in the next few weeks to release a little bit of documentation around the DSR, and in particular, some logic or methodology behind what the starting DSR values might be or should be. I'd be happy to sync with you on that offline, in the chats or somewhere else. |
| # / 01:19:48 | Cyrus Younessi | But yeah. I mean, from an implementation standpoint, we have to start somewhere. I don't think we'll necessarily know what the starting point is going to be, especially because we tried to balance the risk premium, and is the combined number going to remain at 20 and a half percent, and whatnot. But yeah, let's work on it, because I know it's something that you have put a lot of thought into already. And then what's going to happen is, we're going to put a poll out for the community that basically says, "Where should we start the DSR?" And then we will have to iterate quickly to make sure we don't end up with any sort of longstanding in equilibrium states. |
| # / 01:20:38 | Matthew Rabinowitz | Fair enough. |
| # / 01:20:41 | LongForWisdom | Feel like it's inevitable that the peg's going to slip a little bit when we transition. [crosstalk 01:20:43] |
| # / 01:20:43 | Cyrus Younessi | Yeah, something's going to have to happen. We'll see. This is for sure a conversation we'll have a little bit closer to DSR implementation and MCD launch and whatnot. |
| # / 01:21:02 | LongForWisdom | Yeah, for sure. |
| # / 01:21:02 | Richard Brown | All right, [crosstalk 01:21:12] ... sorry. |
| # / 01:21:12 | Cyrus Younessi | Did everybody check out the calendar in the governance facilitation forum post? Maybe two minutes, quickly run through it? Or maybe next week we can do it if we're ... obviously we're pretty over. |
| # / 01:21:28 | Richard Brown | Yeah, I think that we might be waning a bit, but it is a great topic of conversation. I was hesitant to lay that on people in my initial discussion. But yeah, review that calendar. Let me pull up this [crosstalk 01:21:49]. |
| # / 01:21:49 | David Utrobin | Be worth scaring your screen, Rich, or sharing ... just having everybody in the call see the thread as you walk through it. |
| # / 01:21:56 | Richard Brown | I'm considering sharing the secret timeline, actually. Wonder if I should do that or not. It's potentially going out in a couple hours anyway, but here ... actually, no, it's too risky. And I have a pile of other secret windows open as well. I'm actually just going to post them like everybody else has internet as well, because they're here. So, review this. Open it up if you're interested, but the calendar for the next X number of months is going to be divided up into two big chunks. |
| # / 01:22:29 | Richard Brown | The first chunk is a bootstrapping chunk, where we need to get our house in order, optimize a bunch of processes, introduce some roles, and start having conversations. The second major block of that calendar is onboarding collateral, and that onboarding collateral is where we reach sort of MVP for MCD. Or we determine exactly what it is the community wants for additional collateral types, which ones they feel most strongly about, and which ones the community as a whole believes need to be there for MCD. |
| # / 01:23:08 | Richard Brown | The bootstrapping phrase, I'm just going to rifle through these things, but it'll give you a high level overview of the kinds of things we need to figure out. In August, proposal for the interim governance facilitator, then a poll for the collateral asset priority list. Both of those things have gone out. Next week we have the interim risk team mandate, so Cyrus' team is going to publish a doc much like the one that was just published for the interim governance facilitators, so people will understand what risk is all about, what a risk team is all about. And if they want to form one in the future, this is the template for them. |
| # / 01:23:45 | Richard Brown | Week of August the 26th, we're going to be talking about the initial DSR values. We'll also have a poll then to ratify, so that the community will signal whether they support or reject the governance facilitator role and the risk team role. Week of September 9th, we talk about the new oracle teams. We talk about the new oracle providers, because we have this new version of the oracle that's going to be coming out. Talk about ideas around productizing those oracles. Then we talk about the ... we release the MCD white paper, ratify the oracle teams. We start getting into risk models. |
| # / 01:24:25 | Richard Brown | There's a lot to absorb there, and I'm sure that people aren't absorbing it, but it gives you a general sense of what's going on. So, over the next two and a half-ish months, we have this cycle of here's a proposal, here's a debate, and then we ratify or reject, and then we repeat that over and over again until we have all the pieces in place that allow us to do governance at a clip. That means we have our governance facilitator, we have a risk team, we have an oracle team that's well understood. They have roles and responsibilities. They begin to produce work. The community begins to accept or reject that work formally, and then that sets us up for the collateral onboarding cycle, which happens whenever we're finished that initial work. So, this is where it comes back to the rate at which we hit MCD is largely dependent for the rate at which the community gets its work finished in governance. |
| # / 01:25:22 | Richard Brown | The collateral onboarding cycle is divided up into ... conceptually, we have it set up to be about a month long. Each week of that month is devoted to a different task for collateral onboarding, so we'll spend the first week reviewing proposals that have come from the risk teams about these particular asset types, the risk analyses. Spend a week talking about them, iterating and debating them, and then we spend a week polling for which ones the community feels strongly in support, or has expressed assent for. And then week four, we wrap that up and we prepare ourselves for an executive vote to add those collateral types into the system. |
| # / 01:26:10 | Richard Brown | Then potentially we do it again. We do it again, we do it again. That's sort of what the future basically looks like for governance, with the caveat that this is all subject to discussion and feedback and debate, but these are the broad strokes. Definitely worth reviewing, because the interim governance facilitator role is a significant milestone in what it is we're up to. But the calendar is where a lot of the weighty, important stuff is at, so have a look and that'll give everyone an indication of what the future is going to hold for us. |
| # / 01:26:52 | LongForWisdom | Says initial thing like [inaudible 01:26:54] is that ... seems quite cramped for discussion of these things for like two weeks from I'm not sure, proposal to poll, and onchain stuff. |
| # / 01:27:04 | Richard Brown | It does. And so, this is why this is heavily [inaudible 01:27:08] in the term of bootstrapping and defining problem spaces and scaffolding, and collaboration. Because we have to start somewhere, and we have to make up arbitrary numbers and then ... because we need to start with something. So, we could very easily determine that two weeks is just not realistic, or we could determine that the level of interest is not there. Nobody has responded to this thread in a week, and then obviously two weeks is too long, so we might as well just move ahead with this thing. There's so many questions we all need to answer, so this is where we start, and then we start fleshing things out and fixing it as we go. |
| # / 01:27:45 | LongForWisdom | Yeah, for ... |
| # / 01:27:49 | Richard Brown | Okay. We're at the hour and a half mark. Feel like this might be a good place to end the call. I want to thank everybody who presented today. I want to thank me for my lovely governance facilitator proposal. Thanks to LongForWisdom for coming in and giving us an update on what's happening in the forums. Thank you to Vishesh and thank you, Primoz, for reading your article. That was great. And thank you, Matthew, for asking the hard questions, as always. |
| # / 01:28:21 | Richard Brown | All right. Thanks, everyone. The post video, audio, transcripts, comments, everything, summaries, will be coming in the next couple hours, so look out for those. All right, bye-bye. |