Covid Crash: Emergency Governance Summary

In the aftermath of Black Thursday I’ve been spending time creating a summary of the events as they related to governance. What went well, what went badly, etc. I’ve also included a timeline and a brief overview of the executive proposals involved.

Thanks to @scottrepreneur for providing input into some of these sections, much appreciated.

Please bear in mind while reading that these were my interpretations of the events, written between a week and 2 months after the fact. There are likely to be minor inaccuracies and judgements that one or more people disagree with. Please don’t sit on any comments or disagreements! Your input is valuable, and my hope is that this thread can spark conversations and opinions from other sources.

The final section of this document is a Proposed Actions section, some of these are being worked on, others have yet to be tackled. Any comments here would also be great, if you want to help work on any of those Proposed Actions, please comment, get in touch with me, or otherwise just do it.


Timeline of Events

Thursday March 12th
ETH price crash.
$0 Bid Collateral Auctions
Protocol takes on bad debt
Weekly Governance and Risk Call

  • Debt Auctions discussed
  • Keepers discussed
  • Further discussion around risk.
    Continuation of Risk Call over the next few hours
    Contents of first executive determined
    Second Governance and Risk Call (recording not uploaded)
    Friday’s executive moved up to asap.

Friday March 13th
First Emergency Executive (hours earlier than usual) - CC Executive #1
Saturday March 14th
CC Executive #1 passes
Onboarding USDC Collateral discussed in Rocket Chat.

Sunday March 15th
Proposal posted on forums for additional changes.
Onboarding USDC Collateral discussed in Rocket Chat.
Executive containing changes posted - CC Executive #2
CC Executive #2 passes

Monday March 16th
Onboarding USDC Collateral discussed in Rocket Chat.
Emergency Governance and Risk Meeting takes place.

  • Discussion of USDC and Risk Parameters takes place.
  • Moving the executive to Thursday is discussed.
  • Other stuff?
    Proposal for Collateral Onboarding USDC posted in forum.
    Executive onboarding USDC posted - CC Executive #3
    CC Executive #3 passes

Tuesday March 17th
Daily Governance and Risk Meeting takes place.

  • First FLOP Auction presentation
  • Dockerized Auction Keeper Overview
  • Changing Debt Auction parameters discussed.
  • Circuit Breaker discussion.
  • USDC as collateral overview / recap.
  • SCD Migration Overview

Wednesday March 18th
Daily Governance and Risk Meeting takes place.

  • FLOP Auction parameters overview.
  • FLOP Auction manual UI overview.
  • Dai liquidity crunch discussion
  • SCD Shutdown discussion
  • Backstop Syndicate Overview

Thursday March 19th
FLOP Auctions start
Daily Governance and Risk Meeting takes place.

  • Second FLOP Auction presentation
  • Auctions UI Walkthrough

Friday March 20th
Daily Governance and Risk Meeting takes place.

  • PR Response first discussed.
  • ‘FLOP Auctions blocked’ Bug is first discussed.
  • ES discussed in more detail for both MCD and SCD.

Tuesday March 24th
Daily Governance and Risk Meeting takes place.

  • ‘FLOP Auctions blocked’ Described in detail
  • Compensation for Vault Holders is first discussed

Analysis of Major Events

Black Thursday Immediate Response

Successes

  • The Interim Governance Facilitator made the decision to change the regularly scheduled governance call agenda to accommodate the rapidly developing situation.
  • The Interim Governance Facilitator made the decision to leave the call open past the formal close. This seemed to be positive, and resulted in community discussion and information-sharing related to the situation.
  • The Interim Governance Facilitator made the decision to move the regular Friday executive up to Thursday given the urgency of the issues.
  • An emergency response thread was created by the community and used to summarize the situation and start shaping and forming consensus around the desired response using emergency forum polling.

Room for Improvement

  • Maker Foundation personnel entirely disappeared from the governance call at the official end point, only returning later after presumably having had their own internal call. It would have been beneficial if one or more representatives could have been present to communicate what they could regarding the Foundation’s response as it evolved.
  • The recording of the ‘second governance call’ has not yet been posted to youtube, despite containing important discussion and emergency decision-making. It would have been preferable if this was uploaded as soon as possible after the call took place.
  • Emergency response was largely ad-hoc due to the lack of defined processes for how the community and Foundation should respond to emergencies.

Points of Interest

  • There was already a governance call scheduled given the crash happened on a Thursday. It is difficult to predict how the response would have differed had this not been the case.
  • The emergency community polls that took place in the forum closely but did not entirely match the responses proposed by the Interim Risk Team. The most contention appeared around the decision to ignore the weekly polls and instead do a separate set of responses.

First Proposal (Sunday)

Successes

  • Action was taken swiftly which led to the protocol becoming less vulnerable to extreme market volatility.
  • Both the community and the Foundation representatives that commented on the proposal dealt with the disagreement well given the intensity of the situation.

Room for Improvement

  • There was a lack of communication on the Foundation’s part regarding the plan to make further changes to the protocol subsequent to the executive put up on Thursday. These changes appeared to take most of the community by surprise. This could have been avoided by releasing a simple statement saying that additional changes were planned before Monday.
  • There was a lack of engagement from the Foundation with the community feedback that arose from the proposal released on Sunday. This gave the impression that the changes were being pushed through without appropriate feedback.
  • The proposed changes were posted on the forum and one hour later an executive went live with the changes. The forum post asked for feedback on the proposal, but this feedback was ignored by both the Interim Risk Team and the Foundation in general.
  • The community outside the Foundation showed a lack of urgency after the Thursday executive. There was no community push for immediate additional changes prior to Monday. Given the market volatility and the serious nature of the risk, there probably should have been more urgency and initiative shown by the community at this point.

Points of Interest

  • The community response to the Sunday proposal was mixed. Anecdotally, some members were happy that the Foundation was taking a proactive role, whereas others were angry that there was no ‘real’ push for community feedback on the changes prior to the executive being created on-chain. Again anecdotally, it appeared that many of the most invested and active community members were angered by the Foundation’s actions, whereas less involved members displayed were ambivalent or happy with the response.
  • It’s important to try to understand the situation that led to the Foundation’s swift and arguably heavy-handed response on Sunday. At this point, further market extreme volatility was extremely possible and probably would have coincided with the traditional market’s reopening on Monday.

Onboarding USDC

Successes

  • The decision to onboard USDC as an emergency collateral type was made quickly in a situation where it may have meaningfully contributed to the Maker Protocol’s resilience in the event of a further market crash.
  • The discussion in the onboarding USDC meeting was balanced and seemed to consider the meaningful risk parameters and potential objections from members of the community.
  • Forum threads around onboarding USDC saw a great deal of participation and discussion from a wide range of community members.
  • Forum polls to determine risk parameters saw a large amount of participation despite the short-notice timeline of the change.
  • In comparison to the proposal on the Sunday previously, the Foundation took the time for feedback and interaction with the community.

Room for Improvement

  • There was a not insignificant minority of 8 community members making up 18% of the vote that voted against onboarding USDC as a collateral type in this emergency situation. More effort could have been taken at the time, or after the time to try to address these community members concerns in a more comprehensive way.
  • There was a feeling of inevitability around the onboarding of USDC that was largely due to the disregard for community feedback on the previous Sunday. In retrospect this may have been unfair given the efforts the Foundation made to include more time for discussion with this decision.

Points of Interest

  • The initial discussion behind this idea happened in the Maker rocket chat several times over the days leading up to the proposal. It’s difficult to know whether this is what drove the Foundation to prioritise onboarding USDC, or whether it was due to internal opinions. Realistically it was probably a combination of both.
  • Rune took part in one or more of these conversations around onboarding USDC, and appeared to be in favour. This may have influenced some community members to view it more favourably. Ultimately this is unavoidable due to Rune’s status as founder of the project and in my (LFW’s) view is not a reason for him to hold back from contributing to governance. That said, community members should try to avoid giving Runes comments weight due to his position of authority.
  • Ultimately it is impossible to know at this stage whether the Foundation would have created the executive to onboard USDC had a majority of the community (by forum poll) been opposed. It’s also not easy to say whether it would have been wrong for them to do so, given that the final decision should lie with Maker Token Holders.

Debt Auction Parameters

Successes

  • The discussion of debt auction parameters both on the forum and on the governance calls was fairly open and effective. Each of the meaningful parameters was discussed.

Room for Improvement

  • More than most other polls, there was a significant minority in multiple polls (30%, 10/30 voters and 47% 10/21 voters) that signalled ‘I don’t know’ to the questions around changing the FLOP Auction parameters. In the future, further effort should be put forward to explaining the tradeoffs between changing these parameters.

Points of Interest

  • There was something of a sense of urgency around having this discussion that may not have been entirely justified. In the end, governance opted to not change any of the FLOP parameters before the FLOP Auctions started. This urgency may have come from an abundance of caution on the Foundation and Interim Risk Team’s behalf given the pushback from the changes from the Sunday executive.
  • There was no strong consensus towards any changes as a result of these polls.
  • In the end, the FLOP auctions largely ended successfully despite us making no changes. It may be worth noting that when the FLOPs took place, the market price of MKR was greater than the initial bid price. Had this not been the case, this may have had a different outcome (but of course this is speculation).

Emergency Shutdown (SCD)

Successes

  • Given everything going on, shutdown of SCD was not a priority given the greater amount of assets at risk in MCD. Ideally, there would have been time to discuss, understand and mitigate the risks to both systems. But, given the limited time available, prioritisation of MCD was probably the right choice.

Room for Improvement

  • Over the course of the event, it became clear that the process for the shutdown of SCD was not well understood by the Maker community. Complications resulting from the SAI-DAI migration bridge were not well understood at the time.
  • Discussion of SCD shutdown over the various governance calls was not very effective. The issue was largely ignored in favour of MCD. While this is understandable, it’s still an area in which there is room for improvement.
  • Resources such as https://mkr.tools/ and https://makerscan.io/ showed contradictory data. We should communicate when we realize old resources are no longer valid.

Points of Interest

  • It’s worth mentioning that we should have been talking about SCD shutdown more frequently and earlier. Given everything else that was going on this would have required more governance calls. However, having one governance call per day (often lasting close to two hours) is probably the limit of engagement that can be expected from MKR Token Holders (especially on short notice.)
  • It’s unclear how much risk we were accepting for MCD due to the SAI-DAI migration bridge to SCD during the crash.
  • Liquidation of SCD CDPs was delayed during the crash due to the differences between the SCD and MCD liquidation systems.

Public Relations

Successes

  • Given the situation, there were times during the events where the community rose to the occasion in representing Maker in the wider ecosystem. Inter-DAO communication was successful in getting stakeholders rallied with Backstop Syndicate and others committing to participate in FLOP auctions.
  • The Maker Community rallied quickly to take action and disseminate the effect of those actions to the wider community. Different members attempted to provide information about the situation to keep rumors at bay. There is a technical difficulty to understand all the facts of the events as they’re happening, but many community members shared information as it was uncovered.
  • More Keepers did attempt to come online as the situation developed. In some cases this proved difficult due to technical issues, but the messaging ‘We need more Keepers’ successfully made its way to ecosystem participants.

Room for Improvement

  • There was a lack of communication between the DAO and the general public as the events unravelled. Keeping misinformation at bay was difficult as ecosystem participants jumped to conclusions. As actions to mitigate the situation rolled out it could have been much better when updating the community of upcoming changes and changes being discussed.
  • USDC had been discussed in different cases throughout the governance calls, but was a bit of a surprise to the wider ecosystem when added as collateral for Dai, which up to that point had had only trustless collateral. Understanding and actively seeking feedback from the dissenting opinion for onboarding USDC would’ve helped to ensure a happier community and ecosystem.
  • In the Maker community there is a void where a company support team or ticketing system might otherwise be to receive inbound feedback. The governance and community calls, reddit, and chat are common forums for this feedback but can be easily lost in a communication heavy community. A group of community members could seek out specific feedback from all stakeholders for the DAO and ensure it is satisfactorily addressed.

Points of Interest

  • There is a difficulty in speaking authoritatively on behalf of the DAO. There is no swift method to quickly create and ratify a statement for dissemination.
  • Public relations fell lower on the priority list than was perhaps advisable though this is understandable given everything that was going on.
  • It could’ve been great to have more formal feedback mechanisms in place prior to these events but within the events it would’ve been hard to prioritize independent public relations higher.

Emergency Shutdown (MCD)

Successes

  • No one triggered global shutdown, and given the outcome, this appears to have been the correct decision at this time.

Room for Improvement

  • During the course of several of the meetings, it became clear that ES is still not well understood by the community.
  • On multiple occasions both during and after Black Thursday, community members attempted to open up conversations about ES, however these discussions tended to end quickly and without resolution. This led to several community members feeling like the concerns around shutdown were not given enough consideration by the wider community or the Maker Foundation.
  • More in-depth conversations around ES did not take place until days after Black Thursday, and even then, these conversations were without real resolution.
  • Discussion of ES could and probably should have been prioritised more highly during the crisis.

Points of Interest

  • One point that came up was around the importance of focusing discussion on ‘what will not be viewed as a malicious shutdown’ rather than ‘when should we trigger emergency shutdown.’ This comes out of the fact that it’s difficult to coordinate something that involves burning MKR, and it would be more efficient to signal whether we think a potential shutdown would be malicious or not.

Vault Holder Compensation

Successes

  • It would’ve been quite difficult to prioritize Vault Holder Compensation higher at time, given the events that played out with FLOP auctions. Different stakeholders in the community have remained vigilant to see compensation discussion and resolution occur no matter the outcome.
  • Several analyses were produced that helped to determine the extent of the loss suffered by Vault Holders. These are forming a solid basis for a compensation decision.

Room for Improvement

  • Communication with affected Vault holders in the aftermath of Black Thursday could have used improvement. The community could have assured Vault Holders earlier that their concerns were heard and MKR holders were aware of the impact to goodwill in the community due to the issue.
  • Given the outcomes, there is still an incomplete understanding in the community as to how the Maker Foundation, MKR holders, MakerDAO and the Dai Foundation overlap and interact.
  • It would’ve been great to see the analysis for affected Vault Holders appear sooner, but the inertia to start the analysis was moderately high. The calls to remedy the situation were vocal, but there was difficulty in turning talk into action.

Points of Interest

  • Naturally, Vault Holders are anxious to receive compensation. Reassurance that the work is in progress seems to have satisfied most of the affected parties for the time being.

Analysis of Executive Votes

CC Executive #1 - Thursday Night / Friday Morning

Contract: https://etherscan.io/address/0x529b8b4b62b5f32bd47412988a0a66d72f86ba00
Date Started: 2020/03/13
Time Taken To Pass: ~3 hours

Contents

  • Dai Savings Rate: 8% -> 4%
  • ETH-A Debt Ceiling: 150M -> 100M
  • ETH-A Stability Fee: 8% ->4%
  • SAI (Migration) Debt Ceiling: 30M -> 10M
  • Global Debt Ceiling: 183M -> 113M
  • ETH-A Flip TTL (bid length): 10 minutes -> 6 hours
  • BAT-A Flip TTL (bid length): 10 minutes -> 6 hours
  • ETH-A Flip Tau (max auction length) -> 3 days -> 6 hours
  • BAT-A Flip Tau (max auction length) -> 3 days -> 6 hours
  • FLOP TTL (bid length): 10 minutes -> 6 hours
  • ETH-A LUMP (lot size): 50 -> 500
  • WAIT (delay before starting FLOPs): 2 days -> 6.5 days

How effectively did this pass?

  • Passed very quickly given the circumstances. Personally it feels that just over 24 hours is a good result for something that was an unscheduled and unexpected event.

Contents, anything that shouldn’t have been here?

  • It’s questionable whether the regularly scheduled poll results should have been kept in what ended up being an emergency executive. However, this caused no real harm in the end, and was more efficient.

How much risk did we take on as a result of moving quickly?

  • Minimal risk in this case compared to doing nothing.
  • There was some debate on the risk of increasing the ETH-A LUMP and the TTL by such a large amount, with opponents arguing that these could reduce keeper participation due to liquidity requirements and volatility risk. In the end this ended up not negatively impacting further auctions.

CC Executive #2 - Sunday Proposal

Contract: https://etherscan.io/address/0xd77ad957fcf536d13a17f5d1fffa3987f83376cf#code
Date Started: 2020/03/15
Time Taken To Pass: ~1 hour

Contents

  • Dai Savings Rate: 4% -> 0%
  • ETH-A Stability Fee: 4% -> 0.5%
  • BAT-A Stability Fee: 4% -> 0.5%
  • Governance Delay: 4 hours
  • Added ‘FlipperMom’ which allows governance to disable liquidations without waiting for the GSM.

How effectively did this pass?

  • This proposal passed very quickly once it moved on-chain, there was immediate on-chain support.

Contents, anything that shouldn’t have been here?

  • This executive was particularly expedited.
  • While important that monetary policy changes took place quickly, the inclusion of the previously unknown and unanticipated ‘FlipperMom’ took some by surprise.

How much risk did we take on as a result of moving quickly?

  • Minimal monetary policy risk in this update as prices are falling, lower rates are expected to help encourage new borrowing.
  • Some level of Governance risk taken through acting quickly and without sufficient community consultation (from some perspectives)
  • A 4 hour governance delay is quite low when it comes to reacting to a Governance attack quickly enough. At 4 hours, there is a good chance any potential governance attack could happen while the majority of the community were asleep.

CC Executive #3 - Monday Proposal (USDC Onboarding)

Contract: https://etherscan.io/address/0x049e4d10c1b7280cfed5b0d990e39f9c54529a32
Date Started: 2020/03/16
Time Taken To Pass: 4 hours

Contents

  • Set Global Debt Ceiling to 133M
  • Enable USDC-A as Collateral for Dai
    • Debt Ceiling: 20M Dai
    • Collateralization Ratio: 125%
    • Stability Fee: 20%
    • Liquidation Penalty: 13%
    • Auction Lot Size: 50k USDC
    • Percentage between Auction Bids: 3%
    • Max time between Bids: 6 hours
    • Max auction duration: 3 days
    • Liquidations are disabled

How effectively did this pass?

  • Given the potentially contentious contents, and that it was a third proposal in 4 days, this passed very swiftly.

Contents, anything that shouldn’t have been here?

  • Unlike the executives so far, this one was entirely focused on USDC, so no, nothing that should have been separated out under better circumstances.

How much risk did we take on as a result of moving quickly?

  • It would’ve been great to have liquidations active from the onset, but the oracle system is not designed to be deployed on a short timeline.
  • Some level of Governance risk due to unresolvable (given time constraints) disagreement of a minority opposed to the addition of a centralized collateral, even in an emergency situation.

Proposed Actions

Create an Emergency Response Process

Create a MIP that defines emergency response processes and options more fully. This would give everyone in the community a clear framework of a plan to rally around and should hopefully lead to a smoother emergency response process.

Ensure a Clear Foundation<->Community Emergency Communication Channel

Ensure that there is a clear channel of communication between the Foundation and the community that remains as open as possible at all stages of an emergency. The nature of this could be defined in a MIP, and an informal solution could be used in the meantime.

Create a PR / Communications Group

Create a community run PR / Communications group for the DAO that is responsible for dealing with Communications both in emergency and non-emergency situations. Initially an informal solution should be used, but at some point this should become a domain team defined in a MIP.

Create a ‘Governance Intention’ Process

Create a process with which the DAO can communicate its intentions around a certain issue in a consistent way. We already have an informal version of this process in the form of the signaling process. This should be formalized and expanded in a MIP.

Work to develop a better understanding of Emergency Shutdown among the community

More information about this needs to be available, including post shutdown plans, minting MKR as appropriate to refund ES triggers and the governance realities.

Investigate Auction Parameters more deeply

There is some confusion and debate still about the settings of various auction parameters. This needs further investigation and deliberation by the community now that the immediate crisis has passed.

6 Likes

A post was merged into an existing topic: Polls to determine specific Black Thurs vault compensation

While I appreciate the input, I would like to avoid this becoming another Vault Compensation thread if possible and focus instead on the governance implications of the event.

Would you mind moving this to one of the other threads? Perhaps this one: Polls to determine specific Black Thurs vault compensation

Edit: Moved this myself as it was offtopic.

@LongForWisdom Ok if you feel that way. My intention with that post was to highlight the necessary governance implications. And I hope it leads to more discussion and doesn’t result in the topic being buried.

I feel it needed more visiblity so I encourage people to go into that thread and take a look at the comments being made.

I posted it also due to your comment above: “Please don’t sit on any comments or disagreements”.

Thank you.

2 Likes

Thanks for the clarification. There are lots of other points of discussion in this post, and there are already several venues for discussion of Vault Compensation. I would also encourage people to view the post here: Polls to determine specific Black Thurs vault compensation

I’d like to focus more on the general handling of the emergency and some of the other topics rather than diving into the Vault Compensation specifics here. I’d love to hear opinions from those involved at the time.

2 Likes

Wow LFW bows - The above in an absolutely amazing blow by blow aftermath on the event.

I think I do have a few things to add but it will take me time to go back over your really well done event by event analysis and try to add color and additional detail to pieces of the above as I try to ‘recollect’ important details and frame them in a Events, Analysis, successes, room for improvement, points of interest. etc.

Man so well done I won’t have to add too much and will just try to fact check what I can.

1 Like

Sound like a good idea, Great work guys!
During this time, i will bring back the theory of DAO, the art of using and sharing it once again. :m:

This is phenomenal work. Thanks a lot for compiling this LFW.

I think this should read undue weight. Rune is not always right, but it does behoove us to weigh his opinion appropriately.

I’m not convinced this is true given @MakerMan’s report, which would suggest that auction performance with the higher lot sizes has not seen much improvement since BT.

In general, before I go ahead and just do some stuff as suggested, some questions:

  1. Why have a dedicated Foundation<->Community Emergency Comms Channel? Why not specify that this should happen in #general in Rocket Chat? It’s definitionally impossible to agree on what “the community” is anyway, so having emergency comms happen in the most open place seems the simplest and most wide-reaching solution. If you’re worried that important messages might get flooded out, then have it be #governance-and-risk if we agree that all emergencies fall under this ambit (which I reckon they do for now at least). It seems to me the issue this action tries to solve is one of opacity, which dedicated channels, by definition, can only make worse.

  2. The PR group that was beginning to form was being led by @Mitote - has that taken a back seat? If so, where did it end up getting blocked?

  3. ‘Governance Intention’ Process sounds nice, but I am not sure it’s possible. I wrote an Oxford Masters once on communication, so allow me this quick hot take. The DAO should not communicate anything.

As soon as “the DAO” communicates something, it constructs an inescapable linguistic barrier between ‘us’ and ‘them’, which is antithetical to any DAOs goals. DAOs allocate resources, they do not speak.

I find it entertaining when people come here and complain about how ‘you’ should fix this or that broken thing/process. How ‘Maker’ needs to be better etc. Meanwhile, in reality, we are all Satoshi and this is not about building a company with clients/consumers. It’s about building better money. Seeing as putting your money where your mouth is is the strongest kind of speech, this is what “the DAO” should do, not spend time mincing its words. As such, why not incentivise those who complain to find the answer to their own issues and share it where appropriate, rather than try and control PR/comms as a traditional company might? I’m talking about having a running “bug bounty” for negative ecosystem sentiment.

Do you have an issue with the Maker Protocol? If so, that very same protocol, because it’s about better money, will pay you to help fix it!

  1. If you find out that you were wrong, write a public post explaining why, and share it, earn 100 DAI.
  2. If your issue has not been addressed, and you create and manage a signalling thread on the forum that proposes alternatives, earn 300 DAI.
  3. If you you submit a PR based on the results of your thread and Governance votes to make a change based on your input, we’ll give you a further 200 DAI.
  4. If your change makes it into the current hat, we’ll give a last 100 DAI bonus, plus a wizard’s spell NFT (which can be displayed as a badge on the forum for when you next complain).

This way, the only intentions we need signal are:

  1. Constant iteration
  2. Open to criticism
  3. Permissionless, but properly paid, participation (a.k.a. Rich’s major innovation)
  4. Flexible process, not fixed intention
  5. Community consultation, not consumer services
  6. ‘You’ are one of ‘us’

I actually had undue weight originally, and then changed it. Authority is only a valuable metric in the absence of argument. IE if there is no time for argument and Rune and someone else are stating different things, I’d probably go with Rune. On the other hand, if there is time for argument, then the weight and merits of the arguments should be considered without regard for authority.

MakerMan’s report doesn’t touch on FLOP auctions (which are where we print and sell MKR)

This is a misunderstanding based on my used of the term ‘channel’ (my bad.) I was actually not talking about RocketChat here. I was more referring to in an emergency situation there are times when all / most of the Foundation personnel disappear for their own meetings. My assumption for this would be that one or two nominated Foundation members would act as a bridge during these meetings and communicate back to the community. Channel refers to these people rather than a specific communications venue.

No, it still exists and is meeting weekly. @twblack88 could comment.

I disagree, every action the DAO takes is a communication, every allocation of resources is a communication as to values. You can’t escape the DAO communicating through its actions. Realistically there is a separation between people that participate in the DAO, and people that don’t.

Heh, yeah, I feel this. Though honestly it gets a little old after a while.

I agree, I built bounties into the Declarations of Intent MIP for that reason. That said, it’s still useful to integrate an intention process into MIPs without bounties. It happens anyway with signal polls, makes more sense that it be consistent and codified.

I always felt a requirement to pointing out a problem is to point out one or more solutions. Actually implementing those solutions though only falls on a few skilled hands. I am not one of those yet otherwise I would be getting my hands dirty in code and testing new improved Maker MCD code on either a private testnet or public one.

As to putting your money where your mouth is. To me a thorn here is that our huge MKR holders literally seem to be sitting on their big fat money asses doing nothing here. I honestly hope one or more of them would speak up, or have their highly paid agent working their ass of here do so because I don’t really see any of these MKR whales leading any charges here, other than putting up funds other than to buy MKR (so I guess one could argue they have done enough by backing MKR with money). So in the end as one doing work mostly unpaid (yes I was paid once and am highly grateful to foundation/Rich for anything because I expected nothing) kind of chuckle when MKR needs to be flopped. In reality though having a backing value to Maker system is key - which again is a reason to start building system reserves in other asset classes. It would be prudent to have built some system backing value independent of MKR holders just in case our MKR whales lose faith and decide to exit en-masse.

One of the things I find lacking someone isn’t just a list of goals, but an actual practical path to achieving them. What is the real vision for how this system is supposed to work when the goals are achieved. I wish, rather than moan about it, I had not just one but three such visions to present for discussion. I think the ambition of the Maker project captures people, but then when the get your hands dirty hard work pops up only so many can do it, even fewer will try, and less will do.

The Maker as a ‘concept’ is different in everyones mind here and so as a challenge it is really a hard one. I have to say even before coming here I really respected those who worked on this, and are working on it. I can’t imagine a harder project that one could work on that feels - within reach - but also seems so far away from being, and yet DAI is out there, we have Maker v2 with MCD, and are just working out the bugs. In some ways so exciting, in other ways so depressing and all that is left is ‘get back to work’. ugg.

When this whole covid thing comes down Maker needs to have a party celebrating 5 or 10 years of DAI and we all have to get together in a place that has the most lax laws on ‘what one can party with’ and just rip it up together. I hope I am there to hang with everyone and I hope those that left will find some forgiveness, hope, and encouragement to come back and tell old war stories and pass the peace pipes, bottles, or finger foods with each other giving thanks together.

1 Like

We’ve been simmering on a mandate post but defining both the boundaries and scope of a DAO based communications team is tricky/lofty. @Mitote shared a draft of our project plan and we’ve made some progress.

I would like to say that membership in this loose working group is open to anyone with ideas/experience. We meet next Thursday and are happy to include more people. Especially great writers. :wink:

1 Like