Issues and possibile solutions for SuperRare (Space Race) Governance

I will collect some problematic issues about Space Race management and SuperRare DAO governance in general and will sketch some possible solutions.

Issue #1

Some proposals, whitelisted for voting by the council, were in fact very poorly written, nearly blank in some sections, with evident copy-and-paste, no more than a 15 minutes work. These proposals received millions of votes. This witnesses the fact that the proposal content is not central to the race, only the voting power counts.

Possible solution

Set reasonable and transparent standards to whitelist proposals. The SR council should guard on this, independently on the (relevance of the) proposers.

Issue #2

There was evident conflict of interest, for instance members of the SR council that proposed their own Spaces or that participated as artists in the Spaces, or that have close relationships with admins of the Spaces.

Possible solution

Set strict rules for participation in a Space, for instance the same person cannot join more than one Space or council members cannot be affiliated to the Spaces.

Issue #3

Voters were hidden. Contrary to the previous races, the voting process was secret. A weird choice for a supposedly transparent technology. It might be useful for both voters and spaces to see who voted for who.

Possible solution

Make the voting transparent. Address (or domain) of the voter and number of given votes should be visible.

Issue #4

RARE governance tokens should represent your karma in the SuperRare DAO, that is how much you contributed in the positive direction of the DAO. To the best of my knowledge, RARE governance tokens have been airdropped only once to artists and collectors that participated in SuperRare. There is no other way to get RAREs, unless buying them on an exchange or receiving them from someone else.

In particular, RARE tokens are tradable. This is much problematic when they are used for voting, since in principle one can:

  1. buy (or flash-borrow) RARE tokens in advance on a CEX or DEX
  2. vote for a project or a proposal
  3. sell the tokens (or close the loan) soon after, even making a profit if you are smart enough

Possible solution

Use soulbound tokens (SBTs) for governance. SBTs cannot be transferred and traded. They are bound to your soul. Like some items in World of Warcraft. Here I mean soulbound fungible tokens, however, since 1 vote equals 1 vote, and votes are not unique items.

Moreover, your karma, that is the number of governance tokens in your wallet, needs to be connected to the positive actions you took in the DAO. Examples of positive actions through which you can gain karma on SuperRare DAO are:

  • market making, like an artist that sells an artwork and a collector that buys an artwork
  • active participation to the life of the DAO, for instance by making or commenting on proposals, organizing things and events, writing and disseminating useful information

Some related information about soulbound tokens:

Issue #5

The voting mechanism is linear: one token is one vote. This means that whale holders of RARE tokens, like blue chip artists or billionaire collectors, control the voting. The linear voting mechanism is deprecated in most DAOs since it creates centres of power.

Possible solution

Use Quadratic Voting (QV) instead of linear voting. In QV your voting power is the square root of your governance tokens. Hence, 100 tokens is 10 votes, and 1 token is 1 vote. QV greatly mitigates tyranny-of-the-majority problems, since 10 voters with 1 token equal 1 voter with 100 tokens.

2 Likes

Hey @hex6c, thank you so much for sharing this detailed feedback. I am in the process of researching and drafting a proposal for Quadratic Voting now. That said, I really appreciate you raising other issues such as curation, conflicts of interest, and transparency as not all of them have been part of the public discussion so far.

I will address the issues out of order, as I believe Issue #5 and Issue #1 have some downstream effects on the others.

Regarding Issue #5:

I believe this is of critical importance and is a very high priority for me personally. We are now digging into the election results to model how the standings would have been different with quadratic voting. This information will be released along with a Request for Comment and eventual SIP to implement Quadratic Voting for Space Race 5. If you would like to perform independent analysis of the vote results, let me know and we can share the .csv.

While quadratic voting does open the possibility for sybil attacks, which could magnify rather than diminish the voting power of whales, I believe it is worth the risk. There are additional statistical tools, such as pairwise decorrelation (implemented in this repo published by gitcoin, and described in this eth research post by Vitalik) to increase sybil defense, however the ultimate solution would be some sort of proof of humanity or scalar on voting power based on reputation. In future races, I think we (or a grantee!) could develop something similar to the gitcoin passport which magnifies voting power based on marketplace participation, DAO contribution, and proof of humanity / social logins. BrightId is a very promising tool being developed in this arena, but I do not believe we as a DAO are ready to implement it, and it may not be ready for us.

That is far beyond the scope of what we can accomplish before the next race, baring additional contributions from the community, and as such I anticipate recommending a basic implementation of quadratic voting, or one with pairwise decorrelation, for SR 5. Of course, with this going through the SIP process, it will be open to community input and approval before being implemented.

Regarding Issue #1:

We performed an initial screening of applications, removing only those which were artists trying to get on the platform or those which were clearly scammy/spammy/bot submissions. We did not see any submissions in that later group, but did have 10 applications which were labeled as artist self-promotion by a team of four SuperRare Labs employees in charge of administering the Space Race more broadly.

In the future, I would love to see this administration be absorbed back under the DAO in the form of a Space Race working group of some kind. The process for application review, approval, and rejection, could be a streamlined version of what I propose in the grants program RFC ([Request for Comment] Introducing SuperRare Grants Round 1). This would open opportunities for community participation in future races, ensure that a level of transparency is enforced, and that those reviewing the applications are doing so with authority directly delegated to them by the DAO council, rather than only through their position at SuperRare Labs.

I think what I describe above, or something similar to it, is required if we were to perform more restrictive curation. I wouldn’t feel comfortable with SRL making unilateral decisions about whether a space race application is verbose enough, and as such bringing that decision making directly under the authority of the DAO could be a good option to maintain a higher level of curation while keeping things capture resistant (reference: Anticapture — spengrah).

Regarding Issue #2:

I think that resolving Issue #1 and Issue #5 go a long way towards addressing this issue. One of the issues I felt during this race was that DAO insiders were involved, both in application and in voting, and yet that was not clear. If there is a more robust review process for applications prior to race start, there would be an opportunity (and expectation) for high level contributors, SRL employees, and SRL investors to divulge conflicts of interest.

I genuinely do not know whether we should restrict the ability of such insiders from participating in the space race, though I do believe their participation (or the participation of people close to them, such as spouses or close associates) should be transparent and that they should be given no special advantage.

That issue of being given “no special advantage” raises some important questions around the DAO’s current power dynamics. The initial token distribution heavily favored marketplace participants, investors, and employees as of August 2021. As you noted, there has not been additional faucets or airdrops since then. As such, anyone who was involved in SuperRare prior to August 2021 has a special advantage when it comes to matters of token governance. This is a very difficult issue to resolve, but, I hope, a fairly easy one to mitigate.

This is where quadratic voting really comes into play, and can help mitigate the negative effects of the current power structure while we develop new models for DAO participation and token distribution, such as the grants program and NTT implementation.

Regarding Issue #4:

Your assessment here is correct. While the grants program does open a new pathway for $RARE distribution, I believe that any governance system which relies solely on a transferable token will trend towards plutocracy. I have been doing deep research on NTT implementations for DAO credentialing, primarily for access control but also for reputation, primarily via Hats Protocol and Otterspace. The most immediate use case, in my mind, is for token-gated access to multi-sigs, communications, and documentation for members of the DAO Council and Grants Committee. However, I see the opportunity for a broad implementation of these technologies (especially Hats Protocol) across the SuperRare platform so that marketplace participation, SRL employment, DAO Council involvement, sitting on a committee, and even regularly attending discord events could all be credentialed under the same standard. This opens the door for a SuperRare specific “karma” system, as you described it, or more specifically, an engagement based system for proof of humanity and on-chain reputation.

While these systems are unlikely to replace $RARE, they could be used to heavily scale voting power and provide sybil resistance (similar to gitcoin’s passport) - creating a less plutocratic voting environment.

That said, if we start scaling voting power by engagement with the SuperRare platform and contribution to the SuperRare DAO, I feel we reintroduce the issue of conflicts of interest. The people who arguably should have the most “karma” within the DAO are also the ones who would have the most potential for conflict of interest, as they are more likely to be intimately involved in the DAO’s operations.

If you are interested in participating in the planning and implementation of these structures for NTT badging and DAO administration, please do not hesitate to reach out as I am currently trying to identify members of the committee most interested and able to contribute to DAO meta-governance and ultimately form a working group around those problems.

Regarding Issue 3:

I agree with the need for increasing transparency in voting generally, however, I have concerns that being able to prove who voted for what opens the door for vote purchasing and other forms of incentivized or coercive collusion. If I have no way of verifying who you voted for, I have no incentive to bribe you to vote for me, because you could just vote for anyone. This is why voter confidentiality is such a critical component in most democracies.

That said, in the absence of integrity and equity in the voting system broadly, the benefits of voter transparency may outweigh the risks. I just think that we should carefully consider the downsides here and look if there are any ZK solutions or hints from existing democratic systems which could help us find the nuance needed to navigate this issue effectively.

Conclusion

This level of conversation is exactly what is needed for SuperRare to continue to grow and flourish. I want you to know that as DAO Program Manager, I take this input very seriously and am grateful for your contribution. My door and DMs are always open, and I try to make it a habit to check the forum daily. If you are able to attend, I would love to discuss these ideas further in tomorrow’s (November 29th) Meta-Governance call in the discord at 12pm Eastern Time.

1 Like

The initial token distribution heavily favored marketplace participants, investors, and employees as of August 2021. As you noted, there has not been additional faucets or airdrops since then.

I don’t think it changes the thrust of the issue, but I want to note that there have been additional airdrops as a result of different collector incentive programs through SRL.

Additionally, it’s not obvious to me that the design for the token distribution was bad. The airdrop at launch favored participants that had a substantial impact on the growth of the platform leading up to that point. Many that have chosen to hold their airdrop since that period have chosen to forego wealth in order to have a continued impact on this network.

And I think there’s room to discuss other ways of distributing voting power through $RARE to contemporary participants. That can dilute the ability for people to just buy-in and belly flop on any decision process.

Good to see there is already a follow-up proposal exploring the opportunities and challenges with quadroting voting. Big step forward for Superrare.

2 Likes

Would like to chime in regarding increasing sybil defense. Suggesting that SuperRare could in fact implement reputation voting based on a couple of factors, with the first one being a Gitcoin-like Passport.

The implementation should increase the trust of the user and should be reflected through a higher voting weight than someone that is not “verified”- essentially taking SuperRare one step away from plutocracy. I understand that it could be far-fetched before the next Space Race- but would love to see this in the future.

In regards to using NTT badging, I definitely believe that it should be implemented. Like Brennan mentioned, SRL employment, DAO Council involvement, sitting on a committee and attending discord events can be measured. I want to push it one step further, and include examples such as Discord Roles held, POAPs, OATs, other NTTs/NFTs held, length of $RARE tokens held should also play a key part in the Badge.

I believe that on-chain and off-chain contributions should definitely be encouraged especially in a time like now, where participation is super low, and Quadratic voting wouldn’t be able to fix this as it dilutes the votes of “whales”- but doesn’t incentivize contributors in anyway. Yes it may create a more equitable governance structure- but it doesn’t rally the other 99% of non-whales.

In terms of voting power- I’d like to see SuperRare implement a hybrid governance structure, where voting power is based on both reputation/roles as well as $RARE. It’ll be up to the DAO to decide the “sweet spot” as I assume $RARE holders want to hold onto their voting influence, and contributors will also want to increase theirs’.

I’m representing a project that is on the cusp of releasing a product that will be able to create all those things above mentioned- NTT Badging and Hybrid voting (token-based and/or reputation based). I’d love to be part of the discussing/planning and implementation of these structures.

In terms of the SR voting procedures

  • What are additional pain points when it comes to the voting mechanisms?
  • Are there UI/UX requirements that SR would like to see? (Would it help to have a live visual UI of NTT votes vs $Rare Votes?)

Is there a way to reach out to you Brennan?

Hey @slimbokk! Absolutely, I am happy to connect. Feel free to reach out to me on twitter at @bmulligan56 or in Discord. We have informal “meta-governance and chill” chats every tuesday at 12pm ET, and a more structured DAO townhall every other Wednesday at 2pm ET (the next town hall is this Wednesday January 4th).

Looking forward to chatting more.

On the contrary, I think QV will increase participation of small holders, since their small voting power is magnified.

Take this simple example. A has 100 tokens, B has 4 tokens. With linear voting, the voting power ratio A/B is 100/4 = 25. With quadratic voting the same ratio is 10/2 = 5. Hence, with QV, the small holder B has 5 times more power w.r.t. the big holder A.

Sounds good, I’ve given ya a follow however your DMs are closed (@slimbokk).

I see your point, however I believe users need more of an incentive to vote. There needs to be some sort of more contribution = more voting power-type mechanism in a community like SuperRare- I don’t believe that just because a whale has less voting power, smaller holders are more inclined to vote.

This is where I think reputation plays a key part in governance.

I agree in principle, however, one of my biggest reservations around reputation based voting is who gets to decide some key points:

  • Who decides which types of contribution/participation are deserving of increased governance reputation
  • Who decides what the requirements / standards are for evaluating whether someone has adequately contributed/participated to receive increased participation
  • Who decides how much weight is given to each different type of contribution/participation
  • Who decides whether someone has fulfilled the requirements for a certain type of contribution/participation (some may be automatable, but others will not be, and in either case the system would be inherently “gameable”)

Each of the above represents different stages of governance, and present unique opportunities for power capture. We also have to ensure that the system of rules we create accurately track what we want them to track while maintaining the right balance between flexibility and rigidity so that they remain legitimate over time.

This entails a significant meta-governance workload, and also significantly increases systemic complexity.

As such, we need to think very deeply about how this should work, and be very careful in our implementation so as to avoid unintended consequences.

Phillip Sheldrake, a philosopher and identity researcher, recently had an excellent interview with Kevin Owoki on the Green Pill podcast about the opportunities and challenges with digital identity, and why this is an area where careful consideration is required.

All that said, the grants committee will be issuing Otterspace badges to successful grants round 1 applicants for credentialing and general reputation. We are also actively investigating how Otterspace and other NTT frameworks can be used for broader community participation and contribution. Ultimately, I would love to see all of this roll together to provide sybil resistance for quadratic voting and similar governance utility through solutions like gitcoin passport.

Hopefully that helps provide more context about how I’m thinking about this issue, and the incremental steps I’m taking as chair of the grants committee to gather data for the DAO about how NTT badging might be implemented. Ultimately though, this should be a multi-stakeholder effort and we should work together (along with other interested community members and subject matter experts) to form a cohesive strategy and shared understanding