GIP-77: Should the GnosisDAO add delegation & spam reduction?

GIP: 77
title: Should the GnosisDAO add delegation & spam reduction?
author: Gnosis Guild
status: Draft
type: Meta
created: 2023-01-20
  • Let’s do this!
  • Make no change

0 voters

Simple Summary

This proposal aims to introduce a number of changes to GnosisDAO’s governance setup which will:

  1. Reduce low-quality and spam proposals created in the GnosisDAO’s snapshot space.
  2. Make GnosisDAO’s influence distribution less opaque.
  3. Improve vote-weight utilization.

Abstract

Broadly, this proposal aims to introduce two changes:

  1. Modify the criteria for users to post proposals to the GnosisDAO’s snapshot space.
    1. Accounts holding ≥ 1 GNO can create a proposal (current).
    2. Add some trusted moderators that have the ability to remove posts that they deem to be low-quality. (new)
    3. Proposals with less than < 1,000 GNO of vote-weight cast for or abstain are filtered from the main view. (new)
  2. Improve delegation
    1. Improve the delegation UI inside of snapshot to make delegation much simpler and to give an overview of the current delegated vote-weight distribution.
    2. Add a two-year term limit to delegation, to avoid stagnation in the vote-weight distribution; GNO holders must periodically set their delegation, all delegations expire after two-years.
    3. Allow delegates to decline being delegated to / resign from delegation, negating any vote weight delegated to them.
    4. Allow GNO holders to delegate portions of their vote weight to multiple accounts.
    5. Allow for transitive delegation; If A delegates to B and B delegates to C, C inherits A’s vote weight.
    6. Allow delegates to optionally set a cap on the amount of vote weight delegated to them.

Motivation

Popularized by Compound Governor, delegation has proven to be an effective strategy for improving the vote-weight utilization in many DAOs; consolidating vote-weight in the hands of those who have the time and inclination to actively participate in governance by allowing users to delegate their vote weight to others.

Largely due to the constraints of calculating vote-weight on-chain, Compound and OZ Governor’s delegation is somewhat limited. GnosisDAO’s existing delegation strategy on Snapshot already improves on one of these limitations by allowing users to delegate their vote weight and also directly participate in any vote. If a user does not cast a vote, their vote weight will be applied to the vote cast by their delegate. If a user does cast a vote, their vote weight will not count towards their delegates vote weight in that proposal.

However, there is significant design-space to further improve on GnosisDAO’s delegation.

Specifically, we would like to introduce changes that:

  1. Reduce low-quality and spam proposals created in the GnosisDAO’s snapshot space.
  2. Make GnosisDAO’s influence distribution less opaque.
  3. Improve vote-weight utilization.

Another recent forum post also argues for delegation, along with suggesting compensation for delegates. To constrain scope, this proposal does not include delegate compensation. However, if this proposal is successful, and after observing the delegation setup in production, we will likely write a followup proposal to implement some form of delegate compensation.

Specification

TODO

Rationale

TODO

Implementation

TODO

8 Likes

@auryn_macmillan This is an amazing proposal that addresses many issues and adds great improvements to delegate voting!
I’m in support and ofc will vote for the proposal

It is also in line with many improvements that we were considering for CoW DAO. We will definitely be looking to use these new tooling to introduce a delegation program for the CoW DAO, piggy backing on your work.

For this feature, I assume the exact configuration will be exposed in the space settings?
Where users would be able to find the low quality proposals?

Where those settings will be exposed? I assume in the profile section? Or would there be a user-specific settings section for each space?

This should ideally also include information about the term expiration for each delegate, so delegates and users are aware of upcoming expirations.

Few additional ideas I had:

  • Improve the profile page to include a space-specific bio field. This bio will be used if the user was chosen to be displayed as delegate in the “delegate” page
  • Improve the “delegate” page - Add a space defined delegate-list strategy. This will define which profiles are eligible to be displayed to users when they go to the “delegate” page and look for delegates. In addition the strategy will define the order in which the delegate list is displayed. Either top delegates like it is today, or randomly ordered, or any other arbitrary sorting. (I assume this will have a significant bias effect on delegate choice, so at least space owners can control WHO is being displayed and in WHAT order)
2 Likes

Yeah, I think that’s a safe assumption. The value should be configurable.

Not sure about this yet. Admins in Snapshot are currently able to archive proposals, so I’d imagine that low quality proposals are also put in this same state.

I’m honestly not sure yet. I think we’ll probably modify the delegate tab on Snapshot, so perhaps this is the natural place for these settings to live.

1 Like

Thanks for posting this @auryn_macmillan, and to the team that worked on it!

This could be clear for most of all who have been looking at the snapshot space recently, but some examples or general guidelines may help interpret it. E.g. low quality means the GIP is not following the GIP template GIP-0: Template.

Are moderators going to be known, e.g. information added to the space bio, for people to reach out if they are looking for feedback?

I was wondering if a cap on max delegation would help in:

  • Create some additional friction for people that have enough vote power to guarantee quorum
  • Avoid the perception that participating in snapshot voting may be pointless when one wallet is seen to provide quorum and make a decision.

It is granted that such a user could still delegate to more than one wallet in their control. But I thought it may be worth the time considering those two bullets if the effort to implement it is reasonable.

Cheers!

Agreed. The rules around moderation will need to be explicit.

I assume the moderators will be a subset of the forum moderators. Likely it would be @john_szczepaniak and perhaps a small handful of others.

As you said, such a user could trivially delegate to more than one wallet, so I think this is probably a moot point. If we want to make it more challenging for a single entity to reach quorum, we would need to explore different ways to weighting votes; quadratic voting, for example.

3 Likes

Thank you @auryn_macmillan for posting this!

I would suggest some alternatives to this point : Allow for transitive delegation; If A delegates to B and B delegates to C, C inherits A’s vote weight.

As a GNO holder, I’d like to avoid delegated accounts to sub-delegate my vote power indefinitely, I think it’s everyone’s right to decide on whether or not transitive delegation should be enabled, probably we can have an option to allow/block that behaviour.

What do you think?

2 Likes

It’s certainly an option we can explore, but my knee-jerk objections to this are:

  1. It likely adds some complexity to the implementation.
  2. It would likely reduce vote-weight utilization, since A’s vote would not end up being used if B delegates to C, and neither A nor B vote directly.
  3. It feels somewhat redundant. Practically, there is very little difference between B delegating to C and B mirroring Cs vote. In either case, A’s vote ends up being cast in the same way as C. However, the latter required B’s active participation in each proposal, which probably reduces vote-weight utilization (as in 2).
  4. This seems like an issue between A and B. If A does not want their vote weight delegated further, they should pick a delegate who promises not to delegate further. If B delegates to C, then A can always vote directly to override that vote weight and/or change their delegation.
1 Like

I’m aware that I am bumping and older post, but these are great initiatives.

Is there a place I can look for status updates on this?

2 Likes

This took a little longer to implement than we had initially expected. We’re hopefully wrapping up work on it in the next ~month, at which point we’ll make a proposal on Snapshot to update the DAO settings.

You can keep an eye on the delegate registry repo to track progress.

5 Likes

Hey @auryn_macmillan, StableLab is here to support this initiative if you are open for support.
We have a wide range of experience with delegation from both sides. Happy to make your life easier.

Thanks to @coltron.eth for making us aware of this GIP.

1 Like