[BIP-435] Change to Yes based Quorum of 1 Million Yes votes

Governance General Proposal
1

Background

Following on from [BIP-163] Restructure Governance Process - Disband Governance Council, which laid out a clear process for all stages of governance.

This RFC is based on changes made by BIP-300 and the AIP resulting from this AURA Temperature Check changed the way that voting works such that 100% of AURA’s vlAURA is always registered as either (yes, no or abstian). With 3+ million vlAURA locked, this means that the current quorum of 2 million votes will never fail to be met. Quroum has been rendered meaningless.

These changes are being made to restore the Quorum while making minimal changes to the intended voting dynamic.

Yes Based Quorum

This BIP proposes quorum is moved from counting total votes, to requiring a minimum number of Yes votes as well as a majority of Yes votes for governance to pass. As the current quorum is 2 million votes, a new Quorum of 1 million Yes(affirmiative) votes should have the same intended effect, without large wrapper abstaination resulting in the passage BIPs that do not have a quorum of supporting interest.

Specification

Going forward the votes counted towards Quorum will be determined by adding the total number of affirmitive votes. The quorum will changed from 2 million votes in any direction, to 1 million Yes votes.

2 Likes
2

If Abstain doesn’t count towards quorum, it’s unclear to me why it’s necessary as a voting option.

3 Likes
3

Agree with @franklin’s comment. Abstentions traditionally count when it comes to determining if a quorum has been satisfied.

RE: the above, in addition to the “Abstain” option (assuming it has meaning), we may want to change the voting options a bit to something like, (1) “Yes, let’s do it” and “No, this is not the way,” or (2) option voting for different results, similar to what we saw with BIP-161 (Close the Year 2 Funding Gap - Options A, B, and C along with “More discussion is needed”) and BIP-57b (2% v.5% Gauge Framework cap options).

If a change like this is made, the second bullet of the structure would have to also be adjusted accordingly.

1 Like
4

Let’s say that I am a vlAURA voter, and I want to not vote for something. If I don’t vote on the Aura snapshot, then my votes are just lumped in with everyone else.

If I explicitly wish to note vote for for a matter, I can indicate that by choosing abstain. Then those votes will not be counted towards Quorum.

Make sense?

Are you looking to add 4 options? I don’t think this needs to be so complicated, but if there are voters who also wish to abstain and help Quorum, but not vote in a direction we can deal with it. The foundation members said it was fine and they can just stay out of the things they shouldn’t take a position on.

The 3 current options are dictated by [BIP-163] Restructure Governance Process - Disband Governance Council

5

If Abstain has no meaning on a BIP, then imo, it shouldn’t be included as an option. We wouldn’t include Maybe, Maybe Yes, Maybe No, as options on a BIP, for example, and say, these options have no meaning, and don’t count towards quorum. If a voting choice has no meaning on a BIP, then it shouldn’t be included as an option, therefore, it wouldn’t be reflected on Aura’s side. So we would have no Abstain option to allocate votes to. Each voting option should have a purpose.

6

Putting aside the abstention issue for the moment, for the sake of clarity and consistency between this BIP and BIP-163, I would just incorporate the current 3 options referenced in BIP-163 [+ abstain if it goes that way] into the “Structure of Snapshots” part of this post, so the first bullet would read:

  • Going forward, all Snapshot votes shall be posted as Weighted Voting with Options consistent with BIP-163 [+ abstain if this is modified]

And just so we are clear, my use of brackets is to indicate that the text between the brackets isn’t included in the current proposed language, unless this BIP winds up being changed to incorporate abstentions for quorum purposes in the future.

1 Like
7

Abstain means that you do not want your share of the vote to be used to count towards quorum nor that you want to vote in any direction. This makes it possible to indicate as such.

BIP-163 already has Abstain in it. It was just not clear how Abstain was handled, and precident so far has been to count it towards the quorum. This BIP changes that. Am I missing something?

If the Aura Delegate(s) wants the ability to abstain but push quorum we can figure that out by adding a 4th option.

Yes, let's do it.
No, this is not the way.
Abstain - with quorum support
Abstain - without quorum support

It just seems quite complex. We want to make governance accessible and understandable and transparent.

1 Like
8

Apologies - I didn’t reread the whole thing today. It’s a great proposal but is a lot of text. :sweat_smile: I still recommend incorporating BIP-163 by reference for the sake of consistency. If modifications are needed then they can be made by doing something like," The Options are set forth in BIP-163, except ‘xzy…’"

I agree. That’s why it seems odd to me to have abstain as an option without it counting towards quorum. That in and of itself is already confusing to me.

1 Like
9

You’re 100% correct. I added a reference to BIP-163 in the first sentence. I thought it was already there :sweat_smile:

1 Like
10

Would incorporate it here too. Otherwise, the question of whether there is flexibility to vote with other options besides yes, no, abstain, becomes confusing. Per BIP-163:

1 Like
11

The complexity arises when we change the meaning of Abstain.

Here’s the current system:

Yes (All votes are counted, and count towards quorum.)
No (All votes are counted and count towards quorum.)
Abstain (Non-Aura Delegate votes count and count towards quorum. Aura Delegate votes do not count and are disregarded for quorum.)

Here’s what the new Abstain would look like, under weighted voting:

Abstain v2 (All Aura votes, not just Aura Delegate votes, do not count and are disregarded for quorum.)

The difference is subtle–the byproduct of weighted voting. One person voting Abstain will now command a larger share of total Aura votes than if they had voted as an individual. If all Aura Abstain votes (not just Aura Delegate votes) are shoved into Abstain v2, what ends up happening is that you end up disenfranchising a greater portion of votes that otherwise should have been counted. IMO, it’s okay for votes to be amplified and counted, as that gives individuals a greater voice in governance, but it’s not okay for votes to be amplified, then disregarded, as that strips away governance authority, and removes it from a greater number than perhaps may have been intended.

Under the current system, we don’t encounter this issue. Abstain retains its regular, widely-understood meaning–votes count and count towards quorum. We’ve never had this issue before, but if necessary, current Abstain votes can be parsed for non-Aura Delegate votes, and Aura-Delegate votes, so all votes are meaningful. Aura Delegate has authority over its own votes by virtue of its delegations. It can choose for its own votes to be counted or not counted, but cannot disenfranchise others.

Moreover, let’s go back one step, to my original comment. If you have an option such as Abstain v2 where votes are not counted, then imo, it should not be included as an option on a BIP. All voting options on a BIP should have meaning, within the context of the originating BIP/originating proposer itself. It’s the reason why we don’t include Maybe, Maybe Not, as options on BIPs, then disregard the votes later. Currently, votes on every option, including Abstain, do count. If we introduce Abstain v2, it doesn’t seem like that is a meaningful option within the context of the original BIP, and therefore, should not be included. Vote here on this option for fun–your vote will not be counted. There’s no reason to include something like that when the BIP is examined in a non-Aura vacuum.

Aura BIPs are simply a mirror of Balancer BIPs. If an option on Balancer BIP is removed, then the Aura BIP does not have the optionality of adding a new Abstain with its own meaning, or tweaking For or Against to mean different things. The Balancer BIP is controlling, the Aura BIP is just a reflection. So, if a meaningless option is removed on Balancer’s side, Aura should not be able to dictate this option be added back into the BIP, especially, when combined with the analysis above, that option disenfranchises a portion of total Aura votes or perhaps confuses Balancer voters that may end up mistakenly selecting Abstain v2.

The simplest solution seems to be to stick with regular English meanings for For, Against, Abstain. All Aura votes are counted. All Abstain votes count towards quorum.

2 Likes
12

Ok. We can drop this for now.

1 Like
13

Let’s revisit this one. A quorum reduction can also be tied in.

14

Sounds good. What about just changing the quorum requirement to be:

“A vote that passes with less than 750,000 Yes votes will not be considered to have reached quorum”

15

I just flipped through the last 20 BIPs. It looks like we’re ranging between 3.4 to 3.7 M on Yes. Ex-Aura, that’s 300 to 600K Yes votes. With the 750 number, is it your intention to try and get more people involved in governance?

16

Yes, just a little bit/baby steps :slight_smile:
We could do 500=600k yes to meet the quorum as well :slight_smile:

We just saw that if Solarcurve actually votes with his Aura, he can get more than half way there on his own. In the end voters have to be less lazy, and seem to show up when it really matters, and 750k requires less than 10% of the total veBAL electorate (almost all of whom vote from time to time for emissions) to pass.

At best we can get our voters more involved, and move back to 1 million yes votes. Was thinking about adding to this BIP or another BIP, some guidance (social) requiring DAOs and wrappers to actively participate in governance most of the time or delegate to someone who does. This is probably the easiest group to get engaged.

17

We kicked the numbers around a bit, and our preference would be for 1 to 1.5. 750K feels a bit low and a bit arbitrary.

In terms of improving governance, what are your thoughts on a vote cap each week, perhaps 5.

The ideal goal here would be to make individual votes more important, and make BIPs worth noticing and reading.

2 Likes
18

strongly against a vote cap unless we change the system for gauge voting. i.e. optimistic approval or batching them together.

if we cap proposals per week at 5 we’ll run into serious problems with gauges being approved.

19

Disagree with capping as well, and hardly enforceable. Maybe a little clean-up and better organization on Discourse to remove the gauge voting from the landing page would help voters and delegates access the relevant BIPs?

I’m afraid that leaving Aura’s abstentions out of the [AIP-27] weighting might amplify a minority (or even a single voter). So I would support ignoring Abstention votes for quorum calculations.

That being said, as of today, four Maxis delegates together have ~1mio veBAL, so 750k is not good. In case Aura abstains, it will push too much decision-making power towards an entity with great influence in the community already. From that POV, 1,5mio looks better (just because I’m afraid we can’t reach 2mio). Ofc this has nothing to do with competence, but with the risks involved in perceived centralization in governance.

1 Like
20

Ok so how about this:

1: The maxis will remain active and vote on things that are important to them.

2: The quorum will remain as it is but redefined as 1 million yes votes.

3: Votes in which the delegates who contribute to balancer have to abstain for technical reasons will count their abstain votes towards the yes quorum.

We could even move to 1.25 million yes votes under those conditions it seems.

It would also be nice to see a bit of a marketing push from Aura to try to engage large holders to move their delegate to some other delegates and/or be a little more active voting directly on important/contentious governance matters.