Modify abstain so that it does not count toward quorum

Currently, the cosmos hub has an abstain vote option, and currently it has three use cases:

• virtue signaling participation
• avoidance of conflicts of interest
  • Due to the functionality of abstain as a soft yes, it doesn’t really help with the avoidance of conflicts of interest.
• boosting quorum

The way that abstain counts toward quorum is harmful.

Vote options

Vote YES to stop counting abstain votes when tallying quorum
Vote NO to take no action
Vote ABSTAIN to express no opinion
Vote NOWITHVETO to to contribute to a veto tally. If the veto tally reaches one third, then the proposal will fail and the deposit on the proposal will be burned.

I think removing the “Abstain”, option is a terrible idea.

this:

VelvetMilkman🥛

Hush:

removing the “Abstain”, option is a terrible idea.

Sure it seems to me that your sentence is missing a second clause, or are you simply trying to say that it’s a terrible idea but you don’t care to say why it’s a terrible idea and you think that’s sufficient?

Thanks

tom:

this:

He represents a validation organization ought to help make it serious if that’s his point of view I can only write for my own point of view.

I don’t think that removing the abstain option would make it better concerning ‘non-participatory validators’.

If the Abstain option is removed they will blindly choose another option that might not express their behaviour which in my opinion is more harmful than allowing for an Abstain.

Indeed, using Abstain constantly isn’t the best use of participation within governance, however, I would prefer it rather than a validator who might not care to be voting randomly with yes/no.

There’s honest reasons to vote abstain (potential legal ramifications)

But yes, most of the time abstain is just abused.

This is all just governance circle jerk theater though. Ill probably abstain this for the thrill.

Golden-Ratio-Staking:

Ill probably abstain this for the thrill.

I figured that some people would, and I take no issue with your stance good sir.

Damien:

Indeed, using Abstain constantly isn’t the best use of participation within governance, however, I would prefer it rather than a validator who might not care to be voting randomly with yes/no.

It is my belief that because yes and no have consequences, there will be some thought behind the votes. If a validator is voting capriciously and not explaining those votes, then I believe that finally, redelegations will occur.

Golden-Ratio-Staking:

There’s honest reasons to vote abstain (potential legal ramifications)

I agree with this.

Golden-Ratio-Staking:

But yes, most of the time abstain is just abused.

Also with this.

jacobgadikian:

It is my belief that because yes and no have consequences, there will be some thought behind the votes. If a validator is voting capriciously and not explaining those votes, then I believe that finally, redelegations will occur.

Yes and No do indeed have consequences. If an honest validator has no opinion to express on the matter and the abstain vote is removed, (be it for legal or conflict of interest etc.) should they not participate in that governance matter?

I believe that if there is some way to show delegators that even if a validator has a high ‘participation’ rate, there should be a measure of meaningful participation (i.e % of abstain possibly?) - I think that would be better and that would also lead to redelegations if a validator is essentially free-riding governance.

Damien:

should they not participate in that governance matter?

Yep. If that’s how they vibe on it.

Damien:

I believe that if there is some way to show delegators that even if a validator has a high ‘participation’ rate, there should be a measure of meaningful participation (i.e % of abstain possibly?) - I think that would be better and that would also lead to redelegations if a validator is essentially free-riding governance.

I still haven’t seen a time when abstain did good things. Can think of many times it did bad things.

So I guess what I’m claiming is that it’s better to just remove abstain. I understand that we could have new visualizations and stuff but I think that it has led to a number of perverse outcomes.

@jacobgadikian to me, your prop sounds kinda incomplete like that;
it’s a good intention, though; we all know here that governance requires some re-thinking and updating, and the more we will bring that to the discussion - the more chance to figure out something better;
also, it looks like the division only to black\white, which isn’t correct according to society’s nature

1. yep, a high volume of abstaining currently sometimes turns into rofl and abuse
2. no, with those jerking on participation metrics, it’s not very fair to just remove abstain option (e.g., prop 818 for citizen and pumpos, or comm spend prop for those who request)
3. yes, sometimes, as a validator, we are willing to show that we simply don’t care about some specific shit
4. yes, legal matters sometimes matter (when you are afraid to support some comm. spend fearing some governmental persuasion)

so, maybe we should re-think how abstain affects voting through adjustments in quorum % and % of yes\no for the prop to become a decision? or even add one more “fuck you” option, which, vice versa, will be deducted from the quorum?

altbro: yep, a high volume of abstaining currently sometimes turns into rofl and abuse I would say that it’s pretty frequent. altbro: no, with those jerking on participation metrics, it’s not very fair to just remove abstain option (e.g., prop 818 for citizen and pumpos, or comm spend prop for those who request) I thought about these angles too and check it out, I think that removing amstein actually reduces the conflict of interest. I’ll give you some examples: • notional would have voted yes on 104. That was our actual position, anyhow. • @pupmos and @serejandmyself would have voted no on 818 and that was their actual position anyhow. altbro: yes, sometimes, as a validator, we are willing to show that we simply don’t care about some specific shit Then don’t vote. altbro: yes, legal matters sometimes matter (when you are afraid to support some comm. spend fearing some governmental persuasion) Once again, don’t vote. So the issue with abstain is that it really does have an effect on outcomes. Therefore in my opinion it’s better to simply not have it. altbro: or even add one more “fuck you”…

Excerpt (1196 of 1274 characters). Read the whole post on the forum ↗

jacobgadikian:

• notional would have voted yes on 104. That was our actual position, anyhow.
• @pupmos and @serejandmyself would have voted no on 818 and that was their actual position anyhow.

and both of you would have been dishonest doing so.

i agree modifications have to be found.

just “remove” is a rushed solution.

imo

1. abstain vote should be renamed blank vote
2. abstaining is when you’re not voting
3. blank vote should have different consequences when they come from validators and from delegators
4. some amount of blank vote should reboot the discussion/proposal and eventually disqualify the initial proposer

I do think the abstain vote is helpful in case people have a conflict of interest. My concern with the abstain vote is that it could technically mean that any proposal with a 90% Abstain, 6% Yes and 4% No, would pass.

Something about that seems wrong to me.

Imo, either we stop counting abstain to quorum, or we setup something where if abstain votes > quorum threshold (40%), the proposal fails.

How about if voting abstain doesn’t have an effect on anything? For now, abstain was used to fulfill the quorum number.

I am against just simply removing abstain because it most likely lowers participation and increases the challenge around hitting quorum.

There are definite issues about governance but I suspect this just moves how the problem presents itself rather than solving anything.

I’d support that.

Like most here I believe the governance process should be revamped, but simply removing that option does not sound like a solution.

Abstain is generally counted to quorum because historically chains have a hard time reaching quorum in their early stages. But considering the Cosmos Hub has such an active governance, I think we’re well positioned to simply stop counting abstain to quorum, but keeping the option there so validators can signal a conflict of interest.

Especially if we ever more to a governator system (can we please ) where validators who don’t care about governance can opt out and any address can become a governator, this should likely play out well.

We could also consider reducing quorum to ~35% if we’re concerned about reaching the threshold without abstain.

FYI, not counting abstain to quorum, does fix the problems with abstain.

I would still view it as kind of a this schedule organ, but I would also say that probably it makes sense to break off into a little informal working group and have a chat and maybe see if we can set alarm one option or the other.

@jacobgadikian my reading of James’ comment is some on chain analytics would help frame and quantify the problem of Abstain.
For instance, how often has it helped unpopular proposals pass by helping the vote reach quorum, e.g., Prop #89 Ecosystem Vids by Cosmic validator (Y 27%, No 26%, Abstain 47%,)

If you recall, in Prop #82, because abstain is neither in numerator or denominator, “No” votes were working as “Yes” votes. Because voting No just reduced the NWV percentage. Voting No on 82 made it more difficult to reject the proposal w NWV.

I don’t think we can have a simple proposal to remove Abstain. We need to investigate impact to quorum, and consider whether threshold for quorum must be lowered.

We also need to consider whether a change from simple majority to simple majority to win is also needed eg 2/3, 3/5.

Thank you for this comprehensive discussion on the matter of the abstain vote within our governance framework. Recognizing the valid concerns raised by some members regarding the potential misuse and adverse effects of this option, I would like to offer some alternative perspectives in favor of retaining, rather than entirely removing, this feature. After careful consideration, I present the following reasons why the abstain option is valuable to our community: • Contribution to Quorum Requirements : Abstaining from a vote is more than an expression of indifference; it’s a conscious choice with essential implications for our decision-making. By aiding in reaching a quorum, it enables a broader evaluation of proposals, fostering a more robust and inclusive process. • Facilitating Participation for Non-Technical Voters : The Abstain option serves as a vital tool for community members who may not possess the specialized knowledge to grasp intricate technical matters fully. For example, I sometimes use the Abstain vote on technical proposals or technical changes since I don’t know if adding or removing certain technical functionality is good or bad. By choosing to abstain,…

Excerpt (1195 of 3504 characters). Read the whole post on the forum ↗

I know some chains have this option not sure for Cosmos hub has it, but if you were to remove abstain tomorrow there still exist a workaround. I am talking about weighted votes. In theory using CLI you could vote 75% yes and 25% no, or in this case 50-50 and it could count as an abstain in a way.

While I agree that sometimes the abstain option is abused often, removing it would do more harm than good.
For example, constant governance participation is considered a good practice and a good criteria for choosing a validator. If a validator has a proposal that they are legally unable to vote on without getting sanctioned, or they are one of the parties involved so they cannot vote without it causing a conflict of interest, or other cases I am pretty sure might happen. And while not voting for the proposal can basically reach the same outcome (except for not being counted for quorum), it actually shows the validator made their choice and voted, while when the validator hasn’t voted it may just show that it doesn’t care about governance. Not having a separate abstain vote makes it impossible to differentiate a conscious validator choice and ignorance about the governance in general and this specific proposal.
If this proposal would go live on chain as is, I would vote “no”, as it makes more harm than good, but I’m all in for the discussion about how it can be changed without breaking existing mechanics, as there’s definitely a lot to think about and reconsider.

If a validator cannot abstain,
The validator will weight vote 50/50 yes/no and it will be the same
Abstain is not a option to remove

Hello Sir Gadikian - hope all is well with you.

The Top 20 Validators should be voting abstain, no? To further decentralization.

I don’t believe we should be removing abstain for this reason.

Psinn:

The Top 20 Validators should be voting abstain, no? To further decentralization.

This doesn’t even make a little sense. Can you explain?

How about this as a compromise - preserve abstain as a vote option to enable validators to express conflicts of interest, when they exist.

Many have contractual obligations to vote on % of proposals.

Remove abstain from counting towards quorum.

That’s what the changes did.

Cosmos_Nanny:

Remove abstain from counting towards quorum.

that’s not enough.

if blank vote doesn’t count anymore in the quorum, it should have some super power (as NWV has) in some exceptional conditions.

otherwise you nuke it without nuking it.
it would solve an issue by ignoring and making meaningless a fair part of voters.

blank vote from delegators (at least) should be able to weight and mean something when it gets some relative majority.

I thought that blank didn’t have any meaning?

yeah that’s why your whole proposal is wrong and has already been amended

It’s been amended to make it more correct.

Is there a problem with it?

it’s been amended to make it more politically correct.

its purpose remain the same.

Agree that A) 90% of staked ATOM voting abstain with 6% yes and 4% no resulting in a proposal passing is not a good thing B) There are legitimate uses of Abstain such as for conflict of interest, not having a strong opinion but actively considering and following the proposal & developments around it. Contributing to Quorum to help proposals pass (if it would not be hit often without abstain exisiting) Currently I do not see the abstain vote as an issue that can be fixed by simply removing it. @David_Crosnest and others make a good point that without abstain, validators can still vote 50/50 on YES/NO to effectively the same result, so then the question is which is better, voting abstain or voting 50/50? I dont see what removing abstain from the voting option can achieve here. Instead perhaps we can only really hold validators socially accountable to the fact that they often abstain from voting on proposals and encourage them to vote yes or no more often to contribute to proposals in a more meaningful way I appreciate the discussion around the current use of abstain and whether it should be revised, modified in its implications or removed completely. Cosmos governance can…

Excerpt (1198 of 1349 characters). Read the whole post on the forum ↗

Great so check it out, can you give me one example of a time when holding a validators socially accountable has proven to be an effective sanction?

I wasnt necessarily saying it was effective, but I’m saying that removing abstain is quite possibly less effective since a 50/50 split between yes/no votes will mean it has no effect and therefore the best we have is still holding validators socially accountable in this situation.

Thanks for the discussion as always

So okay I guess what I would suggest is that we toss out the idea of social sanction because we both agree that it doesn’t work, and then we work on other things that might have a chance of working

Others have mentioned why Abstain is useful (conflicts of interest, jurisdictional/regulatory compliance, etc.), and I’d add that Abstain votes can be regarded as a proxy for the overall legitimacy of particular policies which are neither spam nor malicious—just like not voting in elections can be a proxy for the legitimacy of the election/electoral process itself. A policy passed with 90% abstain, 6% yes, and 4% no is likely easy to reverse or amend by future governance, though of course this depends. Cutting Abstain’s weight to quorum may have other effects, but I haven’t analyzed the voting patterns to figure out what. What I’d suggest instead of removing Abstain or discounting Abstain to quorum is some sort of (automated, not governance-based) slashing for validators who miss votes or vote Abstain beyond certain thresholds, with an option for those slashed by repetitive Abstaining to appeal (should they have legitimate reasons for it). Slashed funds like this would need to be held in escrow for a duration to allow appeals to be processed. Alternatively/additionally, consider automating a dynamic vote system where the more Abstains are present, the higher the passing…

Excerpt (1194 of 1600 characters). Read the whole post on the forum ↗

many have mentioned why abstain is useful, and what could be working and improving modifications. but i guess the authors don’t care.

Always appreciate engaging with you, sir. Thank you for your response.

Those in TOP 20 have most voting power. Abstain vote helps with balance (for now).

To avoid any hostile takeover via governance - TOP 20 could be required to vote abstain (plug: ATOM One Constitution conversation) in respects to ensuring smaller validators voices to be fully heard/validated, W/O fear of companies with large amount of capital to swoop in and do it’s greedy thing.

*just brainstorming
what if we add 2/3 threshold to abstain (like NWV have), and if it reached by the end of voting prop should either counted as NO, or loop to the new round of voting
seems like main concern among replies mention “no one like when prop with 90% asatain, 6% yes and 4% no passes”

I’m not sure that this solves the problem that you’re discussing, because basically it breaks proof of stake.

The stake of those large validators is also valuable and they paid for it.

100% - large validators stake are valuable indeed

Hey, i’m not sure either - however, if top 20 val set were required to utilize abstain from certain types of prop - abstain could serve the eco well, no? because otherwise, a calculated/malicious group effort by whales are plausible.

As always, thankful to engage with you sir.

Oh and thankful to you as well.

The trouble is that I really don’t think that we should require them to abstain because that invalidates the choices of their delegators.

Delegators to let’s say @Tosch 's sg1 or to @p2porg should have the same voting rights as delegators to notional.

Does that make sense to you?

don’t they have the same right to override?

Makes enough sense to respond with this plug/question:

What are your thoughts on proposing edits to the Atom One Constitution FWD Prop regarding this?

Sounds like the perfect place to resolve.

Yes and the override mechanism is really excellent. However, the purpose of delegation is to allow someone else to take actions on your behalf.

So I don’t think it would be good to prevent validators from voting as suggested here, no matter how well ranked

I agree that the hub needs a constitution. I’m happy to have a look at that one but probably not today, I am working on a number of software patches.

If I were going to take a stab at creating a constitution for the hub, I would be looking to create something that:

• Retains the present economic model
• Encourages the practice of forking in times of political disagreement or turmoil
• Guides the hub itself to the most minimal possible code base
• brings features to the AEZ atomically
• Keeps quorum where it is (it’s a parameter anyhow iirc)
• eliminates abstain
• Preserves veto for only the following governance proposal types:
  • client update
  • software upgrade
  • ICS consumer chain addition

I don’t know if these ideas are interesting or jibe with the a1 constitution but that’s approximately where I’d drive things.

On the ever controversial issue of CosmWasm on the hub, I’ll restate my preference matrix, which I think is pretty well in line with most folks building in cosmos:

1. mesh in go and no contracts
2. CosmWasm in go and only mesh or infra contracts
3. CosmWasm in rust and only mesh

Probably I would vote yes on any of one two or three.

a constitution shouldn’t exist to set economic models or to define governance parameters or to bring features.

a constitution should exist to set limits and to design a path toward a common goal.

Screenshot_2023-08-22-11-27-20-359_com.brave.browser-edit1080×1297 288 KB

circling back after seeing the title changed to “modify abstain so that it does not count toward quorum”

I have very low faith in the cosmoshub governance system to meet quorum without the use of abstain on any, even remotely, controversial proposals.

You a boss, Jacob. Salute. Collaboration ftw. We can do this.

If you don’t mind, i’ll copy paste your thoughts on the Constitution post?

Sure, that’s totally fine with me.

Remember you can always select the text, click quote, and quote anything on this app kinda 4chan style.

If I recall correctly, ATOM one had one thing that I really didn’t like from a design perspective and that is increasing quorum.

I don’t think there’s a reasonable way to increase quorum and remove abstain and between the two options, I would much rather remove abstain because it influences votes without technically even being voting.

thanks for notice, changing title is bad

not counting abstentions in the quorum is not a good solution either.

• If I’m not sure I can vote on a proposal, I abstain.
  => I’m not avoiding the vote, I’m simply indicating that I’m not competent to make the decision.
• If I want to avoid a conflict of interest, I abstain.
  => I don’t want my vote to be misinterpreted.

in both situations, I just step aside to let the right people decide.

If the intent is to use Abstain to express neutrality or perhaps indifference, 50/50 weighted is the correct vote. In the event a validator wishes to express “conflict” by voting abstain, while it’s socially accepted as expressing “no vote,” voting abstain does not recuse a voting validator. Technically and quantitatively it has a calculable effect. In fact, voting abstain in non sequential elections where relative quotas are used as qualifying requirements – tallied abstentions in these contexts carry well known “no-show paradox” effects. The impact of abstain in establishing QUORUM (ABSOLUTE quota, 40%, proportion of network) must not be conflated with its effects on THRESHOLD (RELATIVE quota, 50%, minimum proportion of participating voting power required to pass a proposal). With a relative quota the # of votes required to win varies w # of votes cast; thus w relative quotas, the effect of voter’s decision to abstain REDUCES the number of votes necessary to pass a proposal. The issue with using abstain with current param for threshold, is that the abstainer’s potential support or opposition to the prop is often NOT removed in equal measure – 50/50, as commonly…

Excerpt (1191 of 1321 characters). Read the whole post on the forum ↗

then there should have work done on threshold/quorum parameters, instead of “nuking” anything.

Yes Tom. Think the communications here indicates the majority of us have moved beyond the nuke ABSTAIN position. Language is vital and instrumental in communicating ideas, and I understand how it can be alienating. And upsetting. But when you see the ideas expressed have substantively and substantially moved beyond a semantic thorn, let’s progress the discussion together rather than lite litigating word selection that was perhaps more cheeky than anything remotely malicious.

My updated thinking in modifying abstain is keeping it intact as applied for quorum. However removing abstain from calculating proposal threshold.

This is a more accurate way to express: “no vote,” or conflict of interest without losing participation points (as many validators are contractually obligated to do so).

Should a validator wish to express neutrality or impartiality, 50/50 weighted is the cleanest way to express this sentiment.

As to concerns of removing abstain and its speculated effects on public engagement, please know that much of the research literature (political science/social choice theory/rational choice theory) for 50 yrs suggests abstentions are more often signifiers of voters alienation and apathy.

Not the opposite.

Preserving abstain as a vote option does not promote or enhance voter or civic engagement.

It’s more often an expression of capitulation, while exercising one’s duty of participation (in large public elections and or small legislative bodies).

If you’re interested in this stuff academically, I recommend:

Anthony Downs, An Economic Theory of Democracy (classic)
G Tullock, Toward a Mathematics of Politics (classic)
M Holler, Power Indices and Coalition Formation
“Special Case of Penrose’s Limit Theorem When Abstention is Allowed”, Ines Lindner

i can understand your point. i know i can look like i’m upset, i have to do weird sentences in english ahah, and i must be excessive in a way yeah, but i’m rarely upset, not for virtual stuff. i have no political or professional interest in being soft and academic. btw. anyway. i’m just saying nuking or removing abstain from the quorum count is no different. it absolutely doesn’t take account of the “discussion” and the points made along the road, in my opinion. blank votes should (maybe with a different rate ratio) count into the quorum. they are something. they tell something more than the non-vote. pulling them out of the quorum is like how work most of governance models we have irl. it’s negating the people who are interested, but want better, different, differently structured proposed paths. ok there is a validators behavior issue, the same way there is one with NWV use, sybils, whitelabels, bribes and last minute vote switches, etc. i don’t believe the current proposal, even politically updated, is a solution, for none of those issues. the same way we’re going towards (optionally and modulable) private governance, we should differentiate individuals and…

Excerpt (1193 of 2051 characters). Read the whole post on the forum ↗

I gotta say im fully puzzled to any argument that argues that removing the right to express oneself can be a good thing

I see what a proposal like this would be trying to accomplish, however there is an easy work around for validators. They can just use a weighted vote and do 50%no 50% yes.

And this work around is fine if the intent is to express neutrality and or indifference. A weighted 50/50 expresses this more accurately than voting abstain as currently implemented.

However, if the intent is to express abstention due to COI, voting abstain with current tally parameters is problematic.

Please see

Modify abstain so that it does not count toward quorum Hub Proposals

If the intent is to use Abstain to express neutrality or perhaps indifference, 50/50 weighted is the correct vote. In the event a validator wishes to express “conflict” by voting abstain, while it’s socially accepted as expressing “no vote,” voting abstain does not recuse a voting validator. Technically and quantitatively it has a calculable effect. In fact, voting abstain in non sequential elections where relative quotas are used as qualifying requirements – tallied abstentions in these cont…

Well that is a very different thing from abstain.

I’d like to argue that abstain does not even cover the legal arguments that some people here have been making because abstention does affect outcomes in our current system .

Validators who can’t vote just can’t vote, that’s really the end of it.