That’s a very good point and I’ll use that when writing the code, to make sure that inactive validators do not become governators.
1 Like
Well that’s because there will be an unlimited number of governators, not one
In what sense would this centralise authority differently than in our current scenario?
Because as I read it; we would still have a scenario of different governators with voting power (and hopefully a bit more decentralized at best 
), so bribing might still require more than one party to be swayed successfully.
I’m a bit behind on the whole discussion and haven’t had the time to fully dive deep into the pros and cons of it all.
However, my initial questions that I have would be:
-
Will the governator set be something similar to the validator set where there are a minimum amount of delegations / backers to be considered active and able to participate? I guess this makes sense like how in traditional elections, we vote on council members and only a select few are able to make decisions (if they garner enough backing)
-
If the above is not applicable, will there be countless amount of governators who have like 1 delegate behind them? This is easy to Sybil and create tons and tons of fake governators who can skew a vote if the system will be a 1 person = 1 vote situation.
-
Is this system a 1 person 1 vote type situation or is it also based on VP of the delegator? If not the latter, my previous point will be an issue for the system.
First, let’s get a few things straight about separation of powers in the US federal government.
The legislative branch makes laws. The executive branch administers/enforces those laws. The judiciary interprets the law.
The executive branch does NOT have the power to set the direction.
The closest powers that the executive has to set the direction comes from the ‘bully pulpit’, the requirement set by Congress under the Budget & Accounting Control
Act of 1921 to submit an annual budget, and the ability to make a limited number of appointments with ‘advice and consent’ from the Senate. The direction is very much still set by the legislative branch which holds the ‘power of the purse’.
Bicameralism is also a fundamental element of separation of powers. See John Adams ‘Thoughts on Government’.
(Fun Facts: Nebraska is the only unicameral state legislature in the US. All other states have a bicameral legislature.)
The Judicial branch does not enforce laws.
The US federal government is a very poor model to start with when thinking of how to structure decision making for an organization, especially for a decentralized organization formed for the purpose of operating something like a decentralized exchange. Government models such as school-based decision making championed by William Ouchi, professor of management and organizations at UCLA Anderson School of Management, and successfully practiced across multiple jurisdictions in Canada and the US, serve as much better models in my opinion. There are also nongovernmental models, like holocracy, championed by many today. See https://enliveningedge.org/
Separation of powers is essential to maintaining a functioning government. Doing so carefully and deliberately is paramount to the effectuality of the system as a whole. When forming separations, we become so enthralled with the concept that each division should be distinguishable and isolated that we relinquish care for the totality of the governance system. Specifically, the interdependence between each cameral. To digress momentarily and reform on the need for collective synergistic interdependent government, Polybius theorized that governments function cyclically, beginning at anarchy, then monarchy, tyranny, aristocracy, oligarchy, democracy, and back to mob rule. Monarchy, aristocracy, and oligarchy are the elemental levels of governance, and all have opposing degradations. Polybius defends the Roman governance system positing that, while complex and un-uniform, it unites all the great features of the elemental governments “so that none of the principles should grow unduly and be perverted into its allied evil” (p.g. 317 Book VI Polybius The Histories).
The Roman system of governance functions as a monarchy through the magistracy, an aristocracy with the senate, and a democracy as the Plebian Tribune. Gathering these thoughts back to the original topic, we discover to create multidisciplinary systems that stave off systematic decay, they require cameral interdependence. The current Governator proposal on the forum fundamentally lacks the antecedent to well-functioning cross-affiliation mechanisms called checks and balances.
While I have many ideas, the first I propose works within the current timocratic system:
The validator chamber check on general assembly and vice versa
-
Upon passing a bill through the general assembly or commons (what I’m naming the governator) the vote will enter a cloture period in the validator chamber.
-
The cloture period will last for a finite amount of time in which validator chamber members can submit yea, nae, or abstain ballots. In the event that the cloture period in the validator chamber concludes with a simple majority yea, the hall will move to a cloture motion.
-
The cloture motion initiates a joint session between the general assembly and the validator chamber. Any representative from the validator chamber can propose counter bills, and the general assembly, along with the validator chamber, will come to a decision on the legislation in a set period.
- The first counter-proposal is submitted a period of time after the cloture motion.
- In the case where a house falters in producing an opposing bill in the decided period, the counter-houses bill is accepted.
-
If both assemblies achieve simple majority yea votes, then the bill is formalized and enacted as law.
-
If the counter-assembly fails to achieve a simple majority, then members from that assembly will propose revised bills.
-
If the assembly that motioned for cloture or failed to agree on the counter bill proposes new legislation to the joint session that the proposing chamber itself fails to pass, the joint session is postponed for a period of time and then reconvened. After being reconvened, if the assembly that motioned for cloture or failed to agree on the counter bill proposes new legislation to the joint session that the proposing chamber itself fails to pass, the other house’s bill is passed.
-
If one party desires to cede from negotiations, a walk-out vote is proposed in their respective chamber where a 66% outsized majority is needed. In the circumstance where the walk-out vote is passed, the bill is dead on arrival, and the session halts, consider this a veto.
- Could also be done in a democracy (outside the timocratic system) to ensure equality and get rid of Pareto distribution issues with the veto. For now, I’ll keep the system simple and focus on what I think could get passed. Link to this idea
-
To cease the joint session without passing the bill, either party will submit a desist motion where both chambers require a ⅓ minority yea to forfeit the bill and halt negotiations.
Reasoning behind the bicameral check:
The present design of the check outlined above operates bidirectionally where legislation that is proposed in the general assembly can be under a cloture period, and motion affronted in the validator chamber, and opposingly the validator chamber can be under a cloture period, and motion affronted in the general assembly. Therefore, both houses have influence over the other.
- Why the Cloture period is structured its way:
- Either the period never ends or has a finite time. In the former, the validator chamber could filibuster the general assembly for all time and visa versa. The latter solves the anti-democratic problem of a filibuster and puts a sunset on the ability to act.
- Voting quorums:
- Why do both bodies of Congress need to vote in a cloture motion?: Succeeding the cloture period, any member of the party that has motioned to cloture can propose legislation. The primary reason for many counter bills comes from the attack vector of its opposite. If we were to allow one bill in the joint session, it is possible and even likely that a member from the other committee torpedoes the other by submitting a false flag proposal. The effect of the attack would be multiple failed votes and the bill passing in the attacker’s committee.
- Explaining the chamber that motioned to cloture fails to pass its own bill?: If there is no cap on the number of joint sessions a chamber can call, then it can infinitely filibuster.
- Explaining passing legislation in cloture motions: The concept behind a simple majority across both bodies of Congress simply is the trust assumption that the majority of individuals have the best interest of the chain in mind.
- Explaining Walk-out motions: Ceding is a form of veto and should be subjected to the tyranny of the minority. By making the vote more than a simple majority, the minority takes more precedence in decision-making. 51% of the power in the Validator chamber voted for the motion, and in turn, to cancel the motion and revoke the bill from deliberation, a greater amount of power is required. The foundation behind the supermajority vote for ceding is again the opposite; if the vote to leave negotiations were too low, then the validator chamber would hold too much sway over the other and vice versa.
- Explaining jointly agreeing to walk away: The tyranny of the majority should take precedence in the joint decision-making to forgo the bill. In this circumstance the top validators in the chamber and representatives in the general assembly come to an agreement to forfeit the bill, then the bill is likely not in the best interest of the chain. Noting that each congress member across either chamber represents and holds different values, to agree would be rare.
Other ideas:
- Instead of allowing multiple counter proposals, we could create a drafting period where multiple drafts are voted on, and the proposal with the greatest votes will be proposed as the counter-bill in the joint session.
Misc:
- There are at most 3 joint sessions meaning 1 per chamber after the initial motion.
- There are infinite rounds in each session
- The separations of power between mechanical consensus and social consensus responsibilities should be continued
- The current system of governance in cosmos is timocratic and is closer to aristocracy we need to incorporate democratic elements in the validator chamber.
1 Like
Either just so you know I was not really trying to mirror existing governments. I was trying to build something that’s appropriate for this context.
1 Like
I as a normal delegator vote because I feel it’s my duty and I hope the chain to succeed. I never believe I get any direct monetary incentive for voting. So, why should governators get their piece of the pie for doing basically the same thing? Where there is money to be made, it attracts those who are doing it only to get paid.
Separation of Powers
If we’re splitting powers we should be going for a 3 house system as least so that each can check the others.
-
Validators/Technical (Executive) - has veto power over Legislative by refusing to continue validation.
-
Governators (Legislative) - has veto power over technical by slashing mechanics.
-
? (Judicial), first thoughts are either code reviewers, which perhaps leads to more centralisation if this body is given actual powers rather than guidance. But maybe this is where the “NATO” system comes in. Each chain being able to sanction through blocking IBC channels, imposing taxes on traffic etc.
Impact on current system
-
Validators should not be impacted by this at all in terms of income. That way, there is no financial discouragement for them to vote no on this.
-
Validators typically get Staking Rewards * Commission or about 2.5% of inflation, so the cost to do this would be pretty small. Maybe start with 1% of inflation going to governors, with a fixed 50% commission, so that there is an incentive to choose one, and governors cannot compete on this. This could come from the community pool fraction of inflation to start with since it is relatively small, later eating into other areas.
-
There shouldn’t be compulsory selection; you should be able to vote on proposals yourself still; however, it should be that choosing a good governor is easier. Whether everyone voting should be eligible for the 1% or just as a governor program incentivization mechanism though…
Who can be governors?
- As Jacob said, it should be a case of signing a transaction to say that you understand the role, but beyond that (which should be on an interface), there would be no technical requirement.
- There should be no elections as such. This would be a fluid system where poor voting causes the loss of delegates and is geared towards active participation by all.
- There would be an issue where delegators set and forget rather than monitor their delegate’s votes. The active participants are likely to be attempting to be governors themselves, the inactive participants will likely not care as much as long as the rewards are coming in.
Suggested System
- Your governor inherits 100% of your voting power if you do not vote as Controlled Voting Power.
- Governors get a percentage of inflation based on their Effective Voting Power as a share of the total Effective Voting Power owned by all Governors.
- You can vote at any time and overwrite your governor’s vote on a specific proposal, but it would not impact the inflationary rewards.
- Effective Voting Power is not the same as Controlled Voting Power but can have modifiers.
- Modifier 1 is the inactivity penalty for delegates. For every 1 month that a delegating wallet does not perform a transaction, the Effective Voting Power drops by ActivityWeighting, initially 10%, of the Controlled Voting Power. This ensures that Governors represent the active community and do not end up incumbent due to large inactive delegations. This resets to 100% as soon as activity is detected in a wallet.
- Modifier 2 is an inactivity period for governors. This cannot be based on missing X proposals as these are not regularly timed. Instead, I suggested that this is a background statistic that updates after every proposal completes based on Quorums. If a proposal fails to reach Quorum then it is considered to be poorly promoted or timed and so there are no penalties. If a proposal passes quorum then any governor that did not participate would have their personal modifier decrease by a ParticipationWeighting * % Active EffectiveVotingPower. i.e. the more voting power was active on a vote, the higher the penalty would be. This encourages voting by governors in a kind of prisoners dilemma. The optimal outcome is for everyone to vote on all proposals but an alternative would be for no proposals to reach quorum. However since individual delegators can still vote this is unlikely to happen.
- Modifier 3 would prevent governors from gaming the system by blindly voting on every proposal, and is where I come unstuck. What is a good result? We often talk about the “Stripper Dollars” proposal being heavily downvoted at the time, but in hindsight may have been a good idea to pass.
Finally, should there be a cap in place? A cap would likely promote sybils, which would be far easier to do with the lack of technical requirement. But also there should be a more personal relationship with governors that would be picked up on more easily.
Regarding the impact on the current system; if we would start with a negligible reward because we want to “spare” the current validators we will not see many governators who will forfeit their validating part. And then we still don’t have the separation of powers, where the validating set creates room for more technical people. And note that validators can still be governators, but they don’t need to.
So if the concept would be strong enough that being a governators but not doing so might also hurt their validating business, you would force parties to choose a path instead of being rewarded for everything, while doing only a part.
Regarding ignoring penalties when a proposal does not meet quorum; that would effectively mean that in essence it is most beneficial for governators to have no proposal meet quorum ever. Since then there are no penalties involved. You should go for the simple solution that there is no incentive to not vote, otherwise governators can game the system. Simply make every vote count.
I like the idea of modifiers. One thing to note is that if we want to go this route, it will need some tweaks. Can we punish governators for inactive communities? Because people are not very active in governance anyway (just look at the current system).
Would it be smart to first wait and see how things will progress with the Stride implementation of ICS? Because that will effectively reach the scenario of governators and validators. So there will also be a ton of valuable lessons to be learned for our own scenario.
1 Like
On a more general level: I think it is paramount when forming new systems to look at those that have failed In the past and why.
Tricky brought up to me that we need to somehow separate some of the power away from the validators. I’ll shortly paraphrase:
- he says the problem with forming two bodies is that the validators are the default executor of the chain.
- He claims that the power of Governators would “pale” in comparison to the validator.
- Furthermore, “if there is overlap then the actual diversity of the representation will be an issue.”
I think these are valuable rejections to my former idea, yet the concept of checks and balances still stands.
I have an idea, to solve his proposed objections:
- The concern centered around overlap is not readily tackled. In theory, emplacing a system of blocklisting would stop cross-partisan participation, although a member from either cameral can create a separate address.
- The solution we incentivize the validator chamber and general assembly members alike to cross-identify themselves.
- After doing so, we raise the proletariat general assembly members that are not cross-identified to an external smaller council of governators. In the Council of Governators we require a minimum self-bond that can be burned by the constituent general assembly and validator chamber if the council is malfeasant.
- This Council will function democratically and inherit the sole responsibility of halting the chain along side validators. Now the threat between the aristocratic validator chamber, timocratic proletariat general assembly, and democratic meritocracy is equal.
" Validators/Technical (Executive) - has veto power over Legislative by refusing to continue validation.
Governators (Legislative) - has veto power over technical by slashing mechanics."
I really like this idea, a question for you sir: how do you plan on halting validator participation in the Legislative?
As far as the Suggested system section you have some good contributions here!
Modifier 1: I would ask do we want to punish individuals for not participating in government? If we do it might enforce the system to demagoguery.
Modifier 2: I believe this concept to be more in line with an effectual and well-functioning system because governors should be active members of the administration.
Modifier 3: I think we stay away from measuring good votes or bad votes due to the subjective nature of politics.
I’m passionately against a cap in the governator or general assembly. The assembly should represent the polity, that is to say, the middle-class proletariat, and in doing so needs to be open to all.
I am not sure that we need to prevent this. Currently, validators perform both roles, and this would be a division of specialty. If a validator wants to perform both roles, they can, but would either be a larger validator company or do both below the standard of a specialized validator, leading to those that specialize dominating.
Your self-identifying idea above is interesting, but being a member of the smaller council would itself be an incentive, so I think would be very hard to balance. Assuming that the self-bond would be substantial and burned if it was detected that the council member was raised under false pretenses. Who burns here, the peers? Giving that ability seems ripe for abuse if it is a simpler process than the social slashing of a validator. I think this should be the 3rd global house, not just the internal chain membership.
Modifier 1 wouldn’t be non-participation in governance but any transactions. I originally put governance transactions before posting but had similar concerns.
I do think that some metric is needed. Modifier 3 should, in theory, be driven by active user participation, but as we see at the moment, users are driven by convenience (prominently displayed validators get delegations) and slightly by finance (lower Commission/airdrop/slash protection validators get some increases). The problem is that the optimal governor return would be obtained from blindly voting yes on every proposal - which is a similar issue to what we may have now.
I didn’t mean a cap in numbers here, I meant a cap in voting power that a single governor can bring to bear. Otherwise, you can have a dictatorship by consent, which becomes hard to remove without input from the other house. May as well start off by limiting the power a single governor can hold.
2 Likes