Stake size is not a guarantee of quality, let alone an anti sybil system. It’s been proven on Polkadot with many harmful and even malicious referenda that were proposed and approved by “large holders” regardless of the quality.
This design only accentuates this issue.
1 Like
Hi Saxemberg,
Thank you for the reminder about the challenges faced by validators when they were required to vote on every proposal. Auto-voting, especially systematic ‘Nay’ voting, is indeed a concern if validators feel compelled to vote without genuine engagement due to the sheer volume of decisions required.
The proposal aims to mitigate such issues not by simply increasing the quantity of votes required but by creating a more thoughtful approach where the quality and impact of participation are prioritized. With tiered voting and meaningful rewards attached to significant referendums, the intention is to encourage informed voting on issues that truly matter, rather than forcing validators to vote on every minor proposal.
By reducing the frequency of votes needing active participation and increasing the stakes for more critical decisions, we hope to foster a system where validators feel their time and input are being used effectively, rather than feeling overwhelmed by the quantity of votes.
This is certainly an area we need to get right, and feedback like yours is invaluable in ensuring we develop mechanisms that don’t just shift the burden but genuinely enhance the governance process. If you have specific suggestions on how we can better structure this, I’d be eager to hear them.
You’re right, stake size alone doesn’t ensure quality governance. To prevent large holders from dominating decisions, we’re considering additional checks like more stringent proposal vetting and possibly using mechanisms such as conviction voting to balance influence. These steps aim to ensure decisions reflect broader community consensus, not just the will of a few.
Appreciate your input on refining these systems!
Right. And as an aside - if good improvements get made I’d support adopting them in my community (on another Substrate chain)- because there are similar problems elsewhere. With these changes enabled, I would argue for community (not on-chain) rules that ask pools to stay out of voting or face consequences. I know this may not be possible in other chains, but I plan to advocate this in my community. (One can’t know which node is or isn’t a pool node, but it’s not very hard to find due to the necessity sales/marketing outreach and such).
Secondly, Liberland should be among the (presumably?) heavy and “mission-critical” users of governance/voting and although they’ve started recently I think they must have some interesting feedback or problems they’d like to share. Maybe you could reach out to them to check if you’re interested in their experience.
Centralized pools are the worst. There’s not much to say about large holders - one can be happy or unhappy how some of them vote, but there isn’t much that can be done unless the rules change and if they lose voting rights commensurate to their share they may leave or divest. Maybe that’d be a good thing, maybe it wouldn’t. I wouldn’t call this “gaming” when the rules are 1 coin = 1 vote. That’s why democracy works (or doesn’t, depending on your viewpoint).
But pools recreate the Wall Street ETF problem (investment companies behind ETF participate as shareholders in governance without any mandate from retail or even institutional investors). Even if general large holders aren’t touched, centralized pools ought to be unable to vote even if there are just half-measures. It’s very bad for NPOS systems to have pools that can vote.
“Show me the incentive and I’ll show you the outcome” - Charlie Munger’s wisdom is particularly relevant here.
While incentivizing governance participation has merit, and as mentioned by many in the comments, we must carefully design a system that encourages informed rather than automated voting.
Proposed Solution: Knowledge-Verified Voting (KVV) 
Core Mechanism:
- Pre-vote Knowledge Test
- Voters must demonstrate understanding of the proposal before voting
- AI-generated questions covering key proposal aspects
- Randomized question set to prevent memorization/sharing
Requirements:
- Minimum 80% correct answers on a 10-question test
- Questions focus on proposal specifics, implications, and trade-offs
- Successfully passing unlocks both voting rights and rewards
This approach:
✓ Ensures voters understand what they’re voting on
✓ Reduces blind/automated voting
✓ Aligns incentives with informed participation
So I was thinking about this and using ChatGPT o1-preview to rough out a pseudo-model. Granted, I’ve only given this output a cursory overview and have not verified or tested anything. I’m posting this here for anyone to see a more structured overview and add some potential talking points about the feasibility of a ‘fair’ mandatory voting system for staking rewards.
I also believe staking rewards should not be restricted to validators who provide a critical service to the network. Nomination rewards on the other hand would be subject to governance participation within some time period. I am unsure how this might affect nomination pools, self-funded validators, centralized exchanges, and larger ETF holdings. Someone else may certainly have more experience and better ideas on this.
Apologies for the horrendous formatting of the math outputs. I’m doing a dozen other things at the moment and choose not fiddle with formatting
The remainder of this post is AI-generated, so please read cautiously.
Mathematical Model for Tiered Reward Distribution in a Nominated Proof-of-Stake System
Introduction
This model describes a reward distribution system in a nominated proof-of-stake blockchain where token holders earn rewards by staking tokens with validators and participating in governance voting on referenda. The system incentivizes participation in critical governance decisions by adjusting rewards based on voting participation in various tiers of referenda.
Key Components
-
Total Staked Tokens (T):
- The sum of all tokens staked by all users in the system.
-
Total Rewards (R):
- A fixed percentage (10%) of the total staked tokens:
[ R = 0.10 \times T ]
- A fixed percentage (10%) of the total staked tokens:
-
User’s Staked Tokens (Sᵤ):
- The number of tokens staked by user u.
-
Referenda Types and Tiers:
- Critical System Upgrades (CSU)
- Highest priority.
- Mandatory participation to receive any rewards when available.
- Large Spend Referenda (LSR)
- Second priority.
- Mandatory if no CSU is available.
- Medium Spend Referenda (MSR)
- Small Spend Referenda (SSR)
- Miscellaneous Tips (MT)
- Critical System Upgrades (CSU)
-
Weights for Reward Distribution (Wₜ):
- Percentage of the total reward allocated to each referenda type.
- Adjusted based on the availability of referenda types.
Model Construction
Step 1: Determine Available Referenda Types
- Let A be the set of available referenda types in the current distribution period.
- Referenda types: {CSU, LSR, MSR, SSR, MT}
- Identify which referenda types are available.
Step 2: Assign Base Weights
-
Base weights when all referenda types are available:
Referenda Type Base Weight (Wₜ) CSU 60% LSR 24% MSR 10% SSR 4% MT 2% Total 100%
Step 3: Adjust Weights Based on Availability
-
If a higher-tier referenda type is unavailable:
- Its weight is redistributed proportionally among the remaining referenda types, prioritizing the next critical tier.
- The most critical available referenda type (rtₘ) becomes mandatory for any rewards.
-
Weight Adjustment Algorithm:
-
Identify Missing Tiers:
- Let M be the set of missing referenda types.
-
Redistribute Weights:
- Sum the weights of missing tiers:
[ W_M = \sum_{t \in M} Wₜ ] - Distribute W_M proportionally to the available tiers based on their base weights.
- Sum the weights of missing tiers:
-
Calculate Adjusted Weights:
- For each available referenda type t:
[ W’ₜ = Wₜ + \left( \frac{Wₜ}{\sum_{k \in A} W_k} \times W_M \right) ]
- For each available referenda type t:
-
Step 4: Calculate User Participation Ratios
- For each available referenda type t:
- Nₜ: Total number of referenda of type t.
- nᵤₜ: Number of referenda of type t that user u voted on.
- Participation Ratio (pᵤₜ):
[ pᵤₜ = \frac{nᵤₜ}{Nₜ} ]
Step 5: Calculate Reward Portions
-
Total Reward for User (Rᵤ):
- If user u did not vote in the most critical available referenda type (rtₘ):
[ Rᵤ = 0 ] - If user u voted in rtₘ:
[ Rᵤ = \left( \frac{Sᵤ}{T} \right) \times R \times \left( \sum_{t \in A} W’ₜ \times pᵤₜ \right) ]
- If user u did not vote in the most critical available referenda type (rtₘ):
Detailed Explanation
Mandatory Participation
- Most Critical Available Referenda Type (rtₘ):
- Participation in rtₘ is mandatory to receive any rewards.
- rtₘ is determined by the highest priority referenda type available in A.
Weights Redistribution Example
-
Scenario 1: All referenda types are available.
- Weights remain as base weights.
-
Scenario 2: CSU is unavailable.
-
M = {CSU}
-
W_M = 60%
-
Redistribute 60% among LSR, MSR, SSR, and MT based on their base weights.
-
Calculations:
- Sum of base weights for available types:
[ \sum_{k \in A} W_k = 40% ] - Proportional increase for each available type t:
[ W’ₜ = Wₜ + \left( \frac{Wₜ}{40%} \times 60% \right) ] - Adjusted Weights:
- LSR:
[ W’_{\text{LSR}} = 24% + \left( \frac{24%}{40%} \times 60% \right) = 24% + 36% = 60% ] - MSR:
[ W’_{\text{MSR}} = 10% + \left( \frac{10%}{40%} \times 60% \right) = 10% + 15% = 25% ] - SSR:
[ W’_{\text{SSR}} = 4% + \left( \frac{4%}{40%} \times 60% \right) = 4% + 6% = 10% ] - MT:
[ W’_{\text{MT}} = 2% + \left( \frac{2%}{40%} \times 60% \right) = 2% + 3% = 5% ] - Total Adjusted Weights: 60% + 25% + 10% + 5% = 100%
- LSR:
- Sum of base weights for available types:
-
-
Mandatory Referenda Type: LSR becomes the mandatory referenda type for rewards.
Applying the Model to the User’s Example
Given:
- Total Rewards (R): 100 tokens
- User’s Staked Tokens (Sᵤ): Proportional, so we’ll focus on percentages.
- Available Referenda Types:
- CSU (1 referendum)
- LSR (2 referenda)
- MSR (8 referenda)
- User’s Participation:
- CSU: Voted (nᵤ₍CSU₎ = 1)
- LSR: Voted in 1 out of 2 (nᵤ₍LSR₎ = 1)
- MSR: Voted in 4 out of 8 (nᵤ₍MSR₎ = 4)
Step 1: Available Referenda Types
- A = {CSU, LSR, MSR}
Step 2: Base Weights (All available)
- W₍CSU₎ = 60%
- W₍LSR₎ = 24%
- W₍MSR₎ = 10%
- Total = 94%; Remaining 6% allocated to SSR and MT, which are unavailable.
Step 3: Adjusted Weights
-
Since SSR and MT are unavailable, their weights (6%) are redistributed proportionally.
-
Sum of base weights for available types:
[ \sum_{k \in A} W_k = 94% ] -
Proportional increase for each available type t:
[ W’ₜ = Wₜ + \left( \frac{Wₜ}{94%} \times 6% \right) ] -
Adjusted Weights:
- CSU:
[ W’_{\text{CSU}} = 60% + \left( \frac{60%}{94%} \times 6% \right) \approx 60% + 3.83% \approx 63.83% ] - LSR:
[ W’_{\text{LSR}} = 24% + \left( \frac{24%}{94%} \times 6% \right) \approx 24% + 1.53% \approx 25.53% ] - MSR:
[ W’_{\text{MSR}} = 10% + \left( \frac{10%}{94%} \times 6% \right) \approx 10% + 0.64% \approx 10.64% ] - Total Adjusted Weights: Approximately 100%
- CSU:
For simplicity, we’ll use the original weights as they closely approximate the adjusted ones.
Step 4: Calculate Participation Ratios
-
CSU:
- N₍CSU₎ = 1, nᵤ₍CSU₎ = 1
- pᵤ₍CSU₎ = 1 / 1 = 1
-
LSR:
- N₍LSR₎ = 2, nᵤ₍LSR₎ = 1
- pᵤ₍LSR₎ = 1 / 2 = 0.5
-
MSR:
- N₍MSR₎ = 8, nᵤ₍MSR₎ = 4
- pᵤ₍MSR₎ = 4 / 8 = 0.5
Step 5: Calculate Reward Portions
-
Assuming Sᵤ / T = 1 for simplicity (user’s proportional stake).
-
R₍CSU₎ = R × W₍CSU₎ × pᵤ₍CSU₎
- [ R₍CSU₎ = 100 \times 60% \times 1 = 60 \text{ tokens} ]
-
R₍LSR₎ = R × W₍LSR₎ × pᵤ₍LSR₎
- [ R₍LSR₎ = 100 \times 24% \times 0.5 = 12 \text{ tokens} ]
-
R₍MSR₎ = R × W₍MSR₎ × pᵤ₍MSR₎
- [ R₍MSR₎ = 100 \times 10% \times 0.5 = 5 \text{ tokens} ]
-
Total Reward (Rᵤ):
- [ Rᵤ = R₍CSU₎ + R₍LSR₎ + R₍MSR₎ = 60 + 12 + 5 = 77 \text{ tokens} ]
Result: The user receives 77 tokens, matching the example provided.
Additional Scenarios
Scenario: User Does Not Vote on Mandatory Referenda
- Case: User does not vote on CSU when it’s available.
- Outcome:
- Rᵤ = 0 tokens
- The user receives no rewards for the period.
Scenario: No CSU Available
-
Available Referenda Types:
- LSR, MSR, SSR, MT
-
Adjusted Weights:
- LSR becomes mandatory, and its weight increases.
- Weights are adjusted using the redistribution algorithm.
-
Example Adjusted Weights:
- W₍LSR₎ = 80% (as per system rules)
- Remaining 20% distributed among MSR, SSR, MT proportionally.
-
User’s Participation:
- Must vote on at least one LSR to receive any rewards.
- Rewards from other referenda types depend on participation ratios.
Summary of the Model
- Mandatory Participation: Voting in the most critical available referenda type is required for any rewards.
- Reward Calculation: Based on stake proportion and participation in available referenda.
- Weights Adjustment: Ensures total rewards are fully distributed, emphasizing participation in higher-priority referenda.
- Encouraging Participation: Higher rewards are allocated to critical decisions to incentivize voter engagement.
Note: This model provides a structured approach to reward distribution, ensuring fairness and encouraging active participation in governance, particularly for critical system decisions. Adjustments based on referenda availability maintain the incentive structure’s integrity.
2 Likes
No, financial rewards should only come from staking and validating. That most people not care to vote, including me, is completely normal and natural. If you like to vote, then be happy and vote. But do not start to financially punish those who are just happy to hold and stake, but don’t want, or don’t have time to follow all the voting and politics.
As I said many times before, if holding DOT becomes to much work, I will just sell it into some coin that doesn’t require as much maintainance time & efforts. Preferably even a good fixed supply token. Nothing to vote, nothing to stake, and no valuedilution through inflation.
By forcing people to vote you just make holding DOT an even more unpleasant job, which will simply result in more people selling their DOT for tokens that require less maintanance, and possibly not require any staking at all.
1 Like
I recognize the fact that not everyone is interested in the politics of the on chain governance. I have read numerous responses such as yours across X and Reddit. Surely, it is unrealistic to expect every DOT holder to read every proposal in OpenGov.
However, perhaps it’s more realistic to consider that every DOT holder might select a delegate on a 6-month or annual schedule. You do not necessarily need to be fully versed in the activities on the chain, but you could find a like-minded person who is active in governance to delegate your voting power to. Perhaps there are several people who are small fish with the views and moral compass of a small-scale investor who could have sufficient backing to oppose activities of malicious whales. At the very least, this opposition should improve the quality of excessive Treasury spends by engaging a larger number of total token holders in the voting process.
I support the ideas which push the DOT token further from an investment vehicle or pure cryptocurrency towards a piece of facilitator software for a Web3 future. Incentivized OpenGov voting becomes a social responsibility for the Polkadot DAO.
You likely review your nominations on a regular basis. Reviewing your delegations can be done in much the same way. It does not need to be as difficult as you may be thinking.
1 Like
Yes being able to delegate my voting to someone I trust I would really like and would be happy to do. I guess this also might also come in line with the DOT-Ambasadors programm, or at least could have some common grounds. I would be happy to delegate my votingrights to some of the active communitymembers who I have been following for a long time now.
1 Like
This is a great start for conceptualizing a model to improve the fairness of our staking system. Enhancing the structure to support mandatory participation in significant votes while considering the redistribution of rewards based on voter engagement and referenda importance is promising. Addressing these finer details will be crucial to ensuring that any changes do not negatively impact different participants like nomination pools or self-funded validators. Your input helps shape a more robust and equitable framework for everyone involved.
Thanks for kicking off this detailed discussion. It’s crucial for fine-tuning our approach.
2 Likes
I get where you’re coming from, and your point about not overburdening DOT holders is valid. The idea isn’t to make participation a chore but to encourage a more engaged and informed community. Remember, you can always delegate your voting to trusted parties if you prefer not to dive into the details yourself. This way, you’re not penalized for inactivity, and those who take the time to vote can be rewarded for adding value to the governance process. It’s all about finding a balance that respects both active and passive members of the community.
1 Like
That’s a sensible suggestion. Delegating voting power could indeed bridge the gap between active governance and those who prefer a more hands-off approach, without sacrificing their influence. It could help balance the ecosystem by representing a broader base of DOT holders, enhancing decision-making quality by diversifying the perspectives considered. Regular reviews of delegations ensure alignment with one’s interests, making participation both feasible and meaningful for everyone.
2 Likes
That’s great to hear! Delegating to someone you trust and who is actively involved aligns perfectly with fostering a responsible and engaged community. It’s an excellent way to ensure your views are represented without needing to dive deep into every detail yourself. Aligning this with programs like DOT-Ambassadors could indeed strengthen the network’s governance by leveraging the expertise of dedicated community members.
You are in luck: Delegation Dashboard
1 Like
An additional benefit to polkadot delegation is the fact that the delegations can be set to different people for different referenda tracks. Thereby being more accurate in representing the community of DOT holders.
I have not checked myself, but does the current delegation system allow for fractional delegation? Such that a person could select multiple delegates to vote on the same track with different DOT amounts? This could be like a form of easing or smoothing of a person’s opinions across multiple delegates for a more honest and accurate engagement.