Chapter 5: The Respect Game

Draft v0.1 - 2026-05-25 - awaiting Zaal review

The Respect Game is the weekly ceremony where five-person breakout groups reach consensus on contribution rankings. No votes are cast. No tallies. Just honest peer judgment, encoded on-chain.

I. The Weekly Ritual: End to End

The Respect Game is a 90-minute meeting held every Monday at 6pm EST. The structure is consistent, repeatable, and designed to minimize gaming while maximizing honest deliberation.

Pre-Conditions: Introduction and Registration

Before a member can participate in their first Respect Game:

1. They must post in the #introductions channel (one-time, identifies them to the community).
2. They must run the /register command in Discord, connecting their Optimism wallet address to the bot’s registry.

After these steps, they can join any Respect Game session.

These preconditions serve two purposes: identity certainty (we know who you are in the community) and wallet linkage (we can mint Respect directly to your on-chain address).

Phase 1: Gathering (10 minutes)

Members join the Fractal Waiting Room Discord voice channel. The session begins when 4+ unplayed members are present. (An “unplayed” member has not participated in the past 7 days.)

The official weekly session runs Monday 6pm EST. Ad-hoc sessions can launch anytime 4+ members are available.

Phase 2: Randomization (5 minutes)

A facilitator (typically Zaal or civilmonkey.eth) runs the /randomize command. The bot:

1. Fetches all members currently in the Fractal Waiting Room.
2. Splits them into groups (maximum 6 per group, minimum 2).
3. Auto-moves members into individual Discord voice channels (Fractal Group 1, Fractal Group 2, etc.).
4. Posts a confirmation message with group assignments.

Randomization is cryptographically seeded. This prevents pre-planned collusion. A player cannot predict which group they will join; they cannot pre-arrange to rank each other highly.

This is a critical sybil defense. Randomization forces a fake account to win consensus with different real humans each week. The cost grows exponentially with each new group assignment.

Phase 3: Presentations (25 minutes, 4 minutes per speaker)

In each group’s voice channel, the facilitator runs /timer. The bot:

1. Displays a “Meet Your Group” message listing the speaker queue.
2. Starts a 4-minute countdown for each speaker.
3. Provides interactive controls: Done, Skip, Come Back Later, +1 Min, Raise Hand.
4. Plays a brief audio signal when the timer resets between speakers.

Each member describes their contributions over the past week. Examples: “I shipped X feature, mentored Y person, wrote Z research doc.” The 4-minute limit enforces concision. Members cannot ramble into undeserved visibility.

Audio is streamed via Stream.io (default) or 100ms (transcription-enabled). All speech is preserved for async participants.

Phase 4: Sequential Elimination Voting (25 minutes, Levels 6 to 1)

The facilitator runs /zaofractal [fractal_number] [group_number] to start the voting flow.

The voting proceeds from Level 6 (lowest rank) to Level 1 (highest rank). This reverse order (worst-to-best) prevents anchoring bias - early votes do not unduly influence later ones.

Round Structure (6-person group example):

Level 6 round (finding the 6th-place contributor):

1. Bot posts 6 colored voting buttons (one per candidate).
2. Bot joins the voice channel and plays an ascending-pitch signal.
3. Each member clicks the button for the person they rank lowest.
4. Votes are public - the bot announces each vote aloud: “[Member] votes for [Candidate].”
5. Members can change their vote at any time during the round. A button click updates the vote, not adds to it.
6. Winner detection: Simple majority wins. In a 6-person group, 4 votes are needed (ceil(6/2) = 4). The person with 4+ votes is locked in as rank 6.
7. The rank-6 winner is removed from all future rounds.

Level 5 round (finding the 5th-place contributor):

1. Same process, but now there are only 5 candidates (rank 6 is removed).
2. 3 votes needed for majority (ceil(5/2) = 3).
3. Rank 5 winner is removed.

This continues through Levels 5, 4, 3, 2, and finally 1. The last remaining candidate (no vote needed) automatically gets rank 1.

Why sequential elimination?

Sequential elimination operationalizes the 2/3 consensus requirement from the Fractally protocol. Over 6 sequential rounds with public, changeable votes, a strong supermajority consensus emerges. Members are expected to adjust votes when the group consensus becomes clear. A person cannot win top rank if the group consistently disagrees.

Why public voting?

Public voting prevents vote trading and makes collusion visible. If a member suspects their group is colluding (always voting for the same person regardless of contribution), they can voice objection during the meeting or escalate post-session. Transparency enforces honesty.

Phase 5: On-Chain Submission (20 minutes)

Once all 6 levels are complete, the facilitator opens a pre-filled URL:

https://zao.frapps.xyz/submitBreakout?groupnumber=N
  &vote1=WALLET_OF_RANK_6
  &vote2=WALLET_OF_RANK_5
  &vote3=WALLET_OF_RANK_4
  &vote4=WALLET_OF_RANK_3
  &vote5=WALLET_OF_RANK_2
  &vote6=WALLET_OF_RANK_1

The wallets come from the bot’s /register registry. This link opens a Vite SPA at zao.frapps.xyz. The SPA builds an Optimism transaction to the OREC contract (OREC0xcB05…e532optimism).

The transaction calls:

await orclient.proposeBreakoutResult({
  meetingNum: 100,
  groupNum: 3,
  rankings: [addr_rank6, addr_rank5, addr_rank4, addr_rank3, addr_rank2, addr_rank1]
  // Respect auto-calculated via ZAO's 2x curve: [10, 16, 26, 42, 68, 110]
})

The OREC contract creates a new governance proposal. The proposer’s wallet auto-votes YES with their OG Respect weight (vote weight is frozen at proposal creation time, preventing double-voting).

Respect is not minted yet. The proposal is recorded on-chain but awaits voting and veto windows.

II. The OREC Governance Cycle: Voting and Veto

After submission, the proposal enters a two-phase governance cycle:

Voting Period (48 hours typical)

• Any member with OG Respect can vote YES or NO.
• Vote weight = OG Respect balance at proposal creation block (immutable historical snapshot).
• Cost: approximately $0.02-0.05 per vote on Optimism (cheap, non-prohibitive).
• Process: On-chain transactions via Etherscan or a governance interface.

Veto Period (48 hours typical)

• Voting period has closed; no new YES votes accepted.
• ONLY NO votes are accepted (challenge window).
• Purpose: Allows the community to mobilize opposition if off-chain consensus-building failed.
• Prevents last-minute attacks: an attacker cannot wait until the final block of voting to dump a massive NO vote.

Execution

After both windows elapse, anyone calls execute(propId) on OREC. The contract checks:

(block.timestamp >= proposal.startBlock + votingPeriod + vetoPeriod)
AND
(proposal.yesWeight >= (totalRespect * minThreshold))
AND
(proposal.yesWeight > proposal.noWeight * 2)

If all conditions are met:

• OREC calls the ZOR Respect contract’s mintRespect() function.
• Respect tokens are minted to each ranked member’s wallet.
• Each award is a unique ERC-1155 token with metadata (week number, group number, rank).

If any condition fails (insufficient YES, active veto, or time not elapsed), the proposal does not execute. The group must submit again or dissolve.

The 2/3 Rule: Mathematical Formulation

A proposal passes if:

yesWeight > 2 * noWeight  AND  yesWeight >= minThreshold

Consequences:

• 1/3 of active Respect holders can block any proposal (veto power).
• 2/3 can guarantee passage (supermajority).
• 50-50 splits always fail (YES is not > 2x NO).
• Abstentions are invisible (zero voting power).

This 2/3 rule creates a consent-based system, not a majority rule system. A slim majority cannot impose outcomes on the minority. Real disagreement is respected.

III. The Fibonacci Curve: Mathematical Details

ZAO’s 2x Fibonacci curve (see Chapter 4 for justification) distributes Respect per 6-person group:

Each rank earns approximately 60% more than the rank below (phi = 1.618). This ratio absorbs human judgment error while remaining acceptable under Ultimatum Game fairness norms (see Chapter 4, Section V).

Respect accumulation over time:

A member ranking 1st every week with 2% decay reaches:

R(52 weeks, 1x per week) ≈ 5,720 Respect (with decay factored in)

The same member reaches Elder tier (2000+ Respect) in approximately 50 weeks. Tier thresholds in ZAO are:

IV. Game Theory: Why Honest Ranking Is the Equilibrium

The Respect Game is designed so that honest ranking is the Nash equilibrium. A rational player should rank others truthfully, not collude.

The Payoff Structure

Payoff for honest ranking:

• You rank people by actual contribution.
• Others rank you honestly based on your actual contribution.
• Over time, your Respect balance reflects your real impact.
• You retain voting power proportional to your contribution.

Payoff for collusion (e.g., always ranking your friends 1st regardless of contribution):

• Your friend ranks you 1st (you gain extra Respect).
• But: the group notices the pattern. Consensus fails (not 2/3 agreement on colluders).
• OREC proposal fails due to lack of majority.
• You earn zero Respect that week (removal cost).
• Next week, you are in a different random group (re-randomization prevents sequel collusion).
• Your reputation for dishonesty spreads in a small community (social cost).
• You accumulate lower Respect over time (strategic cost).

The removal cost is decisive. If your group fails to reach consensus because collusion breaks the 2/3 rule, the entire group earns zero Respect. You cannot collude with one person to gain; you must align the entire group. And the group is randomized each week, so alignment is costly.

Ultimatum Game Framing

The Fibonacci curve is designed with game theory in mind. A rank-6 member (10 Respect) receives 9.3% of the group’s total Respect (10/272). A rank-1 member receives 40.4% (110/272). The split is 40/60, which exceeds the Ultimatum Game fairness threshold (30% minimum for acceptance).

In behavioral economics, people accept splits that feel “fair enough” even if unequal. The Fibonacci ratio hits this sweet spot: it is unequal (respects talent), but acceptable (respects equality). The group reaches consensus because the mechanism is perceived as fair.

Sybil Attack Analysis

Attack scenario: An attacker creates 10 fake accounts and tries to systematically rank themselves highly.

Defense layer 1: Randomization. The attacker cannot control group membership. A fake account must join a different random group each week.

Defense layer 2: Peer evaluation. In each new group, the fake account must reach 2/3 consensus with real humans. Real humans will not consistently rank a fake account high if they do not know their contributions.

Defense layer 3: Removal threat. If the group suspects collusion, they can refuse to reach consensus. The entire group gets zero Respect. This creates peer pressure against collusion.

Defense layer 4: Decay. If a fake account somehow amasses Respect, it decays at 2% per week. Within 34 weeks (one half-life), the balance drops to 50%.

Defense layer 5: OREC voting. Even if a breakout group successfully mines Respect for a fake account, the OREC proposal can be vetoed by the wider community. 1/3 of Respect holders can block execution.

Sybil cost estimate: To move one person 10 places in ranking over 15 weeks across 15 different random groups requires approximately 150 fake accounts (10 per group, 15 groups). Each account must have a wallet, Farcaster identity, and human-like Discord presence. This is expensive (time and identity costs) and provides minimal benefit (15 votes on one person’s Respect).

The economic barrier is high. Honest contribution is cheaper than Sybil attack.

V. The Five Voting Criteria: ZAO’s Customization

When ZAO Fractal members rank each other, they are instructed to consider five criteria specific to music and art:

1. The ZAO Vision - advancing music, art, and technology
2. Contribution - impactful work that pushed the collective vision forward
3. Collaboration - teamwork and uplifting others
4. Innovation - creative thinking, groundbreaking ideas
5. Onboarding New Members - helping newcomers join ZAO and Web3

These criteria are ZAO’s adaptation of the generic Fractally protocol. Eden Fractal and other fractals use simpler “contribution to community” language.

ZAO’s specificity matters. It operationalizes what the community values. By explicitly ranking people who onboard newcomers, The ZAO Fractal incentivizes growth. By explicitly ranking innovation, it incentivizes R&D. By explicitly valuing collaboration, it penalizes solo heroism and toxic competition.

In practice, these criteria guide deliberation. During the 4-minute presentations (Phase 3), members explain their contributions relative to these criteria. During voting, members use these criteria to disagree: “I think that was innovative, but it did not advance our vision.” Criteria provide a shared vocabulary.

VI. Visibility Bias: The Honest Critique

The Respect Game has one critical limitation: it ranks visible work higher than invisible work.

Loud social contributions (events, casts, conversation, mentoring) tend to out-rank quiet infrastructure work (code, contracts, internal tools, research). Every fractal community knows this and mitigates it differently.

Why Visibility Bias Exists

In a 4-minute presentation window, members describe their contributions. Visible work (events, social participation, mentoring) is easy to present. You can say “I hosted the Thursday jam session; 20 people came” and the group nods.

Invisible work is harder to present. “I debugged the on-chain voting contract for 20 hours” requires technical context. The group may not understand the impact.

Over time, visible work accumulates more Respect than invisible work of equal value.

Known Mitigations

ZAOstock’s sprint-fractal adaptation (documented in ZAO research Doc 498) prescribes:

• Explicit infra-contribution framing during pre-session briefing. The facilitator highlights the week’s infrastructure work before presentations begin.
• Facilitator pre-briefing on invisible-but-load-bearing work. The facilitator knows what engineering, research, or operations happened behind the scenes and can raise it during discussions.
• Periodic infra-themed rounds where ranking criteria are intentionally tilted toward maintenance work. Every 4-6 weeks, The ZAO could run a session where the criteria emphasize “systems thinking,” “code quality,” or “documentation.”

ZAO Fractal does not yet have a formal visibility-bias mitigation in production (as of May 2026), but it is a known design space for improvement. The limitation is documented and solvable, not hidden or fatal.

VII. 2/3 Consensus in Practice: The Convergence Dynamic

The 2/3 consensus rule is operationalized through sequential elimination voting. Here is how it converges in practice:

Scenario: 5-person group, tight ranking

Initial Level 6 votes:

• Member A: gets 2 votes (needs 3 for majority, fails)
• Member B: gets 2 votes (fails)
• Member C: gets 1 vote (fails)

The facilitator says: “No majority yet. Let us discuss.” Members deliberate: “I ranked A 6th because they were quiet this week, but actually they were dealing with a personal issue. Let me vote for C instead, who was visibly less active.”

After discussion, second vote round:

• Member A: 2 votes
• Member C: 3 votes (majority)

C is locked in as rank 6. Consensus emerges through deliberation, not pure tallying.

This is fundamentally different from voting systems. In voting, you get one vote, you cast it, done. In consensus, you can change your vote based on new information. Members are expected to update as the group’s reasoning becomes clear.

Why this prevents collusion:

If members try to collude (always vote for the same person regardless of contribution), they fail to reach majority quickly. The group notices the pattern. Discussion becomes adversarial. Eventually, real consensus breaks down, and the facilitator may need to reset or abort the round.

Collusion is audible. It shows up in conversation as obvious bloc voting. Real humans catch it.

VIII. Public vs. Secret Voting: ZAO’s Choice

ZAO uses public voting. Every vote is announced aloud by the bot. Members can see who voted for whom.

This is a deliberate trade-off:

Pros of public voting:

• Prevents vote trading (no secret deals).
• Makes collusion visible (can be challenged in real-time).
• Encourages truthfulness (you must defend your vote aloud if questioned).
• Builds trust through transparency.

Cons of public voting:

• Social pressure to conform (members might vote for the loud personality, not the best contributor).
• Intimidation risk (a dominant voice can sway votes).
• Requirement for psychological safety (members must feel safe disagreeing).

Eden Fractal uses secret voting (votes recorded but not announced immediately). The argument: secret voting prevents social pressure and ensures independent judgment.

ZAO’s choice of public voting reflects its values: transparency and conversation over privacy and independence. The small group size (5-6 people) makes privacy less critical; social dynamics are already visible.

Both approaches are valid. The important thing is that the mechanism is consistent, intentional, and tested.

IX. Sybil and Collusion Defense: A Layered Architecture

The Respect Game defends against attacks through five stacked barriers:

No single layer is perfect. Randomization can be broken with enough fake accounts. Consensus can be faked with enough collusion. But the combination is strong.

To move one real person 10 places in ranking over 15 weeks (a modest attack) requires:

• 150+ fake accounts (to ensure group presence across all randomizations)
• Consensus building with 10+ real humans per group per week
• Coordinated voting (public, audible, visible to all)
• Sybil survival through decay (accounts expire if not active)

Cost: High (time, identity, coordination). Benefit: Modest (15 ranking movements, possibly vetoed, decaying anyway).

The economics are unfavorable. Honest contribution is cheaper.

X. The Open Bottleneck: OREC Authority

As of May 2026, OREC minting authority is held by Zaal and civilmonkey.eth. These two individuals review proposals, run the /randomize and /timer commands, and facilitate sessions.

This is intentionally centralized in the short term. It allows ZAO to operate efficiently without complex decentralized governance infrastructure. But it is also an acknowledged bottleneck.

The problem: If Zaal or civilmonkey.eth become unavailable or corrupt, OREC operations halt. The weekly ritual breaks.

The plan: Future ZAO governance should decentralize session facilitation. Possible mechanisms:

1. Rotate facilitators. Each week, a different high-Respect member facilitates /randomize and /timer.
2. Multisig authority. Require 3-of-5 trusted members to approve OREC execution.
3. Automate trivial sessions. If consensus is obvious (e.g., 5-0 ranking), auto-execute without manual review.

This is documented in ZAO research Doc 115 as a medium-priority governance upgrade.

Sources

• /Users/zaalpanthaki/Documents/ZAOfractal/research/whitepaper-foundations/02-respect-game-mechanism.md (step-by-step mechanics, Fibonacci mathematics, game theory)
• /Users/zaalpanthaki/Documents/ZAOfractal/reference/09-respect-game-process.md (the weekly ritual, Discord bot implementation, visibility bias)
• /Users/zaalpanthaki/Documents/ZAOfractal/research/whitepaper-foundations/03-ordao-onchain-architecture.md (OREC voting cycle, 2/3 rule, soulbound enforcement)
• /Users/zaalpanthaki/Documents/ZAOfractal/research/primary-sources/respect-deep-dive.md (Fibonacci justification, Ultimatum Game, equilibrium analysis)
• /Users/zaalpanthaki/Documents/ZAOfractal/reference/07-respect-token-mechanics.md (voting criteria, Gini equality, tier thresholds)
• /Users/zaalpanthaki/Documents/ZAOfractal/research/01-foundations-deep.md (game theory, Nash equilibrium, mechanism design)

Word count: 3,118

Continue to Chapter 6: On-Chain Architecture - ORDAO, OREC