Protocol Evolution Markets

Protocol Evolution Markets

Trit: 0 (ERGODIC - coordinates market equilibrium)
Foundation: Dave White Multiverse Finance + Skill Evolution + Mixing Proofs

Core Concept

Protocol standards evolve through selection pressure. Prediction markets provide:

1. Price signals for which standards will be adopted
2. Incentive alignment for standard development
3. Fork coordination when communities disagree
4. Merge signals when standards converge

┌─────────────────────────────────────────────────────────────────────┐
│                    PROTOCOL EVOLUTION MARKET                        │
├─────────────────────────────────────────────────────────────────────┤
│                                                                     │
│   Standard A ──┬── Fork A.1 ──┬── Merge AB ◄── Standard B          │
│                │              │                    │                │
│                └── Fork A.2   └── Dead End         └── Fork B.1    │
│                                                                     │
│   Market prices predict which branches survive                      │
└─────────────────────────────────────────────────────────────────────┘

Multiverse Finance Integration

From Dave White's paper: split financial system into parallel universes (verses).

Verses as Protocol Futures

# Each verse represents a possible protocol future
struct ProtocolVerse
    spec_hash::UInt64           # Hash of specification
    adoption_metric::Float64    # Current adoption (0-1)
    compatibility::Set{Symbol}  # Compatible protocols
    parent_verse::Union{Nothing, ProtocolVerse}
end

# Example: agentskills.io spec versions
verses = [
    ProtocolVerse(hash("v1.0"), 0.8, Set([:claude, :codex]), nothing),
    ProtocolVerse(hash("v1.1"), 0.3, Set([:claude, :codex, :cursor]), v1_0),
    ProtocolVerse(hash("v2.0-draft"), 0.1, Set([:amp]), v1_0),
]

Push Down / Pull Up Operations

# Push down: bet on ALL forks of a standard
function pushdown!(market, stake, parent_verse)
    children = get_forks(parent_verse)
    for child in children
        market[child] += stake / length(children)
    end
    market[parent_verse] -= stake
end

# Pull up: consolidate bets when standard merges
function pullup!(market, stake, merged_verse, source_verses)
    for source in source_verses
        @assert market[source] >= stake "Insufficient stake in $source"
        market[source] -= stake
    end
    market[merged_verse] += stake * length(source_verses)
end

Resolution

# Oracle resolves which protocol won
function resolve!(market, winning_verse)
    for verse in keys(market)
        if !is_compatible(verse, winning_verse)
            # Non-compatible verses become worthless
            market[verse] = 0.0
        end
    end
end

GF(3) Fitness Signals

From skill-evolution: protocols have triadic fitness:

def protocol_fitness(spec):
    """Calculate GF(3) fitness for a protocol spec."""
    
    # MINUS signals: problems
    validation_failures = count_validation_failures(spec)
    breaking_changes = count_breaking_changes(spec)
    minus_score = -(validation_failures + breaking_changes * 2)
    
    # ERGODIC signals: stability  
    implementations = count_implementations(spec)
    compatibility = measure_compatibility(spec)
    ergodic_score = implementations * compatibility
    
    # PLUS signals: innovation
    new_features = count_new_features(spec)
    adoption_rate = measure_adoption_growth(spec)
    plus_score = new_features + adoption_rate * 10
    
    # Net trit
    total = minus_score + ergodic_score + plus_score
    return Trit(sign(total))

Market Mechanisms

1. LMSR (Logarithmic Market Scoring Rule)

import math

class ProtocolMarket:
    """Hanson's LMSR for protocol evolution betting."""
    
    def __init__(self, protocols, liquidity=100.0):
        self.liquidity = liquidity
        self.shares = {p: 0.0 for p in protocols}
    
    def cost(self) -> float:
        """Current cost function C(q)."""
        return self.liquidity * math.log(
            sum(math.exp(q / self.liquidity) for q in self.shares.values())
        )
    
    def price(self, protocol) -> float:
        """Current price = probability estimate."""
        exp_sum = sum(math.exp(q / self.liquidity) for q in self.shares.values())
        return math.exp(self.shares[protocol] / self.liquidity) / exp_sum
    
    def buy(self, protocol, amount) -> float:
        """Buy shares, return cost."""
        old_cost = self.cost()
        self.shares[protocol] += amount
        new_cost = self.cost()
        return new_cost - old_cost

2. pm-AMM (Prediction Market AMM)

From Paradigm research:

class PmAMM:
    """Paradigm's prediction market AMM."""
    
    def __init__(self, outcomes, initial_liquidity):
        self.reserves = {o: initial_liquidity for o in outcomes}
        self.k = initial_liquidity ** len(outcomes)  # Constant product
    
    def swap(self, sell_outcome, buy_outcome, amount):
        """Swap outcome tokens."""
        # x * y = k (for 2 outcomes)
        new_sell = self.reserves[sell_outcome] + amount
        new_buy = self.k / new_sell
        received = self.reserves[buy_outcome] - new_buy
        
        self.reserves[sell_outcome] = new_sell
        self.reserves[buy_outcome] = new_buy
        
        return received
    
    def implied_probability(self, outcome):
        """Price = probability."""
        total = sum(self.reserves.values())
        return (total - self.reserves[outcome]) / total

Mixing Proofs in Negative Curvature

From prediction_market_proofs.rb: hyperbolic random walks for privacy.

module ProtocolMarketProofs
  # Spectral gap ensures fast mixing (1/4 for Ramanujan graphs)
  SPECTRAL_GAP = 0.25
  
  # Bet anonymization via random walk
  def anonymize_bet(bet, mixing_time)
    walker = HyperbolicWalker.new(bet.commitment)
    
    mixing_time.times do
      walker.step!  # Random step in Poincare disk
    end
    
    # After O(log n) steps, position is uniformly distributed
    MixingProof.new(
      commitment: walker.position,
      proof: walker.path_hash,
      spectral_gap: SPECTRAL_GAP
    )
  end
  
  # Verify bet without revealing identity
  def verify_bet(proof)
    # Check path is valid random walk
    proof.spectral_gap >= SPECTRAL_GAP
  end
end

Protocol Evolution Events

Fork Detection

-- Detect when a protocol forks
SELECT 
    parent_spec,
    child_spec,
    fork_timestamp,
    compatibility_score,
    adoption_delta
FROM protocol_events
WHERE event_type = 'fork'
  AND compatibility_score < 0.8  -- Significant divergence
ORDER BY fork_timestamp DESC;

Merge Prediction

def predict_merge(spec_a, spec_b):
    """Predict probability of two specs merging."""
    
    # Factors favoring merge
    shared_features = len(spec_a.features & spec_b.features)
    shared_maintainers = len(spec_a.maintainers & spec_b.maintainers)
    
    # Factors opposing merge
    breaking_diffs = count_breaking_differences(spec_a, spec_b)
    governance_conflict = measure_governance_conflict(spec_a, spec_b)
    
    # Simple logistic model
    logit = (
        0.3 * shared_features +
        0.5 * shared_maintainers -
        0.8 * breaking_diffs -
        0.4 * governance_conflict
    )
    
    return 1 / (1 + math.exp(-logit))

Example: agentskills.io Evolution

# Current specs in the market
specs = {
    "agentskills-v1.0": {"adoption": 0.6, "trit": 0},
    "agentskills-v1.1-cursor": {"adoption": 0.2, "trit": +1},
    "codex-skills-native": {"adoption": 0.15, "trit": -1},
    "unified-v2-draft": {"adoption": 0.05, "trit": +1},
}

market = ProtocolMarket(specs.keys(), liquidity=1000)

# Simulate betting
market.buy("agentskills-v1.1-cursor", 50)  # Bullish on Cursor adoption
market.buy("unified-v2-draft", 30)          # Bet on unification

# Current prices (probabilities)
for spec in specs:
    print(f"{spec}: {market.price(spec):.2%}")

Tripartite Market Structure

┌─────────────────────────────────────────────────────────────────────┐
│  MINUS (-1)          ERGODIC (0)           PLUS (+1)               │
│  Validators          Arbitrageurs          Speculators             │
│                                                                     │
│  - Check spec        - Provide             - Bet on new            │
│    compliance          liquidity             features              │
│  - Report bugs       - Balance prices      - Fund development      │
│  - Short failing     - Hedge positions     - Long innovation       │
│    specs                                                            │
└─────────────────────────────────────────────────────────────────────┘

DuckDB Schema

CREATE TABLE protocol_specs (
    spec_id VARCHAR PRIMARY KEY,
    spec_hash UBIGINT,
    version VARCHAR,
    created_at TIMESTAMP,
    parent_spec_id VARCHAR,
    adoption_score FLOAT,
    trit INT,  -- -1, 0, +1
    status VARCHAR  -- 'draft', 'active', 'deprecated', 'merged'
);

CREATE TABLE protocol_bets (
    bet_id VARCHAR PRIMARY KEY,
    spec_id VARCHAR,
    direction VARCHAR,  -- 'long', 'short'
    amount FLOAT,
    price_at_bet FLOAT,
    timestamp TIMESTAMP,
    mixing_proof VARCHAR  -- Anonymized via hyperbolic walk
);

CREATE TABLE protocol_events (
    event_id VARCHAR PRIMARY KEY,
    event_type VARCHAR,  -- 'fork', 'merge', 'deprecate', 'adopt'
    source_specs VARCHAR[],
    target_spec VARCHAR,
    timestamp TIMESTAMP,
    market_impact FLOAT
);

-- Query: Predict next merge
SELECT 
    a.spec_id as spec_a,
    b.spec_id as spec_b,
    (a.adoption_score + b.adoption_score) / 2 as combined_adoption,
    COUNT(DISTINCT e.event_id) as shared_events
FROM protocol_specs a, protocol_specs b
LEFT JOIN protocol_events e 
    ON a.spec_id = ANY(e.source_specs) 
   AND b.spec_id = ANY(e.source_specs)
WHERE a.spec_id < b.spec_id
  AND a.status = 'active'
  AND b.status = 'active'
GROUP BY a.spec_id, b.spec_id
ORDER BY shared_events DESC, combined_adoption DESC
LIMIT 10;

Canonical Triads

three-match (-1) ⊗ protocol-evolution-markets (0) ⊗ skill-evolution (+1) = 0 ✓
bisimulation-game (-1) ⊗ protocol-evolution-markets (0) ⊗ gay-mcp (+1) = 0 ✓

See Also

• skill-evolution - Fitness metrics for skills (applies to protocols)
• multiverse-color-game - Dave White's verse operations
• prediction_market_proofs.rb - Mixing proofs in hyperbolic space
• entropy-sequencer - Information-gain ordering for market events

ALife atlas

Part of: alife-commons. Family: protocol-evolution. Canonical: protocol-evolution-markets.