ADR 009: Incentive Model

Date: 2026-03-01 Status: Accepted

Context

The network needs ecosystem-native incentives for storage peers and relays. External subscriptions or tokens create dependency on outside systems. The protocol already has reciprocal storage pacts, follow-as-commitment, guardian pacts (voluntary storage for newcomers), and zaps (kind 9734/9735), but no explicit mechanism connecting contribution to visibility. Relays need sustainable value within the ecosystem beyond being optional accelerators.

See the full design document for detailed problem descriptions and rationale.

Decision

Three-layer incentive model:

1. Pact-Aware Gossip Routing

When a node forwards gossip, it prioritizes content from pubkeys it has active storage pacts with. Priority ordering:

  1. Active pact partners (highest) — nodes you have a live, verified storage pact with
  2. 1-hop WoT — pubkeys you directly follow
  3. 2-hop WoT — pubkeys followed by your follows
  4. Unknown — never forwarded

A user with 20 reliable pact partners has 20 nodes eagerly forwarding their content. Dropped pact = lost forwarding advocate = reduced reach.

No public score — network topology IS the incentive.

2. Relay-as-Curator

Relays have their own root pubkey and publish curated event lists and recommendations. Users follow relay feeds like they follow people. A relay's follower count and WoT position determine how far its curations travel through gossip.

Three relay types:

  • Discovery — curate by topic or quality
  • Infrastructure — fast indexing, high availability
  • Community — serve specific groups

Users benefit from publishing through well-connected relays, gaining a wider audience beyond their own WoT.

3. Lightning Premium Layer

Relays publish a service menu. Users zap to activate services:

  • Priority delivery — faster indexing, wider push to subscribers
  • Extended retention — events kept beyond the default window
  • Content boost — post featured in curated feed, transparently marked
  • Relay-defined services — analytics, custom filtering, etc.

Transparent, competitive (relays set their own prices), optional.

Rejected Alternatives

Blind Contribution Tokens

Pact partners issue blind tokens for passing storage challenges. Rejected: significant protocol complexity (blind signatures, anonymous credentials), gameable via colluding pact partners minting tokens for each other.

Public Reputation Tiers

Coarse-grained public contribution level attested by threshold signature. Rejected: reveals storage participation metadata, requires threshold signature scheme, still gameable.

Consequences

Positive:

  • Self-sustaining incentive loop without tokens or external dependencies
  • Privacy-preserving — no new metadata revealed
  • Relay operators earn sustainable value through curation and Lightning services
  • Storage reliability directly rewarded with content reach
  • Graceful degradation — less contribution means less reach, not exclusion
  • Pay-it-forward loop — today's Seedling becomes tomorrow's Guardian. Users who received guardian storage during bootstrap are encouraged to volunteer a guardian slot once they reach Sovereign phase, creating a self-reinforcing generosity cycle

Negative:

  • Incentive is subtle — users may not directly perceive that reliable storage improves their reach
  • Relay market may concentrate around a few popular discovery relays

Neutral:

  • Lightning layer is fully optional
  • Relay types are emergent (not protocol-enforced categories)