BLUEPRINT
Chapter 7: Social Relationship Graph and Intimacy
7.1 Why GMC Needs a Social Relationship Graph
GMC does not merely record contributions — it also records relationships between people. This is not an add-on feature but the foundation of core mechanisms:
| Mechanism Dependent on the Relationship Graph | Purpose |
|---|---|
| Inheritance mechanism | Determines attenuation ratio (higher intimacy = less attenuation) |
| Stakeholder exclusion | Excludes individuals too closely related to the contributor during voting |
| Anti-fraud | Identifies abnormal relationship patterns and collusion behavior |
| Community governance | Defines community boundaries and membership relationships |
Without the relationship graph, none of the above mechanisms can function.
7.2 Sources of Intimacy
Intimacy derives from Fay-to-Fay interactions and the social relationship network:
- Interaction frequency: Communication and collaboration frequency between two Fay
- Interaction depth: Complexity and duration of collaborative projects
- Relationship declarations: Relationships actively declared by users (family, colleagues, etc.)
- Joint participation: Communities, projects, and votes participated in together
7.3 On-Chain Strategy
Why On-Chain Storage Is Necessary
Conclusion from discussions: Social relationships must be stored on-chain to ensure the authenticity of relationships and prevent fabrication.
If relationship data can be forged, mechanisms such as inheritance attenuation and voting exclusion will all fail.
Layered Storage
| Data Type | Storage Location | Rationale |
|---|---|---|
| Relationship existence | On-chain | Ensures unforgeability |
| Intimacy values | On-chain | Serves as the basis for inheritance and exclusion |
| Intimacy computation proofs | On-chain (ZK proofs) | Ensures computation is auditable |
| Interaction details | Off-chain | Large data volume, involves privacy |
Off-Chain to On-Chain Anchoring
- Interaction details are stored off-chain
- Statistical results are periodically hash-anchored to the chain
- ZK proofs are submitted when intimacy is updated
- Anyone can verify that off-chain data has not been tampered with via the hash
7.4 Intimacy Model
Computation Inputs
Intimacy = f(interaction frequency, interaction depth, relationship declarations, joint participation, time decay)
Properties
- Has a maximum upper bound
- Decays with prolonged lack of interaction
- Computation process is auditable via on-chain proofs
- Symmetry TBD (whether A→B equals B→A)
Intimacy-to-Function Mapping
| Intimacy Range | Inheritance Attenuation | Voting Exclusion |
|---|---|---|
| Very high (> 0.9) | Lowest | Must exclude |
| High (0.7–0.9) | Low | Recommended to exclude |
| Medium (0.4–0.7) | Moderate | Not excluded |
| Low (0.1–0.4) | High | Not excluded |
| Very low (< 0.1) | Very high or disallowed | Not excluded |
7.5 Relationship Types
- Blood relations: Parents, children, siblings
- Legal relations: Spouse, guardian
- Social relations: Friends, colleagues, mentor-student
- Organizational relations: Employment, business partners
Different relationship types may have different intimacy baselines and decay rates.
7.6 Anti-Forgery
- Relationship declarations require confirmation from both parties (bilateral signatures)
- Interaction records are automatically generated by the system, not manually entered
- A large volume of interactions within a short period is treated as anomalous
- Isolated high-frequency interactions between two parties (with no shared social circle) are treated as suspicious
- Relationships must already be on-chain before an event occurs (retroactive recording for inheritance purposes is not allowed)
7.7 Privacy Protection
- Relationship existence is public (used for public functions such as voting exclusion)
- Specific intimacy values can be selectively disclosed
- Interaction details are strictly confidential
- ZKP is used to prove eligibility without revealing specific relationships
7.8 Discussion Notes
Design considerations for the social relationship graph:
- This is the key feature that distinguishes GMC from a pure Token system
- Data volume is the greatest challenge — a global social graph is enormously large in scale
- Layered storage (on-chain relationships + off-chain details + anchoring proofs) is the current balanced approach
- The symmetry question for intimacy requires further discussion
- The relationship graph itself also requires anti-forgery mechanisms