MAP Schema Definitions (skeletal)¶

🧩 HolonTypes¶

HolonType	Description
Agent	Any entity capable of sensing and responding to its environment. It may be biological (e.g., a person, whale, or tree), technical (e.g., a computing process), or social (e.g., a family, cooperative, or commons).A sovereign holon capable of making, evaluating, and keeping Promises. Agents can be individuals, groups, or systems.
Persona	A cryptographic proxy identity for an Agent, with its own keypair. Personas act as the public-facing participants in Agreements and Roles.
Agreement	A shared commitment between Agents (via Personas), usually granting membership and rights within a Space.
Role	A named pattern of responsibility or capability within an Agreement, possibly linked to Promised or Permitted Dances.
RoleAssignment	Links a Persona to a Role within a specific Agreement, indicating participatory promises.
HolonSpace	A general-purpose container holon that defines a boundary for Holons. All Spaces in the MAP are HolonSpaces.
AgentSpace	A membraned HolonSpace for agents and interactions. May be an I-Space (HAS_I_SPACE) or a We-Space (if unanchored) (a.k.a., Space).
Dance	A unit of action that can be invoked in a Space, typically governed by a Promise from one Role to another.
TrustChannel	A TrustChannel is a membrane-bound communication interface, owned by an Agent, that acts in a specific Role within a specific Agreement. It is executed by the Agent’s ComputingNodes and inherits its authority via a RoleAssignment.

🔁 RelationshipTypes¶

Relationship Name	Source Type	Target Type	Description	Inverse Name
ASSIGNED_AGENT	RoleAssignment	Persona	The Persona acting in a Role within an Agreement	ASSIGNMENTS
ASSIGNED_ROLE	RoleAssignment	Role	Indicates the Role the Persona is fulfilling	ASSIGNMENTS
IN_AGREEMENT	RoleAssignment	Agreement	Indicates the Agreement the RoleAssignment is part of	ASSIGNMENTS
CONTEXT_SPACE	Agreement	AgentSpace	Indicates the Space in which the Agreement is valid or applied	AGREEMENTS
PROMISED_DANCE	Role	Dance	Indicates a Dance the Role promises to perform	PROMISED_BY
PERMITTED_DANCE	Role	Dance	Indicates a Dance the Role is permitted to request	PERMITTED_TO
HAS_I_SPACE	Agent	AgentSpace	Declares that a Space is the private I-Space of an Agent	IS_I_SPACE_OF
PERSONA_OF	Persona	Agent	Declares the sovereign Agent a Persona belongs to	PERSONAS
HOME_SPACE	Holon	Space	Declares the canonical home Space for a given Holon	HOME_FOR
DEPENDS_ON	AgentSpace	TrustChannel	The AgentSpace relies on this TrustChannel to communicate with the outside world via a subagent.	SUPPORTS
EXPOSES	AgentSpace	TrustChannel	This TrustChannel is visible and accessible at the AgentSpace's membrane boundary.	EXPOSED_BY
ASSIGNMENTS	TrustChannel	RoleAssignment	This TrustChannel is authorized to play the role assigned via this assignment.	ASSIGNED_TRUST_CHANNELS`

Validation Rules¶

🔒 MAP TrustChannel Validation Rules¶

Rule ID	Scope	Description	Formal Constraint
`trust-channel-assignments-must-share-agent`	`TrustChannel`	All `RoleAssignment`s that assign this `TrustChannel` must reference the same `Persona` as their `ASSIGNED_AGENT`.	∀ ra₁, ra₂ ∈ RA_TC, A(ra₁) = A(ra₂) or equivalently:
`trust-channel-exposed-by-owning-agentspace`	`TrustChannel`	The `AgentSpace` that exposes this `TrustChannel` must be the same as the one assigned to the Agent that the channel’s `Persona` belongs to.	If RA.ASSIGNED_TRUST_CHANNEL = TC, P = RA.ASSIGNED_AGENT, A = P.PERSONA_OF, then TC.EXPOSED_BY = A.ISPACE_FOR

📘 Glossary for Formal Expressions¶

Symbol / Variable	Meaning
`TC`	A `TrustChannel` being validated
`RA`	A `RoleAssignment` that references the `TrustChannel`
`RA_TC`	The set of all `RoleAssignment`s such that `ASSIGNED_TRUST_CHANNEL = TC`
`A(ra)`	The `ASSIGNED_AGENT` (a `Persona`) of a given `RoleAssignment` `ra`
`P`	A `Persona` that is the `ASSIGNED_AGENT` of a `RoleAssignment`
`A`	The `Agent` that `P` is a `PERSONA_OF`
`S_actual`	The `AgentSpace` that `EXPOSES` the `TrustChannel` (`TC.EXPOSED_BY`)
`S_expected`	The `AgentSpace` that `ISPACE_FOR` the `Agent` `A` (`A.ISPACE_FOR`)

🛡️ Privacy Architecture: How Personas Protect Agents¶

MAP introduces Personas as a cryptographic buffer between Agents and their actions. This is essential to maintaining sovereignty and defeating correlation-based surveillance.

🔍 The Problem: Correlation Tracking¶

Without separation of identities: - All RoleAssignments point directly to the Agent - Patterns of participation across spaces can be observed - Behavioral, temporal, and network linkages enable tracking - Agents are exposed to unwanted inference and profiling

🧬 The Solution: Personas¶

Concept	Explanation
Persona	A proxy identity with its own signing and encryption keys
Agent	A sovereign holon who controls one or more Personas
Separation	A Persona is the public actor; the Agent remains private

🔁 Relationship Flow¶

Agent
  └── PERSONAS → Persona
        └── ASSIGNED_AGENT ← RoleAssignment
              ├── IN_AGREEMENT → Agreement
              └── ASSIGNED_ROLE → Role

✅ Privacy Benefits¶

Property	Enabled by Personas
Contextual identity	✅ Scoped to each space or purpose
Unlinkability	✅ No shared identifier between Personas
Behavior isolation	✅ Participation graphs remain disjoint
Revocable presence	✅ Personas can be retired independently
Private agency	✅ Agents stay hidden unless they choose to reveal themselves

🔐 Cryptographic Design¶

Signing keypair: used to authenticate promises and requests
Encryption keypair: used to receive private messages
No link to Agent’s master key or other Personas
All delegation and disclosure is Agent-controlled

🌐 Summary¶

Agents do not act directly in the MAP.
Personas are their emissaries — modular, revocable, and unlinkable.
This makes MAP a platform where privacy and agency are mutually reinforcing.

MAP Spaces¶

Space Type	Agentic?	Owned?	Contains Agents?	Acts as We-Space?
I-Space	❌ No	✅ Yes	✅ Optional	✅ Yes (if it has sub-agents)
We-Space	❌ No	❌ No	✅ Yes	✅ Yes
Root Space	❌ No	❌ No	✅ Yes	✅ Yes (for all agents and spaces)

🧮 Mathematical Functions and Relationships in MAP Schema Context¶

Let:

a be an Agent
p be a Persona
s be a Space
g be an Agreement
r be a Role
d be a Dance
ra be a RoleAssignment
h be a Holon

🧩 Core Functions and Logical Predicates¶

Notation	Meaning
`M(s)`	The set of Agents who are members of Space `s`
`C(s)`	The set of Holons contained in Space `s` (via `HOME_FOR`)
`V(a, s)`	Predicate: Agent `a` is visible in Space `s`
`I(a, s)`	Predicate: Agent `a` has interaction rights in Space `s`
`St(a, h)`	Predicate: Agent `a` is a steward of Holon `h`

These are derived semantics built on schema relationships.

🔁 Mapped to Schema Relationships¶

Persona System¶

p PERSONA_OF → a
(Each Persona belongs to one Agent)
a PERSONAS → p₁...pₙ
(Each Agent may have multiple Personas)

Role Participation¶

ra ASSIGNED_AGENT → p
ra ASSIGNED_ROLE → r
ra IN_AGREEMENT → g

Agreement Context¶

g CONTEXT_SPACE → s
⇒ Derived Membership Rule:

If:
  ra.ASSIGNED_AGENT = p
  ∧ p.PERSONA_OF = a
  ∧ ra.IN_AGREEMENT = g
  ∧ g.CONTEXT_SPACE = s

Then:
  a ∈ M(s)

Space Anchoring (I-Space)¶

a HAS_I_SPACE → s
s IS_I_SPACE_OF → a

Holon Location¶

h HOME_SPACE → s
s HOME_FOR → h

🎭 Role Capability Model¶

r PROMISED_DANCE → d
(Role r promises to perform Dance d)
r PERMITTED_DANCE → d
(Role r is permitted to request Dance d)

📌 Notes¶

Membership, Visibility, and Interaction are derived states, not directly stored
M(s) is dynamically constructed from RoleAssignments
V(a, s) and I(a, s) depend on:
- Membrane rules
- Role-based permissions
- Explicit visibility declarations

🧠 Summary of Key Relations¶

Concept	Notation	Meaning
Membership	`a ∈ M(s)`	Agent `a` is a member of space `s`
Containment	`s₁ ⊂ s₂`	Space or holon `s₁` is structurally contained in `s₂`
Visibility	`V(a, s)`	Agent `a` is visible in space `s`
Interaction	`I(a, s)`	Agent `a` can act in space `s`
Stewardship	`St(a, h)`	Agent `a` is steward of holon `h`

Note: Membership is inherited, visibility is membrane-gated, interaction is Promise-based, and stewardship is explicitly granted.