Membership and Sponsorship Policy¶

Field	Value
`policy-id`	`DIA-MEMBERSHIP-SPONSORSHIP-001`
`type`	Constitutional operational act / entry and sponsorship policy
`version`	`0.2.0-draft`
`date`	`2026-05-27`
`basis`	Constitution Art. VII, XV, XVI; Core Values; Proposal 051; Procedural Reputation Spec

Purpose¶

This policy defines how participants enter shared Orbiplex influence surfaces without turning membership into a moral purity test. The question is not "is this person good?" but "which shared surfaces may this subject affect, under which limits, on which evidence, with whose bounded sponsorship, and with what appeal path?".

Orbiplex keeps local reading and local node operation open, while placing sluices around influence. These sluices protect communication, public memory, marketplace exchange, governance, routing, custody, and public-trust surfaces from Sybil pressure, spam, fraud, factional capture, and careless endorsement.

Canonical Vocabulary¶

Membership and surface-access schemas use doc/schemas/_shared/membership-enums.v1.schema.json as the shared vocabulary source. Normative prose may explain those terms, but implementations should not copy enum lists by hand.

The entry classes are:

guest
contactable-participant
sponsored-candidate
probationary-member
full-participant
public-trust-role

The influence surfaces are:

local-read
contactability
public-comment
public-publishing
unsolicited-dm
broadcast
marketplace
custody
routing
moderation
arbitration
governance
public-trust

public-trust-role is an entry class. public-trust is the surface where high-stakes role authority is exercised.

Canonical Access Matrix¶

There is no single global accepted state that unlocks every shared capability. Entry policy is applied through a matrix of (entry-class, surface) -> decision, captured by surface-access-policy.v1.

Entry class	Surface	Default decision	Additional gate
`guest`	`local-read`	`allow`	local software and public reading only
`guest`	any shared influence surface	`deny`	no common-surface influence by default
`contactable-participant`	`contactability`	`probation+attestation`	contact-channel attestation; anti-spam limits
`contactable-participant`	`public-comment`	`review`	community policy may allow low-rate comments
`sponsored-candidate`	`public-comment`	`n-sponsors`	one scoped sponsor by default; slow-start limits
`sponsored-candidate`	`public-publishing`	`review`	sponsor scope and content-surface policy
`probationary-member`	`public-comment`	`allow`	low-rate or normal limits by local policy
`probationary-member`	`unsolicited-dm`	`deny`	relationship or opt-in required
`probationary-member`	`broadcast`	`deny`	no high fan-out by default
`probationary-member`	`marketplace`	`review`	low value cap, escrow/procurement contract
`full-participant`	`broadcast`	`review`	reputation, rate limits, and anti-collusion checks
`full-participant`	`routing`	`review`	capability passport, reliability history
`full-participant`	`custody`	`review`	capability passport, storage policy, audit
`full-participant`	`governance`	`review`	procedural reputation and COI checks
`public-trust-role`	`public-trust`	`review`	IAL, procedural reputation, COI, audit, revocability

The matrix is intentionally conservative. Federations may relax or tighten rows, but they should publish the result as surface-access-policy.v1 rather than embedding it as hidden runtime branches.

Sponsorship¶

Sponsorship is a scoped accountability relation, not a guarantee of moral quality. A sponsor states:

I know this subject well enough to introduce it to this Orbiplex surface, within this scope and risk limit, and I accept bounded reputational exposure if the sponsorship was grossly careless or collusive.

Sponsorship gives candidacy, not authority. A sponsored subject still needs the required attestations, probation, policy checks, and runtime limits for the target surface.

The default sponsorship templates are:

Template	Meaning	Default use
`light-vouch`	weak introduction, low exposure	contactability or very low-risk local community entry
`standard-introduction`	ordinary scoped sponsorship	basic community entry and low-rate public comment
`strong-vouch`	high-confidence scoped sponsorship	broader publishing or marketplace probation
`mentor-with-liability`	active mentor relation with stronger accountability	high-touch probation or sensitive community onboarding

The sponsorship artifact records the template, scopes, issuance and expiry, probation window, structured due-diligence references, revocability, revocation-tail duration, and evidence policy. It does not carry ad-hoc numeric exposure fields; those are policy projections from the template and evidence.

Derived sponsor liability is direct by default. It may attach to the sponsor when evidence shows negligent, reckless, or collusive sponsorship during the fresh sponsorship window.

Liability is classified ordinally, not by multiplying local coefficients:

Class	Meaning	Typical triggers
`negligible`	no meaningful derived liability	harm was unforeseeable or unrelated to the sponsorship scope
`mitigated`	reduced liability after constructive response	sponsor revoked promptly, reported red flags, or helped contain harm
`moderate`	ordinary bounded liability	sponsor missed weak signals or sponsored too broadly for the scope
`serious`	strong liability	sponsor ignored repeated red flags, mass-sponsored, or exceeded competence
`collusive`	sponsor was part of an abuse pattern	anti-collusion sweep finds sponsor-ring or coordinated capture

Every classification must include triggered conditions and evidence references. This makes the decision challengeable without pretending that five local numeric factors are a portable truth.

Liability does not propagate beyond one level unless an anti-collusion process establishes an organized sponsor ring.

Anti-Clan Controls¶

Sponsorship must not become a private aristocracy. Federations and communities should enforce:

sponsorship only within the sponsor's own scope of reputation and authority,
independent sponsors for higher-risk surfaces,
graph-distance or cluster-diversity requirements,
limits on active sponsorships per period,
automatic review for abnormal sponsorship velocity,
one-hop liability by default,
anti-sponsor-ring detection,
and appeal rights for both sponsor and sponsored subject.

The MVP baseline detector for sponsorship abuse is abnormal sponsorship velocity: too many active sponsorships per sponsor within a policy window, with policy-tunable thresholds.

Newcomer Slow-Start¶

New entrants should begin with narrow capability limits. The canonical defaults are expressed by:

default.surface-access-policy.json
newcomer.participant-entry-profile.json
newcomer.participant-effective-limits.json

Those examples are schema-backed and should be treated as the canonical newcomer profile fixture. Documents should link to them rather than duplicating YAML blocks.

These are defaults, not permanent exclusion. Constructive, independently evidenced action should make advancement cheaper than destructive action.

Sanctions and Return Path¶

Sanctions are expressed as (surface x intensity), not as one mixed ladder of unrelated actions.

Surface group	`soft`	`hold`	`hard`	`block`
`communication`	rate-limit	DM restriction	routing cut-off	federation block
`marketplace`	rate-limit	marketplace hold	escrow-only	marketplace block
`reputation`	downweight	quarantine	freeze	revoke projection
`role`	warning	review	suspension	revocation
`relationship`	warn	sponsor review	sponsor revoke	sponsor-ring action
`routing`	deprioritize	require fresh proof	cut-off	federation block
`custody`	require review	hold writes	suspend custody	revoke custody eligibility
`governance`	reduce weight	recusal	suspend voting/panel eligibility	governance block

The canonical intensity order is soft < hold < hard < block. Sanctions must preserve an audit trace, appeal path, and repair path unless there is an immediate safety threat. They should restrict influence surfaces rather than erase personhood.

Anti-Collusion Baselines¶

The first implementation should not try to detect every collusion form at once. The MVP baseline detectors are:

sponsorship: abnormal sponsorship velocity,
public adjudication: co-flagging coherence, for example Jaccard similarity of flagger sets across objects,
marketplace: closed-loop receipt detection, for example settlement cycles such as A -> B -> C -> A.

Additional detectors should be added explicitly when there is operational evidence that the baseline is insufficient.

This policy is supported by:

membership-invitation.v1
membership-sponsorship.v1
membership-acceptance.v1
participant-entry-profile.v1
participant-effective-limits.v1
surface-access-policy.v1
participant-capability-limits.v1
reputation-signal.v1