# Genesis — 2026-08-16

## Why this project exists

The ecosystem was running two independent canons in the same market domain
(`infotech`) with **no cross-reference in either direction**:

- `info-tech-canon` — InfoTechCanon v0.6.0, `service-baseline`. Kernel, 12
  models, 3 standards, declared per-model concept ownership, an orthogonality
  rule ("standards can import but not redefine each other"), and a live
  CLI/JSON/API surface on `infospace-bench`.
- `identity-canon` — documentation-only research repo. Three `IDENTITY-WP-*`
  workplans, all `finished`; no substantive work since the commercial-identity
  research pass. ~60 concepts in `canon/CanonicalGlossary.md`.

Three findings drove the decision.

**Concept-ownership collision.** `InfoTechCanonOrganizationModel:55` owns
`Actor, Person, Organization, Group, Role, Membership, ...`;
`InfoTechCanonAccessControlModel:106` owns `Subject, Principal, ...`.
identity-canon independently defined all of them. Precisely what InfoTechCanon's
orthogonality rule exists to prevent.

**An unowned gap.** `InfoTechCanonAccessControlModel:214` explicitly pushes
identity provisioning, authentication, identity proofing, and account lifecycle
out of scope, and `:41` does the same for generic organization modelling. So
`Account`, `Identity Record`, `Identifier`, `Credential`, `Claim`, `Persona`,
`Tenant`, `Realm`, `Synonymity Assertion` belonged to no model at all.

**Commercial semantics accumulating in technical canon.** About a third of
identity-canon's glossary is counterparty/commercial modelling, not identity —
hence its `government` secondary domain. Independently,
`info-tech-canon/demand/CapabilityProvisionEconomics.md` (accepted 2026-08-15)
records procurement and economics demand arriving in `ITC-CAP` from consumer
`resource-control`, domain `financials`. Two unrelated donors pushing commercial
semantics into a technical canon is a domain boundary, not a coincidence.
InfoTechCanon's own Purpose/Demand extension names this `ScopePressure`.

## What was decided

`ADR-006` (Custodian canon, status `proposed` at genesis): three federated
canons — Custodian (ecosystem governance), InfoTech (information-processing
semantics), Commerce (counterparty and commercial semantics) — importing but
never redefining each other.

Two shaping choices are worth recording, because both were revisions of an
earlier plan:

**Rename in place, don't found-and-archive.** The first proposal was to create a
new `commerce-canon` repo and archive `identity-canon`. Renaming instead keeps
git history, hub registration, `.repo-classification.yaml` lineage, and the
three finished workplans attached to the commercial content that *stays*. Only
identity emigrates. One repo changes name, one gains a model, nothing is
archived.

**Name it for commerce, not for "business terms".** The working name was
`business-terms-canon`. "Terms" advertises a glossary, and an open-ended
glossary is the failure mode — it invites every business-flavoured word in the
ecosystem to land there with no ownership discipline. The real content is
counterparty and commercial-relationship semantics with a KYC/AML/procurement
driver, so the canon is named for that.

## What the boundary rests on

Two of the collisions were resolved by identity-canon's *own* design principles
rather than by new argument: P1 ("Use Actor As The Participation Root") requires
importing `itc-org`'s `Actor`; P6 ("Keep Authorization Projections Separate")
already treats `Authenticated Subject` / `Authorization Principal` as
projections rather than identity-owned definitions. The boundary largely wrote
itself from principles already agreed inside the research repo.

Six collisions were **not** resolved at genesis and are listed as open questions
in `ADR-006` — `Scope`, `Assurance Level`, `Evidence Source`, `Relationship
Tuple`, `Adjudication Outcome`, and the placement of `Community` / `Family Or
Household`. Resolving them is the first task of `CFED-WP-0001`.

## Risk being managed

CommerceCanon could repeat identity-canon's failure mode: dormant, zero
consumers, drifting. Two mitigations are built into the design — it launches
holding real research-backed content rather than as an empty scaffold, and it
grows only by demand signal with named consumer evidence, the mechanism
InfoTechCanon already runs.