Canon 0.3.0: capability provision economics (ITC-CAP 0.2.0)

Accept resource-control demand: move resource classes onto provisions,
let requirements carry targets and constraints, add human-effort class H
with native-unit consumption. Model stays proposed.

workplans/ITC-WP-0014-capability-model-consolidation.md

@@ -18,6 +18,7 @@ spec_refs:
   - infospace/assimilation/it-capability-canon/proposed-changes.md
   - infospace/assimilation/it-capability-canon/open-questions.md
   - infospace/models/capability/InfoTechCanonCapabilityModel.md
+  - demand/CapabilityProvisionEconomics.md
 state_hub_workstream_id: "18cccc77-5aef-48f2-91d6-124bd439285a"
 ---
 
@@ -32,8 +33,11 @@ closing the items the ITCC v0.1 assimilation deliberately deferred.
 
 Canon version 0.2.0 adopted the capability vocabulary under disposition `adapt`.
 The model is registered, validated, and anchored, but it has no schema, no formal
-mapping artifacts, no profile exercising it, and one unresolved identifier
-question that is blocking because capability ids are durable interfaces.
+mapping artifacts, and no profile exercising it. OQ-2 is resolved (canon 0.2.1).
+Consumer demand `CapabilityProvisionEconomics` (resource-control) refined the
+still-proposed model in canon 0.3.0 / ITC-CAP 0.2.0: requirement targets,
+provision-side consumption, and the H/I/P class set. T02 must encode that
+refined contract, not the v0.1.0 one.
 
 ## Tasks
 
@@ -66,6 +70,10 @@ state_hub_task_id: "1ed8d07a-34ee-4ad4-8e4a-03a513c14836"
 
 Add the capability contract schema under `infospace/schemas/`, register it in
 `infospace/infospace.yaml`, and validate `capabilities.yaml` against it.
+Encode ITC-CAP 0.2.0: no `typical_resource_classes` on contracts; resource
+classes carry `native_unit`, `supply`, and `capacity_behaviour`; requirement
+and provision record shapes live in the model prose until a companion
+requirement/provision schema is justified.
 
 ### T03 — Formal mapping artifacts for each anchor
 
@@ -135,11 +143,54 @@ state_hub_task_id: "455690e3-c39a-4926-a13b-a179b134b2a4"
 Promote the model from `proposed` to `draft` or `release-candidate`, bump the
 canon version, and write the `CHANGELOG.md` entry.
 
+### T08 — Finding A: move resource-class declaration off the contract
+
+```task
+id: ITC-WP-0014-T08
+status: done
+priority: high
+```
+
+Remove `typical_resource_classes` from every capability contract so §4.11 and
+the catalog agree. Consumption is declared on `CapabilityProvision` only.
+
+**Done 2026-08-15 (canon 0.3.0 / ITC-CAP 0.2.0).**
+
+### T09 — Finding B: demand-side targets and constraints
+
+```task
+id: ITC-WP-0014-T09
+status: done
+priority: high
+```
+
+Enrich `CapabilityRequirement` with profile, quality targets against declared
+dimensions, and placement constraints. Keep measurement on ITC-LAND / ITC-OBS.
+
+**Done 2026-08-15 (canon 0.3.0 / ITC-CAP 0.2.0).**
+
+### T10 — Finding C: human effort, native units, intelligence as substitute
+
+```task
+id: ITC-WP-0014-T10
+status: done
+priority: high
+```
+
+Add class `H`, narrow `P`, recharacterise `I`, declare native units and
+supply/capacity defaults, record consumption in native units, add
+`intelligence_intensity`. No `substitutes_for`, no exchange rate.
+
+**Done 2026-08-15 (canon 0.3.0 / ITC-CAP 0.2.0).** Landed here rather than as a
+new workplan: no LAND or commerce owning-model change was required.
+
 ## Out of scope
 
-- New domain models for `commerce.*` / `intelligence.*` (OQ-5) — needs a real
-  consumer demand first.
+- New domain models for `commerce.*` / `intelligence.*` (OQ-5) — the 2026-08-15
+  demand needed consumption structure, which landed in ITC-CAP; invoices,
+  entitlements, prompts, and evaluations still have no owning model.
 - Canon-wide intended/declared/applied/observed/assessed state qualifiers (OQ-3)
   — kernel-level pressure, not a capability-model decision.
 - CARING importing `identity.*` ids — touches a release-candidate standard and
   needs its own review.
+- Booked cost, currency handling, labour rates, or an H↔I exchange rate.