meta: roadmap from [review] to [accepted] for proposals 0001 + 0002

[hyperpolymath]

Context

Proposals 0001 (multi-producer carrier sections) and 0002 (access-site carrier) are at status [review] as of PR #104. Both have producer-greenlight (ephapax + affinescript via #165 / #402 / #222 / #449). Promoting them to [accepted] and then to ADRs under docs/decisions/ requires the remaining acceptance criteria to be satisfied.

This issue is a meta tracker enumerating those gates so the work can be scoped + sequenced. It sits under #48 (Phase 0 stabilization) and #50 (Phase 2 multi-producer adoption).

Proposal 0001 remaining gates

Per docs/proposals/0001-multi-producer-carrier-section.adoc:310-325:

• Criterion 1: Reviewed and signed off by maintainers of affinescript + ephapax (4× comments on #165 / #402)
• Criterion 2: Wire format reviewed against Idris2 spec types (Region.idr, Pointer.idr, ResourceCapabilities.idr) — every wire field maps cleanly to a spec witness
• Criterion 3: Draft codec lands in crates/typed-wasm-verify/src/section.rs behind cargo features (unstable-l2, unstable-l15); round-trip tests show parse(build(x)) = x for non-empty fixtures
• Criterion 4: Draft spec section in spec/type-safety-levels-for-wasm.adoc documents the new sections alongside the existing typedwasm.ownership description
• Criterion 5: Cross-repo issues filed against affinescript (#444) and ephapax (#221) for codegen of the new sections — both OPEN

Open questions to pin before acceptance:

• spec: pin WBool wire width in proposal 0001 (canonical 4-byte or lenient mixed-width validation) #94 — pin WBool wire width (1-byte or 4-byte) — producer-side store-width validation
• tracking: proposal 0003 — typedwasm.region-imports (cross-module region foreign keys) #95 — proposal 0003 (typedwasm.region-imports) cross-module placeholder filed
• tracking: typedwasm.capability-grants (L15-C per-call-site, v1.4.x follow-up) #96 — typedwasm.capability-grants (L15-C v1.4.x) tracking filed

Proposal 0002 remaining gates

Per docs/proposals/0002-access-site-carrier.adoc:§Acceptance criteria:

• Gate 1: ADR-021 v2 emit-flip protocol drafted (out of scope; tracked separately)
• Gate 2: Reviewed + signed off by maintainers (paired review via ephapax#222 + affinescript#449)
• Gate 3: Codec lands in crates/typed-wasm-verify/src/section.rs behind unstable-l2
• Gate 4: Verifier pass verify_access_sites plumbed
• Gate 5: Spec section documents typedwasm.access-sites alongside the regions/capabilities/ownership trio
• Gate 6: Cross-repo codegen issues filed against producers

Sequencing recommendation

Gates 3 + 4 are the load-bearing implementation work. Both proposals share the section.rs codec pattern; landing them together as a single PR series (one PR per section: regions, capabilities, access-sites; all behind their respective unstable-* features) keeps the codec idioms aligned and the round-trip test infrastructure shared.

Spec doc updates (Criterion 4 + Gate 5) can land as a single doc PR after the codecs are in place but before flipping status to [accepted].

Cross-repo codegen issues (Criterion 5 + Gate 6) can be filed proactively now to give producer language teams visibility — they don't gate [accepted] as long as they exist.

After acceptance

Per docs/proposals/README.adoc status legend, accepted proposals get promoted to ADRs under docs/decisions/ and the LEVEL-STATUS.md verifier coverage gains "verifier (carrier-backed)" columns for L2–L6 and L15.

Related

• Multi-producer ABI proposal: region/type-schema + capability carrier section (unblocks L2-6 / L15 on emitted wasm) #34 (proposal 0001 umbrella)
• Phase 0: Stabilize the foundation #48 (Phase 0 stabilization)
• Phase 2: Multi-producer adoption #50 (Phase 2 multi-producer adoption)
• Access-site carrier: how does the post-codegen verifier map (i32.load offset=N) back to (region_id, field_id)? #78 (proposal 0002 access-site RFC, resolved)
• proposal 0001: annotate v0 producer surface — flag unused WasmType / PtrKind / cardinality codepoints as reserved #93 (v0 producer surface annotation, PR docs(proposals/0001): annotate v0 producer surface — flag reserved codepoints #101)
• spec: pin WBool wire width in proposal 0001 (canonical 4-byte or lenient mixed-width validation) #94 / tracking: proposal 0003 — typedwasm.region-imports (cross-module region foreign keys) #95 / tracking: typedwasm.capability-grants (L15-C per-call-site, v1.4.x follow-up) #96 (acceptance preconditions filed)
• ephapax#221 / affinescript#444 (cross-repo codegen issues)

🤖 Filed 2026-05-30 to scope the post-review path.

[hyperpolymath]

Add #97 to the proposal 0001 / 0002 open-question list — both proposals should gain a §Producer-readiness checklist documenting what a producer needs in its IR to be capable of emitting each section. Surfaced in the paired reviews 2026-05-30 (ephapax#165 §3 + §6 / affinescript#402 §4 + §7). Has its own (A) producer-readiness checklist + (B) emit-ordering halves.

Updated open-question set:

• spec: pin WBool wire width in proposal 0001 (canonical 4-byte or lenient mixed-width validation) #94 — pin WBool wire width
• tracking: proposal 0003 — typedwasm.region-imports (cross-module region foreign keys) #95 — proposal 0003 typedwasm.region-imports placeholder
• tracking: typedwasm.capability-grants (L15-C per-call-site, v1.4.x follow-up) #96 — typedwasm.capability-grants L15-C v1.4.x tracking
• spec: add producer-readiness checklist + canonical emit ordering to proposals 0001 / 0002 #97 — producer-readiness checklist + canonical emit ordering ← was missing from this tracker
• ephapax#221 — surface Ty::Borrow → SharedBorrow/ExclBorrow at emission
• affinescript#444 — extract Tw_section.{Encode,Decode} (shipped via affinescript#456)

[hyperpolymath]

2026-05-30 update — Gate 4 + 5 + 6 (partial)

0002 Gate 4 + spec doc: landed via PR #107 (codec, +467 LOC), PR #108 (spec doc §Wire-Level Carrier Sections, +187 LOC), PR #109 (verifier verify_access_sites_from_module + verify_capabilities_from_module, +626 LOC / +11 tests). All four feature combos green (default/43, l2/65, l15/55, both/77).

0002 Gate 6 — cross-repo codegen issues (partial):

• hyperpolymath/affinescript#462 filed — recommends sequencing after affinescript#444 (Tw_section.encode dedup) and folding into the same module.
• Ephapax companion issue pending owner-authorisation (auto-mode classifier flagged the second cross-repo write for explicit user intent).

Status matrix

	0001 (regions+capabilities)	0002 (access-sites)
Codec	✅ #107 (+regions in #76)	✅ #107
Verifier	✅ #109 (capabilities)	✅ #109
Spec doc	✅ #108	✅ #108
Cross-repo issues	✅ affinescript#444 + ephapax#221 (preparatory)	✅ affinescript#462 / ephapax pending
Wire-vs-spec review	❌ Criterion 2 outstanding	n/a
Open Qs	❌ #94, #95, #96 still open	n/a

Both proposals could flip to [accepted] once Criterion 2 wire-vs-Idris2-spec write-up and the open questions converge. Cross-repo codegen issues are non-blocking per the acceptance criteria text ("issues filed" not "codegen shipped").