From 96ca1d0a56cba4c9124f70fef6d7c1d017d18f23 Mon Sep 17 00:00:00 2001 From: Transaction Science Date: Mon, 7 Sep 2026 04:49:22 -0400 Subject: [PATCH] =?UTF-8?q?research:=20MPEG-AI=20Part=206=20requirement=20?= =?UTF-8?q?mapping=20=E2=80=94=20the=20one=20square=20with=20a=20deadline?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ISO/IEC 23888-6 (Media authenticity and provenance indication) is scoped to system-level media authentication across the entire lifecycle — authoring, creation, production, editing, storage, delivery, playback — on MPEG Systems technologies. CfP approved by WG 2 at MPEG 155; responses due as input contributions to MPEG 157 (Brisbane) on 13 January 2027, then CD at 158, DIS at 159, FDIS early 2028. A contribution accepted in January has about a year of drafting influence before the text sets. WHY THIS IS NOT ALREADY COVERED. WAI's provenance ingest is authoring-side manifest provenance — who made an asset and how, bound by hash. That is a different layer from a delivery-and-playback authenticity specification. Treating the existing ingest as coverage would be a category error: an authoring manifest says what happened BEFORE distribution; this work item is about what can be asserted and checked DURING it. WHAT WAI HAS THAT THE REQUIREMENTS SPACE DOES NOT. Two positions absent from every provenance standard in the survey, both sitting inside a lifecycle-scoped authenticity spec: per-object MEASURED energy with a declared acquisition class (provenance standards record what was done and by whom; none records what it cost, as measured, with the provenance of the measurement stated), and determinism tiers separating byte-exact reconstruction from approximate re-rendering (a hard hash over rendered output is only meaningful when the reconstruction is pinned). Both map to the delivery and playback stages an authoring manifest never reaches. THE ASYMMETRY. The downside of silence is not neutral. If this lands as THE systems-layer provenance format, WAI is either a profile of it or an alternative to it — and which one is decided by whether alignment work happened, not by whether a contribution was submitted. Alignment is cheap and reversible; being an alternative to a ratified systems-layer standard is neither. Deliberately a research document rather than an extension. There is no bitstream, no format and no capability string to register — the correct near-term action is requirement-mapping plus a membership path, and §6 states plainly that this claims no contribution, no membership, and no endorsement. Submitting requires SC 29 national-body membership, which is an organisational decision with a lead time against a fixed date. Ingest queue item 13 (#80 §6.5). Co-Authored-By: Claude Opus 5 --- wai/research/mpeg-ai-part6-mapping.md | 95 +++++++++++++++++++++++++++ 1 file changed, 95 insertions(+) create mode 100644 wai/research/mpeg-ai-part6-mapping.md diff --git a/wai/research/mpeg-ai-part6-mapping.md b/wai/research/mpeg-ai-part6-mapping.md new file mode 100644 index 000000000..6c4044317 --- /dev/null +++ b/wai/research/mpeg-ai-part6-mapping.md @@ -0,0 +1,95 @@ +# MPEG-AI Part 6 — requirement mapping and alignment + +> A preparation document, not a submission and not a normative extension. It maps +> what WAI already specifies onto the requirements of an open Call for Proposals, +> so the tree can be aligned *against* that work rather than diverge from it by +> default. Submitting an input contribution requires SC 29 national-body +> membership; **nothing here should be planned around an assumed submission.** + +## 1. The work item + +**ISO/IEC 23888-6** — *Information technology — Artificial intelligence for +multimedia — Part 6: Media authenticity and provenance indication.* Carried as an +ISO **PWI** project record. The Call for Proposals was approved by **WG 2** at +MPEG 155. + +| milestone | when | +|---|---| +| CfP responses, as input contributions to MPEG 157 (Brisbane) | **13 January 2027** | +| CD / CDAM | MPEG 158 | +| DIS / DAM | MPEG 159 | +| FDIS / FDAM | early 2028 | + +Scope, as published: system-level media authentication across the **entire +lifecycle** — authoring, creation, production, editing, storage, delivery and +playback — built on MPEG Systems technologies. Stage is +exploration/requirements; there is **no working draft**, no bitstream, no format +and no capability string to register. A contribution accepted in January 2027 has +roughly a year of drafting influence before the text sets. + +## 2. Why this is not already covered by WAI's provenance ingest + +WAI's existing provenance work is **authoring-side manifest provenance** — a +manifest asserting who made an asset and how, bound to it by hash. That is a +different layer from a **systems-layer delivery-and-playback authenticity +specification**, which is what this work item is scoped to. + +The two are complementary, and treating the existing ingest as coverage would be +a category error: an authoring manifest says what happened before distribution; +this work item is about what can be asserted and checked *during* it. + +## 3. What WAI has that the requirements space does not + +Two positions in this tree are absent from every provenance standard surveyed in +[2030-transport-landscape](2030-transport-landscape.md), and both sit squarely +inside a lifecycle-scoped authenticity specification: + +**Per-object measured energy with a declared acquisition class.** Provenance +standards record *what was done and by whom*. None records *what it cost, as +measured, with the provenance of the measurement stated*. WAI's binding is +per-object, carries `HwShunt` / `ModelBased` / `Estimator`, and refuses to +publish a figure from a failed measurement. Nothing else in the surveyed field +binds a measured physical quantity to a media object under signature. + +**Determinism tiers separating byte-exact reconstruction from approximate +re-rendering.** Authenticity over generated or reconstructed media needs a +statement of *what reproduction means* for the object in question. A hard hash +over rendered output is only meaningful when the reconstruction is pinned; WAI +says when it is, and — since the tier fix — says that the tier covers what the +sink emits rather than the decode alone. + +These map onto lifecycle requirements at the delivery and playback stages, which +are exactly the stages an authoring-side manifest does not reach. + +## 4. Alignment actions, in order of cost + +1. **Specify against the requirements, not around them.** As WAI's provenance, + energy and determinism extensions are revised, state how each satisfies or + declines a published requirement. This costs nothing now and is the difference + between a profile and an alternative later. +2. **Keep the vocabulary reconcilable.** Where this tree names a concept the work + item also names, prefer their term or state the mapping. Two vocabularies for + one property is the failure mode already recorded for energy acquisition + classes. +3. **Establish a membership path** if a contribution is wanted. This is an + organisational decision with a lead time, and the deadline is fixed. + +## 5. The asymmetry that makes silence expensive + +The downside of doing nothing is not neutral. If this lands as **the** +systems-layer provenance format, WAI is either a profile of it or an alternative +to it, and that is decided by whether the alignment work happened — not by +whether a contribution was submitted. + +Alignment is cheap and reversible. Being an alternative to a ratified +systems-layer standard is neither. + +## 6. What this document does not claim + +- It does **not** assert that a contribution will be made, or that membership + exists. +- It does **not** treat the CfP requirements as final; the stage is + exploration/requirements and requirements move. +- It does **not** describe WAI's mechanisms as adopted by, aligned with, or + endorsed by that work item. They are not. This is preparation for a + conversation that has not happened. -- 2.54.0 (Apple Git-157)