Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
45 changes: 45 additions & 0 deletions docs/dnn-localization/457-document-tier-translation-workflow.md
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,51 @@ From the Phase 1 inventory §3, the document-tier content is **2 standalone FR H

Total: ~37,300 chars of FR prose across 2 files. These are **standalone FR content** — translating them is a document/prose task (like Scenarii narrative fields), **not** a key-value dictionary task (like `ui.*`/`res.*`).

> ### ⚠️ VERIFICATION UPDATE (2026-06-22, po-2024) — the §1/§4 structure figures above are PRE-VERIFICATION ESTIMATES and are INCORRECT
>
> The table + §4 selector design were written from the inventory's character-budget estimate, **not** from
> parsing the files. An actual structural probe (`grep` tag counts + content sampling) on 2026-06-22 by
> po-2024 (taking over this lane from po-2023) shows the real structure differs materially, which changes
> both the extraction cost and the value of this lane:
>
> **`fallacies.html` (81 KB) — verified tag counts:** `2 <h2>` + `2 <h3>` + `8 <li>` (jQuery-UI tab labels:
> "Manipulations de l'esprit", "Appel aux émotions"…) + **`56 <h4>`** (fallacy NAMES — "Argument d'autorité",
> "Appel à la pratique courante", "Raison de la majorité"…) + `114 <p>` (almost entirely boilerplate labels
> "Code Html à copier:" / "PhpBB Html Replacement:" preceding code blocks, **not** fallacy descriptions) +
> `55 <textarea>` + `55 <form>` (HTML code samples to copy-paste — **code, not prose, skip**).
>
> - **It is a developer integration charte** (how to embed the fallacy infographic PNG into a page/phpBB),
> NOT a fallacy glossary. Fallacy *definitions* are not in this file (they live in the infographic + the
> taxonomy CSV).
> - **The 56 `<h4>` fallacy names DUPLICATE the taxonomy CSV** (`Cards/Fallacies/...Taxonomy.csv`, already
> translated 8-language). Translating them here = **duplicate work** — they should be **joined from the
> taxonomy CSV**, not re-translated. Genuine NEW translatable surface = ~4 section/tab titles (h2/h3/li) +
> ~3 boilerplate labels (p) ≈ **7 short strings**.
>
> **`MariagePourTous.html` (38 KB) — verified tag counts:** `103 <div>` + `102 <a>` + **`0 <p>`, `0 <h2>`,
> `0 <h3>`, `1 <li>`**. The §4 assumption ("prose, selectors `p, h2, h3`") is **invalid** — there are no
> `p`/`h2`/`h3` elements. The prose is **bare text between `<div class="fallacieContainer">` blocks**, and
> each argument line (e.g. "01. Parce que dans le dico, le mariage c'est l'union légale…") appears
> **duplicated** — once as bare text, once inside a `fallacieContainer` div — interleaved with dictionary
> definitions and source citations. Correct extraction needs **text-node walking of `<div class="texte">`**
> and **dedup** of the bare-text/container pair, not a `p`/`h2`/`h3` selector sweep.
>
> **Net effect on the lane (honest assessment):**
>
> - `fallacies.html` is **low-value + partly duplicate** (7 new strings; 56 h4 joinable from taxonomy CSV;
> role = deprecated-looking dev charte — po-2023 flagged it "almost certainly not in v0.9.0 nav").
> - `MariagePourTous.html` is **1 essay of uncertain scope** with a **non-trivial bare-text dedup**
> extraction (no clean selectors).
> - Combined with the **unanswered scope gate** (§8 Q3: are these 2 pages even served in v0.9.0?), building
> the §4 extractor + spending gpt-5.5 translation budget now risks duplicate/possibly-deprecated work.
>
> **Recommendation (po-2024, surfaced to ai-01):** **defer the §4 extractor build and any translation**
> until jsboige resolves §8 Q3 (live-nav scope). If `fallacies.html` is confirmed out-of-scope (likely),
> only `MariagePourTous.html` remains — a single essay, where a bespoke segment pass is proportionate.
> The §4 selector design below must be corrected against the verified structure before any build. **This
> lane is NOT a blocker for v0.9.0** (string tier A/B is complete; fallacy bodies are already 8-lang via
> the taxonomy CSV); it is a nice-to-have pending a scope decision.

**Out of scope of this doc (covered elsewhere):**
- String tier (`ui.*`/`res.*`) → [tools/dnn_i18n/README.md](../../tools/dnn_i18n/README.md) + DatasetUpdater config #487.
- Fallacies Explorer / Rules Explorer → already CSV-driven via 2sxc queries (taxonomy CSVs already localized 8-lang).
Expand Down
Loading