You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
## TL;DR — **Plan B is recommended** for an AI-driven translation workflow.
3
+
The i18n architecture is based on **externalized UI strings and full-replacement content files** (see `plan-b-externalized-strings-full-translations.md`).
4
4
5
5
---
6
6
7
-
## Why Plan B wins when translations are AI-generated
7
+
## Why this approach works well with AI-generated translations
8
8
9
9
When a new slug is added and AI generates the translations automatically, the
10
10
pipeline becomes:
@@ -13,23 +13,10 @@ pipeline becomes:
13
13
New English slug → AI prompt → Translated JSON file → Commit to repo
14
14
```
15
15
16
-
### Plan A (overlay) is awkward for AI
16
+
Key properties that make this AI-friendly:
17
17
18
-
- AI must know *which fields to include* in the overlay and which to omit
19
-
(`oldCode`, `modernCode`, `docs`, `related`, `prev`, `next` must all be
20
-
absent).
21
-
- If the AI accidentally includes any of those fields the build silently uses
22
-
the AI value instead of the canonical English one — hard to detect.
23
-
- Prompt engineering must explicitly say "only output these seven fields":
0 commit comments