Twinned from https://github.com/cfsh/ninefold/issues/444 by tools/beads/import_github.py's reconcile pass.
Suggestions from ninefold-reviewer's review of #443. All advisory, none blocking — #443 covered every player-visible string it set out to. These are the places the *word* survives as a noun in text a developer reads, after the PR's new CONTROLS.md "The nouns" section declared it dead in the UI.
1. games/tonapse/CONTROLS.md — two rows the rename missed, in the file that now defines the vocabulary
- Gesture table: | Sound-editor **dial** | — | set absolute value | … | — names the panel the UI now heads CHANNEL EDITOR.
- R3 row: "restore the parameter's default (#323) — the value the sound was created with (authored base sound / instrument manifest) … disk-loaded sounds carry authored mix values". This is "sound" meaning *timbre*, which is exactly the ambiguity #441 set out to remove; the intended word here is *instrument*.
- (Playtest-capture row's "song/pattern/sound into the session journal" names a journal field, so it is arguably internal — judgement call.)
2. games/tonapse/CUJS.md still reads ## CUJ 4 — Tweak a sound mid-song and ## The sound editor as a "3rd dimension" (#257 — for discussion). Not player-visible, but it is the other tonapse design doc and it now disagrees with CONTROLS.md.
3. Probe _intent prose mixes the two vocabularies. tools/probes/focus-grammar.json was updated to "change-instrument row" but the same sentence still ends "…one down from any sound chip", and later "switched from the SOUNDS list". tools/probes/grid-scroll.json: "with 8 sounds … Plus the SOUNDS strip". Both name a strip the UI now heads CHANNELS.
4. Code comments naming the strip by its old on-screen label — TonapseHome.cs:429,431,464,759, ui/PatternPicker.cs:71, ui/SongGrid.cs:624 ("entering the editor through the SOUNDS list"), ui/SoundEditor.cs:9 ("The SOUND EDITOR section"). #443 deliberately left type names alone, which is right; these are different — they quote a label that no longer exists, so a future reader greps for a strip they cannot find. Cheap to sweep in whatever PR next touches each file.
5. No probe drives the rename ring or the channel-delete confirm. #443 flags this itself: RENAME CHANNEL, TYPE A CHANNEL NAME and Delete channel 'x'? are verified by reading the diff only — delete-confirm.json deletes a *song* and never reaches the RenameKind.Sound branch of BeginDelete. #442 builds a creation flow through that same ring and is the natural place to add the coverage.
None.
No comments.