Absorbs the small/related items from next-plan-carry-forwards.md that didn't fit Plans 7–9: log-level adjustments, prose spelling, doc cross-references, test floor assertions, and a test-update-count trim. Leaves sketch-specific items (Flame/Dots/Cymatics/Waves) and pre-release distribution work in the doc for their respective plans. Items landed: - hellochar#13: `info!` → `debug!` on `restart_on_settings_change` trampoline log calls (noise in release once the cycle proved stable). - hellochar#20: stray "vetos" → "vetoes" in `idle.rs` doc-comment prose (field is `vetoes`). - hellochar#34: `LineSettings` module doc now calls out the deliberate absence of `#[serde(deny_unknown_fields)]` for forward-compat with TOML written by older v5 builds (`drag` / `attractor_radius` keys still deserialize cleanly). - hellochar#35: lifecycle test floor tightened from `> 0.0` to `>= 0.1` (the documented `#[setting]` min). Catches a future typo that drops the default below the floor. - hellochar#42: `MouseAttractorState.power` field doc cross-references `line_idle_veto` (which reads it to keep the sketch `Active` during decay). - hellochar#49: `enter_line()` reduced from 4 to 3 `app.update()` calls (the fourth was headroom; three reliably cover fold + leafwing tick + nav handler + OnEnter). Items confirmed already done in earlier phases (no-op here): - hellochar#14: structured `node = "LineComputeNode"` trace tags shipped in Plan 8 Phase 0. - hellochar#28: `MAX_ATTRACTORS > 16` TODO note already at `particle.rs:56`. - hellochar#43: prerequisite comment in `update_sim_params_does_not_run_when_idle` already sits immediately before `let dt_before`. - hellochar#50: `EXPECTED_POST_DECAY_POWER` already derives from the production `MOUSE_POWER_PRESS` const. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

add video of sketches when webgl won't work #34

Description

Activity

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions