typewriter/docs/spikes.md

Rendering & UX spikes

Scope: bench experiments that retire display/UX risk, run after the v0.1 hardware bring-up spikes (1–7) in v0.1-mvp-technical.md. Unlike 1–7, these are not a v0.1 integration gate, and each feeds a different release (noted per spike).

They ride on the epd::Frame DrawTarget and the Spike 5 partial-refresh path. Project overview: ../README.md. Vocabulary: ../CONTEXT.md. Release sequence: macroplan.md.

These prove display/UX risks, not stack risks, so they sit outside the 1–7 "prove before integration" gate. Run Spike 8 first: it partitions the panel into the writing column and the side panel that the later spikes (scrollbar, line-number gutter, status readouts) draw within and around. Each feeds a different release, but the bench experiment is worth running now to retire the risk early.

Spike	Feeds
8 — Layout: side panel + writing column	v0.1
9 — Boot splash bitmap	v0.1
10 — Dark / light theme	v1.0
11 — Transient panel (help / config)	v0.4 : · v0.5 palette · v0.9 settings (mechanism)
12 — Scroll position indicator	reading / View-mode UX
13 — Line-number gutter	v0.2
14 — Multi-file navigation	v0.5

8. Spike 8 — Layout: side panel + writing column. Split the 792×272 landscape into a full-height writing column (~60 cols, left) and an always-visible side panel (~150 px / ~20 cols, right), per the Screen layout. Today the editor fills the full 79-col width with text and a cramped 12 px bottom status (firmware/src/editor.rs, COLS/ROWS + draw_status); this spike narrows the text to a legible measure and moves all metadata into the panel, retiring the header/status bands entirely. Two driver facts shape it. First, narrowing does not speed up typing: update_part already drives both controllers full width, windowed only in Y, because "the waveform time dominates, not the data clock-out" (firmware/src/epd.rs) — the win is line length and persistent info, not latency. Second, since every partial refresh spans the full width, a keystroke's windowed-Y band repaints the panel's pixels on that row for free (redrawn identically), but a panel field that changes on a different row than the cursor costs a second windowed-Y band — burning the "20 partials → forced full refresh" ghosting counter twice as fast (render module). That sorts what the panel may hold: static (filename, dirty), event-driven (mode, Wi-Fi, keyboard-disconnect, publish state), and throttled (clock, word count, session) fields only — nothing that repaints per keystroke, so no live cursor column. The panel sits entirely in the master half (right of the x = 396 seam), so its glyphs never split the seam; the writing column still straddles it, as today. Prove: render column + panel; type and confirm the windowed-Y refresh stays a single band (panel redraws in place, no second region); then update a throttled field (word count on a typing pause) and a discrete one (mode flip) and confirm each is one isolated band that forces no extra full refresh. Decides the column/panel split in pixels, the panel's field layout and update cadence, and whether the ghosting counter needs per-region accounting. (Feeds v0.1's screen layout — the writing column + side panel the product doc now draws.)

9. Spike 9 — Boot splash bitmap. Embed the Typoena logo as a 1-bit asset and blit it, with the build tag, at boot before the editor opens. Spike 2 already proved vector + font rendering, so the only new risk is the image-asset pipeline (embedding + blit) and that a large near-full-frame image takes a clean full refresh. Mostly a feature — kept as a spike only to prove the asset path. (Feeds v0.1's "e-ink shows Typoena splash + boot log".)

   Built 2026-07-11 as a vector splash, not a bitmap. The frame is display::Frame::splash — the typoena wordmark centred inside a stroked Circle, drawn with embedded-graphics, one clean full refresh. It is shared by two callers: the splash bench binary (just flash-splash) and main.rs's boot path, which shows it right after EPD init — replacing the old white-clear baseline — while the SD mounts and the note loads, then a second full refresh brings up the editor. Nothing is embedded, so the image-asset pipeline named above is deliberately left unproven — an acceptable trade because Spike 2 already covered vector + font rendering. Proposition, deferred to end-of-project polish (v1.0): replace the vector mark with an embedded 1-bit raster logo, which is when the asset-embed/blit path would finally be exercised. Confirmed on the panel 2026-07-11 — the splash renders cleanly at boot, then the editor comes up. This closed the last v0.1 display gate; v0.1 shipped the same day.

10. Spike 10 — Dark / light theme. Invert the 1-bit framebuffer (white on black) and refresh. The invert is trivial (XOR at blit); the real unknowns are (a) ghosting and refresh time on a predominantly-black panel — full-black waveforms stress the panel differently — and (b) whether partial-refresh ghosting accumulates faster in dark mode, forcing a more frequent full refresh. Prove: legible inverted text, acceptable ghosting, and partial-refresh latency unchanged from Spike 5. (Feeds v1.0's light/dark theme.)

11. Spike 11 — Transient panel (help / config). Prove a modal full-screen view that swaps in over the editor and restores the buffer exactly on exit — the shared primitive behind a help screen, the v0.9 settings screen, the v0.5 file palette (Ctrl-P), and the v0.4 : command line. The risk is e-ink-specific: entering/leaving with partial refresh leaves ghosting, and the editor viewport underneath must return artifact-free. Prove with a static help screen (trivial content): a full refresh on enter and on exit restores the prior text exactly; measure in/out latency. Decides overlay-box vs. full-screen swap — recommend the swap, since a dimmed partial overlay is both slow and ghost-prone at ~630 ms. Panel content (which config keys, help text) is feature work deferred to the owning release; this spike proves only the mechanism.

12. Spike 12 — Scroll position indicator. The buffer already scrolls (View mode j/k/space); this spike proves how to show position without wrecking latency. Measure two affordances on the bench: (a) a right-edge scrollbar repainted each scroll step, and (b) a compact side-panel readout (34/128 or 27%). Risk: the scrollbar is a tall pixel column repainted on every scroll — measure its partial-refresh cost against the small side-panel field, on the windowed-Y path, during held-key scroll. Prove: pick the affordance whose refresh cost is acceptable and confirm it forces no extra full refreshes. (Feeds the reading/View-mode UX; the a-vs-b choice is a design decision this spike hands data to.)

13. Spike 13 — Line-number gutter. Draw a fixed-width digit column left of the text area. Absolute numbering only — relative numbering was dropped (2026-07-11). It renumbered the entire visible gutter on every j/k, so a single cursor move became a partial refresh of a tall digit column, straight into the Spike 5 windowed-Y path and the "20 partials → forced full refresh" ghosting counter (firmware/src/epd.rs, render module) — real e-ink cost for no proportionate gain on a distraction-first panel. Dropping it removes the genuine e-ink risk this spike existed to prove: absolute numbering is cheap — only the rows below an inserted/deleted line renumber. What remains is a layout decision, not a latency one: the gutter steals horizontal columns from the render-time soft-wrap, and wrapped continuation rows have no number (blank vs. tilde). Prove: reserve the gutter width, then confirm a single-line edit repaints only the rows at/below the change and forces no extra full refresh. (Feeds v0.2's line-number gutter; now low-risk after the relative drop.) VERIFIED on the panel 2026-07-11 — a single-line edit repaints only the rows at/below the change, no extra full refresh; closes the last v0.2 gate.

14. Spike 14 — Multi-file navigation (open / switch / new / delete). The panel mechanism is already Spike 11 (which names this v0.5 file palette), and filename entry reuses the v0.4 : command line — so this spike proves neither, and depends on both. What is genuinely new is persistence and buffer lifecycle, not rendering: enumerate a FAT directory across both /sd/repo and /sd/local, keep ≤ 3 ropes resident, and on switch swap the active rope + cursor + soft-wrap after an atomic save-of-current (persistence module). Two sharp edges. New file (:enew) must prompt for scope — tracked (/sd/repo, pushed) vs. local (/sd/local, Ctrl-G disabled). Delete extends v0.5 scope (the roadmap lists :enew but not delete): removing the currently-open buffer must close it and fall back to another resident buffer or an empty one, and — the load-bearing check — the FAT unlink must reach the next Publish's staged set, which the git module's git add .-equivalent staging may not catch for removals (needs git rm / add -A semantics — git module). Prove: open/switch/new/delete across both roots with a correct ≤ 3-buffer lifecycle, a delete that propagates into the following commit, and an artifact-free full-refresh swap on every buffer change (leaning on Spike 11). Latency is not the risk here; buffer-state and git-staging correctness are. (Feeds v0.5 multi-file, delete included — now authorized in macroplan v0.5.)