·timeline
First prototype reads a paper end-to-end
Used the prototype to read Karpicke & Blunt 2011 cover-to-cover. It survived. The gesture set didn't.
Used the rough prototype to read Karpicke & Blunt (2011) end-to-end. The software survived — nothing crashed, all marks were preserved on re-open. Three discoveries:
- The double-tap-to-note gesture conflicts with the system’s own text-selection. Will need a long-press, or a held modifier.
- I made seventeen marks in a 12-page paper. Three turned into notes. This ratio feels right; the others were the kind of breadcrumb that doesn’t deserve prose.
- The “promotion” ritual (Notes vs. marks) is the right thing — it felt like a small, satisfying gesture.
Next: redo the gesture set in Gesture set v2 and re-read.
Pointed at by
Gesture set v2 — a note — “…er the first prototype. --- After the lessons in first-prototype-end-to-end, here's the next attempt at the ge……” read