Target/codes: Attribute and share separate source/trial accounts while bounding comparisons; AC9HSFS01, AC9HSFS02, AC9HSFS03, AC9HSFS04 and AC9HSFS05 partial. Actual route/codes: AC9HSFK03 conditional actual place/source account opportunity. Ready: P: F/E, second layout and separate log panel; R: first record and continued permission/willingness for optional second arrangement.
Child product: Two separate report/try panels or honest second-trial gap; source-labelled earlier/later account where genuine.
- Open · 3 min. Ask “Whose words are these? Were the people and steps the same?” P: names stay separate. R: renewed reader invitation/child choice; no retry assumed.
- Read source · 4 min. P: read F/E exactly; original question/conditions beyond words aren't recorded. R: confirm actual source/feature/care scope; a second try needs fresh willingness.
- Model · 5 min. P: place JORI/SIDE-BY-SIDE and MELL/STACKED on separate panels. R: show how time/people/layout/question/help are named separately; different conditions aren't controlled comparison.
- Child inquiry · 7 min. P: child represents F's reported question without inventing answer/consent, compares it to E and shares short source/place trail with one open question. R: record optional second actual response/pass under conditions; compare exact records without inferring a causal best layout or everyone. Keep actual particular-person meaning distinct from response to display.
- Check/respond · 4 min. A question isn't proof of all-user failure; a successful point isn't all-user benefit/fairness. If no second record, keep gap; don't invent it for comparison.
- Exit · 2 min. Save first account/help and gaps. Next move: improve one instruction against source, then ask whether reader wants another try.