This is a desk simulation, not classroom testing or a report of child outcomes. The author rehearsed the educator script, child-facing page route and evidence review against the actual files.
| Lens | Concrete rehearsal and finding | Adjustment or open gate |
|---|---|---|
| Educator, Day 12 | Put an oval card fully in a wide frame, then direct it partly under a fixed broad mask. The B page visibly shows whole versus partial outline. Timed six stages total 25 minutes. | Explain “hidden part” without implying the oval itself changed; cards must be broad/fixed. |
| Child route, Day 13 | Follow the two equal-shape panels and point to the broad open region. A child can choose a different open region rather than duplicate model. | Do not require words “negative space” or exact left/right; keep prompt concrete. |
| Educator, Day 16 | Compare printed dashes/dots, then try a school-approved broad dry tool route on fixed paper only if authorised. A paper proxy can be planned in the same time slot. | No classroom material, allergy or tool approval has occurred. Physical-process evidence stays open. |
| Learner, Check I | Read the page without key: both frames begin small and equal, so a larger second five-corner form must be child-created. The first remains unchanged. | A neutral scribe records the actual child decision, even if non-target. |
| Learner, Check J | The blank frame and separate whole token do not display a cropped answer. A child can choose any edge and leave a distinct clear area. | Assess crop and open region separately; do not grade beauty/neatness. |
| Assessor | Compare explicit code/evidence boundaries with QCAA: paper direction gives idea evidence; actual making, authorised artwork response and real audience need observed events. | Real maker source, material handling and actual audience remain held until school approval and observation. |
| Funder/public claim | Seven original practice pages, two check sheets and a separate key can be audited without paid resources. | No claim of measured learning gain, a full-year course, verified translation or real community endorsement. |
Visual print QA: A4 SVG/PDF pairs were rendered and inspected at page scale; see verification receipt at handoff. The diagrams must show the contrast independently of their labels. Exact text/tactile routes name layout and information in each panel. Human classroom rehearsal with educators/children, local safety approval, authorisation of real art, translation review and actual audience consent remain future work.
Asset maintenance: Rebuild original A4 aids with python3 generate_assets.py --write. Audit the pack without edits using python3 verify_pack.py. After an intentional source edit, inspect the changed materials, run python3 verify_pack.py --write-manifest, then run the read-only check again.