This is a desk simulation of the supplied material, not a classroom pilot. It tests whether a new teacher, child, access-support adult and evidence-conscious funder can follow the pack. No child was observed.
Teacher with 25 minutes and no animal available
Read materials, do the adult gate, find no approved animal source and record REAL OBSERVATION NOT AVAILABLE. Use N1–N4 and the four figures, sort, rule board and scripts. The six steps on each daily page total 25 minutes. Keep the actual log blank rather than converting paper figures into field data. This is a complete reasoning lesson, but does not complete actual biological observation. Improvement made during review: the gate and fallback are repeated in the index, Day 143 and crosswalk so the teacher can find the boundary at the decision point.
Child who notices that a point is a point
The child places N2 and N3 together by a visible pointed outline, then says one is a plant and one is an animal. The teacher reads “plant drawing” and “animal puppet drawing” as the invented author's labels, points out leaf and ear are differently named parts, and asks what the shape rule really shows. The correct model revision is smaller, not a guess about species. Day 145/150 use unseen new edge/spot cases, so rehearsing N2's answer does not supply a check answer.
Access-support adult using no touch
Read the exact linear descriptions, offer large text, speech, sign, AAC, gaze or adult scribing. On Day 147 say “lower view not supplied”; never make the child feel an unfamiliar plant or handle an animal. A tactile aid is a secured paper diagram with the same evidence limit. A real photo described only by an adult is logged ADULT REPORT for that child. The accepted target is the child's feature/source reasoning, not hand movement, sight or speech.
Evidence-conscious partner/funder
Open the crosswalk: exact seven ACARA source rows and QCAA Prep placement are separately cited. Open source/rights: all packaged visual/model content is original; live/photo sources are teacher-supplied with rights/provenance; audio is transcript-only. Open assessment: two formative model checks and a distinct key, no measured gains. The read-only verifier checks structure, code/source hash, links, A4 geometry/text and asset hashes. It cannot prove classroom effectiveness or accessibility in every environment.
Remaining human review: educator and local syllabus judgment; safe source selection; photographic rights; visual/tactile/AAC use with learners; cultural consultation if future local content is added; recording/translation review if audio or languages are produced. No pilot or endorsement is represented.
Print and verifier desk receipt
All six PDF pages were rasterised and visually inspected at A4 scale. The first notebook render let plant strokes touch the label line; the geometry was revised, the six pairs regenerated and the first figure inspected again with its labels clear. The sort visibly begins unsorted, the four-source gate has four empty labelled panels, the two-pair log has no observation prefilled, the revision board has three blank stages, and the share strip leaves the listener field blank. PDF text extraction and SVG title/description checks supplement visual review; they do not replace learner accessibility testing. The manifest excludes temporary visual previews and distinguishes the separate DejaVu font notice from original assets.