# Desk run-through · teacher, learner and funder lenses

**Scope of claim:** This is a human-in-the-loop **desk simulation** on 29 September 2026, not an observed class, child usability study, state approval or evidence of outcomes. The [verifier](verify_pack.py) gives reproducible mechanical checks; the notes below capture editorial decisions and remaining risks.

| Lens | Run actually performed | Finding and repair/next action |
| --- | --- | --- |
| Teacher on a busy morning | Started at [README](README.md), traced Day 11 Card A/B and sorter, then Day 18 Card F/G/planner; checked that each material is reachable, the spoken model and sample responses are explicit, and 2+5+6+7+3+2 totals 25. Followed Days 15/20 to the sealed learner sheet and separate staff key. | The core is teachable with paper and voice; the teacher still needs a school-approved independent text, local plan and confidential evidence system. Day 15/20 cannot provide individual reading checks for a whole class in 12 minutes; rotate or schedule these separately. |
| Learner needing low-reading-pressure access | Followed Card E in its linear text order and clicked/read the named question anchor; used the text alternative to sort report/procedure/viewpoint/story without colour, then composed a summary with three tactile-card trays. Checked A/B/C route prompts for the same target and absence of a required purchase/device. | The adult-read versus self-read distinction is explicit. Braille, AAC vocabulary, image interpretation and actual accessible web code still require local learner testing. A child may change routes, work quietly and keep sensory/motor demands out of the English judgement. |
| Funder or curriculum reviewer | Matched the 15 used code tokens to the pinned official Year 4 English workbook rows; checked source claims, licence boundaries, original images/texts, hash manifest and the absence of invented class outcomes. Compared story/model claims against the cards and checked fresh H/I are unused in practice. | This is a reviewable **fortnight** with authored evidence and clear holds, not a claim of full year, a validated classroom effect or an inclusive translation catalogue. Public copy must retain that distinction. |

**Factual desk checks:** Card A consistently has 1 shelf, 2 low tables and 3 baskets. Card F top-view and side-view make different details salient, and neither gives scale. Check A's new box has 2 trays, 1 handle and 1 RETURN pocket; Check B's three-panel design has 2 front picture cards and no dimensions. The keys do not turn a drawing or conditional note into observed human outcomes. Word-part examples are analysed as meaning/reading opportunities; a teacher checks the child's taught phonology and morphology before treating an error as a learning gap.

**Next field test:** Ask a Year 4 teacher to time Days 11, 15 and 18 with real class constraints, and an access reviewer/learner to use the text/tactile and mock-page route. Record friction, revise, and keep consented private pupil data outside this repository.

**Original resource rights:** © NeuroForgeIO Pty Ltd 2026, SubjectNest, [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Credit, link and indicate changes.