# Desk run-through · teacher, learner and reviewer

**Honest status:** A three-role desk simulation on 29 September 2026, not a classroom observation, user study, local-authority approval or outcome claim. The [verifier](verify_pack.py) provides reproducible checks. These notes record what was tried and what still needs human field use.

| Lens | Run performed | Finding and next move |
| --- | --- | --- |
| Teacher with a full class | Followed [README](README.md) into Day 11's 48×19 model, then Day 13's 128×23 mat and Day 16's invented $16×28 decision. Recomputed the products and Day 15/20 keys. Checked six phases sum to 25 and the staff answer file is separate from learner cards. | The core can start with paper, no device/purchase. Days 15/20 allow twelve minutes of response; some pupils will need a recorded continuation, and a teacher cannot hear every pupil individually within that station. Rotating checks and local scheduling remain necessary. |
| Learner using a different route | Read the [text equivalent](print/TEXT-ALTERNATIVES.md) in linear order, used two labelled trays for tens/ones, then selected the oral/tactile route on Day 13 and quiet private budget card on Day 16. Compared one estimate and exact value without needing colour or grip speed. | The number relationship remains visible in words and objects. Actual Braille, AAC vocabulary and contrast/size need testing with learners. No fixed “style” is assigned and no family income question is present. |
| Funder/curriculum reviewer | Traced five exact Year 5 code rows to the official pinned workbook, checked the public/private assessment split, original CC BY art and text, DejaVu rights and manifest. Sampled `247×18=4,446`, `213×14=2,982`, `178×24=4,272` and the stated bundle constraints independently. | This is a reviewable **fortnight**, not Year 5 coverage or validated learning impact. Public copy should preserve that scope and the remaining teacher/access/local-syllabus holds. |

**Arithmetic spot checks:** `48×19=912`; `128×23=2,944`; `247×18=4,446`; `28×$16=$448`, leaving $22 of $470; `24×36=32×27=864` but only 27 bundles fits a maximum of 30; fresh A `213×14=2,982`; fresh B `178×24=4,272`, leaving 228 to a 4,500-label limit. The code audit checks all 30 core numerical results and numeric optional tasks, not just these samples. No assessment number appears in practice cards or print aids.

**First field test to schedule:** a Year 5 teacher times Days 11, 15 and 19 with actual class transitions and captures anonymous usability notes; an access reviewer/learner tries both text/tactile mats; a local curriculum lead compares the pack with the school's adopted sequence. Any named response remains in school-approved systems.

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