# Author desk run-through · teacher, learner and reviewer

**What happened:** On 30 September 2026 I simulated teaching from all ten scripts, responded to both fresh checks before comparing with their key, read every source against the printed aids, rendered eight A4 PDFs, inspected their combined preview, checked Markdown list semantics and ran the local verifier. This is an **author desk review**, not a child/classroom trial, external educator or cultural review, accessibility certification, safety result or funder validation.

## Teacher desk pass

- **Preparation:** a teacher can read A–E from [Materials](MATERIALS.md) or print four source pages and two broad mats. Ten scripts use 3/4/5/7/4/2-minute phases, with a source, model, child response, corrective move and exit. The optional HASS blocks need placement in the school's actual timetable.
- **Day 31 rehearsal:** Source A answers what the invitation plans. It cannot answer whether anyone came or the show occurred. The “future statement versus event report” contrast is explicit enough to teach with the source page in hand.
- **Day 33 rehearsal:** A and B both belong to Story Day 1; C is an attributed Story Day 2 note. The teacher should not force an order between A and B within Day 1 or say the class directly observed the show. The planned table box remains unknown in C.
- **Day 38 rehearsal:** Tao's view can inform “could consider a large card on table” as a fictional proposal. Source B planned an empty box, not a displayed card; C reports wall panel only. No physical arrangement or effect was tested.
- **Time/access risk:** Day 35 and 40 each have four short tasks within seven independent-response minutes. A teacher may split a check across another session, mark prompted/independent parts accurately, and avoid interpreting speed as HASS attainment. For a child's AAC, visual or tactile system, use the [full linear source](print/TEXT-ALTERNATIVES.md) and locally reviewed supports.

## Learner desk pass

- **Day 35 first-response simulation:** Sources 1/2 are Day 1 invitation/plan; Source 3 is Sol's Day 2 report of **one triangle card on blue rail**. The table box is only in the plan, not in Sol's note. A valid unknown is whether the box appeared. The [separate public key](teacher/KEY-AND-NEXT.md) agrees and does not silently infer attendance.
- **Day 40 first-response simulation:** Source 1 draws a wide table left and label strip right, no box. Source 2 says a box *could* be added. Umi likes the strip; Dex wants a box at table. Message A is a careful proposal, while B invents a completed box and benefit to everyone. The public key gives the same distinctions.
- **Routes:** all ten cards have three choices reaching the same day target. A child can change routes mid-lesson; no route requires personal disclosure, a device, small cutting, a real event visit, music performance or family purchase. Ten optional home routes use invented paper/toy situations.

## Reviewer/funder desk pass

- [The pinned crosswalk](CURRICULUM-CROSSWALK.md) reproduces all nine ACARA v9 Foundation HASS rows and Queensland Prep context. K03/S01–S05 are partial opportunities; K01, K02 and K04 remain open. The fictional paper-card show cannot complete K02, and original marks carry no First Nations cultural meaning.
- [Source/rights ledger](SOURCE-AND-RIGHTS.md) distinguishes original CC BY 4.0 text/media, Apache-2.0 code, the separately licensed font, and official curriculum source rights. Real-place belonging, celebration, Country/Place, safe-sign action, Arts pattern understanding and physical benefit all need evidence outside this pack.
- Teacher keys are public by URL, so the checks are fresh compared with lessons but **not secure exams**. An implementer would need educator/child feedback, authentic source permissions and local cultural authority before a coverage or impact claim.

## Render and verification receipt

- Eight original SVG/PDF pairs are one A4 page each with high-contrast broad regions, SVG title/description and selectable PDF text. The PDFs are untagged, with [full text and tactile-equivalent directions](print/TEXT-ALTERNATIVES.md). All eight were rasterised and visually inspected together: no visible clipping, overlap, missing source block or unreadable hierarchy in the rendered A4 preview.
- MarkdownIt parsed learner pages as **10 bullet lists / 30 items** and checks as **2 ordered lists / 8 items**. The two public keys were read only after independently answering the corresponding source cases.
- The [read-only default verifier](verify_pack.py) checks exact source rows/hashes, timings, route and swap counts, held-case leakage, PDF/source consistency, local links and a deterministic SHA receipt. I regenerated the eight PDFs twice with fixed build metadata and confirmed the receipt still passed. A passing receipt is authoring evidence only.

## Next live reviews

1. Ask a Prep educator and children to trial the source sequence and both check lengths; record what they actually understood and where prompts were needed.
2. Review AAC, tactile and visual use with the relevant learners and specialists; adapt routes without assigning fixed learning styles.
3. Obtain school-approved familiar-place/family/event sources and locally authoritative First Nations cultural review before teaching or claiming K01/K02/K03/K04 authentic content. Keep actual child evidence private.