Your CBA build or portfolio is well under way. Today is a checkpoint: you document what you have made so someone else can understand and judge it, note honestly where AI tools helped, check any ethical or legal issues, and package your evidence cleanly. You are not finishing the whole CBA today. You are making the work ready to be read, and ready for a partner to open and try in about 30 seconds.
Must-finish today: a short overview, your plan in the package, one test note, an AI note, an ethical note, and one package that opens in a single try and is saved to the class location. Extra polish (full renaming, multi-cycle tests, partner open-test detail) comes if time allows.
This is a single-period checkpoint on ongoing CBA work, not a build lesson. Students should already have a project or portfolio in progress from the planning and building sessions. Lead with their live project (open and run), then frame every doc task as unlocking the next presentable moment: the overview exists so a partner can open the work cold; the package exists so a marker can judge it fairly.
Have the class AI ground rules (from the earlier AI ethics lesson) visible, and a student-friendly version of the Features of Quality on the board or shared screen. Display the suggested package layout (or your adapted version) once at the start. Keep must-do versus nice-to-do on a progress board so on-screen student text can stay short.
If anyone is missing their work, use the class recovery route: open the last saved version from the class folder, or rebuild the missing piece from their plan. Do not assume browser storage survived on shared devices.
A judge or classmate who never saw you build this should still be able to follow it. Good documentation means:
- Clear names for files, pages, sprites, variables and functions so a reader can tell what each one is for
- Comments or short notes on the tricky parts: what a section does and why you chose that approach
- A short written overview: what the work is, who it is for, and how to run or open it
- Honest AI notes: where a tool helped (drafting, explaining, suggesting), and what you changed or checked yourself
- Evidence of process: plan, tests, feedback you acted on, and any ethical or legal point you considered
Good enough looks like this:
- Someone else can open the main work in one try
- AI help is named, with what you changed or checked
- At least one test is noted, plus what you fixed
- One ethical or legal point is written (or a clear reason none major applies)
Example overview: "This is a MakeCode Arcade sorting game for first years learning recycling. Open recycling_game in MakeCode Arcade and select Run. Arrow keys move the sorter; score rises when plastic hits the right bin."
Example AI note: "ChatGPT suggested a fix for my sprite overlap bug. I tested the change, kept only the overlap block, and rewrote the comment in my own words."
You will also work against the Features of Quality checklist your teacher shares. Package means one clear folder or portfolio layout a marker can open without guessing. Use the class package layout shown on the board (or your teacher's adapted version of it).
Keep this tight. Put a short Features of Quality checklist on the board in plain language, for example: the work can be understood; decisions are explained; testing is evidenced; AI use is noted; ethical issues are considered. Stress that documentation is part of the assessed work, not an add-on after the fun bit. The four "good enough" bullets and the two written examples give a visible bar even if the slide is late.
Show the default package layout once now so students know where the hands-on is heading. If you can, flash a 20-second before/after of a messy folder versus a clean 01–04 package so the payoff of packaging is visible, not abstract.
Common misconception: students think more comments always means better docs. Prefer clear structure and a few well-placed notes over walls of text.
Open your CBA project or portfolio from where you last saved it. Keep the class package layout on the board in view (01_Overview, 02_Project, 03_Process, 04_AI_and_ethics, or your teacher's adapted version). You will move your files into that layout as you work today.
Must finish this block: overview, plan in the package, one test note. (AI note, ethical note and final package come next.)
1. Open and run (about 2 minutes). Run what you have once. Start from the thing you made, then document it. Before you run: what do you expect still works, and what might break?
2. Overview so a partner can open it. Write a few plain-English sentences: what you made or collected, who it is for, and how to open or run it. Read it aloud to a neighbour in about 30 seconds. They say one clear thing and one confusing thing. Fix the confusing line so they could open it cold.
3. Process evidence. Put your plan in the package. Note one test in this shape: Test: I tried ___. Result: ___. I changed ___.
If time: clearer names, comments on tricky parts, one feedback note.
Done: project run once, overview shared once, plan and one test note in the package.
Before the work
Students open from last save, not from a finished package layout. Show the class package layout once (default below, or your adapted version) as the target they will fill during the lesson. Circulate with the Features of Quality checklist. Put a simple progress board on the wall or shared screen with must-do items marked clearly: Overview (must) / Plan (must) / Test note (must) / Names (if time) / Comments (if time) / Feedback (if time) / AI note (must, next block) / Ethical note (must, next block) / Package saved (must, next block).
Default package layout teachers can adopt or adapt:
- Software project: /Surname_Firstname_CBA/ then 01_Overview.txt, 02_Project/ (main file or link at top), 03_Process/ (plan, tests, feedback), 04_AI_and_ethics.txt
- Portfolio: same four sections as pages or clearly named files in that order
- Main entry must open in one click from the overview
Predict beat (PRIMM, keep short)
Before they run their own work, ask once: "Before you run it, what do you expect still works, and what might break?" Twenty seconds only. This ties the checkpoint back to the live build without becoming a coding lesson.
Sentence starters (display or hand out for support)
- Overview: "I made ___ for ___. To open it, ___." / "The main thing a marker should try first is ___."
Teaching tips
Frame the overview as the key that lets a partner open and try the work in 30 seconds, not as admin. After the overview, run the fixed neighbour share (about 1 minute total). Teams on the software-project approach should agree who writes which section so nothing is duplicated or missing. Portfolio students need curation notes that show growth across strands, not a dump of every file.
Watch for vague overviews and missing test evidence. If someone is behind on the build, prioritise overview, plan and the one test note for what exists. Do not invent finished features.
Differentiation
Support: use the overview sentence starters above. Stretch: a short testing log that shows more than one test cycle.
Partner open-test (about 2 minutes): swap. Your partner tries to open your main entry from your overview in about 30 seconds. Before they try, predict: will they find it in one go? They name one thing that is clear and one thing that still confuses them. Fix the confusing part, then continue.
Must-do next:
- AI note. List any AI help (drafting text, explaining an error, suggesting a design). For each, say what you checked or rewrote yourself. If you used none, write that clearly. Tell a neighbour one part you can explain fully in your own words.
- Ethical note. Note one issue that applies (privacy of data or photos, ownership of images or code, fairness, safety online) and what you did about it. If nothing major applies, say so and why.
- Package and save. Move everything into the class package layout (01_Overview, 02_Project, 03_Process, 04_AI_and_ethics, or your teacher's adapted version). Check you can open the main entry in one try. Save to the class shared location and confirm it is there.
Done: AI note written, ethical note written, one clear package that opens cleanly, save confirmed in the class location.
Mid-point beat (fixed, not optional)
Start this block with the 3-minute paired open-test described in the student copy. Keep energy high: this is the payoff for the overview they just wrote. Before the partner opens, prompt the owner with a quick predict: "Will they find the main entry in one try? Why or why not?" Update the progress board so stalled students are visible early. Keep feedback to one strength and one fix so time stays on packaging.
Park nice-to-do items only on the progress board (full Features of Quality tick-off, extra renaming or comments, multi-cycle test log). Do not load them into the student step body.
Sentence starters (display or hand out for support)
- AI note: "AI helped me with ___. I checked or rewrote ___ myself." / "I did not use AI tools on this work."
- Ethical note: "One issue I considered was ___. I handled it by ___." / "No major ethical or legal issue applies because ___."
Teaching tips
Mirror must-do versus nice-to-do on the progress board so "done" is obvious under time pressure. After the AI note, prompt the 20-second verbal "I can explain this bit" beat to a neighbour so honesty stays tied to the live project.
Watch for copy-pasted AI text left unchecked and packages that only work on the author's machine path. End of this block: every student should have a saved, openable package with AI and ethical notes. Note anyone still missing core evidence before the presenting session.
Differentiation
If someone is behind on the build, prioritise AI note, ethical note and a clean package of what exists. Stretch: a one-page "how to judge this" guide for the marker.
Documentation is how your thinking becomes visible. Clear names, a short overview, process evidence, an honest AI note and a clean package mean a marker can judge the work you actually did. Packaging is not tidying for its own sake: it is part of making the CBA fair to assess.
Pull the class together for a brief whole-class beat. Invite two quick examples: one strong AI note phrasing and one ethical issue someone handled well. Keep names positive; do not single out weak packages publicly.
Unlock the full learning experience
You're previewing this lesson. Get full access to this lesson and hundreds more — each one ready to teach, with interactive activities, printable resources and pupil progress tracking built in.
This lesson is copyright of Coding Ireland 2017 - 2025. Unauthorised use, copying or distribution is not allowed.