Session 43: Capstone Building, Testing & Iteration

At a glance: teams turn their v0.1 skeleton into a working v1.0, run a silent user test with another team, record fixes in a Change Log and save an improved v1.1 ready for Demo Day. · Time: 90 minutes (adjust the timings if your period is shorter) · Workbook pages: 6

Learning outcomes

By the end of the session, pupils can:

Explain iteration using the clay potter at the wheel: build, user test, feedback, refine.
Run a silent user test, watching without explaining or defending (the Zip-Lip Rule).
Sort feedback into Fix Now, Upgrade and Ignore buckets, and accept critique calmly.
Record at least two fixes in a Change Log.
Save Capstone_v1.1_[ProjectName] in the Final_Showcase folder.

Before the session

Check each team's v0.1 file is in BIH_Practice > Capstone_[TeamName] > Drafts on the same PC they used in Session 42.
Pair up teams for testing in advance, ideally different tracks, and write the pairs on the board.
Have a visible clock or phone timer for the 25-minute build, 5-minute silent test and 20-minute refine.
Check internet if Track D teams need research, and have backup printed notes or fact sheets if it is down.
Open the Build the Whole House 3D lab on each team PC; it should remember their Session 42 charter on that computer.

Running order

Time	Activity	Workbook	What the mentor does
0–10 min	Typing warm-up	–	Pupils open the typing trainer (Typing in the menu, or the warm-up banner on the session page) and do the 10-minute warm-up. Walk round checking posture and home-row fingers.
10–18 min	The clay potter, the iteration cycle and vocabulary	Page 1	Go through the four vocabulary words.
18–28 min	Silent user test, red flags, feedback triage	Pages 2–3	Model the Zip-Lip Rule with a volunteer. Explain the three buckets.
28–48 min	Phase 1: build working prototype v1.0	Page 5	Check formulas, loops, citations and the 6x6 rule. Save as `Capstone_v1.0_[ProjectName]`.
48–58 min	Phases 2–3: the swap and peer evaluation	Pages 5–6	Swap PCs; 5 minutes of silence. Testers fill in the Page 6 form.
58–63 min	Triage the feedback	Pages 3–4	Teams sort the notes into the three buckets.
63–80 min	Phase 4: refine to v1.1 and fill the Change Log	Pages 4–5	At least two concrete improvements, written in the Change Log.
80–85 min	3D lab: record fixes and tick the polish checklist	–	Teams enter fixes and tick the polish checklist.
85–90 min	Phase 5: final save, readiness checklist, clearance	Pages 5–6	Check saves in `Final_Showcase`; complete the clearance.

Key ideas and how to explain them

Iteration: the potter spins, checks the thickness, smooths the cracks and trims the rim. Software gets better the same way, cycle after cycle.
You cannot test your own work: creators are biased. A confused tester proves the design needs to be clearer.
The Zip-Lip Rule: do not explain or defend. Watch where the tester hesitates and write it down.
The same four tracks as Session 42: Track A is a Word notice plus a 3-slide pitch, Track B a spreadsheet with SUM/AVERAGE plus a column chart comparing items (or a pie chart showing each item's share of the total), Track C a Scratch game, and Track D a researched report with citations plus a photo/slide guide. Tracks A and D test both of their files.
Feedback triage: Bucket 1 (Fix Now) is broken formulas (such as #VALUE!), crashes or tiny text; Bucket 2 (Upgrade) is a chart title, better contrast or punchier bullets; Bucket 3 (Ignore) is "add 10 more levels", "make it 3D" or personal colour gripes.
Change Log: never "we fixed some stuff". Write what was wrong and exactly how it was fixed, like reselecting the chart data so it covers only the item rows and leaves out the total row.

The 3D lab

Use Build the Whole House, continuing from Session 42:

"Step 3: record at least 2 fixes from user testing." Teams copy Fix 1 and Fix 2 from their Change Log. The roof appears when two fixes are recorded.
"Step 4: tick every item on the polish checklist." The lab's seven items are: spell check done on everything; every team member's name is on the title; formulas start with = (spreadsheets); pictures resized from the corner handles; sources cited (in a footer or last slide); no real private data, only fictional names and numbers; final file saved as …_v1.1. The workbook's Pre-Showcase Readiness Checklist (Page 5) has the same seven items. Teams should tick only what is truly done.

The lab saves progress on one computer only. If two teams share a PC, only one team's house is stored; let the other team use a different PC or rely on the paper workbook.

Common mistakes and quick fixes

Pupils often…	Try this
Talk the tester through the project	Remind them of the Zip-Lip Rule. Ask them to write notes instead.
Write "it was nice" on the peer form	Ask the tester for one specific thing that confused them, as the form asks.
Try to fix every suggestion	Fix Now first, then Upgrade. Ignore anything too big for one session.
See `#####` in Excel	That is not an error: the column is too narrow, so widen it. If totals are wrong, check formulas start with `=`, such as `=B2*C2` and `=SUM`.
See `#VALUE!` in Excel	A formula is using a cell with text in it. Check the Qty and Price cells hold only numbers (no "GH₵" typed in).
Save v1.1 over v1.0 in Drafts	Use Save As, rename to `Capstone_v1.1_[ProjectName]` and choose `Final_Showcase`.

When things go wrong

No internet: only research and citations need it. Track D teams (and any Track A team checking facts) can use printed notes and add citations later; the rest of the session is offline.
Not enough PCs: test in turns while other teams keep building.
A PC freezes or the power goes: this is why there are versions. Reopen the latest saved version and carry on. Save every 10 minutes.
A team's file is missing: search the PC for "Capstone" and check Downloads.

Answer key

This session has no questions with fixed answers. Use this guidance for the written parts.

Change Log (Page 4), Fix 1 and Fix 2: Look for: a specific problem the tester found (for example, "the total was in the chart", "the score didn't reset", "the heading was too small", "a fact in the report had no source") and a specific fix ("reselected data", "added set [Score] to (0)", "made the heading 20 pt"). "We fixed it" is not enough.

Peer Testing Evaluation Form (Page 6): Look for: an honest tick in each row, one real piece of praise and one specific, constructive suggestion. Ask testers to rewrite any comment that is unkind or vague.

Pre-Showcase Readiness Checklist (Page 5): each item should be checked on screen, not just ticked. Not every item applies to every track: for example, a Scratch game may have no formulas. Mark "not needed" rather than ticking falsely.

Support and stretch

Needs more support: cut v1.0 back to the minimum (for example, 3 sheet items, or just the loop and score). Sit with them for the triage.
Finished early: make a third Change Log fix, test with a second team, or start writing who says what in the 3-minute pitch for Session 44.

Evidence for sign-off

A v1.1 file in Final_Showcase that opens and works without crashing.
A Change Log with at least two specific fixes linked to the peer feedback.
A completed peer evaluation form from the testing team.
Tick the Milestone 2 boxes only when all of the above are true; otherwise give a clear to-do list.

Safety and wellbeing

Feedback must be kind and specific. Stop any mocking straight away. Check again that projects use only fictional names, prices and data, with no real phone numbers or addresses. Testers must not change or delete the other team's files. Photos of people need their permission.