Start with a Scratch flappy bird — English carousel cover for this classroom case

Parents ask it often: what if kids copy a tutorial and still don't get it? In our Markham classroom the honest answer is simple — rebuild the why before you change a block. A working copy is not the same as owning the steps. Four questions on paper beat one more blind edit.

A real classroom case: Scratch flappy bird that flies with no why

Last month at afterschool in Markham we ran a Scratch flappy-bird tutorial. The bird flew. Then a classmate could not change flap strength or say why it falls. We paused and wrote four questions on paper before touching another block.

Once those four answers are clear, the copy stops being the finish line. The typing (or the clicking) is the fast part. Our 2026-09-24 Xiaohongshu note used this exact classroom beat: rebuild the why first, then change one number — do not treat a working copy as owning the steps.

Why this still matters when AI can write code

Claims below are grounded in that teaching note — not an external product case with published accuracy metrics.

1. AI does not remove the need to rebuild the why. “Change flap so a classmate feels a clear difference, and say why it falls” is a different job from “make any bird look like the tutorial.” If a kid (or an AI prompt) muddles those, the next edit that comes back solves the wrong problem.
2. The useful state is tiny on purpose. You only hold each key block's job → one number change → predict → play. That is abstraction: keep what the next edit needs, drop the rest.
3. The done rule and edge cases are the algorithm. The same four questions transfer to a Lego booklet then a taller tower from memory, baking from a recipe then adjusting sugar, or desk assembly then explaining steps to a sibling. Contest-style questions inspired by CCC Junior, CCC, BCC, and school contests reuse those moves — kids who want to sit those register on their own; CodeSky is not an official contest program and is not affiliated with Waterloo or CEMC.
4. Languages are tools on a path, not the answer. Early work often uses Python so the steps stay readable; JavaScript when the same idea becomes something clickable on a screen; C++ later when speed and structure matter. Tools change; rebuild-the-why does not.

So AI’s ability to type code does not cancel a thinking-first start. It raises the value of kids who can name each block's job, change one number with a prediction, and check the edge cases before they ask a model to type — and before they copy the tutorial again.

Three steps kids can name

We practice the same shape on the bench:

1 · Write — four questions before changing a Scratch block

2 · Transfer — same rebuild-the-why to Lego, baking, desk teach-back

3 · Own — change one number before copying again

• Decomposition — goal → state → done rule → special cases
• Abstraction — keep only each key block's job → one number → predict → play
• Algorithms — name jobs → change one → predict → play → handle missing block / duplicate names / broken flight / “just copy again”
• Pattern recognition — same four questions in Lego rebuilds, recipe sugar tweaks, desk teach-backs, contest-style I/O

Follow CodeSky

Instagram @codeskyca  ·  小红书 / RedNote  ·  WeChat 公众号: search codeskyca

Sources: CodeSky Xiaohongshu note pack 2026-09-24-孩子照着教程做完了却还不懂怎么办 (Scratch flappy-bird tutorial case; Python → JavaScript → C++ pathway; contest-style disclaimer).