The brief kept changing.
At first I noticed it as a project-management problem. A mechanic would work, then one of the kids would decide the vacuum needed to move differently, the cat needed a sound, or the entire point of the game had quietly changed since breakfast.
Then I realized they were doing something adults are trained out of doing.
They were protecting the feeling before protecting the plan.
They had no attachment to the scope because the scope was never the thing they cared about. They cared about whether the game felt funny, surprising, difficult, fair, or boring.
Scope is a compression.
Adults need briefs because coordination is expensive. Scope turns an intention into something teams can estimate and execute. That is useful. But once the compressed version exists, it becomes easy to mistake it for the original idea.
Children have not learned that confusion yet.
The danger starts when the brief becomes more authoritative than the thing it was trying to describe.
Fast iteration changes the power balance.
AI-assisted coding makes this especially interesting. When a revision takes minutes instead of days, the cost of listening drops. A child can say, “No, it should feel more slippery,” and we can test the sentence instead of debating it.
The game becomes part of the conversation.
The lesson travels.
This is an argument for remembering what scope is for. I try to keep the reason attached to the rule, and to notice when reality is telling me the plan stopped representing the intention.
Sometimes the most sophisticated thing you can do is let the player tell you that your beautiful mechanic is boring.