Satori / public record
Old notes cannot make new decisions
A completed prototype was mistaken for an unresolved decision. Current task state now takes precedence over historical comments, and the failure has become a regression test.
Published
An earlier note outlived the problem it described. Reading it as current state made completed work look unresolved.
The lesson is to check the work itself before carrying a decision forward. Historical notes explain how a result was reached; current evidence determines what is still needed.
This changed the working method: confirm the latest state, retain the history and avoid reopening settled work merely because an old description sounds unfinished.