Notes on engineering risk reviews, validation planning, design assurance, and the judgment that connects them.
A good specification does not choose the solution. It fixes what the solution must satisfy, within which limits, and how it will be judged. Concept work that starts without it is exploring in the dark.
Validation asks whether the product does what it was needed to do, in the conditions where it will be used. Practice quietly replaces that question with whether the product matches its specification, and the two can have different answers.
Most lessons learned records are written to close a project, not to help the next one. A lesson only teaches when it is written at the level of the principle, carried to the point of decision, and able to change a check.
A maintenance schedule is the list of tasks and dates. The strategy is the reasoning that produces it. Most organisations hold the schedule, assume it contains the strategy, and let both stay fixed while the asset ages.