The summary.
REQUIREMENTS BECOME CLEAR THROUGH USE
Which is why iterative approaches suit most software, and why sequential ones suit regulated or physically constrained work.
Never adopt iterative language while keeping sequential practice — it produces the costs of both and the benefits of neither.
PLAN TOWARD A GOAL, NOT A LIST OF FEATURES
Without a goal nobody can judge which features matter.
Prioritise by value, by risk reduced, and by what blocks other work. Building the uncertain part early reveals whether the approach works.
AN ESTIMATE IS A PREDICTION; A COMMITMENT IS A PROMISE
Confusing them causes most schedule disputes. Give ranges, state assumptions, and track actual against estimated — the only way estimates improve.
Include testing, review, deployment, documentation and coordination, which are frequently more than the coding.
ANSWER SCOPE CHANGES WITH WHAT THEY DISPLACE
"Yes, and that means this moves" converts an argument into a prioritisation decision.
MAINTENANCE IS THE MAJORITY OF A SYSTEM'S TOTAL COST
So optimising for initial delivery at the expense of maintainability is usually wrong. And most maintenance is change, not repair.
