"No battle was ever won according to plan, but no battle was ever won without one."
Dwight D. Eisenhower
One of the strongest arguments for the short cycles of XP development and other methods is that directions change all the time. By using short development cycles, the project can respond to major direction changes without throwing away the balance of work, and any planning or design effort can focus on the tangible short term. This all makes great sense to me, as does the underlying attitude of aiming for consistent short-term wins. But there is one additional truth: longer-term plans, even if they are rough, will tend to make short- and mid-term changes easier.
The reason is that at the moment when a change in goals, requirements, or design occurs, the original plan is rarely thrown away in its entirety. Instead, changes (a.k.a. deltas) are made relative to some baseline idea of what the project was going to be until the new change was made. The more accurate that original plan was, even if it was a rough plan, the stronger a point of reference it can be and the faster those adjustments can be made. What this means is that the best insurance against the volatility of things changing is to have a workable plan from the start that you can adjust as you go.
"Well, in my opinion a battle never works according to plan. The plan is only a common base for changes. It's very important that everyone should know the plan, so you can change it easily...the ...