The significance of the first question has been expressed well by Agile methods pioneer Kent Beck in his book Extreme Programming Explained [Beck 1999]:
Figure 10-1 is “the technical premise for XP” (eXtreme Programming). If the cost of making a software change “rose slowly over time...you would make decisions as late in the process as possible...and...only implement what you need to.”
However, “if a flattened cost curve makes XP possible, a steep change cost curve makes XP impossible.”
Figure 10-1. The cost of change may not rise dramatically over time
We can therefore ask whether XP or other Agile methods can ensure a flattened cost-of-change curve. We can also ask what other strategies can flatten the curve—so that late fixes do not threaten a project’s costs and delivery dates—as well as how to reduce fixes for projects that have a steep cost-to-fix curve. This chapter will provide some answers to these questions.