From Shack to Skyscraper

So what is extreme in XP? It is the assumption that analysis and design can be done incrementally; the assumption that a complex system can be grown incrementally with hardly any up-front work. XP detractors liken it to "building a skyscraper out of a shack."

This assumption is in no way trivial or obvious. Where this assumption does not hold, we will not be able to apply XP successfully.

Which XP practices depend on the incremental assumption?

• The planning game grows specifications story by story, expecting each story to deliver business value.

• Simple design solves today's problems, assuming that we will be able to solve tomorrow's problems when they arise.

• Small releases assumes that we can deliver regular, useful increments of the system to the customers.

• Customer in team assumes that we can refine the specification of the system gradually, when we need to.

Working incrementally delivers some benefits.

• We learn all the time from the customer, from the system being developed. If we can make decisions later, they will likely be better.

• We keep the system as simple as possible, making it easier to understand, easier to change, less likely to contain errors.

• The customer quickly gets useful output. The system can be used to generate value and to guide further specification, planning, and development.