Fact versus Fantasy

Reading the XP method practices can create the impression that they are a silver bullet, but of course, they are not. As always,

Process is only a second-order effect. The unique people, their feelings and qualities, are more influential.

One fantasy regarding XP adoption is found in groups that believe by just adopting iterative and evolutionary development and avoiding up-front specifications, they are "doing XP." Likewise with unit testing, working in a common project room, and so forth. Although data is still sketchy, it seems that many of the projects claiming to be doing XP are simply applying some iterative and evolutionary practices common to many IID methods (such as short iterations), and the group mistakenly believes these are unique XP ideas.

Probably the most common XP fantasy is getting onsite customers. It seems to be rare as hen's teeth to achieve this. Also, there is no shortage of so-called "XP" projects one investigates that could not arrange pair programming, which Beck considers one of the basics of an XP project.

Resistance to pair programming is perhaps the most common issue among developers. Some just don't want to do it.

It is also rare to find a common project room, or enough whiteboards.

Test-first development, early acceptance tests defined with customers, constant refactoring, and continuous integration are all widely confirmed as sustainable, excellent practices.