Use XP wisely

Xtreme programming, or XP, is getting lot of attention these days. Some bill it as the definitive solution for implementing business solutions quickly and effectively. Like most big ideas and technology movements, however, believing the more ambitious claims made about XP will provide a very nasty surprise indeed.

Xtreme programming does embody some superb ideas and goals. Pairing up developers really can result in an astonishing productivity boost. But it rather depends on which developers are matched with which. And the healthy obsession with refactoring can result some amazingly slick and efficient code. But it can sometimes descend into a strange quest after nirvana that takes a ridiculous amount of time.

A minimalist approach to documentation can certainly increase the likelihood that documentation has a chance of remaining up to date. But strike the balance wrong and new additions to the team can take an age to get up to speed.

XP certainly is powerful medicine, but it needs to be administered in carefully measured doses. XP works best when you have a very able project leader, a relatively small team that is highly committed and very skillful, and a well defined problem domain.

XP can be applied to larger teams, and to much less defined problems. But in the same way that most anyone can drive a car quickly in a straight line, the curves require a significantly greater degree of skill.

My advice to people considering the adoption of XP is simple. Go ahead and try it, but make sure you properly pave the way. Create a sensible framework for Xtreme programmers to operate. Be clear about the documentation that's needed when it must be completed to ensure maintenance of the application after all the original developers have moved on. Make sure a strong project manager with the proper skills to keep a team motivated and the ability to pair the right developers is selected.

And most important, don't lose sight of the fact that your goal is not the 'perfect subroutine', it is the business solution that your customer needs.

For more Ovum, go to:

About the Author

Gary Barnett is IT research director at Ovum Ltd., a United Kingdom-based consulting firm.