The problem with a lot of new ideas, particularly software development methodologies and management trends, is separating the wheat from the chaff. Until recently, test driven development was another bit of hype I didn’t buy, but I few recent experiences have changed my mind to the point where I’m actually using it at home on personal projects.
What really made a difference was seeing it in action: writing a test, writing some code, and seeing that it really worked. Repeat that for a number of scenarios and you start to feel pretty good that what you have is solid and you’ll know if you break it in the future.
The second feature is that someone sold me on what it could do for me as a developer - free me from the tedious, ad-hoc, labor-intensive testing process that tends to go along with my rapid style of development. At work, it’s allowed me to bypass all the clicks of a web application to test the underlying logic, and at home I can test my photo management scripts without manually shuffling around files to be modified.
The real test of my commitment and realization of its value will probably come the next time I modify Gallery, a very large PHP photo gallery with an equally large testing suite.
del.icio.us/mbotos
