Updated: Following an incremental development pattern, I've updated the title to reflect the bug that Nick pointed out. Now to find a way to unit test and automate these updates. :)

Nick Boldt suggest a good summary of the last couple of blog postings.

Small green test-driven builds good; big red broken-test builds bad.


This is a good summary, but the problem is that it doesn't explain the whys. Unless you know the details of Unit Testing and Continuous Integration, you don't know the whys and what causes the problems. With out the context of the larger posts, this summary doesn't help much. It also doesn't suggest ways to get from the big broken tests to the small green tests. How can one move away? What is causing the big broken tests?

For a good balance between the detail that is necessary and the summary picture. I highly recommend people view the Extreme Programming website. It is a model that I would eventually like to emulate with the completely documented process I'm flushing out now.