I mentioned some time ago that when I restarted work on WhaTBI? it would be on a test-driven basis. Well, that time has finally come (I'm currently at 'green' on the first iteration, so will post the code and commentary as soon as I've refactored) so I thought I'd set out my ideas here, both for comments and as ground rules for myself as development progresses. So here goes:

1. Keep everything as simple as it can be, but no simpler.

As you can see, I've even applied the set of rule(s) to itself! The 'no simpler' bit probably means I should add some explanation though.

Basically what I've decided to do is put to one side all my preconceived notions of what the architecture and code could/should/will look like so I can concentrate on adding one feature at a time. To do this I'm using one of the ideas that David Allen advocates in 'Getting Things Done', namely writing them down. The reasoning behind this is that unless your subconcious is sure that they are captured safely somewhere it will continue to feel responsible for them.

To this end I will complete the WhatbiDesign articles as soon as time permits (there's not too much left to do) and also start WhatbiFutures for less fully-formed thoughts and ideas.

Hopefuly this whole process will also generate a number of more general posts, especially as I delve into the new features of the V2 framework, and have a go at implementing some of Fowler's 'Patterns of Enterprise Application Architecture'.

Well, that's the theory, anyway...