And perhaps I didn't make my points clearly enough in the article; I was really trying to differentiate between the Test-Driven Development dogma as a panacea for all development problems ever, and developers writing tests first where it makes sense to do that.