uitgewis

There seems to be a crowd of sw people, never unanimous, that tend to place tests on a very low ranking order when considering their application/framework/library.

In my experience, having good test coverage gives one confidence that an algorithm will behave in the manner intended. It assists tremendously with finding root-causes of problems and tends to lead to well thought out applications.

I therefor tend to question the reasons behind the decision not to write tests?