Track Talk F3

Good Enough Testing: Finding the Goldilocks Zone

Evgeny Tkachenko

Goldilocks was a spoiled little girl and a home invader. Fortunately, that’s not the moral of that story. The value of the fairy tale is that there is a point at which something can be just right. Just right is a porridge that is not too hot or too cold, a chair that is not too big or too small and so forth. This parable is relevant to software testing. I love the idea that there is a point where our test efforts are just right or in the Goldilocks Zone.

The tricky thing is that there is no steady state in which to define that Goldilocks Zone.

As Graham Freeburn states: “The (software) world continues to change at an incredible pace. We are faced with new technologies, unprecedented problems and increasing complexity. If we want to shape what we do now, and the direction for the future, our testing needs to evolve as well.”

The shape that testing should take is one where test effort input while producing good software is just right, that is to say, not too big or too small. The just-right point thus is one that we always need to be readjusting. It is common to become too heavily invested in our work and its importance, to the point that we become blind to the art of moderation. Making the hard choices to delete test cases, not build features, and even cancel entire projects feels like failure. However, these actions are necessary for maintaining a balance where quality is good enough to provide value to your business without overburdening your teams.

In this presentation, I will discuss seven maxims or prevalent thoughts in IT and testing that could use some moderation:

  • Testing is dead
  • Build fast, fail fast
  • The customer is always right
  • Automate all the tests
  • Bugs found early are cheap to fix
  • Never Agile Enough

In each example, I’ll show you how to find that Goldilocks Zone.