Tester as Editor Metaphor

Sometimes when I’m asked to explain how I see the role of testers on development teams, I use an (admittedly simplified) editor role as an analogy. For example, I might be asked to write an article on a particular topic, say test automation, and the publisher asks me to touch on three areas: a test strategy, a list of test tools and an example. These three requirements serve as the acceptance tests. I will write several drafts on my own, and with each draft I will remove spelling errors and grammatical mistakes. These are my unit tests. I move iteratively with article drafts until I feel I have expressed the ideas that are needed for the article. Once my unit tests have passed, and I’m confident I’ve have touched on the three areas I’ve been asked to write about (my acceptance tests seem to pass), I send the article to an editor for feedback.

The editor checks the content. If I haven’t met the obligation of the acceptance tests, they will let me know quickly that I missed a requirement. Good editors point out areas where I haven’t expressed an idea as clearly as I could. Good editors really know the audience, and point out areas that the audience may have problems with. They may also point out spelling mistakes and grammatical errors I have missed. They help draw out ideas and help me make the article all it could be given the time and resources available.

The editor doesn’t merely point out flaws, they can also provide suggestions to help overcome them. A cluster of writing errors may indicate to an author that an idea is malformed. Editors can spot problems that the author might miss because they are a skilled “second set of eyes”. They can provide constructive criticism and help encourage the author prior to publishing. It isn’t necessary for all articles to have a formal editor, but when I work with a good editor I realize how much better the article is than if I did it on my own.

In many ways a software program is also an expression of an idea. An idea may be technically correct and may meet the requirements of the customer, but may not be expressed clearly given the intended audience. My role of a tester is not an adversarial one, but like the editor my goal is to encourage and help make the program all it could be prior to its release. Like a good editor, a tester has an ability to know the audience of the program and the context in which it will operate.

In this post, Brian Marick talks about writer’s workshops which look like a good idea for doing software testing:

I’m drawn to the analogy of writers’ workshops as they are used in the patterns community. Considerable care is taken to prepare the author to be receptive to critical comments. At its worst, that leads to nicey-nice self-censorship of criticism. But at its best, it allows the author to really hear and reflect on critical ideas, freed of the need to be defensive, in a format where teaching what should change doesn’t overwhelm teaching what works and should stay the same.

Software development teams could learn a lot about constructive criticism from the good writer’s workshops.
In another post, Brian sums up the idea I’m trying to get across here:

… I offer this picture of an Agile team. They – programmers, testers, business experts – are in the business of protecting and nurturing the growing work until it’s ready to face the world.

Good testers are not the adversaries of developers, they are part of the team who works collectively towards creating the best software they can. Bad testing, like bad editing does not seem to be in the business of nurturing a growing work.