This year’s theme at CAST was multidisciplinary approaches to testing. Tying my experience in psychological research, legal practice, and testing, I gave the keynote address:Ã‚Â The Value of Checklists and the Danger of Scripts: What Legal Training Suggests for Testers. At its core, the talk says this:
In the hands of professionals, checklists facilitate the exercise of judgment by the human professional. In contrast, scripts attempt to mechanize the task (whether a machine runs the test or a human being treated as a machine). Professionals often use checklists and, in my experience, rarely use scripts. When I went to law school, professors spoke often about reliance on script-like tools–in essence, they described them as paving stones on a fast highway to malpractice. My training and practice in psychology and law laid a foundation for my skepticism about linking scripting to “professionalism” in software testing.
Checklists (of many kinds) are fine tools to help testers prepare for exploratory testing or to document their work. Scripts are fine tools for bug reporting (“do these things to get the failure”) and necessary tools for dealing with regulators who demand them, but as test planning tools, they are an anti-professional practice.
Along with the overview, the talk provided several examples of different types of checklists, with thoughts on how to apply them to testing.