Play is the highest form of research.

Neville V. Scarfe

Dr Merritt Moore used the above quote in her Keynote speech at the 2023 WISE Conference to talk about transferrable skills. Dr Moore is a quantum physicist and a professional ballet dancer, and she excels in combining the creative and the technical through her performances with robotic dance partners. She got me thinking about the importance of creativity and play in the world of technology, and testing in particular.

In recent years, we have seen a sharp rise in test automation to. Continuous Integration and Deployment practices enable software to be tested regularly, accurately and quickly without taking any time away from the development or testing teams. Python libraries can generate constrained random inputs for test data. Static analysis tools can identify potential causes of run time errors.

And even more recently, we’ve seen AI tools such as ChatGPT that are now being used to automate test data creation and develop test case scenarios. Some would say that eventually all testing will be automated, and the role of the tester has an end date.

But there is one key difference between people and machines that Dr Moore highlighted in her talk:

Robots [and machines] can answer questions, but not ask them.

Even the most sophisticated AI models available today need to be given a prompt by a person in order to produce a result. What makes people, and particularly testers, different from our AI counterparts is about ability to ask questions, and to explore new scenarios. This is where the beauty and benefit of play and creativity come into testing: Testers are at their best when they are able to be creative, experiment with technology and ultimately have a play.

I’ve often heard testers (and other people working in tech) say that as a child, they often took things apart to see how they worked and played with things around them. These testers are now grown ups, and they need to capitalise on this child-like sense of exploration to get the most value from their testing activities.

Automation is great for removing administrative and repetitive tasks from a tester’s workload, but only if we give that saved time back to testers to allow them to learn, explore and play with the technology they are testing. Exploratory testing can be one of, if not the most, powerful and valuable form of testing that a tester can perform. It allows testers to ask new questions, discover new edge cases, and produce a higher quality product.

Machines and AI are not going to take test roles from human testers as long as we encourage those testers to ask questions, get creative, learn new techniques and explore technology. Creative testers ask the best questions.