We can gain by learning about other industries such as aviation and healthcare, and studying other disciplines, argued Conor Fitzgerald, software tester at Poppulo, at RebelCon.io 2019. Aviation has a history of continually learning from its mistakes, whereas in healthcare, culture and bias seem to challenge learning and continuous improvement.
Fitzgerald’s interest in aviation stemmed from a coaching course; the instructor referenced the changes in aviation cultures several times, for example, the shift to a blameless culture. He mentioned how he was intrigued to learn how checklists had played a part in the aviation industry’s excellent safety record.
During a first aid course, Fitzgerald noticed that healthcare used heuristics, mnemonics and oracles. He decided to explore what else healthcare used that could be of value to software testing.
Fitzgerald also investigated a variety of disciplines including economics and yoga. His reflection on learnings from economics came from studying a postgraduate course in business and from books and podcasts on economics. He has been practicing yoga for some time. Fitzgerald mentioned that "yoga is very different to the other areas covered, but can provide hugely valuable insights."
The aviation industry is fascinating, said Fitzgerald. It has a history of continually learning from its mistakes through the use of checklists, black box recorders, blameless culture, cockpit re-design and Crew Resource Management (CRM). These learnings have lead to its excellent safety record, argued Fitzgerald.
He mentioned two books that contrast the differences between healthcare and aviation, two of the most important safety-critical industries in the world today: The Checklist Manifesto by Atul Gawande, and Black Box Thinking by Matthew Syed. These industries have many similarities, said Fitsgerald; mistakes can lead to serious injury or loss of life, both deal with huge complexity, and traditionally surgeons and pilots have had a hero status.
Healthcare, unlike aviation, has not had a culture of learning from its mistakes and continually improving; this has lead to preventable medical errors being the third biggest killer in the United States – behind only heart disease and cancer, argued Fitzgerald. Based on 2013 US healthcare data, this is the equivalent of two jumbo jets falling out of the sky every twenty-four hours, he said.
Fitzgerald said that there seem to be multiple reasons why healthcare does not learn from its mistakes, including a culture of blame, although it seems the core issue is cognitive dissonance, where dissenting evidence is reframed or ignored. It’s not only healthcare professionals who suffer from cognitive dissonance; it can also be seen in a wide variety of other disciplines such as politicians and economists.
Essentially, the more we have riding on our judgements, the more likely we are to manipulate any new evidence that calls them into question. Cognitive dissonance quashes feedback, improvement and innovation, said Fitzgerald.
InfoQ interviewed Conor Fitzgerald about his learnings from other disciplines and industries.
InfoQ: What made you decide to explore other disciplines and industries? What did you set out for?
Conor Fitzgerald: Initially my curiosity was piqued by various books I have read over the years that covered different disciplines and industries, for example, I really like listening to the Freakonomic podcast and have read a number of popular books on economics, such as Malcolm Gladwell’s Outliers, Tipping Point and David vs Goliath. Additionally, my previous test roles involved working in a variety of interesting industries including telecommunications, finance and gaming, each of which were very different.
Although the real catalyst to further explore was the fact that two large software projects I worked on failed. After the second project finished I began to reflect on the factors of a successful project.
On reflection, it seemed that different cultures in different industries played a critical role in both the quality of the software and ultimately the success of software projects.
InfoQ: How are culture and quality related?
Fitzgerald: Henry Ford is famously quoted as saying, "Quality is doing it right when no one is watching". We could view culture as a habit, the habit of "doing it right," and the aviation industry provides a great template for what a quality culture looks like.
Using Checklists so we don’t forget critical steps. A blameless culture, where there is no finger pointing, no successes or failures, just learning. Crew Resource management focuses on open, honest communication and focuses on teamwork, rather than heroics.
InfoQ: How can we use heuristics and oracles to become better in testing?
Fitzgerald: A heuristic is a fallible method of solving a problem or making a decision, in simple terms a "Rule of Thumb". Heuristics not only improve you as a tester by enriching your testing knowledge and experience, but more importantly, heuristics can help you to communicate to others how you know where potential issues may be unearthed in a feature, and the rationale behind your thinking.
Elisabeth Henrickson’s Test Heuristics Cheat Sheet is a great starting point for those who wish to learn more about test heuristics.
Oracles are simply the principle or mechanism by which we recognize a problem. Katrina Clokie says, "Oracles help me to discover the real reason that I think there is a problem." Just like heuristics, there is a two fold benefit: if you consider your oracles before you begin testing, you will have a wider variety of risks to consider and more interesting areas to explore.
Micheal Bolton’s FEW HICCUPPS mnemonic is useful when considering oracles.
InfoQ: How have the things you learned impacted your role as a tester?
Fitzgerald: The biggest impact has been to have a greater focus on collaboration and communication through the whole team testing.
A recent blog on the Five stages of test collaboration that I wrote covers practical ways to improve communication and collaboration through the software development process:
- Ensure the problem that is being solved is understood by the team, and that discussions about risk and the user are had early on in the project.
- If possible, take customer feedback on-board during feature development.
- Focuse on Pairing, Mobbing, Team Walkthroughs and Bug Bashes.
- The whole team deciding when a feature is ready to release for beta testing by customers.
- Testers learning more about coaching, and using these skills to coach others on the craft of testing.