Exploring the Testing Bubbles of Misunderstanding
??Andrei Contan
Testability advocate | co-founder Testingscool | Humans in Learning | Romanian Testing Conference
Have you ever felt like you're living in a bubble, where the world outside is a vast, uncharted territory of insights and discoveries? Well, I stepped out of my testing bubble, and let me tell you, the findings were nothing short of concerning. It seems that in the ever-evolving landscape of software testing, some things remain, shall we say, a bit "bubbly."
Killing Regression ??
One of the intriguing trends I've observed is the push to kill regression testing by removing it from the end of the development cycle. The focus? Shadow traffic, beta testing and a bit of ketchup. It's an attempt to ensure the quality of software without the elementary understanding on the psychology of software development. While those concepts, taken individually, have its merits it's a bit like saying, "I'll fix my bike by removing the pedals." Can we achieve "zero bugs" and create "bulletproof software" by sidelining regression testing? What does that even mean?
"Traditional Testing" or "Manual Testing"? The (wrong) Dilemma
In the quest to modernize testing terminology, we've moved from "Manual Testing" to "Traditional Testing." But really, it's like deciding whether you prefer "couch potato" or "sofa spud." The problem is not (only) in the label, but in the meaning.
Team of Testers: Are They the Solution?
There's still the notion that employing teams of testers can solve quality problems without bothering to identify the origin of the problem. It's as if we're saying, "I don't know what's wrong, but let's hire a bunch of people to figure it out." Can we do better than this?
Test Automation Simplified
The term "test automation" is now casually used to describe testers who know how to code. It's a bit like calling a baker a "flour master." While coding is undoubtedly a valuable skill for testers, reducing the craft of testing to this one facet is, well, a bit simplistic.
领英推荐
Empathy and Understanding: The Missing Links
In our pursuit of testing excellence, it seems we've overlooked the essential ingredients of empathy for customers, understanding stakeholder needs, and the awareness required to identify risks. It's like baking a cake without sugar – something is missing!
Mindlessly Automated Testing
I've witnessed the mindless automation of testing scenarios without a second thought. Asking a tool like ChatGPT to provide 10 test cases for timezones and blindly incorporating them into a test plan without considering the context is a recipe for poor/misguided/irrelevant testing.
The "Security Testing Pyramid": A Top-Heavy Mystery
And then there's the enigmatic "Security Testing" inside the Test Automation Pyramid, added mysteriously at the top of the hierarchy. The explanation? "Tbh, I didn't know where else to put it." Well, that clears things up!
So, where do we go from here? It's clear that some of these ideas are shaping business decisions and leading to unrealistic promises. We're at risk of recreating the same problems we've been trying to solve for the past decade, all under the new label of "genAI."
In the end, it's a bit like finding out that, despite all our advances, we're still floating around in bubbles of misunderstanding when it comes to the craft of testing. But hey, that's the world of tech as I experience it today – full of surprises.
Let's hope that as we burst these bubbles, we can land on solid ground, armed with a more nuanced and thoughtful approach to software testing.
QA Analyst | Cybersecurity
1 年Interesting article. I'll give my input here. Killing regression - I didn't notice this trend. Maybe is because I'm working by myself many times. But with DevOps I am fairly convinced that people may want to kill it for another sweet release. However, with regression, you can have automated checks, or maybe retesting, testing after some change, or just to probe whether the quality got worse. Semantics - I have never considered the term Manual Testing. I always thought that by using that term one is automatically diminishing himself, cutting himself short, and yes, I have mentors who still use the terms QA and Manual, but that didn't change my respect for them, it's just my personal preference. Probably the first book on testing was by Gerald Weinberg and Herbert Leeds in 1961. Weinberg talked about testing as a matter of imagination and being suspicious and critical, not automation. Something that is fundamentally a non-algorithmic process. Then in 1973, Hetzel published a book in which he talks about Manual testing and how to automate the testing process. I currently use my own terms, and I have created a glossary for that. Semantics matter, one can read Daniel Kahneman or Dan Ariely....
Qualisoof, senior (quality) consultant, speaker and trainer @OrangeCrest
1 年interesting post Andrei, shows clearly that we as testers are fallen into a 'If you do what you always did, you're going to get what you always got". Which at least is not something the organization needs... So a lot of creativity was missing and the urgency of knowing the purpose of what is important.