Why don't we test in other activities in life?

(nilanjan) #1

If testing is, ‘How do I know my work is good?’, why don’t we ask this question in other activites, e.g., design, business. Of course, we do use techniques such as ‘5 Whys’. However, I don’t think we do as much as when we test software.


(Kate) #2

The really cynical answer: Because there are fragile egos at stake and testing dents those egos.

The (slightly) less cynical answer: Because it takes time and energy, and most of us have limits on how much time and energy we’ve got. Personally I apply testing techniques to my testing job and my writing. I apply a more limited set of those techniques to anything that comes from the news media. I don’t have the energy to do more than that, although I’ve got to admit my natural chaos magnet effect has a tendency to do it for me.

(Chris) #3

Because of the definition of software testing being the testing of software, and “testing” being in the context of software or other items under test, and within the context of a tester role.

We perform activities related to testing every second of every day. Your brain is a set of tacit heuristics and hidden processes constantly working, even while you sleep. Driving a car is an exploratory performance, where feedback from sight and sound are fed into our mental models that affect our choices and behaviours based on the immediate context and that of our environments and social contracts. We use focus/defocus to find solutions to problems and identify things that matter. We change our attention to where it’s required at the expense of something else.

For the question “how do I know X is good?” - I’m entirely with Kate on that one. Skillful skepticism comes with limits to our time and energy. From Kahneman’s now-classic “Thinking Fast And Slow” we simply cannot maintain a consistent stream of System 2 thinking (slow, tiring, requires attention). That’s not to say that we don’t perform System 2 thinking at all - we just have to “decide” when it matters.

Specifically in the realm of business I’d say that testers are best deployed when testing design and ideas. A good tester, if well armed, in my opinion (who else’s would I be using?) can test a discussion, a document, a business proposal, pretty much anything. “Shift-left”, or whatever we’re calling that now, was a well-meaning attempt to push testing towards the start of the process. Slowly we move towards the idea that testing is a platform on which we rest the entire lifecycle, one that pervades and supports all elements of it.

Another interesting take is what our skills and tools are capable of doing. Can we “test” HR? What would that look like? Who would have the skills to do it? What would we examine? What tools might we use? What does a bug look like? Software exists to solve a problem. We can think of ways in which it might not solve that problem, or solve it in a poor way. I’m sure it’s someone’s job to gain insight that support are supporting, ops are opsing, HR are hrrring, and ways in which they might not be - or ways they could be doing it better or cheaper. We just don’t call those people testers, I guess.

I’d say dedicated testers exist for software because it’s closest to the money. It’s used to test what a customer will most associate with the money they’ve handed over. That quality translates to a bunch of stuff (positive brand association, uptake, low turnover, word-of-mouth advertising, portfolio, etc) that translates to cash. It’s the answer to the question “is the thing we’re selling any good?”.

(nilanjan) #4

@kinofrost Thanks for the answer. I’ll respond later in more detail. Just a quick note: I didn’t mention ‘software testing’ in my question.

Edit: I did mention ‘software testing’. However that was just to make a contrast.

(nilanjan) #5

@kinofrost @katepaulk

To give some context.

Why don’t we ask this question in most professions/trades:

  • carpentry, construction, medicine, business, design, recruiting

Why don’t we ask this question in our individual activities:

  • when you cook, when you drive, when bringing up children, when we read, when we write, make a sandwich…

(Chris) #6

We most definitely ask that question in construction. Buildings regulations approval, building inspectors, safety officers; the industry is rife with legal mandates and the people who uphold them, constantly asking the question of whether the work is good. Those roles exist to answer “how do I know my (their) work is good?”. Within each role each person takes responsibility for identifying good work, but this is checked by a dedicated team. This is because failures to create good work in construction can lead to property damage, injury and death.

We also definitely ask the question in medicine. Journal clubs, research presentation Q&A, hospital grand rounds, clinical trial transparency and reporting, paper peer review and so on. The question “are we doing a good job?” is one sewn into the fabric of medical practice and medical science because we know that not asking that question can cause, and has caused, mass harm and death (e.g. http://www.jameslindlibrary.org/articles/therapeutic-fashion-and-publication-bias-the-case-of-anti-arrhythmic-drugs-in-heart-attack/). The answer to “how do I know I’m doing good work” is science itself - the only workable mechanism for determining truth about the natural world despite human biases. We only know medicine to be effective because we asked “how do I know my work is good?”.

I think we also ask that question of ourselves all the time - am I living a good life? Could I be doing this better? What could improve this? Cuisine, driving and childcare are all realms of human experience that have improved over the years because people looked at the context and the performances and asked the question.

I think there are lots of places where, as a tester and therefore a professional skeptic, we look at a poor situation and wonder why nobody’s doing something about it and why they think it’s okay to not improve - the assumption being that nobody’s asked the questions “is this good enough?” and “am I good enough?”. I find that the reason behind that tends to be that it’s nobody’s job to do it. Somebody cares about it, but not the people around it, and they are not paid to care. This is why when a decision is made to do something at a meeting, unless it’s assigned to someone, and unless it’s going to be checked that they did it, it will not ever be done. A tester is hired to be devoted to the cause - the business wants to understand the quality of a product, and its potential problems, because it has a direct business value. It’s a risk mitigation strategy. If you can get someone to find business value in the quality of your sandwiches you’ll find people asking a lot more questions about your sandwich-making output.

(Jesper) #7

BTW: more and more of the IT projects Im involved in is not about testing software. … https://jlottosen.wordpress.com/2016/10/04/less-software-more-testing/

(Brian) #8

Seriously? If we tested software one tenth as much as we tested medicine, software testing would be dead from the high costs. And there would still be a rallying cry of “Did anyone even test this?!”

I don’t know how you bring up children, but I just might feel offended that you would even imply that many parents don’t address their ability to raise their children with a healthy dose of critical thinking or, as you put it, testing.

(Finny) #9


We certainly do this, but not as a conscious, separate activity that is labelled as ‘testing’. You can say this is built into the system and as we gain experience, we get better at doing the activity based on the feedback from ‘testing’.

To quote Terry Pratchett - “Sometimes I really think people ought to have to pass a proper exam before they’re allowed to be parents. Not just the practical, I mean.”

There are all kinds of people, some are really good at parenting, some not so much. Some people might prefer getting feedback after all.

(nilanjan) #10


Thanks for the detailed response. I am going to spend some time thinking about your response and get back.

One quick note: I don’t think business value and defects are either-or. I also don’t think businesses understand the problems customers face. If a problem causes harm, then it gets attention. For others, there is customer service and customer experience. To demonstrate this, I can list many problems that customers face and complain about on social media. Business have the option to address common trends or even specific problems.

To me, one approach that I find reduces defect is superior design. I am thinking about the product design in Silicon valley type companies - Twitter, Facebook, etc. However, even with product designers, I don’t think they really understand how to test their ideas (similar to experienced testers who follow context driven testing).

More later…

(Kate) #11

They don’t understand.

One of my favorite books about software is Why Software Sucks - because it explains why so much software is difficult or confusing to use. I’m looking forward to the sequel, because the principles haven’t changed, and neither has the general usability of most software.

What it comes down to is that software geeks (including testers) think differently than most people. The skills that make us valuable to create software also make us lousy at assessing how easy it will be for others to use. The same applies to product designers: they have skills that make them sufficiently different from their target audience that it’s next to impossible to think like their target audience.