Matcher Equivalente ao AsserEquals do JUnit
Vaga QA Automação
[Dúvida] -Inspecionar acordeom
CTAL-TAE 2019 - Simulado e FlashCards
Sobre o Gherkin
Dúvida: Validar link de e-mail - Capybara/Ruby
[Duvida] - Gente é possivel pegar um valor que é retornado na tela usando o Capybara?
Testes e2e na pipeline do CI ou em um CronJob separado?
Dicas de curso
Erro ao encontrar um elemento no internet explorer
[Duvida] Automação mobile visualizar itens da lista
Lint para Selenium no visual Code
revista assert(QA) agora faz parte do Agile Testers :)
Python + pytest
Pessoal, o AT precisa da sua opinião :)
Como Facilitar os Testes Regressivos de API Usando REST-Assured
Artigos e Tutoriais1
[Dúvida] - Vale a pena documentar casos de teste?
Automação de Testes Mobile com Appium
Qualidade de codigo em Delphi
Episode 009 - The Why? Special - The Evil Tester Show
“Why?” is an incredibly popular question.
I think it has problems. I try to avoid it when I’m consulting.
In this show we will examine “Why?” and under what circumstances I do use it.
“Why?” is almost a default question because it can take the place of every other question. It is easy to ask. We don’t even need to know what answer we want. It’s also a very ambiguous question, and when we ask why there is a risk that people try to read into our motivation, and that people might take offence.
When I’m working with clients and I come in for a week, I have a short period of time to build a model of the system of development, the interactions, the reasons, the issues, etc.
I could ask Why? a lot.
- Why do you do things this way?
- Why do you use this tool?
- Why do you not do things that way?
I don’t usually do that because:
- I get a lot of conflicting answers
- I don’t often get evidence, I get a lot of beliefs
- I don’t get as much actionable insights
If I want to long consultancy engagement that build dependency on me as a consultant then Why? could be a great question because I can chain whys and get a really long discussion and lost of vague concepts to follow up.
When I consult I want actionable insights, next steps, obvious alternatives, evidence.
If you want software testing or development consultancy then check out EvilTester.com/consultancy
“Why do we ask why?”
“Why is the sky blue?” …answer… “Why?”
As children it is almost a default question, it is easy to ask, we get answers, and its fun to see people get frustrated when we keep asking it.
In IT, Management, Engineering we often taught that it is a good question - the Five Whys Technique.
- by asking “Why?” 5 times we get to the root of a problem
It’s a default. The reason I say that is because I’ve been working hard on my own problem solving processes to avoid it being a default but when I’m not keeping a tight rein on my thinking, it pops up.
“Why?” is ambiguous
But when we ask “Why?” we don’t know what question people will actually answer. It is quite a vague question.
- Why did you go out? - “because I went to meet Bob” answered the question “who did you meet?”
- Why did you go out? “because I wanted to see the new Marvel film?” “What film did you want to see?”
- Why did you go out? - “because I went to see a film in the cinema?” “Where did you go?”
- Why did you go out? - “because I went to see a film at 8pm” “When did you go out?
- Why did you go out? - “because I had to catch a bus to see the film” “How did you get to the cinema”
Anytime we ask why, we do no know how people will interpret it. Will they interpret it as a What? How? Where? When? Who?
Or will the interpret it as:
- a teleological why? “What was the motivation behind this?”
This might be evidence supported or belief supported.
- Lean Tool
- An actual problem to be solved
- Usually a physical (to people problem) to be solved
- Everyone knows what the problem is
- Why did the machine stop? - because the power went off
- Why did the power go off? - because the supplier shut it down
- Why did the supplier shut it down? - because we didn’t pay the bill on time
- Why did we not pay the bill on time? - because it is a manual process and the person processing it was off sick
- Let’s automate that and assign it to a team instead of a person
This is very focused on a specific problem. With evidence based answers at each step.
In software or consultancy…
- Why do you write test scripts?
- because Bob said we have to
- because we are mandated to do so
- because that is the right thing to do
- because expert X said we should
- because that is the best thing to do
- Why are you asking?
- I don’t know
- What is the problem?
- Do we know what answer we want?
- Do we know what we are going to do with the answer?
- Do the people know the reasons for asking the question?
Five whys could easily be rephrased as the “What caused this?” chain to five levels of direct relationships.
“Five Whys” might be targeting people’s beliefs.
“What do we believe is the reason for writing test scripts in this process?”
Is there is risk in chaining beliefs?
Sometimes beliefs aren’t based on evidence.
Humans are very good at having and maintaining beliefs despite evidence.
People are also very good at defending their beliefs.
If we keep asking why then people might go into defensive mode which impacts our ability to work with them.
And when I’m consulting I can’t afford the time that that resistance creates.
Easier to avoid it in the first place.
Why is a very good question for working with people’s beliefs. And if you are trying to effect change in an organisation you will probably have to clarify those beliefs and work with them on an individual basis.
I don’t know
People often avoid saying “I don’t know”
If we ask “Why” and someone doesn’t know. We risk asking them to “make something up” to avoid saying they don’t know.
That then becomes a belief.
And we put them in the position where they have to continue to justify that statement that they just made up.
Why are you asking?
Without a proper context to questions people may read motivations into the question.
“Why?” can trigger a lot of negative motivation.
“How”“, “When?” etc. are more neutral and trigger less resistance.
But do you know why you are asking?
We have to have more conscious direction over our questions if we are problem solving or building specific models.
If we are building a model from nothing, and all information that comes back is valid then it doesn’t really matter what we receive in response then it might be a valid default.
But at some point we have to decide what information we want to receive in the answer. And if we create questions that target that answer they probably won’t be “Why?” questions.
What is Why? useful for?
- general unspecified answers
- how do people interpret ambiguity
Usually about the person your asking, rather than the thing you are asking about.
But try it for yourself
When you ask a “Why?” question - think:
- what could you have asked instead?
- were you clear on the type of answer you wanted?
- did you get the type of answer you wanted?
- did the person respond as you hoped?
If this interests you then try and avoid “Why?” ask “What”, “How”, “Who”, “when”, “where” type questions instead.
and I’ll leave you with a couple of quotes.
This is from Fritz Perls Book The Gestalt Approach on page 77
“The why questions produce only pat answers, defensiveness, rationalizations, excuses and the delusion that an event can be explained by a single cause. They why does not discriminate purpose, origin, or background. Under the mask of inquiry it has contributed perhaps more to human confusion than any other single word”
And… he goes on to suggest that if we pursue inquiry with a why that we
“join the group of worrying grandmothers who attack their prey with such pointless questions as ‘Why did you catch that cold?’ and ‘why have you been so naughty?’”
If you want a quote for test consultancy contact me via eviltester.com/consultancy
And if you found value in this or any of my podcasts, do let people know or like this on itunes or youtube or wherever you found it.
Proactively tell people about good stuff rather than have them ask you “why are you listening to that?”
Additional Related Reading