As the slightly out of date reference to a killers song that is the title suggests, I’m talking about humans and testing.
First I want to flag up a great paper called The Irrational Tester. It goes through biases that humans are known to have and how this applies to testing and how to be aware of these biases and the effect they can have.
There’s a sketchnote that was drawn at TestBash which inspired this episode, I’ll drop a link to it in the show notes, but pretty much the first line got me: ‘Testing is fundamentally human’. So good, so true.
Testing is fundamentally human. I’ve spoken before about how QA can be a bridge between users and devs, and this isn’t to imply that devs aren’t human, but that they interact with the system in an entirely different way to manual testers and users.
There’s also a note about quality assurance not being about correctness, but being about quality and I love that nuance there. Things don’t have to necessarily be correct, but they have to be right. And by right I mean right in various different contexts – whether that be meeting code standards, accessibility standards, and, most importantly, fitting the needs of the user. This might mean some inconsistent choices, ones that we might not agree with sometimes, but they still have to have quality, they still have to be right. Not correct.
Its all about delivering value – the end users are the people who decide what quality is for them, though the team can and should guide and assist them in defining quality and goals, the client or user is who sets the end goal.
There was a testing conference called Stareast at the beginning of May, and one of the keynotes was Blunders in Test Automation, by Dorothy Graham. One of the blunders she covers is the misconception that testing tools test. There’s a great quote, that automated tools check, they don’t test, testing involves thinking.
Also, following on from something I mentioned last week, when I’m testing, I know what I’m doing, I can say, I did x, y, I expected Z, but I got Q, and I can write that out. When an automated check fails you generally just get a ‘Test #29 failed’, which isn’t very context heavy.
That was a bit of a tangent, but what I’m trying to say is that, testing requires human interaction with the product.
This leads me to exploratory testing, which I want to talk about a little bit. I generally try to get some time boxed off to go do some exploratory testing on a project, especially as a lot of work I do is around existing projects that have new features, or bug fixes. This can lead to interactions between new code and existing code, and so I want to have some time to have a click around to check things. I also love doing this on new projects, its great to get to grips with the project, I very much learn by doing, and so I don’t feel entirely comfortable with a project unless I can get down to manual testing and clicking about.
There’s a great video from TestBash this year on bug detection by Ian Mccowatt. It covers implicit and tacit knowledge, and knowledge sharing, and it’s a really good talk. There’s some interesting stuff on how testers interacts with systems, and how to share knowledge, and how tacit and implicit knowledge works. Ian also gives a couple of book recommendations which I’ll definitely be checking out.
There’s some interesting thoughts on communication and how big formal knowledge transfers may not be the best way of sharing, and involving testers in the social life of the project and it’s something I’ve been thinking about a lot, as communication is such a huge part of my job. I’ve got a few episodes coming up on communication where I’ll be expanding my thoughts on this.
Before I send you to the outro, I am after a recommendation or two: I really want to head to a testing/QA conference, but there seems to be a few about, and I have never been to a testing/QA conference before! So I’d love to get some opinions on what’s good. I know I’ve missed a few this year already, so even if your favourite has been and gone, please let me know and I can plan for next year. Ideally, I’m looking for UK based ones, as that’s easiest for me, but happy to get recommendations for ones further afield.