Thursday, August 10, 2017

We don't test manually at all

We sat in a room, the 7 of us. It was a team interview for a new candidate and we were going though usual moves I already knew from doing so many of these in the last few weeks. And as part of the moves, we asked the programmer candidate on how they test their code.

It wasn't the candidate that surprised me, but one of my own team's developers, who stated:
"We don't test manually at all".

My mind was racing with thoughts of wonder. What the hell was I doing if not testing? How could anyone think that whoever was figuring out scenarios, very manually wasn't doing manual testing at all? Where has my team's education failed this much that any of them could even think that, let alone say it out loud?

At the team room, I initiated a discussion on the remark to learn the meaning of it.

What I was doing wasn't included (I do a lot of exploratory testing and find problems) because I refuse to test each build the same way.

What the developers were doing wasn't included because manual testing targeted for a change is just part of good programming.

Figuring out scenarios to automate and trying them out seeing if they work when turned into code and debugging tests for failing wrong (or not failing right) wasn't included because it is part of test automation.

So I asked what then was this infamous manual testing that we did not do? It is the part of testing that they consider boring and wouldn't label intellectual work at all. The rote. The regression testing done by repeating things mindlessly without even considering what has changed, because there could be things that just magically broke.

We test manually, plenty. We are no longer mindless about it. So I guess that's what it really means. Not manual, but brain-engaged.

I can just make sure that people who matter in recruiting make sure someone is particularly well brain-engaged when joining the teams. That someone sometimes is not the tester who specializes in automation.