The testers who taught us how to explore

International Women's Day is a good moment to say something we've been meaning to say for a while. A lot of what we believe about testing, and a lot of what we're designing into Testomatica, comes from books written by women who spent their careers taking testing seriously when plenty of the industry didn't.

Three of them have been on our desks, physically, for most of the past two months.

Elisabeth Hendrickson and exploration as a discipline

Elisabeth Hendrickson's "Explore It!" came out in 2013. Before reading it, one of us thought of exploratory testing as "clicking around and seeing what happens," which is roughly how it's still described in a lot of job postings.

The book makes a strong case that it's nothing of the sort. Exploration, done well, is structured. You go in with a charter that says what you're exploring and what you hope to learn. You look for the variables in the system, the things you can change, and you change them on purpose. You take notes, because a session you can't describe afterward is a session nobody can learn from.

That framing ended up in our design almost directly. Our agents don't get a vague instruction to "test the site." They get a focused charter tied to what changed in the release, a time and cost budget, and a requirement to record what they did. When an agent finds something, its notes are what the referee replays. The idea that exploration should be purposeful and accountable is one we got from reading Hendrickson, not from reading model documentation.

Lisa Crispin, Janet Gregory and the whole team

Lisa Crispin and Janet Gregory's "Agile Testing" was published in 2009, and it's the book we hand to anyone who asks where testers fit on a team that ships constantly.

Its central argument is that quality belongs to the whole team, not to a separate department that receives code at the end and hands back a list of complaints. Testers work alongside developers from the start. Testing happens continuously. Everyone owns the result.

We think about this a lot, because it would be easy to build an AI testing tool that recreates the old wall. Code goes over, a report comes back, nobody talks. That's the opposite of what we want. Findings should land in the same tracker the team already uses, in a form a developer can act on in minutes. Humans keep the final say on whether to ship. The tool is a member of the team with a narrow job, not a gate that replaces the conversation.

What we're trying not to forget

Reading these books alongside a pile of recent AI testing papers was a little humbling. Many of the hard questions the papers wrestle with, like how an explorer decides where to look, how to tell a real problem from noise, and how testing fits into a fast-moving team, were thought about carefully years ago by practitioners who did the work.

It's tempting, when you're building something new, to act as if the field started with you. It didn't. The agents are new. The questions mostly aren't.

So, thank you. To Elisabeth Hendrickson, Lisa Crispin and Janet Gregory, and to the many women testers whose work never made it into a book but shaped the teams they were on. A lot of us learned how to test from you, whether we knew your names or not.

If there's a tester whose work shaped yours, today's a good day to tell them.