And There's More

When new staff join Linguamatics they'll get a brief intro to each of the groups in the company as part of their induction. I give the one for the Test team and, as many of our joiners haven't worked in software before, I've evolved my spiel to be a sky-high view of testing, how we service other parts of the company, and how anyone might take advantage of that service, even if they're not developing software.

This takes a whiteboard and about 10 minutes and I'll then answer questions for as long as anyone cares to ask. Afterwards we all go on our separate ways happy (I hope) that sufficient information was shared for now and that I'm always available for further discussion on that material or anything else.

I mentioned the helicopter perspective that I give to Karo Stoltzenburg when she was thinking about when and how to draw a testing/checking distinction in her Exploratory Testing 101 workshop for TestBash Manchester. I was delighted that she was able to find something from it to use in her session, and also that she produced a picture that looks significantly nicer than any version I've ever scrawled as I talked.

Karo's picture is above, her notes from the whole session are on the Ministry of Testing club, and below is the kind of thing I typically say to new staff ...

What many people think software testing is, if they've ever even thought for a second about software testing, goes something like this:

an important person creates a big, thick document (the specification) full of the things a piece of software is supposed to do
this spec gets given to someone else (the developer) to build the software
the spec and the software are in turn given to another person (the tester) who checks that all the things listed in the spec are in the software.

In this view of the world, the tester takes specification items such as "when X is input, Y is output" and checks the software to see whether Y is indeed produced when X is put in. The result of testing in this kind of worldview probably looks like a big chart of specification items with ticks or crosses against them, and if there are enough crosses in important enough places the software goes through another round of development.

While checking against specification can be an important part of testing, there's much more that can be done. For example, I want my testers to be thinking about input values other than X, I want them to be wondering what other values can give Y, I want them to be exploring situations where there is no X, or when X is clearly invalid, or when X and some other input are incompatible, or what kinds of things the software shouldn't do and under what conditions ...

That's all good stuff, but there's scope for more. I also want my testers to be wondering what the motivation for building this software is, who the users are, and whether the specification seems appropriate for a software product that meets that need for those users. I'd also expect them to think about whether the project, and the team — including themselves — is likely to be able to create the product, given that requirement and specification, in the current context. For example, is there time to do this work, is there resource to do this work, do the team have sufficient tooling, expertise, or other dependencies, ...?

Even assuming the spec is appropriate, the context is conducive to implementation, and the team are not blocked, you won't be surprised to find that more possibilities exist. I'd like the tester to be alert to factors that might be important but which might not be in the specification much at all. These might include usability, performance, or integration considerations, ...

For me, one of the joys of testing is the intellectual challenge of identifying the places where there might be risk and wondering which of those it makes sense to explore with what priority. Checking the specification can certainly be part of testing, but it's not all that testing is: there's always more.
Image: Jimmy Cricket, Karo Stoltzenburg

Meet Me Halfway?

The Association for Software Testing is crowd-sourcing a book, Navigating the World as a Context-Driven Tester , which aims to provide responses to common questions and statements about testing from a context-driven perspective . It's being edited by Lee Hawkins who is posing questions on Twitter , LinkedIn , Mastodon , Slack , and the AST mailing list and then collating the replies, focusing on practice over theory. I've decided to contribute by answering briefly, and without a lot of editing or crafting, by imagining that I'm speaking to someone in software development who's acting in good faith, cares about their work and mine, but doesn't have much visibility of what testing can be. Perhaps you'd like to join me? --00-- "Stop answering my questions with questions." Sure, I can do that. In return, please stop asking me questions so open to interpretation that any answ...

How do I Test AI?

Recently a few people have asked me how I test AI. I'm happy to share my experiences, but I frame the question more broadly, perhaps something like this: what kinds of things do I consider when testing systems with artificial intelligence components . I freestyled liberally the first time I answered but when the question came up again I thought I'd write a few bullets to help me remember key things. This post is the latest iteration of that list. Caveats: I'm not an expert; what you see below is a reminder of things to pick up on during conversations so it's quite minimal; it's also messy; it's absolutely not a guide or a set of best practices; each point should be applied in context; the categories are very rough; it's certainly not complete. Also note that I work with teams who really know what they're doing on the domain, tech, and medical safety fronts and some of the things listed here are things they'd typically do some or all of. Testing ...

The Best Programmer Dan Knows

I was pairing with my friend Vernon at work last week, on a tool I've been developing. He was smiling broadly as I talked him through what I'd done because we've been here before. The tool facilitates a task that's time-consuming, inefficient, error-prone, tiresome, and important to get right. Vern knows that those kinds of factors trigger me to change or build something, and that's why he was struggling not to laugh out loud. He held himself together and asked a bunch of sensible questions about the need, the desired outcome, and the approach I'd taken. Then he mentioned a talk by Daniel Terhorst-North, called The Best Programmer I Know, and said that much of it paralleled what he sees me doing. It was my turn to laugh then, because I am not a good programmer, and I thought he knew that already. What I do accept, though, is that I am focussed on the value that programs can give, and getting some of that value as early as possible. He sent me a link to the ta...

Reasonable Doubt

In Your job is to deliver code you have proven to work Simon Willison writes: As software engineers we ... need to deliver code that works — and we need to include proof that it works as well. He is coming at this from the perspective of LLM-assisted coding, but most of what he says applies in general. I think this is a reasonable consise summary of his requirements for developers: Manual happy paths: get the system into an initial state, exercise the code, check that it has the desired effect on the state. Manual edge cases: no advice given, just a note that skill here is a sign of a senior engineer. Automated tests: should demonstrate the change like Manual happy paths but also fail if the change is reverted. He notes that, even though LLM tooling can write automated tests, it's humans who are accountable for the code and it's on us to "include evidence that it works as it should." Coincidentally, just the week before I read his post I told one of my...

Great Shot, Kid

This week I've been playing with altwalker , a model-based testing tool. To get the hang of it, I attempted to build a very simple model of a workflow that is supported by the service my team owns. Hacking away at the example code, and looking frequently at the docs, I was able to get up and running in a few hours, creating: a basic model: nodes for system states, edges for operations simple assertions: mainly consistency checks on the states client: HTTP client to implement the operations against the service's API I configured this so that altwalker will perform a random walk of the model, starting state data is randomised, and the client will choose randomly whenever offered an option. Why so much randomness? Because it means that, over successive runs, more of the infinite space of possible workflow executions will be covered. Once I had that basically working I wrote a shell script that would run this loop a number of times: call altwalker ...

Notes on Testing Notes

Ben Dowen pinged me and others on Twitter last week , asking for "a nice concise resource to link to for a blog post - about taking good Testing notes." I didn't have one so I thought I'd write a few words on how I'm doing it at the moment for my work at Ada Health, alongside Ben. You may have read previously that I use a script to upload Markdown-based text files to Confluence . Here's the template that I start from: # Date + Title # Mission # Summary WIP! # Notes Then I fill out what I plan to do. The Mission can be as high or low level as I want it to be. Sometimes, if deeper context might be valuable I'll add a Background subsection to it. I don't fill in the Summary section until the end. It's a high-level overview of what I did, what I found, risks identified, value provided, and so on. Between the Mission and Summary I hope that a reader can see what I initially intended and what actually...

LLEWT 2024

This weekend I was at LLEWT 2024, a peer conference on Anglesey , north Wales, discussing communication. Given the day jobs of the participants, it was no surprise that the experience reports and the conversations that followed them mostly focussed on software development contexts. Notes from my presentation are in Express, Listen, and Field . I made sketchnotes (below) for each presentation and a mindmap (above) to try to summarise the whole. Without much reflection yet, I guess I would pull these common high-level threads from the day: There are multiple reasons that communication fails ... like, duh! ... but having multiple strategies for framing a message can help ... and having multiple tactics for delivering a message can help too. Understanding what you want from an interaction is key ... so setting the context to make that more likely is wise ... which might mean meta-conversation, being transparent, or changing your approach...

PR Coup!

My team uses Dependabot to keep software dependencies up to date. The tool submits PRs against our repositories and we use a checklist/decision tree to help us judge how deeply to review and provide an audit trail of the decision. Patch-level updates of company-internal packages might be just waved in if the tests pass, for example, but a major update of an external library could require us to review or redo a risk analysis. For reasons , the Markdown source for the tree is not automatically added to the PR so we pick it up from another repo, copy it, and paste it into the PR we're looking at. This has got sufficiently irritating to me that I looked for another way. I know about bookmarklets and I've made them before, so I first tried to make one that would paste the text into a comment box on the PR. The checklist is long and I needed to escape various characters and retain line breaks and it was tricky to edit and debug in a simple URL field, so I aborted. I wondered if the...

The Love of a List and a Loop

I attended Maaret Pyhäjärvi 's Intro to Contemporary Exploratory Testing webinar for Codementor this week. In it, she demonstrated exploration with automation on Alan Richardson 's E-Prime checker web site using Playwright to provide control and visibility of the page and PyUnit to make assertions about its state. Nothing unusual there, you might say. There are probably millions of test suites with a similar setup running right now in companies all over the world. And you're right. What's different about this approach is what Maaret did with the tooling, and why. She created a simple skeleton test case, parameterised it to take an input and expected output, and interactively called it with test cases from a list. What she had now was a lightweight testing rig where: Adding a new test idea is adding a new item to the list. A note, bug report, or to-do is a comment above an item in the list. Variant ideas can be created quickly and easily with copy-p...

Exploring It!

This week the test team at Linguamatics held our first internal conference. There was no topic, but three broad categories could be seen in the talks and workshops that were given: experience reports, tooling, and alternative perspectives on our work. (The latter included the life cycle of a bug, and psychology in testing.) My contribution was an experience report looking at how I explore both inside and outside of testing. I've tidied up some of my notes from the prep for it below. There are testing skills that I use elsewhere in my life. Or perhaps there are skills from my life that I bring to testing. Maybe I'm so far down life's road that it's hard to tell quite what started where? Maybe I'm naturally this way and becoming a tester with an interest in improvement amped things up? Maybe I've so tangled up my work, life, and hobby that deciding where one starts and another ends is problematic? The answers to those questions is, I think, almost certai...

Hiccupps

Search This Blog