Iris Pinkster O'Riordain

Professional Testing

This episode was recorded LIVE at the EuroSTAR Conference in Stockholm.

Thank you to our Community Hosts for Season One, Russell Craxford from the UK Department of Work and Pensions, and Gek Yeo of AIPCA.

On this episode, Russell interviews Iris Pinkster O’Riordain, test advisor at Professional Testing and supporting organisations with complex testing issues since 1996. Iris talks about  the importance of sustainability in the present and future of testing and AI.

Iris delivered many successful tutorials at EuroSTAR and was our 2017 Programme Chair. Check out the current programme of talks on the EuroSTAR website.

Episode 2 Transcript – Sustainability in Testing with Iris Pinkster O’Riordain

Russell Craxford: Hello and welcome to another episode of the EuroSTAR podcast. I’m Russell Craxford, your host today. With me, I have…

Iris Pinkster O’Riordain: …and now it’s my turn. I’m Iris Pinkster O’Riordain. That’s always a good name to start with.

Russell Craxford: Multicultural that one, yes?

Iris Pinkster O’Riordain: Multicultural? Yeah definitely. I’ll explain. I’m married to an Irish guy, so that’s where the O’Riordain comes from. But that’s a difficult name in Holland, so Pinkster is the easiest one to say. So anyway, I’m from the Netherlands.

It’s always very confronting to say that I’m working in testing since 1996. So that makes you one of the old people walking around EuroSTAR. Doesn’t make a difference, brings a lot of experience and still learning, I suppose, still after all these years when you walk around, here at the conference.

Russell Craxford: Testing is one of those things, always growing, always changing, always learning, it never stays still. Technology changes, we have to change with it, don’t we?

Iris Pinkster O’Riordain: Exactly, how can you use that in your day to day life?

Russell Craxford: So, that leads nicely into a good question, really. So, you know, what do you, think the future holds? So you’ve lived the last 28 years in testing, probably should have done the math, sorry, what do you think the next few years or the next decades of testing might bring to us?

Iris Pinkster O’Riordain: Well, obviously as you can see walking here at the event, but probably also when you’re reading articles or see where trade is going.

Obviously there has been a lot of automation going on, right? And that is, I suppose, this time and age when you, when you would have that because you have the possibilities, so you would look into how can you use that in your testing job. But I think with all the automation, I’m a test consultant, so I’m at different customers, and what I notice nowadays is that all the tools brought a lot of efficiency. But what we’re missing now is maybe sometimes a bit of thinking. So let’s say my role is now test manager and five years ago, we didn’t need test managers anymore and now there’s a big request for test managers and whatever the name of the role is, you know, that doesn’t make a difference.

But people are looking for whatever end to end or how do we do this? Because I mean, just marking all the boxes with your tooling doesn’t make it necessarily a good test. It’s more like, did we build a system now that actually does what we expected it to do, And that’s maybe where some thinking. I’m definitely sure AI is a big topic.

I’m sure we’ll use more and more of AI because it makes, let’s say, boring jobs easier. So you have time for more challenging jobs.

Russell Craxford: More thinking, as you say, really, the leadership stuff and so on.

Iris Pinkster O’Riordain: Yeah, so I definitely more testing, more tooling, but, I’d say a lot more human interaction again.

Russell Craxford: Yeah. You mentioned AI, one of, I guess things, I’ve noticed about AI in particular is if you see the research on it, it’s the amount of energy it consumes. Both CPUs, and also power. I think it was Microsoft talking about copilot and saying that adding copilot to the thing consumed about 30 percent more energy.

So I guess things like AI might lead us to more questioning our sort of sustainability models and how we can do things more effectively. What’s your take on the sustainability side of it?

Iris Pinkster O’Riordain: Yeah, well, that’s a good question. Because probably that is definitely something not only as testing, but as a whole I T industry, right?

We have to think about and, we know we have to save our planet and not to sound really, you know, getting really into that. I want to change you into becoming vegetarian or, you know, really doing all the things. But from an I T perspective, we do use a lot of, Energy, CPU uses because we have the cloud.

I mean, sky is the limit. It’s easy. You can put everything to the clouds. You can upscale how many companies are actually downscaling. We just scale up somehow. So I suppose also from a testing perspective, we can actually look into, what can we contribute to it? Let’s say one of the things I’m thinking of lately is I just mentioned test automation.

Russell Craxford: Yep.

Iris Pinkster O’Riordain: And then sometimes when I bring the topic sustainability up to a company that I’m working for, they say, Oh, you know what? You have it. That’s great. Why don’t we go back to manual testing? And I was like, OK, but is that the solution? Because what if I’m doing a test where I need 10, 15, whatever concurrent users, let’s say testers? I mean, they all have to drive into work that we all need computers to keep them going. It’s probably difficult calculations, but I mean, I don’t know what’s more beneficial. It’s not like you have to stop all your tooling. You have to be smarter in using it.

Russell Craxford: Efficiency.

Iris Pinkster O’Riordain: Efficiency. And when do you really need it? Maybe if you start thinking about tooling, you can maybe look into, is there any information available? Whether Python is more sustainable than maybe another tool that you, that you’re thinking of.

Russell Craxford: Makes sense, yeah.

Iris Pinkster O’Riordain:  I think those could be the, the starting points anyway.

Russell Craxford: I think there’s a lot of interesting things there, because like, automation frameworks, there are hundreds. There’s probably four or five that are, the key ones, the main market share. And obviously then if you look into the performance of them. So, for example, if you run a hundred tests in one tooling and it takes a hundred CPU units.

And then if you do another tool and it takes 50, then there’s an efficiency of that tooling. It’s not just in time, it’s also environmental impact and coding language is a classic one as well. Compiling it. Do you even need that test? You know, you talked about concurrency and we often, I’ve been in worlds where you write a test for a bug, like something escaped. So you just write a test for the sake of it to make sure that specific bug doesn’t happen. And sometimes you don’t put it in the most efficient place. Maybe it should be unit testing. But actually you write an end to end test. So you’ve got to start thinking about putting the test in the right place.

And it’s for time, for money. And as you say, the sustainability element is becoming more and more, interesting. And more companies are going for sustainability badges to show it. So I guess as testers we’ve got to look at what we do. And, you know, how many servers we need. What environments we need.

Do they stay up forever? Do we take them offline?

Iris Pinkster O’Riordain: Those type of questions. The easiest is give me for testing a production like environment. Do you need it? I have no idea. What testing, how many tests do we actually execute double? Because it’s easy. We have a tool. We can do it all the time. Do we really need to do it? So those are definitely good questions to think about and one of the things that’s playing in my head is we have a lot of quality attributes. Like whether it’s performance or functionality.

Would it be an idea to maybe add one for sustainability? So at least when I’m a test manager or someone who creates a test plan or a test strategy, that at least it makes you think about, do I have to ask some, maybe some questions on sustainability?

You know, is there something that we can, sustainability just like performance and all the other quality attributes, right? It also has to be contributing to the business goals.

Russell Craxford: Yes.

Iris Pinkster O’Riordain: You know, you can’t just force sustainability in some company, let’s say, who is not thinking about it at all. But if your company is thinking of it, it might be maybe something to do, you have to find a definition of what is exactly sustainability. And if you read some articles on it, you will see that sustainability has a broader range than green IT.

Russell Craxford: Yes. Yeah, yeah I’m with that.

Iris Pinkster O’Riordain: IT is just green IT. Sustainability goes from the idea of the added value that a product should have to actually, the value that you have at the end, right? So the product, the use of the product. How is that for, for you as a person? It could be social impact, economic impact, environmental impact, right? So it’s broader and I don’t know whether we can always kind of influence maybe the things around IT. So I think it’s a good thing to start with IT.

But then I think as a tester, you can ask questions about a product. I just give an example.

Russell Craxford: Sure.

Iris Pinkster O’Riordain: I don’t know which Tesla model it is, but one of the biggest ones, the most expensive ones, has this feature that you press a button, there is, I think it’s Beethoven the song is playing, and on the beat of that sound, the doors open and close and there’s lights flashing on the car.

And it’s pretty cool. You might look at it once.

Russell Craxford: Yeah.

Iris Pinkster O’Riordain: Was that really necessary, you know, to build? Does someone really need that in their car? Otherwise, and it’s pretty cool for your friends to show off, let’s say.

Russell Craxford: It’s a gimmick, yeah.

Iris Pinkster O’Riordain: Yeah, it’s more like a gimmick. So maybe, from a tester perspective, you can ask when you see some requirements, maybe you could ask if sustainability is on the list, of a company.

Do we really need this? What does this add to the value of the car, otherwise than a gimmick, for example?

Russell Craxford: Yeah, that makes a lot of sense because sometimes we do do things for the sake of it. And actually that’s human endeavor. That’s, you know, getting up in the morning, that’s eating, drinking. It’s the energy of your laptops, it’s the hardware you’ve got.

It’s that testing that makes sure it doesn’t break every time forevermore for the rest of history. So you’re adding quite a lot of burden, carbon, other things to this world that may not be strictly necessary or may not actually align to your company values. It’s one of the things. I do see a trend that more companies are putting sustainability as one of their kind of missions, values and so on.

So yeah, we as testers, we as engineers, we as software advocates, voices, whatever we want to call ourselves, should be challenging just as much on bugs as on green. And I’ve seen a big movement towards sort of FinOps, financial operations, where certainly with the cloud, you’re trying to maximize cost efficiencies.

It’s some forms that does help with green. You know, you’re trying to cut down the server size, the memory size, all these other things. So there’s some synergy I guess with that, but green is a big push. I work in UK government. And it’s one of the aspects we’re looking at to make sure that we are greener.

We’re trying to teach our engineers what it means, like having an environment up 24/7 when you only run tests for a period of time is not as green as it could be. And, you know, shifting towards, continuous deployment models, using environments, short lived, kill them when you don’t need it. Using tools to help us though, because we want to keep the logs, so therefore we have to upload the logs somewhere else.

So it’s all about minimising impacts, not, not having an impact. Again, efficiency.

Iris Pinkster O’Riordain: Exactly. Yeah, because I mean, we, in this world, let’s say, we can’t go without tools, without products, without software, right? So I mean we will need it. But, it’s just, I think this, why the topic is actually interesting for this podcast is maybe to, give some awareness to people to think of it.

It’s so easy to just go for more and more and more and more, but

Russell Craxford: yeah,

Iris Pinkster O’Riordain: you don’t need it. And in, well, actually from this year onwards, I think there is an regulatory directive CSRD. Right. So I think it’s. For now, it’s only like big companies. I don’t know exactly where the trade off of an amount of people are probably when you’re stock exchange or something, they have to report in a certain format on what you do based on sustainability.

So I’d say more and more companies do have to look into this. And I think from a test, from the testing perspective, because we ask a lot of questions, we see a lot of stuff passing our hands from requirements to actually the software to whatever the software is built into. I think we have a role in that as well to actually question and look into what we’re doing and how can we do it more efficient and well, things like that, like, legal things would help, right?

Because if I’m in a company and I just drop sustainability and people say, yeah, yeah, yeah, they nod and then they go, but the AI stuff that we’re looked into that is a lot cooler. That is you see something straight away. But I hope that, more and more companies do get some awareness and look into what you’re doing and how can you make it climate efficient?

Russell Craxford: Well, yeah, there’s multiple angles of it. Obviously, there’s the activities of testing itself, making them more efficient, effective. Energy is one, but obviously there’s a shift towards green energy now, so that starts becoming a little bit, less impactful, but still impactful. But it’s, the big thing is generally around the hardware, I think.

I did some research a while ago and it was talking about CPUs for computers are one of the worst things environmentally there is, because the process to actually generate them and make them is highly expensive and efficient. So making better use of processes, equipment, using it more efficiently and effectively.

You know, I’ve worked in places before where you end up with three laptops for different situations. It’s just like if hardware is bad for the environment, Yeah. If there’s a way that we can narrow that down by a third, just get two laptops, it’s an improvement. One, even better.

Iris Pinkster O’Riordain: Exactly. Yeah, exactly.

And sometimes maybe you have to make the trade off as well, between the short term where it’s maybe easy to say, okay, I have 20 test environments and I just ask one all the time because it’s easier, but I never end the life of the ones that I don’t use anymore. You know, there could be really simple things on a short term that I noticed, like test environment is really a good one because we keep spinning them up and always forget to take them. We just do a lot of testing and we never check do we test things double or not? So there’s easy things. On the other hand, if we’re testing software that is part of, let’s say, let’s use the car example again, right.

We all think, electrical cars are better for the environment than, non electrical cars. So that’s more the use of the sustainability perspective, right? So not the IT, but just the use of the product. So maybe in the longer term, that would definitely be a good idea on the shorter term, well, you might’ve read these articles as well. That maybe some, maybe if you make the calculation at this moment in time, it’s actually more expensive and creating all these batteries, it’s maybe not as good for you, but on the longer term, it might be interesting, you know. So I think it’s also not everything can be solved on a shorter term.

Russell Craxford: Yes.

Iris Pinkster O’Riordain: That makes it more difficult as well to sometimes think of these topics.

Russell Craxford:  You’ve also got to think about not being perfect sometimes. Like you mentioned electric cars. I think, you know, it said batteries are particularly bad for the planet, lithium and other things.

Iris Pinkster O’Riordain: Exactly, yeah.

Russell Craxford:  But actually the drive towards electricity to use that as a more sustainable energy source, because that’s what it’s trying to utilise, a better energy source, actually has a consequence. And actually that drives the technology though, potentially to look at it. So we’ve used petrol, diesel car engines for eighty, a hundred years now. And they’ve got to a point they haven’t really evolved too much. Now we’re getting to a point where we’re like, let’s try energy. Let’s try electric batteries. Let’s try electric cars, motors and things.

And I think we’ve only touched on the power of batteries and the more we need it, the more motivation there is to improve it. And there’s already been leaps in battery technology, and there’ll be, continue to be, hopefully, and we’ll all benefit from that, but is oil worse?

I guess is sometimes the question. The answer is, I don’t know, but they’re both, mining lithium and mining oil, to me, is both bad for this planet, but which one weighs up? And I think we have to listen to the information that’s out there. We have to be more researching, more looking, not just trusting.

I guess that’s a long story. Leaning to AI as well, rather than just taking the answers that you get given, actually. Do some investigation yourself and look at the impacts of what you do. But yeah, drive your teams, encourage your teams. There’s a good theory on nudging people towards better behaviours.

Changing things so that actually you make it easier for them to do the right thing. And one I’ve worked with in the past is Environment to Environments, which is by default they were always auto destroyed after three hours. So you could never have anything living more than three unless you chose to go into the environment and then change it.

So by default, our environment never lasted that long. And so people couldn’t accidentally leave them running for weeks, months, years, which has happened. You lose them in sometimes in infrastructure. And that helped, I think, because it made people more conscious about extending them. You always got the person that went in and changed it, but we made it hard to change. You couldn’t just go, right, I’ll change this to three weeks. Actually, you had to press a button that would extend it by three hours. And then you had to go through a check process to make it a bit longer. So you can put engineered solutions in.

Iris Pinkster O’Riordain: Exactly.

Russell Craxford:  To help you think more conscientiously about choices you make. We as testers have to think about, is this test unique enough? And do we need a unique test versus things? Can it be tested lower? Yeah. And I think. We do have a part to play in the overall energy consumption of our products. We run more tests, we do load tests of 10, 000 users.

That’s got to have how we do that, the efficiencies of these tools. Yeah. Makes a big difference.

Iris Pinkster O’Riordain: And that is just really how we can start, right? We can’t save the whole planet just by doing, but little processes or little, you know, thinking of processes that you just described.

Just think of the things that you can do. If the business you’re working in wants to change, there has to be an urge from higher management, whatever to say, we’re going to do something about our products in, in total, the whole life cycle of our product. What do we, because if, if the, if it’s the end of life of a product, I mean, how can you recycle it?

You know, those types of things, nothing from a testing perspective, we can really do with that. But in the influence that we have, I think it’s really worthwhile to look at it. Because it maybe it was too easy, to have all your, just ask another one because it got cheaper and cheaper, right, to do everything. And, it’s good to it’s good to rethink anyway, right. Because of test process improvement, we all try to do that. And this is just another topic of looking at it. I think the only difficulty is to sometimes quantify it, you know,

Russell Craxford: Yeah

Iris Pinkster O’Riordain: How much energy is it now to run this test if I do it, let’s say if you want to the example I gave before, like if you have all your automated tests and then the jerk reaction would be “Then we all do it manually”. I mean, it’s very difficult to make the calculation of these ten people. Where do they live? How much energy do they use to come to work?

How much you know, so it’s sometimes difficult to make the calculation, but I think we can feel there is a difference there. Whatever the exact calculation is, we can have an efficiency there.

Russell Craxford: We know it’s worse or better, but we may not know the exact math.

Iris Pinkster O’Riordain: Yeah, you get, right? When you propose a change in a process, people go, OK but what’s the benefit?

Russell Craxford: Yeah, show me the math, show me the numbers. How much less CO2 is this process going to actually contribute to the environment?

Iris Pinkster O’Riordain: Yeah

Russell Craxford: And it’s not as easy to do it. There are quite a few vendors out there at the moment selling tooling to help you model your inputs impact. I’m not sure, obviously, it considers testing as a whole, like environment, things like that. But it looks at travel to work. It looks at equipment costs because they have calculated how much energy a laptop takes, how much so on and so forth. So, you know, technologies getting out there to help make these calculations easier. Yeah, but again, it’s not common, but you’re right. It should be part of our strategies, really.

Iris Pinkster O’Riordain: Yeah. So what I make now, cause I’m doing an assignment at the pension fund. They might be a bit more interested maybe in these type of things because they have a lot of money to spend and know. So that is from that perspective, but then also as a company, they might look into it.

So I’m working on this test strategy now for a project. And I actually put it in, you know, the sustainability, like I just pretend it’s a quality attribute, between all the other things and we have to make choices. If we think it’s important, you have to maybe look for different stakeholders, but that’s for me personally, right?

You just look at, ok, I write this test set strategies, same chapters as I always do. I just add another topic and it means that we have to make choices just like we do. I have only an X amount of time, so much functionality, so much security, so much performance. Okay, there’s another attribute at it, but if there’s requirements for it or someone says we have to be more, whatever, energy efficient, you have to do it.

So I just put it in the strategy and if they say, yes, this is what we’re going to do. It’s just a trade off that you have to make, there’s only an X amount of time. If we think this is important. We have to put time and effort in it. And I can help in my testing to see whether things are more energy efficient or the strategy could be, we’re using tool X now, maybe we can look into another tool or we’re still having this tool selection or whatever your strategy is.

I think that could be definitely something that we can contribute. Also on a written, but not only at the coffee machine, just mention, Hey, what do you think about sustainability, but actually do something with it in your, in your project.

Russell Craxford: Take action

Iris Pinkster O’Riordain: Yeah, exactly.

Russell Craxford: You just reminded me of a couple of things, like where I work, we have six principles of engineering, and one of them is actually about sustainability.

One of the things that we’re trying to guide people to do is consider the environment. We don’t force upon it, you know, principles aren’t kind of forced mission things, but it’s trying to make it more public, trying to make it more conscious. That’s one of the big things, and test strategies are a classic one, that if you’ve got someone in your project team asking about the efficiencies, asking about the cost.

Asking about the impact of the environment. Not just to the user. Then it starts to advocate for that. It makes people think a bit more. It’s a journey. We don’t, we don’t have the solution necessarily, but we start asking the questions. People start considering it. We start making better choices and every little step, every 1% adds up.

And especially in big companies. Big things. And certainly enterprise level software that’s gonna be ran in the cloud for thousands of customers and things like that. The energy consumption does remarkably add up. Cloud has abstracted some of it away from us because we just pay a unit fee type thing and we just, if we want bigger, we just go bigger until you get to a point where you can’t go bigger.

So we do forget a little bit more, I think than once it was running on our own kind of thing. Once we got electricity bills, it becomes less, we can’t associate it. So easily costs impact. But I think it is something we need to do more awareness of, and certainly I think it’s worthwhile we encourage our test communities out there to look into sustainability, how testing impacts it, how software impacts it, how us as humans impacts it, so that we make those conscientious choices.

Iris Pinkster O’Riordain: Now, let that be a good outcome of the podcast, right? That there’s a bit more awareness of these topics, and that maybe the people who listen to this podcast, start looking at their projects and their testing and their test strategies in seeing what they can do little steps and then you see what the outcome is and you can go to the next step and Build it up

Russell Craxford: maybe the EuroSTAR 2025 you might be doing a talk on sustainability. Perhaps who knows?

Iris Pinkster O’Riordain: I’m aiming for that

Russell Craxford: I’m gonna say thank you very much for joining us. I appreciate the talk and it’s lovely to hear more about sustainability. It is, I think, a really important thing, as you said, legislation and other things. Just our environment is so critical, really, isn’t it?

So, it’s good to highlight it. Thank you very much for joining us.

Iris Pinkster O’Riordain: Yeah, thank you for having me.