There’s a lot in the press focusing on the negative consequences of the Internet of Things — and surely it raises some privacy and security concerns — but there are positive results besides how it will integrate and automate our world. IoT is changing the way teams design, build, and test. IoT is helping to build better, stronger, more efficient and thorough teams.
IoT further eliminates what Michael Bolton, founder of Developer Sense and co-creator of Rapid Software Testing classes, calls the “Hot Potato Theory of Testing” — the you-had-it-last syndrome. “To suggest that a tester is responsible for a bug, that suggests that the tester is responsible for every aspect of the product,” he explains. “Responsible people understand the premise that a product is a result of work by many people [and] problems in a product are the result of many people.”
He says this is a part of a greater shift in software development, where programmers are also taking responsibility over more of their work.
European QA testing expert Paul Gerrard says that the tester should be treated like a trusted advisor. He compares the relationship between developer and tester to a young king and an older advisor or a pilot and navigator system. “In some ways, the developer’s the hot shot. The developer just wants to write code, he’s not thinking beyond. Now the tester must ask: How will it work in isolation? In an integrated system?”
No Room for The Waterfall Method in IoT
Gerrard echos Bolton that, in the Internet of Things, there is just no room for what he calls classic waterfall: “It’s not my problem, just pass it down the production line” because in IoT that’d be simply too much of a financial risk. IoT forces testers to move further left and enables more cross-team collaboration than we’ve ever seen before. Like in scrum framework, and in the increasingly popular microservices trend, the roles between developers and testers will become more blurred and testers will be sitting in on development meetings and even joining development teams.
“Testers are there to protect the interests of the stakeholder but also to protect the interest of the developers” Gerrard says.
Conversely, developers for the IoT space have to design more with the testing in mind. Gerrard says to “Design for testability because the problem won’t be in building them but in our abilities to test them,” pointing to the fact that it’s not hard to design sensors, but to build them for a smart city, it is a bit like an operating system — immensely complex.
“I see the role of testers shifting from people who design and run tests of existing software to becoming Worriers and they just spend their time asking kind of awkward questions,” even joining DevOps day-to-day.
Diwakar Menon, CEO of Last Mile Consultants, observes that “the horrendous complexity that’s on the way is promoting a new model of testing as a way to get into the heads of what testers do. It’s a picture of the different thought processes in your mind — ‘If that’s the way testers think, crikey, now on left hand side that’s exactly how developers think too’.”
And it’s not just developers and testing coming closer. “Marketers are getting the budget to build digital transformation projects, going from bricks and mortars to online. Their mentality and approach to life is quick market testing. The implementation cycle is very short.” Menon says that the whole business world will experience a change with rapid release cycles and continuous delivery, “shifting left and bedding with developers and joining their world.” And testers won’t just bed with developers but designers too.
Vlad Trifa, co-founder and VP of research and development, EVRYTHNG IoT platform and of WebofThings.org and the book of the same name, argues that, at least in the IoT space, quality analyzing starts with testing the design. “Don’t read the spec at the end but you want to make sure the QA guy that’s going to be testing is in the same room as the designer up front. Equally, the QA person understands better who the people are who are going to use it. It’s all about a much better collaboration between the two.”
He went on to say that it’s not just going to be QA at the end but it’s going to be many smaller cycles throughout the process, starting with QA on-board with prototyping: “It’s going to be many more smaller cycles, literally in a giant scrum process of design,” putting QA into the early phases of design and development. “Iterative design and improvement will just make things so much better.”
Bruce echoed his contemporaries saying,
“For the Internet of Things, testers are your superstars. They’re the ones that are going to be identifying what went wrong.” He continued that, “getting that feedback from those testers is going to be vital to be sure that the dev and the ops processes are truly rapid release.”
Even at giant Microsoft, there’s movement away from trying to monopolize the market and toward everyone participating. “We don’t care what devices you’re using, and if we can get Windows on your devices, that’s really cool. We just released Windows for Raspberry Pi 2. That’s cool too — we can meet you in the data analysis in the cloud even though we are a device company,” Mulcahy said. “I think in the long haul, the Microsoft story is ‘Yes, we are very much aligned with enterprise and we have the ability to support thousands of devices without messages being able to be dropped, support[ing] security’.”
But companies like Microsoft have to be aware of what those devices that connect will be and how people will use them. Only then can they test for almost every situation. This is where testing doesn’t seem so changed at all.
In the Internet of Things, it’s never been more important to have the right tools for testing your API
Whether you’re looking to automate functional testing, load testing, or security testing — SmartBear’s suite of API testing tools will ensure your team is ready for the challenges and opportunity of IoT.
Security in the Age of IoT
How the IoT Will Change Security Testing
Automation & IoT Testing
Automated Testing is Key to IoT Success