In this on-site interview from STARWEST 2016, Jonathon Wright, the director of software engineering at CA Technologies and a speaker at the conference, joins Josiah Renaudin to discuss the Internet of Things, artificial intelligence, scaling for load, and virtual reality.
Josiah Renaudin: So many people in software worry that when something new like the Internet of Things or artificial intelligence debuts, they have to undo all the work they’ve done up until this point. They need to change how they look at testing, development, QA, and everything else along the lifecycle. For those people who are dealing with change, what advice would you give them?
Jonathon Wright: Don't panic. Digital is going to change things, but now, those challenges are going to be the same as what you did before. You'll still be using classification trees; the difference is you'll be applying them in a sense of testing an IoT platform or a new banking API. You have to still use your skills.
Josiah Renaudin: Same concepts, but in a different way.
Jonathon Wright: Exactly. People will be scared. People will go, "How do I test something I can't physically see, or how do I know the machine learning algorithm is right? How do I train it so that I can use huge amounts of data in a big data set scenario?"
Josiah Renaudin: So what you can do now is prepare them for that, so when it happens, they won't freak out, because they realize it's similar.
Jonathon Wright: It's exactly the same. That's the great thing. That's why I was trying to bring things down to do the node left thing. I was told when I started college that everything was just a node. I thought they were just talking a bunch of crap. Then I thought, actually, if you take it down to just nodes, it's very simple. Whether that node's an IoT device or sensor, or it's just an API, it doesn't matter. Part of it is you're proving a path through an application with some data and you're getting confidence, the more you learn, the more you repeat through there with different sets of test data and try different negative exhaustive testing mechanisms. You're getting more confident.
That's where part of me talking a higher scale of whether it be smart homes, solar power, nuclear power stations, it doesn't matter. The challenge sounds difficult, but in actual fact, it's the same as what you've always ... Business as usual. Just don't panic, is the big message. A lot of the tools that we've got now are really to support how you go about doing this in a very simple way. That's a lot to do with visualization.
If you're able to visualize something, you can understand the model. I think the days of coding have disappeared. Someone from the business has got to understand the value of what you're doing, and understand the model and what you're proving. It has to be in a very simple format, whether it is a very complex mission critical system, or it's just another app. You need to visualize that.
A lot of data visualization and pipeline visualization is making it simpler for people to interact with very complex systems, but in a way that everyone can understand what we're doing and what value we're adding, not just build another thousand tests. Why another thousand tests? What does that mean? What confidence do you get from doing that? This then leads into the whole digital assurance side of things.
There's a whole stack of stuff I could talk about for days and years and months. It is really exciting. It's been an eye opener for me for the last few years, of how much of big digital has suddenly changed. Like you said, it's when someone starts trying to come into and disrupt in an industry. Take FinTech for an obvious thing. I spoke to somebody here who's attending STARWEST who said that in the US, they're thinking about this, creating banks which have a coffee shop built into the bank, so that people can hang out and there's a really good vibe. It's like a coffee shop and a book store.
I was thinking, wow, that's very different culturally to what they're doing in the UK, where we're actually closing banks down, and we're replacing them with BlockChain and FinTech organizations that let you use ... Mining blocks of data. It is very different to encouraging people to come in and socialize, and as a new disruptive way of getting into your bank.
Josiah Renaudin: The culture of it, yeah.
Jonathon Wright: Maybe that is the way, is that you interact in a social environment, and you're able to consume services in a more secure environment. That's maybe the way banking will be in the future. We're seeing a definite cultural shift. If you're not able to respond to a disruption, then you're going to get these phoenixes which are these new companies, these lean startup companies, that will consume larger organizations that can't pivot their business model in time, and it takes them two years, three years to respond to disruption.
I spoke to a thought leader the other week in Chelsea who is doing the response to the Airbnb app. This is Hilton and IHG coming back and saying, "Yes, guys, it took us two years to come back with a response to Airbnb, that will then take some of that market share back." That's them realizing they have to evolve and change their systems, but it's taken them two years to come back. By that time, they've already eaten such a large percentage of their business. Paris Hilton must be downsizing her mansion or whatever.
In the meantime, the challenges which we get with testing, which we've had for the last few decades, which is do more with less, and now do more with less, but faster, more rapidly, and prove hypothesis quickly with levels of confidence. These are new types of challenge. I think people will panic if they don't have the same amount of time.
Josiah Renaudin: Yeah, if they wave it off, if they wait and see if it's going to happen, instead of actually being proactive and being anticipatory about that.
Jonathon Wright: This is why what I talked about blazed me, the whole performance engineering aspect. Performance engineering is about bringing it shift left, is performance engineering from day one. Someone comes up with an idea; we're going to do the response to Airbnb. Okay. What are we expecting as far as scale? Airbnb's got 20 billion transactions a day, then we need to be able to build it for 20 billion transactions. Let's not wait until we've designed the app, we've created the app, and then see whether or not we can scale it on AWS or Azure or whatever the platforms or service is. Let's start building this in from day one, and think about how we've got these challenges.
This means testers have a very different thing. I call it DesignOps, which is taking it away from dev. Dev's too late. Design is the first stage. You're going to have to start doing very quick from design to production, very quick, prove hypothesis, fail fast, learn rapidly. This won't scale, this won't scale, we need to change this. Performance, integration, API, whatever it may be, those challenges are going to be right at day one, not six weeks before someone's built an app and then decided they've got the wrong framework, the wrong architecture, they've designed it wrong, they've had the wrong requirements. It's too late.
I think this is the new shift, is that we're going to see every part, whether or not it's mode three, which is these old systems that are now going to be dragged over to this lean UX, lean startup world where they go, "What can we do with the data we've got in our systems, and how can we add value to our customers?" They need to be able to innovate at every single stage, not just in this new idea concept, this incubator area. You've got to be able to scale, evolve your current systems at the same pace. Otherwise, you won't be around in ten years.
This is the scary thing. I know people at AMEX, who are a customer of ours, who come to this event, they were asking these questions ten years ago. How do we deal with the Apple Pay, the new generation? We've got to build new services that make us desirable. They've been thinking about that for ages. So has everybody else in that area. You've got to be proactive about it.
Jonathon is a strategic thought leader and distinguished technology evangelist specializing in emerging technologies, innovation & automation with over 15 years of international commercial experience within global organizations, and is currently Director of Digital Assurance at CA in Oxford in the UK. His practical experience and leadership in the area of DevOps and Digital Assurance has seen him in demand as a speaker at international conferences including Gartner, Unicom, HPE Discover, Oracle Digital Forum, STARWest, STAREast, and EuroSTAR, where he was awarded Innovator of the Year (2014). He is the author of several books on test automation as well as numerous online webinars, podcasts & training courses on Automation and API Testing. With Jonathon’s practical insights into real world application of the core principles and methodologies underpinning DevOps and Digital Assurance, his presentations are not to be missed.