How can you reduce risk when implementing agile development in your SDLC?
Jan: Why do you think every company isn’t doing crowdsourced testing already?
Phil: One reason is that I think people do have this perception that it’s a lot of people with no skills, clicking through predefined test plans with no skill or human insight -- basically automation with people, or Turk-work. That’s clearly not true in the case of our testers, but there’s a lot of noise in the crowdtesting market that creates that confusion.
Jan: Can you tell us a little bit about what you’re planning to do at test IO?
Phil: I have a deep respect for what the founders of test IO and the team have achieved.
I think the simplest way of saying what I want to do is that we do so much here so well, and I want to make sure first of all that we’re continuously directing all that we’re learning into every customer or prospect engagement, and second of all that the world knows about this.
I don’t want to be the greatest company that nobody’s ever heard of!
Jan: What does that mean for the product?
Phil: One of the most important things we can do in the product is provide more analytics about how our customers’ development is tracking over time. Because provide continuous human software testing with a wide variety of testers, we can provide our customers with a unique set of insights about their apps -- not just how many bugs they have in this test, but also how the testers’ insights about the app have changed over time, and how those track with real-world outcomes.
We’re in the business of testing apps, but the value we provide is not just a list of bugs -- it’s really the confidence to know that you are where you need to be.
Jan: How do you see the testing business changing over the next few years?
Phil: I see two big trends that are particularly important.
The first is the rise of machine learning systems that enable computers to more accurately mimic human behavior. With Google’s recent announcement that they are open-sourcing their TensorFlow libraries, I think we’re going to see a tremendous increase in the number of products that incorporate machine learning in some way. And per Alan Turing, these systems need to be tested and judged by humans. So there are big opportunities there.
The second is the Internet of Things, which, as I mentioned earlier, poses a systemic challenge for software developers and people who depend on software working -- which at this point is everyone in the developed world.
Between those two trends there will be plenty of work in crowdsourced testing, and for test IO, in the coming years.
Jan: You come from the Bay Area but you’re working in Berlin now. How big of a difference is it, and how do you like Berlin?
Phil: That’s a long conversation. When I first walked into test IO during the interview process, it was bizarre to me how similar it seemed to all the other places I’ve worked: whiteboards with branching/merging diagrams on them, TV monitors with system uptime statistics, that slightly-funky informal tech start-up vibe. Silicon Valley really is a state of mind in some ways.
But of course, the Bay is also a real place and a place that’s different from Berlin. Berlin has two things that we really, really need in the Bay Area: great transportation infrastructure, and a reasonable supply of housing that people can afford. So the two biggest expenses -- housing and transportation -- that we worry about in the Bay Area aren’t nearly as big a factor here in Berlin. And they’re also building more housing throughout the city and more subways in the denser areas. Even though Berliners complain that the rents are too damn high and the S-Bahn doesn’t work, they have no idea how good they have it!
Berlin is also a very cosmopolitan place, and the German government is doing a good job of encouraging in-migration of talented people. The local universities are world-renowned and they are much cheaper than their American counterparts. So while the Bay Area (and the United States more generally) has an enormous historical lead in attracting the best technical talent from all over the world, we cannot rest on our laurels.
Jan: How about the Germans themselves?
Phil: I’m loath to engage in ethnic or national stereotyping, so I’ll mostly leave this one alone.
But one thing I will say is that Germany has a reputation for precision in engineering, and the fact that test IO started in Germany and has been extremely successful with German companies should tell you that the quality and precision of what we offer is very high.
Our US business is growing and we are bringing that experience with us. It’s going to be great!
This is part 2 of our interview with Phil. Please see part 1 here.