An Interview with Justin Hunter @Hexawise
Current Role/Designation: Founder & CEO
Location: Chapel Hill, NC, USA
Justin Hunter, Founder and CEO of Hexawise, is trying to change the way that people test software. He is an enthusiastic advocate of using test design methods that improve the efficiency and effectiveness of test case selection approaches. The improbably circuitous career path that led him into the software testing field included working as a securities lawyer based in London and launching Asia’s first internet-based stock brokerage firm. The Hexawise test design tool is a commercially supported web-based test design tool currently being used by more than 100 Fortune 500 firms and thousands of other companies.
1. What did you do before entering the software testing field?
My father was a statistics professor and consultant to businesses with an expertise in the field of Design of Experiments. He was passionate about his job; dinner table conversations at my house when I was growing up frequently involved strategies my dad was urging companies to adopt to improve the quality of their products and service offerings.
After I graduated from university (where I had studied International Economics and Chinese), I went to a joint degree graduate program in law and business and earned my JD/MBA. Leaving graduate school, I had an interest in working internationally. I worked as a lawyer for a couple years in London and Hong Kong, worked an entrepreneur for a couple years in Hong Kong and Ireland, and then worked for 8 years at a large global IT consulting and systems integration firm, first working as part of their venture capital group, then (when that shut down), I worked as a “special strategic projects guy” for a couple of the firm’s global leaders in its Financial Services practice.
2. How did you get into software testing? Was your path planned or did you end up here by accident?
Does anyone in the software testing field wind up here as part of deliberate career progression plan? If so, I sure haven’t met many of them yet! Like most, I got here through a weird and winding road. I was already 35 years old before entering the software testing field.
In my last major project I had while working at the large systems integration firm, I was asked to do a competitive analysis of the firm’s software testing offerings and make some practical recommendations that would make the firm’s offerings more competitive.
Some of my suggestions were, in retrospect, pretty mediocre and were rightly ignored. The best idea I proposed, though, made a big positive impact on teams. I suggested making major changes to the process by which tests were selected. The process changes I recommended were based on “Design of Experiments” methods like my father used to teach his Chemical Engineering students about. It turned out that adopting this more rigorous and scientific approach to test case selection led to breakthrough improvements at the firm in every project it was implemented on. We were regularly more than doubling the number of defects teams were finding per tester hour with this approach. As results of the test design changes I helped implement continued to roll in from project after project, the firm realized they had developed a powerful competitive differentiator and they presented me with a global “Innovation Award” for my contributions toward developing it. I’m proud of that award. I still keep it in my office.
3. When did you realize your passion was software testing?
After a while, my boss at the large IT systems integration firm asked me to wrap up my software testing project and move on to other projects unrelated to software testing. I found myself unable to do that. I was having too much fun and the people I was helping were continuing to report that the test design changes we were putting in place were generating dramatic benefits. I had become obsessed with “spreading the word” of these test design approaches within the firm and working with teams until they understood how these improved test design methods could be applied on their projects. Without any formal company approval or mandate, I began to host global conference calls with any and all project teams who would listen. By that point, there was no turning back.
The experience of collaborating with software testing teams during that time changed the course of my career. I wound up leaving the big systems integration firm and founding Hexawise. As a company, Hexawise is focused on two things. First, we have created a powerful, easy-to-use test design tool that allows testers to quickly create unusually powerful tests. Second, we provide clients with test design training, tool customizations, and live test design support by phone and screen-sharing (available on demand via a “chat button” within our test design tool).
4. Design of Experiments-based test design is not a common term. What does it mean?
Design of Experiments is a field of Applied Statistics that is almost 100 years old. For decades, extremely smart people have been coming up with progressively more clever ways to design experiments in ways that will lead to the discovery of actionable information in as few experiments as possible.
Software tests are experiments. Software tests generated using Design of Experiments-based approaches are dramatically different than most software tests that are documented, one-at-a-time by tester. That’s because Design of Experiments-based tests scientifically maximize variation in tests, and scientifically minimize the amount of wasteful repetition. In addition, Design of Experiments-based sets of tests are structured to include a greater variety of interactions between different elements of the System Under Test designed to uncover defects more thoroughly and efficiently than standard tests.
The most common Design of Experiments terms in you’ll hear in software testing are “pairwise”, “all pairs”, and “orthogonal array test design” (AKA “OATS”). When software testers create their first few sets of pairwise tests / orthogonal array-based tests, they’re likely just starting to scratch the surface of the potential benefits that Design of Experiments-based test design methods offer. As testers get more experience designing tests using these and more sophisticated Design of Experiments-based methods, and develop more experience thinking through how different parts of a System Under Test might vary in interesting ways, the testers gradually realize that this powerful test design approach is applicable in far more places than they initially thought it would be.
The software testing field is still in its infancy. In the coming years and decades, software testing practitioners will incorporate lessons from other industries and modify them as necessary to fit within the software testing field. I am confident in predicting that Design of Experiments-based test design methods will be much more broadly adopted 20 years from now than they are now.
5. How can you be so confident that adoption of these test design methods will grow?
All of the evidence I’m aware of points to that conclusion. If you just take a look at us, we’re slightly over 5-years old as company. Our test design tool is already used at more than 100 Fortune 500 firms and 1,000’s of other firms. Charts I’ve seen of the number of practitioner experience reports, the number of conference talks, and the number of academic papers on the topic all go up and to the right.
And besides, large efficiency and thoroughness benefits are readily apparent if teams simply take the time to execute 50 “regular tests” and 25 “optimized tests” in parallel. When benefits are so large and so consistent, increasing adoption naturally follows. That has been the consistent trend with Design of Experiments-based methods in other industries as well; it’s just that software testing is a couple decades behind in this area. A final data point is that, Gary Gack conducted a survey of software testers and found that 96% of the testers who had used these test design methods found them to be useful.
6. Do you regret being associated with software testing today? Given a chance would you move from testing to any other field in IT?
That’s crazy talk. As a general response, there has never been a better time to be in the software testing field. There are more interesting developments and challenges than ever before.
And for me personally, I’ve got a mission here that is decades away from being realized. I’m very happy doing what I’m doing.
7. What kind of a software testing tool is Hexawise?
There are a lot of tools out there, so let me first clarify what Hexawise does, what it doesn’t do, and how it fits into other tools and processes. Hexawise is a test design tool. Hexawise helps software testers quickly select and document sets of unusually powerful tests. It is not a test management tool like Quality Center; testers will quickly design powerful tests in Hexawise then import them into Quality Center / ALM. Hexawise is also not a test automation tool like QTP or Selenium; people can export test data tables from Hexawise and import them into test automation tools.
The reasons that testing teams like to use Hexawise vary quite a lot. Ask 10 Hexawise users and you’re likely to get 10 different answers. Common responses include:
— It helps me select and document my test scripts faster by partially automating the selection and documentation process.
— It saves me test execution time; Hexawise-generated tests scientifically minimize the amount of wasteful repetition which allows me to execute fewer tests (when fewer tests are what I want).
— I consistently find more defects; Hexawise-generated tests scientifically maximize the coverage of interactions between test inputs in my system that help me to find defects I would not otherwise have found.
— When we use Hexawise, my colleagues and I tend to ask more and better questions because we’re required to develop a clear understanding of the moving parts of a System Under Test before we can generate tests.
8. What makes the Hexawise tool stand out from other test design tools?
Our customers consistently tell us two major things stand out about Hexawise. First, as a tool, Hexawise is famously easy-to-use and we customize it for enterprise clients to integrate it into their tools and processes. Second, we give clients much more than just a tool; we also provide test design training and live test design support (with chat and screen sharing) that companies often find is necessary to introduce and successfully rollout new test design processes across multiple of teams of testers.
9. You were one of the co-authors in the “How to Reduce the Cost of Software Testing” book. What made you want to co-write the chapter on “Science-Based Test Case Design: Better Coverage, Fewer Tests”?
Software testing, as a field, remains in its infancy. Other fields, such as manufacturing, agriculture, and advertising, have used scientific methods to design tests ways that reveal as much actionable information as possible in as few tests as possible for decades. At this point, in 2014, far too few software testers are aware of this extremely powerful approach to designing tests. I want to help change how software testers test software.
10. We noticed that Hexawise is your third start-up. What drove you to get into start-up companies?
My relative ineptitude as a corporate lawyer had something to do with it. I’m good at a few things but being a corporate lawyer is not one of them. The first “real job” I ever had was working as a corporate lawyer for the Hong Kong office of a UK-based law firm. The job required a skillset and temperament that I didn’t have. When this realization began to dawn on me and the partners who had brought me to Hong Kong, I began to look around at more entrepreneurial opportunities.
I quickly found one of the best jobs I’m sure I’ll ever have in my life. I joined a pre-launch team and helped build and launch Asia’s first internet-based stock brokerage company. I helped the CEO raise a couple million dollars from venture capitalists and we built a stock brokerage from the ground up. It was a fairly insane plan given how many challenges we had in front of us but we actually managed to pull it off somehow. Perhaps we succeeded because we were too young and naïve to realize we shouldn’t be able to. That company is still around and customers can execute trades on more than a dozen different stock exchanges.
11. Do you have any advice for our readers on what to look for/out when creating or joining a start-up?
Get ready for a roller coaster ride. In my experience, good days at a start-up are much more fulfilling than good days at a huge company and bad days at a start up are worse. While the hours tend to be longer, many people find it’s worth it given the huge impact individuals at startups often have on building exciting new products. If you’re at the right place at the right time, your startup may even create an offering that defines an entire new product category or you might help transform an industry.
It’s hard to put into words how satisfying it feels when you hear people say complementary about something you’ve worked tirelessly to create.
12. What is your next big idea?
I’m not like most people on my third start up. I’m not looking for some other big idea to move on to. This idea is more than big enough to keep me occupied for years and I’m personally attached to it because of my family background. This feels like what I should be doing with my life; it feels like “home.”
Let me share a quick story to put that into perspective. My father believed deeply in the power of Design of Experiments methods to transform the quality of people’s lives in addition to simply helping improve efficiency and the quality of products and services. As a result of his beliefs, my father, mother, brother, and I moved to Nigeria in the mid 1970’s. My father taught at the main university there for a year motivated by a strong desire to help people and make a real difference in their lives. He helped Chemical Engineers design far more efficient experiments. He didn’t do it for the money; he did it because it was personally rewarding to help people become much more efficient and effective. I too get deep satisfaction out of spreading awareness of better test design method and helping software testers design far more efficient software tests. It’s really satisfying to think that I’m following in the footsteps of my father in my own little way.
13. According to you, what is lacking in today’s commercialized training industry, especially in testing?
Officially recognized tester certificates for “Rapid Software Testing” testers would be good. Maybe I’ll suggest that idea to James Bach.
I’m kidding. I don’t follow the commercialized training industry very closely but I think that additional course offerings on software test design would probably be useful in particular. I can tell you that we have clients who are eager to hire testers with strong test design skills.
I respect Cem Kaner enormously and have been impressed by his recent comments on commercial software testing training. One of the things he’d like to see emphasized more is testers working through hands on exercises vs. simply filling out multiple-choice answers. It will be interesting to see how things develop in this area.
14. What qualities will you look for in a candidate when you want to recruit someone for software testing job?
Curiosity, intelligence, initiative, communication skills, and an ability to quickly pick up advanced test design skills.
15. What will you suggest to people who want to join IT industry as software testers?
Expose yourself to a wide variety of perspectives by reading and participating in the software testing community online. It is shocking to me how many testers I meet have not taken the time read any software testing books or blog posts on software testing.
Read a lot of different books and blog posts by software testers. After a short while, you’re sure to find a few authors who say things that resonate with you and begin to make you think about software testing in new and different ways.
— Follow testers on Twitter; it is a surprisingly active community and you can easily find testers who share thought provoking software testing topics. Many testers post links to interesting blog posts, articles in testing magazines, etc.
— Follow conversations on online testing forums like softwaretestingclub.com.
Finally, if your goal is to maximize your earning potential in the short term, learn Selenium; if your goal it to maximize your long term satisfaction as a software tester, learn test design skills, and start learning about Context Driven Testing and Exploratory Testing.
Why these three things in particular?
— Selenium is a test automation skill currently in very short supply.
— Well-developed test design skills will always be in demand and improving those skills is a fun, satisfying endeavor for many.
— Learning about Context Driven Testing and Exploratory Testing will help make you a better tester regardless of whether you agree with everything CDT leaders express. And the leading thinkers in both of these areas are unusually good at challenging testers to look past superficial issues and think critically about important software testing questions in fresh new ways. Many testers who feel burnt out and frustrated at their jobs find new enthusiasm for their careers when they discover the Context Driven Testing community.
16. Name a few people you would like to thank, people who helped you directly or indirectly in your career as a software testing professional.
Wow. Great question. Too many to list, but here’s a few:
— My father, William G. Hunter and his friend and mentor George Box for exposing me to the field of Design of Experiments
— Cem Caner, James Bach, and Michael Bolton, as a trio, for thoughtfully challenging traditional, entrenched, concepts of software testing.
— Elisabeth Hendrickson because she’s awesome, her book “Explore It!” is terrific, and her testing heuristics cheat sheet is inspired.
— Jon Bach because his modest, inquisitive manner is a good reminder to us all that we’re all still learning.
— Rosie Sherry for building an impressive community at softwaretestingclub.com and for publishing the Testing Planet.
— Authors of impressive test design books, Lee Copeland and Rikard Edgren.
— Aaron Hodder who, for being as low key and unassuming as he is, gave one of the most sensational presentations I’ve ever heard at CAST.
— Shmuel Gershon, another friendly unassuming guy who is extremely talented and knowledgeable (not to mention the fact that he created Rapid Reporter).
— Harry Robinson, great speaker, great thinker, creator of my favorite software test of all time, and model-based-testing guru.
— Alan Page, both for his astute insights and his curmudgeonly quips on Twitter.
— Jason Huggins, for his role in creating Selenium and his quirky robot-o-phile ways.
— And the thousands of Hexawise users who have given us the motivation to keep improving our solution (especially the ones who care enough to complain when they experience problems.
17. One last question – Do you read Testing Circus Magazine? If yes, what is your feedback to improve this magazine?
I do. I like it.
I would suggest getting more detailed experience reports if possible. Easier said than done, I know.
Thanks for the interview.
Blog – http://hexawise.com/blog
Twitter ID – @Hexawise
* Interviewed by Jay Philips. This interview appeared in our April 2014 edition.https://www.testingcircus.com/an-interview-with-justin-hunter/https://i2.wp.com/www.testingcircus.com/wp-content/uploads/Justin-Hunter-1.jpg?fit=971%2C1024&ssl=1https://i2.wp.com/www.testingcircus.com/wp-content/uploads/Justin-Hunter-1.jpg?resize=150%2C150&ssl=1Interview with TestersInterview with TestersJustin Hunter Organization: Hexawise Current Role/Designation: Founder & CEO Location: Chapel Hill, NC, USA Justin Hunter, Founder and CEO of Hexawise, is trying to change the way that people test software. He is an enthusiastic advocate of using test design methods that improve the efficiency and effectiveness of test case selection approaches. The...Testing CircusTesting Circus[email protected]AdministratorThis article was posted by Testing Circus Editorial Team. Testing Circus is one of the world’s leading English language magazine for software testers and test enthusiasts. The magazine is read by thousands of software testers worldwide.Testing Circus