We’re thrilled to have Rex Black as our latest guest for Testing the Limits. A widely-known figure in the testing world, Rex is one of the industry’s most accomplished authors, speakers and practitioners. In fact his latest book “Managing the Testing Process” has now sold over 25,000 copies worldwide. His 25 years of experience include stints as the president of organizations such as the ISTQB and ASTQB. Rex is currently the president of RBCS, Inc., a leader in software, hardware, and systems testing. To learn more, you should check out the RBCS blog, or follow them on Twitter, Facebook and YouTube.
In this interview, we discuss the last 25 years of software testing; the lack of evolution in test automation; Agile hype vs. agile reality; the benefits of tester certifications; the plight of today’s testing manager and more. Be sure to check back tomorrow for Part II.
uTest: You’ve been in the software testing business for over 25 years now. Assuming the industry did not evolve exactly as you predicted, what has surprised you the most during that time?
RB: Thanks for giving me the opportunity to talk to you and your readers, Mike. I hope this interview is the beginning of a productive conversation between not just you and me, Mike, but also with the readers of your blog and mine.
To answer your questions, here are two bad surprises and three good surprises.
Biggest disappointments: First, slow progress in test automation. The widespread adoption of GUIs, while making life easier for users, lead test automation onto a path where progress slowed after 1990. Keyword-driven architectures and open-source tools like Selenium are starting to get us back on track, though. Second, the continued gap between best practices and common practices. This gap exists across software engineering, but especially in testing.
The gap between best practices and common practices is about 25 years, though we have some real improvements in this area, which leads us to the pleasant surprises: First, the adoption of best practices in dynamic new regions by dynamic and eager new test professionals. In spite of a general problem with best practices adoption in our testing profession, it’s great to see emerging software testing powerhouses—China, Sri Lanka, Malaysia, and India, among many others—quickly and seriously adopting the best strategies and techniques of software testing. Second, the rejection of the mission of the tester as a “quality cop.” When I entered testing, it seemed the reigning view testing (by developer and tester alike) was that testers were there to “sign off,” to act as a conscience of the project, to “stop” bad releases, and, if necessary, to act as an enforcer. This adversarial role always struck me as a bad idea, and over the last ten years I’ve been really pleased to see how we are redefining ourselves and our mission as enablers and supporters of quality.
Finally, the success and longevity of my consulting business, RBCS. Most small businesses and boutique consultancies are gone within five years or so, and most that survive labor in obscurity to eke out a living. RBCS is 16 years old and still going strong. I have the on-going privilege of addressing my professional colleagues with books, articles, webinars, videos, and speeches, and many of these colleagues, in a widening international circle, pay me the compliment of reading, watching and listening. I’m pleased–and inspired.
uTest: In 100 words or less, what’s the difference between agile and Agile? In your view, is agile really the best way to improve the development process? Or is it just hype?
RB: Yes, there’s hype—and dogma. Hucksters gull the gullible with tools and services promising instant agility. People ignore Fred Brooks’ warning, buy (and sell) the silver bullets, and insist that process alone will fix everything. However, agile includes many proven best practices—continuous integration, automated unit testing, code coverage, and more—though many were proven best practices before people spelled agile with a capital “A”. A growing number of our clients transcend the hype, discard dogmatic and foolish ideas pushed by a few agilistchiks, and intelligently and successfully adapt agile best practices. As dogma and hype recede, process improvements remain.
uTest: We’ve heard from a number of testing gurus who are critical of tester certifications – you’re the past president of ISTQB and obviously an advocate. In your opinion, why does the certification debate always seem to get so heated? And what is the biggest misconception about these certifications?
RB: Let me start with a simple and reasonable request to all of the open-minded readers of your blog, Mike. Please read the ISTQB Foundation and Advanced syllabi, and then ask yourself whether the testing world would be a better place if everyone who earned a living as a tester knew the techniques, strategies, and concepts discussed in those documents. I think so, and clearly, a large number of companies and individuals agree with me, as the ISTQB has issued over 160,000 certifications in the last ten years.
I’m not asking people to accept those documents as revealed truth, or to agree with everything we wrote, but rather to adopt the good ideas that are there. I’m also not saying that all certifications are necessarily good, as I made clear in my blog entry on choosing certifications.
As for the “heated” part, I haven’t been in a “heated” debate about the ISTQB certification for approaching ten years, though I enjoy reasonable, open-minded, and frank discussions with supporters and skeptics alike all the time. These conversations occur with mutual respect, as part of a fair debate, in a civilized and mature tone, with an equal amount of listening and talking and no yelling. Regarding those who engage in other sorts of discussions, well, you’ll have to ask them about their motivations for doing so, rather than addressing the facts, and you’ll have to excuse me for reserving my time for other, more productive endeavors and conversations. We can disagree without being disagreeable, and the disagreeable will have to disagree with someone other than me, at least if they want a dialog.
uTest: If you could change one thing about the tester training/certs process, what would it be?
RB: We’re constantly looking for ways to make our training—whether certification and non-certification, for testers or business analysts or managers or programmers—more accessible to people around the world. This is one reason why we have so many webinars, podcasts, videos, articles, templates, and other resources available for free download from our web site, from iTunes, and from YouTube. Yes, we do have to make money, so we have our for-profit offerings like our exam prep guides, our e-learning courses, our on-site courses, and our live courses. Given our emphasis on making as much material accessible to as many people as possible, with so much of that material available on-line for free, I like to say that RBCS is a “not just for profit” company.
uTest: What’s the biggest difference between a good tester and a great tester?
RB: Mike, this is really a central question, and I’m glad you asked it. Obviously, skills, tools, environments, and so forth are tremendously important to testers. I don’t want the answer I’m about to give to detract from the importance of competent teams, well-equipped, well-lead, and given adequate time.
However, the most successful testing teams I know, around the world, have one thing in common: their core mission is to serve. In other words, they recognize that testing only has value when we deliver value, usually to a diverse set of stakeholders. Those stakeholders can include developers who value timely, accurate, objective bug reports. Those stakeholders can include product managers who want to balance schedule risk and quality risk. Those stakeholders can include customers who want good product at an affordable price. Successful testers, as part of successful test teams, identify their stakeholders, understand their stakeholders’ quality and testing objectives, and commit themselves to satisfying their stakeholders’ objectives.
uTest: If you could debate the topic of certs with one person on the main stage at the Testing Olympics (which we clearly just made up), who would make for the best discussion (Bach, Bolton, et al)?
RB: While such a debate might provide some drama, past attempts to have fact-based public discussions with some critics of ISTQB certification have produced much more heat than light. If I want drama, I’ll choose Tilda Swinton or Edward Norton in a nice cool movie theater; I’m afraid I’ll pass on the offer of drama on a hot stage with either of the alternatives you offered, Mike. Besides, the real debate happens in the free and open international marketplace of ideas, and the ISTQB certification is winning, 160,000 certifications and counting.
uTest: Has the life of the testing manager improved at all in the last ten years? Why or why not?
RB: Well, certainly everyone was happy when the big tech downturn ended, though the tech recovery that started in 2005 was pretty weak. We just had a few years—and not exactly luxurious ones—before we got another big helping of pain in 2008, of which we have yet to clear our plates. So, I’d say the 2000s decade was not an easy one, fiscally, and test managers like all managers suffer when money is tight.
It’s been a mixed bag on other fronts. Technology has improved, but to some extent it has also made the systems under test more complex. Virtualization is a perfect example of a technology that both helps test managers by making test environments easier to set up and maintain, while also making production environments more complex and thus harder to test effectively.
At least in terms of people skills, the expanding avenues for training and certification of staff have been a net positive. Back in the 1990s, if you wanted to train a team of testers, you had two options: public courses or on-site courses. Given a good trainer and good materials, each option can provide good training to the attendees, but time away from work and the costs of training put training out of reach for some teams, at least when budgets were tight or projects were intense. One of the reasons we’re so bullish on e-learning, webinars, podcasts, and videos is that it provides the test manager with a way to stretch her training dollars and make training available even during intense periods of project work.
Editor’s note: Continue to Part II of our interview with Rex Black.