DevelopsenseLogo

Why I Am Not Yet Certified — EuroSTAR Presentation

Today, December 4 2007, I gave a presentation at EuroSTAR on “Why I Am Not (Yet) Certified“. James Bach was originally slated to give a different presentation with the same title, but I got the nod due to the untimely illness of James’ wife Lenore, which caused him to cancel his fall schedule (she’s much better now).

Stuart Reid, the chair of the conference, strongly supports the notion of certifications in their current forms. I disagree with that, but I have considerable respect for people who are willing to provide a platform for opposing views, and I therefore thank him for providing the opportunity to speak. I think the controversy opens up the discussion, and thereby strengthens the conference and the craft of testing.

As I said as I finished the presentation, I felt a little like Martin Luther nailing 42 PowerPoint slides to the screen. The talk was generally well received, but there were several conversations that I found rather sobering.

At least two people to whom I spoke–one a former ISEB instructor–told me that they had wanted to effect change in the multiple choice Foundation exams, but their experience was that that couldn’t happen unless the ISEB/ISTQB Syllabus were to change–and changing that proved an insurmountable obstacle for them.

Almost everyone who approached me afterwards said that they were glad that I had said the things that they had been thinking privately for several years. They tended to be enthusiastic but they also tended to check to see whether they were among friends before they spoke freely. The latter is a tendency we need to break. As it was, it felt like revolution and insurrection were in the air–but nobody was quite brave enough to speak up. I encourage people to talk about this stuff, out loud and in public. Open criticism of things that are damaging to the craft is a form of self-certification in my community.

The complacence and chill were disturbing, but once a group of people were together, the complaints started to flow. Many had taken the ISEB/ISTQB certifications. All but one found little to no value in it. They complained about the triviality and the one-and-only-one-answer nature of the Foundation Level exam. Saddest of all, they noted that in Britain and in several countries on the continent, almost all businesses that are hiring testers require applicants for entry-level jobs to have the ISEB/ISTQB certification. I’m pretty certain that this will have several nasty effects. First, it is likely to discourage people from entering the testing field the way many of our best testers have done–by accident and opportunity. In turn, this will make the profession more insular and less diverse. In turn, this will prevent new ideas from reaching the craft. This is very bad.

We’re already learning this business slowly enough. If you attend conferences–especially the major commercial ones–you’ll hear near endless repetition of the same themes: heavyweight planning and estimation for a task that should be nimble, rapid, and responsive; bloated approaches to test documentation and artifacts; relentless focus on confirmation, verification, and validation, and very little talk of investigation, exploration, and discovery. It’s narcotic–the conferences seem addicted to these talks, and they make the craft sleepy. If we’re going to repeat anything, let’s repeat Einstein’s notion that the we can’t solve problems by using the same level of thinking that we used when we created them.

10 replies to “Why I Am Not Yet Certified — EuroSTAR Presentation”

  1. The cognitive dissonance is overwhelming.

    Literally, overwhelming.

    Not only can I not imagine ever working for an organization that required certification, I can’t imagine such an organization valuing my actual skills highly enough to pay reasonable wages.

    I’ll stick with the (uncertified) crazy dreamers, for now at least. They’re more fun and they pay better.

    Reply
  2. Michael,

    As always you have amazing presentations full of great Ideas, and I think this one represent the real problems in the certification projects. I haven’t think about the Cisco ones, and yeah, if they do it, why we doesn’t think on it…

    Thanks a lot.

    Reply
  3. That was a great presentation. Let me share some info from my country.

    The software development industry in Bangladesh is fairly competitive and making good progress. Although it is miniscule compared to India.

    There are no tester certification boards and so no one can take the tests. So although many employers are hiring testers, there are never any prerequisites for certification.

    Also there are no testing training institutes and so at least there is no bad training. Finally, there are literally no books on testing, when you can find abundance of books on programming and software design. So testers have no choice but to collaborate on the internet and with other testers and gather as much relevant information as they can regarding testing.

    I am curious of where it goes from here.

    There was an attempt by a local company to initiate ISTQB certification, but they are so slow and inexperienced in testing that I doubt they will make much progress. Lucky us! Well, it has been about two years and still they are at square one, i.e. starting to register a company 🙂

    A blessing in disguise?

    Reply
  4. Hi Michael,

    I think that the ISEB/ISTQB certifications will not make you a good tester, but I also believe that is good that some enterprises
    require it to hire people because in some countries of the world, testing is seen as a worthless activity into the development process and the certification is a good beginning for the derstanding of what testing is for the applicants that look for certain testing job position. I hope that those enterprises use this requirement just to scare some applicants that believe that doing
    testing is a simple job and they are going to be hired without a problem, and to attract the attention of those who are really interested and have experience in testing eventhough not having any
    certification. What do you think??

    By the way I am not certified and I not planning to do so. 😉

    Reply
  5. Sajjadul: A blessing in disguise?

    I’d guess so. By the way, folks: Sajjadul is someone whose work and whose writing suggest to me that he has the mind of a skilled tester. He articulates his process, his observations, and his thinking very well. That’s not a certification–I haven’t watched him testing–but I would be very surprised if he weren’t a good tester.

    Carlos: I think that the ISEB/ISTQB certifications will not make you a good tester, but I also believe that is good that some enterprises require it to hire people because in some countries of the world, testing is seen as a worthless activity into the development process and the certification is a good beginning for the derstanding of what testing is for the applicants that look for certain testing job position.

    Hi, Carlos…

    I’d contend that it depends on what kind of certification you’re talking about. If you’re referring to the Foundation level certification, I disagree. It’s a toy test (40 multiple choice questions) of trivial nomenclature (not of knowledge), and it uses the approach of one and only one right answer. Skilled testing doesn’t brook that kind of thinking, to me. If you’re talking about a Cisco-style certification, then I might be inclined to agree with you.

    (In the Cisco program, you’re given the task of fixing a broken router. If your process is reasonable, and if you can articulate it, then you pass the test, maybe even if you don’t fix the router. This emphasizes that even a skilled network engineer and appropriate heuristics can be fallible; you might not have fixed the thing because you were unlucky. On the other hand, if your process is not reasonable or you can’t articulate it, then you’ll fail the test, even if you fix the router. The issue is that even an incompetent engineer and an unreasonable process can be lucky, but we’d just as soon not depend on that.)

    There’s no need for organizations to scare incompetent or unqualified applicants. When testing is seen as a worthless activity, it’s largely because worthless work is being done by worthless testers. A worthless exam is hardly a way to break this cycle.

    A company that insists on an ISEB/ISTQB certification as a requirement for employment attracts my attention, too–just not the kind of attention that I infer they’d like to attract.

    What do you think?? By the way I am not certified and I not planning to do so. 😉

    I think, on this point, your actions speak far more clearly and accurately than your words do. 😉

    Reply
  6. Here’s my own not-so-brief 2 cents. I have seen (very recently) attempts by non-qualified people to quickly gain certifications (CSTE, etc.) to justify an internal power-grab of (insert_testing_role_here). Instant test expert, just add water with a spoon of cluelessness and a pinch of talent – shaken, not stirred?

    In my opinion, it is far more valuable at the organizational level to value internally developed certifications. Preferably multiple types based on various disciplines in place at your company. For example, at the early start of my career at a company called Quartedeck (where my career in SQA began) we had internally developed “certifications” for the technical support of our products. Each product had one or more study guides and tests, created by the more experienced techs (Michael?). These “certifications” were specifically designed to prove your subject matter knowledge of our applications. They were often difficult yet not impossible and truly provided a vehicle for knowledge growth (both personally and departmentally). Other benefits were feeling a firm sense of accomplishment gaining some respect amongst your peers and managers for the effort. All in all it made for a damn fine tech support team, fully capable of dealing with supporting our software over the phone. And, since the more “talented” techs tended to eventually move to QA or Development, made for better testers and developers. Which should be the primary goal of any certification program, homegrown or otherwise.

    So, consider creating your own certifications for your team or organization. You’ll be happy you did.

    Reply
  7. Thanks for commenting, Will.

    I don’t remember having created study guides or tests–except in the sense that I wrote a goodly number of technotes at Quarterdeck. These were written mostly to help provide quick answers to our customers, but I was very conscious of the fact that they served to improve my own understanding of the products, and that they were valued internally. I also was a relentless participant in the internal support bulletin boards. Those discussions and the richness of the learning available are missed by everyone who was involved with them, at least in my circles.

    I don’t think I ever took any of Quarterdeck’s internal certifications, but I got a fair amount of recognition for technotes and internal and external support. These were things that the bosses saw every day, so in my case, certification was continuous. I agree with Will, though, in that, all other things being equal, internal certifications stand a somewhat better chance of being relevant to some constituency and context. On the other hand, my recollection was there wasn’t a one-to-one correspondence between expert techs and certified techs, Will’s certification and his expertise notwithstanding.

    My recollection is that the diversity of skills, interests, and backgrounds was the principal asset of the technical support department, certainly when I arrived. Another thing that helped was the diversity and complexity of the problems that we had to solve. Yet another was a culture that valued all that stuff. I’d be prepared to argue that we had better techs before the certification movement. That’s a matter of history; I’m not arguing that the certifications did much harm and they may have helped some people to some degree.

    Reply
  8. Thanks Michael for sharing this presentation with us. You made me think about it and gave me some new awareness.

    I have to admit that I’m certified. The initial reason was to express to the world that I believe in Software Testing as a profession were you need skills. And the certificate was a sign that I know/knew the terms.

    Over the years the meaning of having the certificate changed. It changed from a sign of my self to be proud of into a tool for account managers as customers are asking for testers with certification without guarantee of the level of skills.

    I think in your presentation (and off course J. Bach) you express very well that we shouldn’t rely much on the note having a certificate.

    Only it helped me also personally. As said initially, showing that I take my profession seriously by showing that I take courses to increase my knowledge. As starting tester it is hard to attend seminars as speaker to show you have something to say or write articles. So organizations are calling to hire you.

    It also helped to give me some borders of software testing and make me think beyond those borders. It also helped me identifying other information which might be useful in our occupation.

    Is certification useful? Depends on for whom. It might be useful for:
    – Sales to obtain assignments.
    – Organizations to make the initial selection, only they have to admit the risk that there are also very good testers without certification.
    – Testers to learn the basic of testing and identify borders

    And it will be more useful for those who start the certification program to read your presentation and this blog to learn that having a certificate doesn’t turn you into the wizard of test.

    Reply

Leave a Comment