Certification Wars
I started diving in the late 70's, and in1985 I became a PADI certified open water scuba instructor. In those days, scuba instruction involved a lot of classroom time discussing various concepts that impacted human physiology such as Boyle's Law, Dalton's Law, Haldane's Principle, and also immediate responder first aid for scuba diving related maladies (most doctors are not trained in hyperbaric medicine and divers were generally more aware of potential symptoms of DCS as compared to most doctors). The training at the time emphasized in-depth knowledge of the theory of diving as well as repetitive practicing of fundamental and critical skills in controlled environments (usually a swimming pool) and 'real-world' environments (the open water). Certification entailed successfully completing multiple written examinations as well as demonstrated competence in the water.
I remember those early days when PADI and NAUI instructors would compete for students, and many instructors would ridicule other certification agencies while claiming the agency they supported to be superior as compared to the others. Some instructors would completely denounce other certifications as inadequate or claim that those certifications were not accepted worldwide, or make some other malicious attacks based on ignorance and their own personal motives.
Of course, we know that some people are incapable of thinking on their own, or they are easily persuaded with scare tactics, and some simply blindly follow the bloviated jabberwocky of charismatic people. But, most intelligent people who are capable of processing cognitive rational thoughts are able to see through the prevarication and realize the hypocrisy of people who lambaste some certifications while selflessly pander their own certification.
My personal views on certifications in software testing haven't changed. I can see both the perceived benefits of certifications by employers as well as the limitations of certifications. For example, certifications in software testing or other professional disciplines are generally based on knowledge of the discipline rather than on the ability of a person to perform a particular task.
But this is really no different than other professional organizations. For example, it is possible to get a certification (Juris Doctor) to practice law in California without ever having gone to law school or presenting a case before a judge under the tutelage of a mentor. And a person only has to drive Interstate 5 through Seattle to know that the Washington state certified engineers who completed their requirements for certification succeeded in concocting a major transportation boondoggle. Perhaps the hundreds of successful major corporations in Europe and elsewhere around the world see the value of a well-established testing certifications such as the ISTQB and ASQ for their ability to help professional testers effectively communicate using a common discipline jargon rather then constantly coming up with confusing neologisms. (Of course, all these successful organizations could be wrong...but I tend to think they are successful because the influential decision makers in those companies make the right decisions most of the time regardless of what some external person with a limited perspective or an idealistic neophyte thinks. Perhaps that's why they are successful!)
Of course, it would be ideal for certifications in our industry to also include practical skill-building exercises that taught testers how to test a product using various techniques and approaches, and certification required both practical demonstration and in-depth knowledge of the discipline. This doesn't simply mean that we teach people to find bugs by banging on the GUI, or asking 'probing' questions such as should a button control enlarge when a user mouse's over it. (Please...finding bugs is really not that hard, and if I wanted to know if a button control should enlarge or can enlarge I would look at the button properties for that control rather than sit there and ponder the question for 5 minutes.)
While I would like to see certifications include skill based learning along with teaching in-depth knowledge, we should also realize there are potential limitations with practical skill-building exercises. For example, in our own training we can assess if an individual learns to correctly apply a systematic procedure to design effective tests after in-depth analysis and logically decomposing a feature area or data set for a simulation used in our training. The concepts and application of some approaches and techniques such as combinatorial analysis are applicable across multiple software projects within appropriate contexts. So, we instruct our SDETs to identify the well-defined contexts in which certain approaches or techniques are appropriate and when they are not, and we also explain how they are sometimes misused so they don't also fall into the same traps that untrained people fall into. However, current certification schemas can not yet accurately assess how well a person will perform on a real project until that person is put into that situation. This is true of software testing just as much as it is true of scuba diving.
Fortunately in scuba diving the certification wars are mostly in the past. But, it seems the testing certification wars are heating up. Today, certifications are valued in some business sectors for various reasons. I suspect some new certifications will come along and claim some great benefit beyond the others. And perhaps they will, or perhaps they will be isolated communities of zealots who want to simply be different. I just find it rather Pecksniffish for any person to claim all certifications are bogus; of course except for the one in which he or she has some personal vested interest.
Comments
Anonymous
June 01, 2008
The comment has been removedAnonymous
June 01, 2008
The comment has been removedAnonymous
June 01, 2008
The comment has been removedAnonymous
June 06, 2008
The comment has been removedAnonymous
June 11, 2008
The comment has been removedAnonymous
June 11, 2008
The comment has been removedAnonymous
June 11, 2008
The comment has been removedAnonymous
June 11, 2008
The comment has been removedAnonymous
June 11, 2008
Shrini, As I said before, I usually choose my words very carefully, so there is no reason to rephrase, or reframe my statements. The word evaluate infers your first statement. How we evaluate something often involves several approaches and perspectives, and I think that point has been stated frequently. Sorry to inform you, but hacker attacks are not "random (and guessing kind of) keyboard inputs." Good hackers have in-depth knowledge of the systems they attack, and those attacks are generally very targeted and methodical. Getting back on track to certifications.... Your statement regarding no middle path simply exposes your biased and rather narrow perspective. Many of us also see this same sort of viewpoint when you and others talk about scripted testing (although no one has really stated what that means) versus exploratory testing. Again, I hate to say this but there is often a middle path, and people who are capable of opening their minds and studying many sides of an argument soon understand there is often not a binary option. Clearly Rahul has taken the initiative to persue a certification because he finds personal value in that endeavor. And, if you read his words carefully you will see that it exposed him to new information and led him to investigate additional areas and expand his knowledge directly related to the discipline of software testing. And of course, a person's ability for rational and critical thinking and logical questioning largely depend upon that person's in-depth and broad knowledge of the subject. This is why I read and promote reading books and papers on software testing and computer technologies to grow professionally as a software tester.Anonymous
June 12, 2008
The comment has been removedAnonymous
June 17, 2008
@Shrini I guess you expected a Yes/No response. I think a little different from this. If I have to talk about certifications, it is not compulsory to support either "certifications" or "no certifications". For clarity sake, following is what I think:
- Certifications are useful. Based on interest, one should certainly take up certifications.
- Many a times, studying for certifications makes you study thoroughly about the subject rather than touching a few aspects of it and getting away to some other thing.
- Certifications need improvement. This is true for any other thing in the world. That does not mean that certifications shouldn't exist at all.
- If one feels certifications are not useful, he or she can suggest ways on improving upon the existing ones or come up with a sample certification programme as an example.
- When one says "No certification", it means that no certification is useful or at worst no certification has been or can be designed which will be of use. I guess that is too generic a comment. You might choose to say "Not in favor of existing certifications" rather than "No certifications". Probably, in future you might come up with a certification programme that is very practical and useful for the community. So, making such generic comments is not fruitful. Now that's my thought process. I hope you are able to relate to it now. Regards, Rahul Verma.
- Anonymous
June 25, 2008
The comment has been removed - Anonymous
June 25, 2008
The comment has been removed - Anonymous
June 25, 2008
The comment has been removed - Anonymous
June 25, 2008
Hi Shrini, "Why certifications in the first place?" is a good question. The answer as to why some people value certifications is not going to be understood by viewing certifications in a myopic context, or as something unique to software testing. Instead, we must look at why various industries and professional disciplines have and value certifications to answer that question because I suspect the goals of any certification body are very much similar. Instead of saying that we "need" certifications, I would say some benefits of certifications are...and items 1, 2, and 3 on your list are good. I would add that certifications (or degrees) can also be used as filters by hiring agencies. WRT to learning and improving skills certification programs can do those things if they are designed and administered correctly. I also disagree with your assumption that "certification breed "One world" and "one way" of doing testing." Some certifications teach the theory and concepts of commonly practiced techniques (systematic procedures used to help solve complex problems), but I don't know of any certification that promotes "one way" of testing. Can you please provide a qualified reference to a certification board that specifically promotes only one way of designing and executing software tests? Professional certifications in any discipline are very different from philosophical schools of thought that exist within some professions. Since only people in the 'context-driven' school seem to identify as a member of a particular philosophical 'school', I don't think you will find "a member from "Factory school" to offer a counter-point. So, I am sure that it comes as no surprise to you that I also disagree that those who oppose the concepts of schools are in "a state of "continuous" confusion of "where they belong" and "what to promote and what not". I don't agree with the concept of separate philosophical schools in our discipline, and I understand and promote the value of multiple perspectives in professional software testing and harmoniously use each within the appropriate context rather than continuously approach testing from one myopic view. And, I am not in a continuous state of confusion; I am only occasionally confused by the lunacy of people who refuse to expand their perspectives and seem to be locked into a single train of thought or fixated on some non-universally accepted philosophical school.