Doubts have been raised over the value of certification programmes for software testing, after the visit of an international software testing expert.
Rex Black, president of the International Software Testing Qualifications Board, told an audience of testers in Wellington last month that he distrusts qualifications constructed by a single vendor or a few experts.
They look like an attempt to equip trainees to work with only one particular software suite or, even worse, a scheme to enable the sole source to sell training courses, he said.
Black went so far as to mention one purveyor of a testing qualification he doesn’t think cuts the mustard, QAI and its founder, William Perry. Missing parts of the syllabus showed this organisation was unacquainted with some decades of work in the field, Black said. “It basically represents Bill Perry’s ideas.”
However, Perry disagrees. “The software-testing certification is based upon the collective knowledge of the 1,000-plus IT organisations affiliated with QAI,” he wrote in an email to Computerworld New Zealand.
“This knowledge is combined into a software testing CBOK (common body of knowledge) under the direction of an independent certification board. The current CBOK is now under revision and an updated version will be issued during mid-2008. Our CBOK has been compared independently with other software testing standards and includes almost all topics in these standards.
“I am flattered that you think I alone was the only person involved in this effort. That is unfair to the many contributors.”
Testers qualified on the basis of a limited pool of expertise would probably not acquit themselves well in a job where people have to collaborate with others trained using a more representative syllabus, Black told the Wellington testers.
This sparked a discussion on how a trainee could judge the merit of a particular qualification. Look for openness, said Black. By this he meant a large body of people with mixed skills — from software developers and trainers to academics — who converse with and listen to one another, and are sometimes persuaded to change their opinions. This is a “marketplace of ideas” that is involved in the continuous evolution of the body of knowledge.
A poor qualification, on the other hand, usually reflects only “two or three people’s opinions”, he says — people who may be close associates and mutual admirers too.
Naturally, openness is a virtue of the still-evolving ISTQB syllabus, says Black. The qualification has two established levels (foundation and advanced), with the advanced level being split into three specialist streams: technical testers; functional testers, and test managers. A third, expert, level is still being defined.
But, suppose two qualifications look equally open, with equally large numbers of experts involved. Which would be the best choice?, asked an audience member. Look to the reputation of those devising the qualifications, said Black.
Reputation includes practical career success as well as, perhaps, the publication of books (Black has three to his name) and scholarly papers and articles.
An Australia-NZ chapter of the ISTQB already exists and, according to a chart Black acknowledged was somewhat out of date. It has seven specialists who tailor the syllabus for the local environment. It has qualified at least 1,307 testers at foundation level and eight at advanced level.
Black’s talk was organised by the NZ Computer Society’s test professionals division, which took the opportunity, when he was visiting New Zealand recently, to ask him to speak at a Software International testing seminar.