I once had the misfortune to make my living as a waitress. I was neither a particularly talented waitress nor an enthusiastic one, but I was occasionally capable of being either efficient or pleasant and, in extremely rare instances, of both. So rare was the latter that it happened exactly once and with very good timing indeed.
On that serendipitous eve, a couple and their two childrenundercover patrons hired by the restaurant to surreptitiously rate the quality of their food service experiencewere seated in my section. The result, I would find out several days later, was that I was the first server in the restaurant's history to receive a perfect score in the secret shoppers' grade book.
This resulted in embarrassingly effusive accolades from management, who, despite my reputation as a mostly average and perpetually surly food server, lauded my fictional perfection loudly and publicly, much to the chagrin of far more experienced individuals whose tableside manner was the filet mignon to my Happy Meal attitude.
The point of which is to say that academic metricsin non-academic settingsare wantonly, and exasperatingly, unreliable. In the IT industry there exists several score of academic credentials to be had, and in the majority of cases I believe they carry about as much weight as my freak waitressing success. They may serve as evidence that the person has been exposed to the knowledge needed to do the job well, but it doesn't, a priori, make them a good practitioner of the trade.
Software certifications are inventions of the formidable marketing machines of the software vendors, who profit directly from the adoption of certification-selective hiring criteria by the clueless minions of human resources departments the world over. Education is never a bad idea, but in the same way that a university degree is never a reliable indicator of the true value an individual's ability to apply knowledgeor even to have retained it at allin many cases, a software certification amounts to little more than the rubber stamp imprint on the back of the cancelled tuition check.
Certifications Are for Public Protection
The litmus test for determining the necessity of any professional accreditation programregardless of the industrycomes down to civic responsibility. We certify doctors because people require a minimum standard of competence and a way of enforcing that standard to protect their physical well-being. The FAA certifies commercial airline pilots. We like that; it's good for us, individually and collectively. Contractors need licenses so we don't screw our children out of their inheritance buying bad real estate or get crushed under our own stairwells.
I'm not saying that all certifications should be government-issued. I think sometimes private industry certification might make sense. But the reasoning behind the certification should still lie in direct customer or consumer benefit. Software does have an incalculable value in all aspects of societyeven public safetyand that will only continue. But to conclude that we have reached a point at which the software industry, as a whole, carries such an influence over the public's well being that we must implement stringent, centralized control of its practice is sheer hubris.
What's more, if a particular certification is truly necessary to protect the public good, it follows that those who work in the industry should be able to easily comprehend what that certification involves, and why it's necessary. I find it amusing and ludicrous that a page like this one (an explanation of the sundry "benefits" of the Microsoft Certified Systems Administrator) even exists. I'm resisting the urge to be derisive here (oooh, an online magazine subscription and a lapel pin? Sign me up, quick!).
Who Wins in the End?
So who benefits from IT certifications? There are a couple of answers to that question, but first, here's who doesn't benefit: end users. In all honesty, it wouldn't help me sleep any better at night if I knew that the software I used was written by people with some fancy acronyms after their name. Would software quality improve if we formalized and certified? Possibly. But that doesn't pass the litmus test. Better software, like better automobiles or grocery store scanners or paper clips or whatever, is a business goal. Whether you work for a software manufacturer or a VAR, whether you're a consultant or in-house corporate developer, improving the quality of your work is just part and parcel of the work itself. Smart, conscientious workers will strive for it and be rewarded with more job opportunities and higher salaries. So, too, will employers who use real and not invented criteria for hiring and promotion.
So then, who's the real beneficiary of IT certification? The software companies who invent and ordain such proprietary designations as MCSE, SCSA, and CCNP, stand to profit both directlyvia fees for training and examinationsand indirectly as time wears on and practitioners, faced with either a real or perceived advantage in the job market, keep coming back to re-up their credentials. Your human resources department may argue that employers benefit, too. By weeding out job candidates who don't possess the prerequisite, they can ensure the job will be filled well. It's specious reasoning; serving only to arbitrarily segment the number of r