You're making a leap here. The chance for someone to fix a bug does not make developer certifications useless. The chance for someone to catch a bug during re

You're making a leap here. The chance for someone to fix a bug does not make developer certifications useless. The chance for someone to catch a bug during review, before it harms anyone (which is what I was referring to) still does not make developer certifications useless. The latter does, however, decrease their marginal utility. This is why it's an apt comparison to engineering: an engineer can't "patch" a defect in his design once it's been produced, but he can have other engineers look it over beforehand. Now, what I mean by "certified" varies a bit across disciplines. The professional engineer cert is a general certification, which is generally much less useful than specific certifications for particular disciplines. The considerations in electrical engineering are considerably different from those in, say, structural engineering. While there is also significant overlap, you're not going to get a general-purpose "engineer" who can verify both.