"I blame the educational system" says, like so many, a leading computing pundit, and legend John C. Dvorak in his latest column. This is the core of his argument:

It's Time for Universal Computer Education - Columns by PC Magazine

People know nothing of the dangers inherent in natural gas, static electricity, lightning, or anything else. None of this is taught in school; instead we have diversity studies and self-esteem lectures. When I was in school, we were taught how to do income-tax returns, how to balance checkbooks, and how to fix cars. There were ethics classes, health classes, home economics, shop, and everything in between. All this was in addition to physics and science classes.

This is not an uncommon complaint but it invites critical commentary that goes to the very heart of our understanding of education. The central question is: How well do we know how particular changes in the curriculum content will be reflected in populations? Requests that particular social ills (sometimes hard to distinguish from the pet peeves of cranky elders) be cured through this or that educational change are common. I wonder what the curriculum would be like if we added all of them up. But the data for how such a change in education would actually be reflected in society is missing. Intuitively, we can guess that if we teach a whole generation of children that Pluto is a planet and intersperse this fact among pop-culture icons, there is a good chance a lot of the population will know that Pluto is a planet. But that doesn't mean we've done much about the education of astronomy and that that knowledge is any more valuable to today's youth than the fact that Zeus was a God was to the Greeks. What amount of science education, for instance, would be necessary to prevent the rather pointless debate surrounding Pluto's recent reclassification? Indeed, an argument could be made, that (at least in some cases) the real causality goes the other way. A change in society occurs which is then reflected in the content of education. For instance, we could say that computers are taught more because more people use and know how to use computers. The general increase of the computer skill pool in the population is the cause of its introduction into the curriculum rather than the other way around. (Of course, some sort of a feedback loop would be a less radical explanation.) Literacy is another example where a similar scenario has been postulated.

Although, this (not quite sufficiently articulated) concept of educational causality, is on the surface in sharp contrast with the following:

Education is the key, and it is never too late to start. At least a single year of computer training—the sophomore year would be my pick—would benefit high schoolers no end.

But it need not be. It's simply possible that there needs to be a certain critical mass of skill and knowledge in the population before educational efforts alone can sustain their maintenance.

And, to put my computer expert hat back on, Dvorak is absolutely right in the following:

Yes, many of them [students] can put up a Facebook or MySpace page, but what else can they do? The nerds in the school can do a lot (up to and including hijacking public transportation using a TV remote), but the rest need to be taught, and even a single course would be valuable and fun.

The level of computer competence of young people is grossly exaggerated by those who did not grow up with computers a still approach them with a certain amount of awe. True, today's teenagers have none of the timidity around technology that people in their 30s and 40s still occasionally experience but when looked at closely, they still exhibit remarkable levels of ignorance even of fairly basic facts about IT and today's world.