Well age groups do have a developmental psychology theory going for them. A year in the life of a child is huge, someone might be slightly above or under their year's average, but how do you judge that?

As for other assumptions of learning, it's true that we decide what is important and that might not be the best way. But again it is really hard to say what would be better. Anyway, I totally agree that changes need to be made. But how do we make them so that we don't fuck over any age group or generation, since the results will be seen years after only. Kind of ethical issue already. Same as with science, we can't test everything because of ethical limitations like giving a cure to someone and not the other. :)