Interesting classification. I can't help but feel that it's overly simplistic to think that each "genius" is only a genius in one of those areas. Back in Oxford, I met many people who were clearly good at several of these areas, amongst both other students and professors.
I think that swombat's intended point was that the areas of genius correlate very highly with each other's presence -- to the point that treating their respective occurrences as independent events (possibly) damages our actual understanding of the situation. Which has been my experience also.
Actually, for me, the most interesting thing about this article was the short comment about the high-school to college transition. The classification is very simplistic indeed.
Something I find interesting about this post is that the author is an accomplished theoretical physicist, who has known and worked with many leading theoretical physicists and computer scientists. It gives his classification some extra credibility.
The problem is, if you abstract any amazing person into one word, it is usually "genius." Which elicits the natural tendency to find patterns and make comparisons (since we are comparing "apples" to "apples" now, aren't we!?). Amongst the different flavors of humans, there is a common flavor of romanticism and bias.
Imho, the "dots and spaces" metaphor is still the best one to describe the diversity of brilliance around us: http://www.stephanietolan.com/self-knowledge.htm