Patrick Cox over at TCS Daily has an interesting media take on things. However, he has fundamentally committed an error by cleaving too closely to Brooks' Red-v-Blue hypothesis.
First off, there is some truth to the "density is destiny thesis." But it can be carried too far.
When Brooks first wrote, in order to find a conservative space, he had to leave the DC area and drive out to the sticks. (Fine with me, I like the sticks.) But this skewed his writing, on the assumption that liberal and conservative areas were merely "urban v. retro," as the saying goes, completely missing the fact that there are large urban areas that are quite conservative, Dallas being my home-town example, and, perhaps more importantly, that the number of medium-sized cities that are "red" without having anything at all in common with the countryside, is huge.
Therefore, just because big cities are starting to fester less, doesn't mean that the red-blue divide is in any danger of vanishing.