I’ve just finished a book I picked up at the Menlo Park library sale a few months ago (any venue that sells books by the grocery bag is lethal to me). Metropolis is an urban studies reader from 1964. The first lesson I have gleaned from this is that I really need to break the habit of buying generations-old textbooks for a quarter; they’re interesting, but it’s not really a time-efficient way to learn things.
The second thing I notice is that urban theorists, from the earliest days of technocratic America up to the modern day, seem to dwell in a curiously non-geographic world. (I’m not just drawing from Metropolis on this; it just crystallized something that has niggled at me about various contemporary pieces I’ve read about reimagining cities.) I’ve seen any number of discussions of how to lay out cities, optimizing density and accessibility while preserving an appropriate amount of open and agricultural space, which completely elide the question of what happens if there’s a mountain or a river in the way of, say, your concentric rings scheme.
Perhaps I’m simply oversensitized to the question from living in a part of the world deeply at the mercy of its geography when it comes to urban development. I suppose that would explain why a lot of urban visionaries are Dutch; the Netherlands are pretty much flat, and heavily modified by human intervention already.