Here is Nicholas Lemann talking about his New Yorker review of recent books by Richard Florida, Ed Glaeser and Elijah Anderson and John Kasarda. The audio is not as interesting as the print review. (Lemann likes to talk about his old days in NYC and the where is various old girlfriends lived.)
To his credit, Lemann steers the interviewer from cities to a more useful conversation of the much larger metropolitan areas. He describes them this way: "When you fly in at night, it's where the lights start." I thought that I had coined this one, but it is more likely that many others had the same idea. But more to the point, Lemann is mainly right (in my view) that there is plenty to think about outside of Manhattan and similar enclaves.
Look at it this way. There is remarkable stability of the size distribution of U.S. urbanized areas at the top. These (not the metropolitan areas, which are based on county boundaries) are the ones made up of functional boundaries that are redone every ten years and where the lights really do start as you approach at night.
At demographia, there are population data for these for each of the census years, 1950-2000. So there are five rank changes possible for each area. But for the top five urbanized areas, there are only three rank changes over the fifty years (out of a 25 possible); for the next five there are 19 (out of 25 possible). The biggest places find ways to get bigger. They build on their agglomeration advantages by growing outward -- and without losing their competitiveness via "sprawling." That's the big news. Can it be that sprawl is good?