View Single Post
Old Posted Sep 22, 2019, 2:48 AM
llamaorama llamaorama is offline
Unicorn Wizard!
Join Date: Oct 2008
Posts: 3,832
Originally Posted by RavioliAficionado View Post
The monocentric model is just much less efficient. Having both housing and jobs spread out requires lower average commute distances than having all the jobs in one area and all the housing in a ring surrounding it. There's really no good reason for cities to remain the way they are given modern communications technology. De-centralization makes way more sense and is only opposed by the structural inertia of the major cities existing built forms.
Is this really true, mathematically? The average of a set of points is the centroid. You can use whatever measure of central tendency you wish but its all doing the same thing.

When jobs are located in a suburb on the north side of the region, people on south side have a longer commute. By definition, the average ends up washing out whatever advantage northsiders have because of the southsiders. It would be the same if everyone met in the middle, in the city center.

There is no escaping the reality that as a city gets larger, citizens can only access the whole of its jobs and points of interest if either average commutes grow longer, density increases, or commute speeds somehow increase. Otherwise, as a city gets larger, the actual amount of places citizens can reach in a reasonable time frame(isochrone) stays the same and the expanded metro in reality is a conurbation of many cities.

What seems to happen in urban geography is that as a city gets really big, its core reaches a limit. Then, neighboring centers start to grow due to their proximity(because transportation and economic benefits are bidirectional) and reach their natural limit, and their neighbors grow, and so on and you get this fractal looking thing similar to the way slime molds grow(read about how cities and mold grow the same way, its fascinating). Even though there is a tendency for us to want to classify these places as unified giant megalopolises, realistically very few people are going to commute the entire distance across them on a regular basis. A resident of Campinas or Riverside is not really living their full life in Sao Paulo or LA. If it becomes impossible to objectively separate where one city begins and the other ends, maybe we could look at cities as being an individual experience - the personal local geography and abstract social and economic networks that each of us lives in that overlap and entangle massively with others. At some point us city nerds have to stop being borderline autistic about population stats and labels and boundaries and just realize that in the real world it runs together, its fluid.

Yuri's question is a real one. Our present modes of transportation mean you can only go so far and so fast and as density increases that effects what transportation we can use. Autonomous cars might increase the potential size of a city, so would high speed regional rail. If more people worked in virtual reality rather than in person the attraction to population centers would revolve around access to things which people's travel time tolerances are higher, like cultural and natural amenities, or access that's infrequent like health care services, then there would be less traffic and people would go further so if people still accepted higher densities you could have very big cities.
Reply With Quote