4 ms·

It seems to me city planning can be quite easily supported through simulations. Like a Sim City on steroids. You want to optimize "quality of life" -"as manifes

It seems to me city planning can be quite easily supported through simulations. Like a Sim City on steroids. You want to optimize "quality of life" -"as manifested by a bunch of factors - walkability, low pollution, short commute times etc. Partially these targets can be formulated in some differentiable manner. Partly one might need to do some more black box optimization. Then overhaul zoning rules...

masklinn 10y ago
The biggest problem in the US is the wide-spread application of "Euclidian" zoning which strictly segregates use (often down to single versus multi-family dwellings), which leads to dead suburbias with houses everywhere but not a grocer or convenience store in place, which makes for absolutely dreary and dreadful environment. You don't even need to run sims to get something better, you can just look at "organic" countries and cities and how they implemented their zoning. Here's a comparison of Japanese and American/Euclidian zoning for instance: http://urbankchoze.blogspot.be/2014/04/japanese-zoning.html http://urbankchoze.blogspot.be/2014/04/japanese-zoning.html (the first link leads to an article on Euclidian zoning)
- cmarschner 10y ago
  When I was a kid and played Sim City for the first time it was also the first time I was confronted with zoning. The whole concept of having a square grid and strictly separate residential/commercial/industrial zones seemed foreign to me. I understood slowly that there was plenty of cultural bias in the game. For me the most natural form of a city was that of a circle surrounding some central area. That would be the european bias I guess... We do have zoning as well of course, but in fact new laws are coming in place that aim for more of a mixture of spaces within a city. I think this is a great development.
jnordwick 10y ago
Those aren't really quality of life metrics though. They are things you think might lead to quality of life, but the isn't the same thing. Walkability might not be the important thing, but ease of commute might be and separate cars might win out on that (convince a majority of LA natives to give up theirs cars in exchange for the subway and see how far you get). Your most likely scenario is going to be a simulation that optimizes the preferences of the person seeing it up. This is a problem and we haven't even come to the part about knowing how goals and actions interact with each other. This would literally win you a Nobel prize and change the world.
- cmarschner 10y ago
  You put it like finding out people's needs is complete voodoo... but in fact there are well established disciplines in the social sciences that do nothing but that, and quality of life can be put into indices and even ranked. And, yes, at this point European cities with good public transport, like Copenhagen, Zürich, Vienna or Munich come out on top in many cases in these surveys.
bbarn 10y ago
Except, you could build out said algorithm, in theory, put it into use, and an election cycle later the community is polarized in some new direction and everything changes. Urban planning is as much prediction as it is response to changing circumstance.

masklinn 10y ago

The biggest problem in the US is the wide-spread application of "Euclidian" zoning which strictly segregates use (often down to single versus multi-family dwellings), which leads to dead suburbias with houses everywhere but not a grocer or convenience store in place, which makes for absolutely dreary and dreadful environment. You don't even need to run sims to get something better, you can just look at "organic" countries and cities and how they implemented their zoning. Here's a comparison of Japanese and American/Euclidian zoning for instance: http://urbankchoze.blogspot.be/2014/04/japanese-zoning.html http://urbankchoze.blogspot.be/2014/04/japanese-zoning.html (the first link leads to an article on Euclidian zoning)

cmarschner 10y ago
When I was a kid and played Sim City for the first time it was also the first time I was confronted with zoning. The whole concept of having a square grid and strictly separate residential/commercial/industrial zones seemed foreign to me. I understood slowly that there was plenty of cultural bias in the game. For me the most natural form of a city was that of a circle surrounding some central area. That would be the european bias I guess... We do have zoning as well of course, but in fact new laws are coming in place that aim for more of a mixture of spaces within a city. I think this is a great development.

jnordwick 10y ago

Those aren't really quality of life metrics though. They are things you think might lead to quality of life, but the isn't the same thing. Walkability might not be the important thing, but ease of commute might be and separate cars might win out on that (convince a majority of LA natives to give up theirs cars in exchange for the subway and see how far you get). Your most likely scenario is going to be a simulation that optimizes the preferences of the person seeing it up. This is a problem and we haven't even come to the part about knowing how goals and actions interact with each other. This would literally win you a Nobel prize and change the world.

cmarschner 10y ago
You put it like finding out people's needs is complete voodoo... but in fact there are well established disciplines in the social sciences that do nothing but that, and quality of life can be put into indices and even ranked. And, yes, at this point European cities with good public transport, like Copenhagen, Zürich, Vienna or Munich come out on top in many cases in these surveys.

bbarn 10y ago

Except, you could build out said algorithm, in theory, put it into use, and an election cycle later the community is polarized in some new direction and everything changes. Urban planning is as much prediction as it is response to changing circumstance.