#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Options

# Interesting presentation on the math of cities

Check out this TED presentation on the math of cities and corporations.

Probably most here already know this, but its nice to see someone evangelizing these concepts. Does anyone know the people doing this research?

• Options
1.
edited July 2011

Not directly. I pointed out this popular report and also suggested that you need a really high standard of evidence to declare a power-law based on fitting (rather than having a model which you then show gives a power law). (This isn't a new point, people like Cosima Shalizi have been saying this for years.) Note that I'm not saying they're wrong, just that I'm not convinced that what they've done necessarily shows that it's a power law. Also I'm not criticising the aim of understanding the causal dynamics (rather than the predictive dynamics) of cities, just that doing it purely by curve fitting is perhaps not the way to do it.

I spent a while trying to show how many different kind of models can fit quite well to a "looks like it's a power law" data set, but I couldn't find any publicly data set that looked like remotely like a power law in the first place. It's interesting that a key point of their papers is their discussion of the collection, and cleaning, of datasets.

Comment Source:Not directly. I pointed out [this popular report](http://www.math.ntnu.no/~stacey/Mathforge/Azimuth/comments.php?DiscussionID=310) and also suggested that you need a really high standard of evidence to declare a power-law based on fitting (rather than having a model which you then show gives a power law). (This isn't a new point, people like Cosima Shalizi have been saying this for years.) Note that I'm not saying they're wrong, just that I'm not convinced that what they've done _necessarily_ shows that it's a power law. Also I'm not criticising the aim of understanding the causal dynamics (rather than the predictive dynamics) of cities, just that doing it purely by curve fitting is perhaps not the way to do it. I spent a while trying to show how many different kind of models can fit quite well to a "looks like it's a power law" data set, but I couldn't find any publicly data set that looked like remotely like a power law in the first place. It's interesting that a key point of their papers is their discussion of the collection, and _cleaning_, of datasets.
• Options
2.
edited August 2011

Curtis: I've recently been working with a former postdoc of Geoffrey West's, Chen Hou. He has worked on metabolic scaling during ontogenetic growth. He hasn't been involved in any of the work on cities, but if you're interested in this type of work in general I could put you in touch.

David: I've always been skeptical of this work myself. Do you have any more references on the points you mentioned? Thanks!

Comment Source:**Curtis**: I've recently been working with a former postdoc of Geoffrey West's, Chen Hou. He has worked on [metabolic scaling during ontogenetic growth](http://biology.unm.edu/jhbrown/Documents/Publications/2000s/2008-2%20Energy%20uptake%20and%20allocation%20during%20ontogeny.pdf). He hasn't been involved in any of the work on cities, but if you're interested in this type of work in general I could put you in touch. **David**: I've always been skeptical of this work myself. Do you have any more references on the points you mentioned? Thanks!
• Options
3.

Curtis wrote:

Check out this TED presentation on the math of cities and corporations.

Probably most here already know this, but its nice to see someone evangelizing these concepts. Does anyone know the people doing this research?

I don't know them. Geoff West gave a talk on this over at the Long Now Foundation, and there's a nice summary of it by Stewart Brand over there.

Perhaps some of the less quantitative conclusions are more robust than the precise "power law" idea:

Cities create problems as they grow, but they create solutions to those problems even faster, so their growth and potential lifespan is in theory unbounded.

(West pointed out that there is a bit of variability between cities worth noticing. On the plot of crimes/population, Tokyo has slightly fewer crimes for its size, and Osaka has slightly more. In the U.S., the most patents per capita come from Corvalis, Oregon, and the least from Abiline, Texas. Such variations tend to remain constant over decades, despite everyone’s efforts to adjust them. “Exciting cities stay exciting, and boring cities stay boring.”)

Are corporations more like animals or more like cities? They want to be like cities, with ever increasing productivity as they grow and potentially unbounded lifespans. Unfortunately, West et al.’s research on 22,000 companies shows that as they increase in size from 100 to 1,000,000 employees, their net income and assets (and 23 other metrics) per person increase only at a 4/5 ratio. Like animals and cities they do grow more efficient with size, but unlike cities, their innovation cannot keep pace as their systems gradually decay, requiring ever more costly repair until a fluctuation sinks them. Like animals, companies are sublinear and doomed to die.

What is the actual mechanism of difference? Research on that continues. “Cities tolerate crazy people,” West observed, “Companies don’t.”

Here in China it's interesting that some of the biggest, most important cities now, like Beijing and Shanghai, were not the big important cities in the Tang Dynasty: Changan and Luoyang. As far as I know, Changan and Luoyang are just gone.

Comment Source:Curtis wrote: > Check out this [TED presentation on the math of cities and corporations](http://www.3quarksdaily.com/3quarksdaily/2011/07/geoffrey-west-the-surprising-math-of-cities-and-corporations.html). > Probably most here already know this, but its nice to see someone evangelizing these concepts. Does anyone know the people doing this research? I don't know them. Geoff West gave a talk on this over at the [Long Now Foundation](http://blog.longnow.org/2011/07/26/geoffrey-b-west-%E2%80%9Cwhy-cities-keep-on-growing-corporations-always-die-and-life-gets-faster%E2%80%9D/), and there's a nice summary of it by Stewart Brand over there. Perhaps some of the less quantitative conclusions are more robust than the precise "power law" idea: > Cities create problems as they grow, but they create solutions to those problems even faster, so their growth and potential lifespan is in theory unbounded. > (West pointed out that there is a bit of variability between cities worth noticing. On the plot of crimes/population, Tokyo has slightly fewer crimes for its size, and Osaka has slightly more. In the U.S., the most patents per capita come from Corvalis, Oregon, and the least from Abiline, Texas. Such variations tend to remain constant over decades, despite everyone’s efforts to adjust them. “Exciting cities stay exciting, and boring cities stay boring.”) > Are corporations more like animals or more like cities? They want to be like cities, with ever increasing productivity as they grow and potentially unbounded lifespans. Unfortunately, West et al.’s research on 22,000 companies shows that as they increase in size from 100 to 1,000,000 employees, their net income and assets (and 23 other metrics) per person increase only at a 4/5 ratio. Like animals and cities they do grow more efficient with size, but unlike cities, their innovation cannot keep pace as their systems gradually decay, requiring ever more costly repair until a fluctuation sinks them. Like animals, companies are sublinear and doomed to die. > What is the actual mechanism of difference? Research on that continues. “Cities tolerate crazy people,” West observed, “Companies don’t.” Here in China it's interesting that some of the biggest, most important cities now, like Beijing and Shanghai, were _not_ the big important cities in the Tang Dynasty: Changan and Luoyang. As far as I know, Changan and Luoyang are just gone.