The quality of TED talks is in free fall and has gotten only worse since the last time I mentioned this. So it is worth to point to a rare good presentation whenever one comes along.

Geoffrey West’s The Surprising Math Of Cities And Corporations shows nicely how other than biological systems, namely cities and corporations, undergo similar evolutionary shaping by natural selection, even though they do not necessarily die or go through generations. Therefore, they can be described - and their development predicted - with simple scaling laws, which physicists like yours truly always find endearing (since that is all we ever really do).

His main result is basically:

Biology: Bigger system = more energy efficiency, less metabolic rate (sub-linear scaling)

Cities: Double the system size = 15% savings on infrastructure but also increase in metabolic rate, crime etc. (super-linear scaling)

The talk also shows very clearly what should be obvious for a long time: Via super-linear scaling, we are headed for a collapse! Geoffrey shrugs it off with “A major innovation takes place”, but I wonder whether he himself believes that innovation will somehow magically come along every time it is critically needed (maybe that part and his repeated mentioning of super-creative people was pandering to the TED audience).

Of course, some day, magic innovation will come too late, and since the collapse opportunities occur faster and faster, as he also points out, innovation not coming along once may happen pretty soon, and once is enough. With the cities as big as they are already, shit is going to hit the fan big time. I just hope I won’t be around.

Anyway, agree with this or not, I think this is a TED talk worth watching for its scientific content, though the information density could have been higher, but you have to take the audience he speaks to into account. ;-)