 |
 |
One of the key differences between systems which display lognormality - which is the log of a normal distribution - and where the fat tail can be approaxiimated by a power law, is between systems that are organically changing and those which are manufactured. This is not the same as systems that are physical versus systems that are animal. Biological systems show very good lognormality where you get competitive growth like city populations, growing trees, crystal growth and so on. Manufactured systems like buildings and integrated circuits and streets (insofar as we can think of a street as somehting which is manufactured!) show somewhat cruder lognormality and hencecruder power laws. Another way of saying this is that manufactured systems are likely to be over a much lower orders of magnigtude in terms of scaling than biological systems. This means that we should not expect brilliant power law approaximations when we measure the geometries of cities in contrast to measuring the growth of them in terms of population, in my view. The systems in our 1994 book Fractals Cities and in my Cities and Complexity (2005) book are all about populations, not about geometries Mike Batty At 14:30 01/06/2007, Joao Pinelo wrote: >Was it chance that power laws were first identified in a 'non spatial' (non >geographical) environment? In virtual space where the physical constraints do >not rule topology? >Power laws were identified in a series of networks. Including social >ones. But, >how much are these networks influenced by the physical space, its >infrastructure? Is it everything? Should we expect perfect power laws to be >embedded in environments where physical space is an obvious restrictor? > >The point is: Are physical networks organized by the same principles >as networks >which, although supported by a physical set of components, are not the same as >the infrastructure? > >Joao Pinelo > ---------- Michael Batty | CASA | University College London | 1-19 Torrington Place London WC1E 6BT UK | Tel 44 207 679 1782 | Mobile 44 7768 423 656 | email: <mailto:[log in to unmask]>[log in to unmask] | web: <http://www.casa.ucl.ac.uk/>www.casa.ucl.ac.uk