News
NANIA has now finished. Our application for continued complexity funding was unsuccesful and the collaboration has largely return to our day jobs. There are still some nice applications on the site, in particular BallViewer is continually being updated.The main outcome of the project was, probably the repeated discovery of a type of oscillation in evolving systems which has no direct parallel in physics, and results from the explicit treatment of inhomogenous space. Broadly, it comprises the following steps:
1/ There is plenty of free space to grow. The fastest growing autonomes spread and grow. Cooperation is beneficial, altruism evolves, there's plenty of resource and the population expands.
2/ Space fills. The autonomes which are best at outcompeting others spread and grow. Cooperation is not beneficial: traits which reduce the "fitness" of an autonome, but reduce its neighbour's fitness even more evolve. Eventually the population becomes disfunctional and collapses.
3/ Some regions of space have, by chance, survived the population collapse. There is plenty of free space to grow. The cycle repeats.
This kind of oscillation was found in models of ecosystem population dynamics, in daisyworlds, in economic system models and in interacting social systems. Unfortunately, it also seems to apply in the real world to financial institutions. After a long period of chasing ever smaller, and ultimately negatively, profitable investment the system appear to have collapsed taking NANIA with it. Not to worry, we predict it will come back.
New papers
Income and poverty in a developing economy Chattopadhyay AK, Mallick S and Ackland GJ Europhysics Letters 91 (2010) 58003. http://ideas.repec.org/p/arx/papers/0905.3803.html
We show that the distribution of incomes in India for the last 40 years can be modelled as a set of autonomes whose income fluctuates proportional to their current income. This give the usual power law tail of rich people's incomes that everyone knows about since Pareto. But it also describes the poor. Combined with food price inflation, we can define a measure of poverty independent of the "poverty line" which is normally used to fix(sic) the level of poverty. We find that things are steadily getting better, but identify a previously-unrecognised(except by hungry people) mid-eighties peak of poverty. This was due to food prices rising faster than mean incomes.
Emergent patterns in space and time from daisyworld: a simple evolving coupled biosphere-climate model Ackland, GJ and Wood, AJ Phil.Trans.Roy.Society A 368 1910 161-179 (2010)
A further Daisyworld study showed that two species with different preferred habitats would evolve either to regulate their habitat at a "compromise" condition, or one would drive the other to extinction and optimise the habitat for itself. The determining factor for which behaviour was seen is the total amount of life: if the combined amount of life in compromise conditions exceeded that of a single species in ideal conditions, the compromise occurred.
This may be related to maximising entropy, but it's unclear to the authors how to count entropy!
Boom and bust in continuous time evolving economic model Mitchell, L and Ackland, GJ European Physical Journal B 70 4 567-573 (2009)
This model showed how a simple economic model went into the kind of oscillations described above. The banks just showed us how to do it. We reckon the airlines will go next.
Richard Blythe's work with Alan McKane, Gareth Baxter and Bill Croft on New Zealand English has recently been featured in various media sources around the world. There are two papers associated with the work Fixation and consensus times on a network: a unified approach (Baxter, Blythe, McKane) and Modeling language change: An evaluation of Trudgill's theory of the emergence of New Zealand English (Baxter, Blythe, Croft, McKane). Richard also has additional information.
Ian Main and Mark Naylor describe how the 2004 Sumatra earthquake has changed the facts surrounding frequency-magnitude distributions: Nature Geoscience 1:142 (2008)
About NANIA
This project is funded under the EPSRC Novel Computation Initiative and aims to bring together researchers from a wide variety of disciplines to investigate generic computational techniques to study and understand complex dynamic systems. It is one of a number of research clusters set up in this area.
Many systems can be thought of as local objects – autonomes, or agents – which interact with a spatially distributed set of neighbours and with some overall external environment or forcing. It is envisaged that novel computational approaches will be particularly important in areas where the external environment is currently treated by mean field approximations. This project will bring together researchers from different disciplines with the aims of (a) transfering experience of novel computing and (b) allowing exchange of ideas and analysis techniques. See the project overview for more details.