All posts by Gerard Weisbuch

Contribution to a Research Program for Global Systems Science

I fully agree with the idea p7 to go from big challenges to small questions.

And I think that part III, Knowledge Technologies needs more than Computer science and its present sections. We should not start from Tabula Rasa as more less implied in many arguments heard during Open-GSS2. If we adopt a State of the Art/ Beyond the State of the Art analysis as requested when answering European calls for projects, the state of the art could be Complex Systems approach and beyond SoA GSS.


Following foundation works in the late 40′s in discrete maths and network of automata, the idea of a search for generic properties, (equivalent to classes of universality), was a step-back from the pretention of full predictions about a complicated system under study (such as the Brain, the Earth climate, or predicting the phenotype from the genotype). SFI research e.g., although not unique, was a search for generic properties, stylised facts, scaling laws etc. to be observed in different systems such as the brain, the immune system, societies etc. In this quest, computer simulations played an important role, and this might be the reason for the confusion about Complex system research = Computer simulations.

Beyond SoA

It as already be proposed that GSS research implies:

- Systems of Systems, and obviously their couplings;

- Global systems, at the level of the Earth Globe;

- A science for policy making. Because of the difficulty to deal with such challenges I would also add smaller horizontal questions such as:

- The Man in the loop, i.e the coupling between the dynamics of human individual and social cognition (representations, learning, paradigm shifts etc.) and the dynamics of natural systems such as Climate, natural resources, Energy and so on. I would also lift the obligation to use Big Data, HPC, Global Challenges as compulsory request to start Research.


The astronomy/telescope metaphor was used to refer to
the coupling of a challenge, Policies for Global Systems
with an instrument New Tools in Computer Science:
Big Data, Simulation tools , Visualisation.

The Challenge and the Tools might not be sufficient;
if we carry on the metaphor, the success of 16/17th century Physics
was certainly made possible by Tycho Brahe, Galileo and Copernic,
but it was fully exploited by theoretists such as Kepler and Newton.
Thanks to theoretists, the discoveries in Astronomy were generalised
and transferred to the much larger domain covered by Mechanics.

Theory is an important part of our understanding of the world.
Theory does not necessarily imply a comprehensive new World vision,
such as Quantum Physics or Relativity.
But it certainly includes unifying concepts and new methods.
In the case of Complex systems, new concepts were for instance percolation,
avalanches and SOC, classification by attractors, classes of universality,
dynamical regimes and transitions…
while new methods include renormalisation, replicas, damage spreading
and Liapunov exponents, search and learning algorithms.

GSS needs not only tools but also Theory.