The World as Computation presents an approach to physics based on a concept of finite precision computation, which opens to digital simulation of physical processes viewed as analog computation, in general in the form of time-stepping relaxation towards satisfaction of certain equations expressing conservation of mass, momentum, energy et cet.
In this world of finite precision computation, turbulence as a main mystery of classical mechanics, can be understood as the result of a computational relaxation process aimed at producing an exact solution of Navier-Stokes equations of fluid mechanics, which however misses this goal because every exact solution of Navier-Stokes with small viscosity is unstable and thus is not reachable by relaxation.
Turbulent flow as fluctuating, complex and locally unpredictable flow, or physical reality in all its complexity, thus can be seen as the result of repeated failure to compute an exact solution:
- The ideal (exact solution) is unstable and thus cannot take real form.
- What can take real form is an approximate solution resulting from a finite precision computational process aimed at producing the exact solution.
The complexity of the world can thus can (maybe) be viewed as a result of massive repeated failure, rather than the creation of a creator of infinite fantasy.
It is maybe illuminating to realise that (according to Kevin Kelly in What Technology Wants): The closer a face hews to an ideal average human face, the more attractive we find it.
It is natural to compare with machine made copies of a mechanical part of a car or a microprocessor, which are constructed with the explicit goal to make them all identical and interchangeable, a goal that is less obvious when nature is constructing (although ans to us seem pretty identical and interchangeable).