Rapid developments in the technology of electronic computers in the latter half of the twentieth century have led to dramatic changes in the ways that consequences of a scientific theory can be calculated and presented.

There does not seem to be a difference in kind between the way that the detailed motions of the planets might have been calculated in the nineteenth century, and the way that five-day weather forecasts are calculated today. In both cases, for example, approximation methods need to be introduced at some stages in the calculation to keep the computational task within reasonable bounds. Atmospheric circulation is clearly a three-dimensional problem, but one-dimensional and two-dimensional models are and were commonly used to reduce the size of the computational task to reasonable bounds. In an exactly analogous way theoretical chemists of earlier times reduced problems to over-simplified one-dimensional or two-dimensional models to make them possible to solve. It was only in this way that quantum mechanics could be applied to chemical systems prior to the computer age. But although there are no differences in kind between manual and machine computational models, there do seem to be important differences of scale.

Firstly, a modern computer model can be sufficiently complex for there to be a danger (or an opportunity, depending on your point of view) that it can become an end in itself.

The large computer models developed and used in the last few decades have somehow taken on a life of their own.