point

 

 Remember me

Register  |   Lost password?


 

Complexity Digest Blog Header


Models: From exploration to prediction: Bad reputation of modeling in some disciplines results from nebulous goals

Fri, 11 Sep 2015 22:08:31 GMT

Until the second half of last century, science was progressing on two legs, theory and experiment. Karl Popper built his Logic of Scientific Discovery [1] on the dichotomy of these two pillars. He said in a nutshell: The established theories reflect the state of the art in science, theories are falsified by new experimental data, and new theories emerge that can explain the new findings together with the established body of knowledge. The two examples par excellence for Popper's epistemology are (i) Einstein's theory of relativity and (ii) quantum mechanics. The advent of electronic computation in the middle of the 20th century changed the situation and brought a new player on the stage: scientific computing. The very modest possibilities, computational speed, and storage capacities of the early electronic computers allowed for handling highly approximate models only and the prediction made by the pioneers in numerical research were commonly ridiculed by hard-nosed experimenters. By now, the situation has completely changed because of the breath-taking development of electronic facilities, and computational science has indeed become the third leg on which gain in scientific knowledge rests. Although the computational approach has become a well-established research tool there are still serious misunderstandings and wrong expectations in the significance of the results derived from computer models. This essay makes an attempt to illustrate some of the common problems.


Models: From exploration to prediction: Bad reputation of modeling in some disciplines results from nebulous goals
Peter Schuster

Complexity

http://dx.doi.org/10.1002/cplx.21729



, , , , , , , , , , , , , ,