Computational Science “Same Old Silence, Same Old Mistakes” “Something More Is Needed … ”

Abstract

Today it is fashionable to portray computation as the third leg of science, the other legs being the classical disciplines of experiment and theory. But in the rush to promote computational science's strengths, a blind eye is often turned to its weaknesses. This paper aims to increase awareness of a number of key deficiencies in the hope that the community can galvanize itself and tackle the identified issues head on. Specifically, the thesis to be developed here is that software automation could be used to package worked examples - in the form of dynamic electronic documents - that would allow interested parties, from different backgrounds, to communicate more effectively than at present. The hope, by making work easily repeatable, is that practical expertise can be properly archived. Currently, many avoidable mistakes are repeated time and time again as the mistakes do not lend themselves for journal publication and so go unrecorded.