Output details

The case for open computer programs

<09> This highly cited paper uses both published and previously unpublished research results to illuminate the many sources of ambiguity in computer programs in support of the following central premise. ‘If a piece of scientific research involves computation, then the results of that research cannot be considered scientific unless the complete means of reproducibility are supplied with the research.’ This must necessarily include source code. It closes by making the point that current source code strictures for journals and funding bodies are now largely obsolete and suggests a method of classifying papers according to the degree of source code disclosure.