Reproducibility for the Computational Scientific Method
This project is maintained by Collegeville
The goal of this workshop is to bring together a small representative group to discuss competing reproducibility taxonomies associated with computing and computational science, and to accurately characterize them by illustrating their use with a set of concrete scenarios. Workshop results and subsequent discussions will be captured in a workshop report.
The purpose of this workshop is not to create a single taxonomy. The hope is that, with an accurate characterization of existing reproducibility taxonomies, possible next steps will emerge.
Reproducibility is at the heart of the scientific method. In order to differentiate approaches and degrees of reproducibility, research communities use synonyms such as replicate and repeat to convey specific kinds of reproducibility. In these efforts, incompatible definitions can and have emerged. While differences are not desirable, with proper context we can communicate within and across communities, if competing definitions are clearly described.
It is well known that there are at least two competing taxonomies that have significant communities and significant overlap within computing and computational sciences. The situation leads to confusion among researchers who work across these communities.