Reproducible Research in Computational Science — Montevideo 2013-04-11
Presentation of IPOL in the context of "Reproducible Research in Computational Science" on April 11th, 2013, at the Institude of Electrical Engineering (IIE), Faculty of Engineering (Fing), University of the Republic (UdelaR), Montevideo, Uruguay.
Software is an essential component of applied and computational sciences, but not included yet in the research publication workflow. Without implementations, one can not know all the fine details of an algorithm and accurately reproduce and verify previous scientific results. Therefore, publishing algorithms without their implementation doesn't satisfy the scientific method.
For a repruducible computational research, software used in the elaboration of a research paper should, at least, be usable by referees, and if possible be publicly available in source form. Likewise, software development in research needs to be recognized and time and effort must be rewarded by scholarship standards.
IPOL answers these problems for image processing. In this research journal, each article contains a text describing an algorithm and source code, with an online demonstration facility and an archive of online experiments. The text and source code are peer-reviewed and the demonstration is controlled.
After a short introduction to the Reproducible Research and Open Science background, we will explain why we need new publishing models like IPOL, how IPOL works, its benefits, and our plans for future development.
presentation derived from
Software as Science — Bruxelles 2013-02-02. Nicolas Limare. figshare. http://dx.doi.org/10.6084/m9.figshare.155649
with parts from
IPOL: Image Processing On Line. Gabriele Facciolo, Nicolas Limare. figshare.