Best publishing practices to improve user confidence in scientific software

Abstract

The practice of science is becoming increasingly reliant on software — despite the lack of formal 2 training (Hastings et al. 2014; Wilson et al. 2014), upwards of 30% of scientists need to develop 3 their own. In ecology and evolution, this resulted in several journals (notably Methods in Ecology 4 & Evolution, Ecography, BMC Ecology) creating specific sections for papers describing software 5 packages. This can only be viewed as a good thing, since the call to publish software in an open 6 way has been made several times (Barnes 2010), and is broadly viewed as a way towards greater 7 reproducibility (Ince et al. 2012). In addition, by providing a peer-reviewed, journal approved 8 venue, this change in editorial practices gives credit to scientists for whom software development 9 is a frequent research output. 10