There is currently a crisis in scientific software development. The vast majority of scientific software is buggy, undocumented, and developed by self-trained people who either ignore or are ignorant of best practices used by professional developers. This talk will discuss why this is a symptom of both a larger crisis in the scientific community and fundamental problems with how software writing skills are currently taught. After briefly summarizing the progress made in software education, we conclude with a vision of how to improve scientific software writing practices.
David is a systems biologist working at the Institute of Technology at the University of Tartu. His daily work involves writing software to analyze various types of biological data.