Best Practices: Automated Testing for Research Software

Print Friendly, PDF & Email

C. Titus Brown from Michigan State University writes that most research scientists need to think more critically about their code, and should adopt at least some of the basic coding hygiene used by virtually every modern practicing programmer. As outllined in his co-authored paper, Best Practices for Scientific Computing, that means you should use version control, optimize later, and plan for mistakes.

But, at the end of the day, it’s not ok to be a computational scientist and ignorant of good practice in programming any more, just like you can’t do data analysis and be completely ignorant of statistics, even if you have no formal training. As a researcher it’s your responsibility to do a good job on your research, period. If that means learnng something new, well, you’ve presumably had to do it before, and you’ll have to do it again!

Read the Full Story.