To Be Reproducible or Not To Be Reproducible — That is so Not the Question
An interview with Christoph Treude
Audio is streamed directly from the publisher (traffic.megaphone.fm) as published in their RSS feed. Play Podcasts does not host this file. Rights-holders can request removal through the copyright & takedown page.
Show Notes
Listen to this interview of Christoph Treude, Open Science Editor at the Journal of Systems and Software, and also Associate Professor of Computer Science, Singapore Management University, Singapore.
Christoph Treude : "One good heuristic for deciding whether the research is reproducible is this: Have the authors given others a fair chance at reproducing the results? Because, for me now, particularly in my role as Open Science Editor, I feel that the papers I push back on are the ones where the authors don't even given others a chance to reproduce the results. So, I am not saying that reproducibility has to happen at the push of a button. Of course that would be great. But I also acknowledge that the incentives we have in place now in research publishing and in the academic career do not really favor that approach. On the other hand, if researchers aren't even being given a chance at reproducing something because the data simply aren't available or the algorithm isn't available or there's absolutely no documentation — well then, that is just no good, and it is the kind of scenario where I, as Open Science Editor, will push back on the paper."
This interview is a collaboration between the NBN and the Journal of Systems and Software.
Link to FSE-C paper about Reproducibility Debt
Link to JJS paper about paper links to GitHub
Learn more about your ad choices. Visit megaphone.fm/adchoices