Open Source, Open Science, and the Replication Crisis in HCI

Abstract

The open-source model of software development is an established and widely used method that has been making inroads into several scientific disciplines which use software, thereby also helping much-needed efforts at replication of scientific results. However, our own discipline of HCI does not seem to follow this trend so far. We analyze the entire body of papers from CHI 2016 and CHI 2017 regarding open-source releases, and compare our results with the discipline of bioinformatics. Based on our comparison, we suggest future directions for publication practices in HCI in order to improve scientific rigor and replicability.