Open Source Is A-Changin': How Qualitative Research Can Help Us Adapt

In the past five years, industry has overwhelmingly turned to open source as a primary means to build products and services atop of, with around 80% of surveyed companies acknowledging that they run on open source. 66% of companies say they would first consider open source options before proprietary ones. A decade ago, this wasn't the case at all; these numbers were flipped, with a tiny minority of companies preferring open source solutions. This has changed the dynamics of open source community markedly. In the early days of the open source movement, most people using a project were also contributing back to it in some way. However, the number of users versus contributors has changed by likely uncountable orders of magnitude. With unicorn startups being built on top of this open source shared infrastructure, the incentives and motives behind open source contributors building this valuable infrastructure has also begun to shift. Meanwhile, software engineering studies focused on open source have been overwhelmingly quantitative; quantitative analyses on codebases, issue trackers, surveys, etc. In this essay, we argue that with these changing tides comes dynamics that we can't easily quantify. Why then do rely predominantly on quantitative methods when we attempt to understand the dynamics of open source communities? We lay out a number of research questions and qualitative techniques from the social sciences that can help us better understand these trends, and how to adapt to them going forward.