Self-selection by, for example, inexperienced users is probably a valid criticism of using SO data for inferring a ranking of languages–although my impression i

Self-selection by, for example, inexperienced users is probably a valid criticism of using SO data for inferring a ranking of languages–although my impression is that they are at a scale where this is less and less relevant. In this case, the point the article is trying to make is actually something different: "What languages do people use for fun", and I'm having a really hard time coming up with any plausible alternative explanations for their data.