> Opinion: a log like decrease in happiness vs income is basically consistent with "more money doesn't make you happier".
I think you mean log-like increase.
Assuming so, then more money does make you happier (although at a decreasing rate per dollar per year.)
Another one from the article: I'm arguing without empirical evidence of the superiority of good type systems (Note: I try to stay away from claims about what those are). I also think it's not possible to make controlled studies on this area - which is why it's such a contested issue. Studies use either students and/or toy problems, or irrelevant datasets such as bugs reported against github repos etc. The only way top make a valid study is to have N teams implement and maintain a product with the same spec over K years and then compare the results and try to factor out developer skill. I say it's not a feasible problem to solve. We'll have to stick to arguing type systems without evidence to back up the argumebts. But that is the best kind of arguing, after all.