I don't know, the point is that we would need statistics to judge better than with the anecdotes we have now. Amazon might be an especially egregious case of s

I don't know, the point is that we would need statistics to judge better than with the anecdotes we have now. Amazon might be an especially egregious case of soulcrushing manual labor and exploitation of poor people who don't have many options. Or they might be a far less headline-worthy case of having an incident rate no worse than industry-average but with above-minimum wage pay (as TFA claims) and the reports we're seeing are mostly due to their size amplifying any tiny signal. Anyway, your initial post about the mere existence of incidents at X being proof that working conditions at X are appalling. Which is incorrect because it doesn't tell us anything about whether the incident rate is above the base rate outside X or not. Of course one can still find those incidents appalling, but if X is nothing special compared to its environment then we would have to improve things everywhere and can't uniquely blame X.