The author mentions those at the bottom of the article, but two problems highlighted still remain: * There's another intermediary concept (kernel density estim

The author mentions those at the bottom of the article, but two problems highlighted still remain: * There's another intermediary concept (kernel density estimation) between the audience and the data * They're still likely to misrepresent tight groupings and discontinuities, which will be smoothed out