I was wondering whether someone can explain the difference between Pearson's coefficient of skewness: $$ \mu_3=\frac{E[(X-\mu)^3]}{(E[(X - \mu)^2])^{3/2}} $$
and its nonparametric version: $$ \mu_3=\frac{3(\text{mean - median})}{\text{standard deviation}} $$
Probably my confusion comes from the fact that I don't see any "parameters" of $X$ in the first definition. Any reference to literature is appreciated.