Suppose we have a sufficiently nice scheme or say we are working with an abstract nonsingular variety $X$ over an algebraically close field. In this setting, one can study (Weil) divisors. It is then said that two divisors $D,D'\in\operatorname{Div}(X)$ (i.e. formal finite linear combinations of prime divisors) are linearly equivalent $D\sim D'$ if $\operatorname{div}(f)=D-D'$ for some function $f$ inside the function field $K(X)$.
I have pondered this choice of terminology for some time. Even in the case of curves, I do not precisely see why we call them linearly equivalent instead of anything else?
Any insight into this would be greatly appreciated. I have searched the internet and this site for potential answers and have not found a satisfactory answer.