In numerical analysis, elementary function membership, like special function membership, is ambiguous. In many circumstances, it’s entirely reasonable to describe the natural logarithm as a special function.

Then this is a good math paper. Everyone asking for "elm gates" (if such a thing even is possible or efficient) ought to relax a bit.