A mathematical proof is a better guarantee that something works than a demonstration in code. The higher math researchers usually pay for it being less accessi

A mathematical proof is a better guarantee that something works than a demonstration in code. The higher math researchers usually pay for it being less accessible, because the first engineering paper that uses their works gets cited a lot more. But that also leads to cargo-cult engineering, where the first implementation of the paper just gets copied around without understanding (e.g. if there are limitations/assumptions).