I think mathematicians favour abstractions more then programmers, opposite as to what the OP suggests Its like when you're coding this solution in haskell and

I think mathematicians favour abstractions more then programmers, opposite as to what the OP suggests Its like when you're coding this solution in haskell and start to wonder if the GHC is going to optimize all the intermediary data structures with it's di-forestation algorithms as apposed to doing it in a imperative language and including an assembly file with the floating point routines implemented using SSE instruction.