Enough? It is rather the observation that if you use the same parsing approach, more or less, in a language which looks like C, then the possible improvement is

Enough? It is rather the observation that if you use the same parsing approach, more or less, in a language which looks like C, then the possible improvement is rather small. There are two things at work here: how small is the parser? And second, how fast is the parser? I know you can write some extremely fast highly idiomatic parsers in Haskell though. But I doubt they will be faster than hand coded C.