On Sun, Feb 22, 2026 at 08:06:18PM +0100, Salvo Tomaselli wrote:
> A human needs to see 0 quicksort implementations to be able to do it,
> an LLM needs to see millions of them.
[ Citation needed ]
First of all I doubt there are literally millions of quicksort
implementations out there in the 'net.
Secondly, I will agree that there is at least *one* human being that
came up with an quicksort implementation without seeing a prior
example --- specifically, Tony Hoare who invented quicksort in 1959.
But what *percentage* of human beings who have emitted quicksort
implementation have never seen an quicksort implementation before?
Maybe it's in a different language, or in pseudo-code that a professor
has written on a chalkboard. Or maybe they went to Stack Exchange, and
just copied and pasted from there.
The qustion of whether a human *can* implement a quicksort from first
principles is quite different from whether most humans *do* --- in
actual practice, in real life.
Cheers,
- Ted
