The for loop runs at most len times each iteration of while, it is geometric progression, so the runtime is actually O(n).

This snippet is bad, and the code should be read using the fs module.

Well, if you can wing it that easily you're definitely much better than average ;-) So I don't think it's that overranked.

Whilst this proof is indeed not a difficult one, I consider the 2kyu rank to be a reflection of the need for advanced knowledge of the Haskell type system. I do not understand why you have characterised this knowledge as 'basic'; it is far from it.

In any case the ranking is totally beyond my control. As such I'm resolving this suggestion.

The successor characterization of the natural numbers, plus some basic knowledge of the type system.

This is not trivial for most people though ;-) So the entry barrier is pretty high. That's why it was ranked quite high.

I ranked this 3kyu but some people (hint: it's two of the people below this thread in this discourse) ranked it purple, dragging the average up to 2kyu. So it's stuck at the average.

Algorithm problems belong to completely different category from academic (sort of, this kata is related to formal proof) problems, so you can't compare their ranks directly.

@Johan

You seem to be the only one that is contactable and because you approved it, I'm asking you for help. Could you please have a look at my question I posted up top? The performance test is assuming things about time and it is causing some loop to miserably fail.

And after 3½ years, it's out of beta. It deserved that.

Fixed this.

I was able to edit the mean/stddev-issue and republish just now.

Closing this issue for now.

Could you try again to fix it?

Could be a CW issue. Things look as you say - that the corrections haven't been made. I went back and attempted to edit the kata, but re-publishing times out.