We have to work around this issue. The Haskell tests give the error:

Falsifiable (after 27 tests):
  ([[0,0],[1,1],[1,0]],0)
expected: [[0,0],[1,1],[1,0]]
 but got: [[1,1],[1,0]]

In my undersanding "The return value should be a 2d array cropped around all of the living cells." means that the first list [0,0] should be removed.

Needlessly complicated. Not Kyu 8. Why have both None and Zero for division by 0?

Investigated; the kata is mishandling this in JS and the tests are not actually testing anything at all.

wow, very nice, another option that I learn from you!!

the Date object is no longer tampered with by the tests

fixed there

hmm perhaps. I'll re-review this at a later time
I can't delete the original comment sorry

For javascript? No, it hasn't, you're misunderstanding the instructions. Division by 0, means the divisor is 0, not the dividend.

which ones exactly?

This kata has issues.

Did you try it? From time to time I remind people to return to their katas

I thihnk it would be better to test all the numbers: all positive and all negative shuffled. Thus getting rid of random test at all.

Approved.