I seem to be unable to publish the one labelled as haskell as it is a fork of the "coffeescript" one, and it complains that the solution code is not unique from parent. I'll try some renaming to see what I can do. Also, the unpublish button does nothing.

• 1 and 3 were leftovers from me trying to debug the quickcheck (error was that I used == instead of ===); that's fixed in the most recent fork
• 2 and 4 are now fixed.
• 5 is only in the name, and I think it was due to a connection problem while trying to save it. When clicking "add translation", Haskell is now unavailable, but CoffeeScript is available.

I did indeed test it, though there were indeed a few mistakes based on my inexperience with QuickCheck. I've responded further to your comment on the translation directly. No clue why it filed as coffeescript, though! I had some server connection issues while trying to save it the first time, which will not have helped.

Translated into Haskell, if you would like to approve it

How should the state of (x=3) + (x=2) look after evaluation? {'x':2}? Or just left as undefined behaviour as it is in C?

Unknown error is normally caused by a website error, rather than an error with your code found by the test suite. Try resubmitting!

I found it kind-of tricky, but only because I had trouble coming up with an intuition for the type of foldExpr. Once I got past that mental barrier, it was quite simple, but so interesting that I definitely think it was worthwhile. Thanks for sharing this functional pearl with us on here!

The only issues I've found are in the description: it should mention that you have to implement rep, and maybe be a little nicer on the description of interp. For such an easy kata, I would mention what the first parameter actually is (the dictionary lookup function)

I definitely didn't need the ViewPatterns extension, but I found out about it a few days ago and wanted to try it out :)