.. some haskell implementations apparently had (practically) limited stack size, at least as of 2007, so as long as "pure" functions still allocate stack frames

.. some haskell implementations apparently had (practically) limited stack size, at least as of 2007, so as long as "pure" functions still allocate stack frames, it's possible for the result of the pure function call to be "stack exhausted" in one case and not another. https://mail.haskell.org/pipermail/haskell-cafe/2007-November/034017.html https://mail.haskell.org/pipermail/haskell-cafe/2007-Novembe...