Re: Curry

From: Simon L Peyton Jones <simonpj_at_cse.ogi.edu>
Date: Wed, 18 Dec 1996 07:55:34 -0800

| Sorry Herbert, but I don't think that the current proposal honestly
| tries to keep close to Haskell. It deviates in *many* annoying little,
| *unnecessary* details, and that is -- as far as I understand --
| Simon's critic.

That's exactly what I meant. *Very* many!

| Just to give some examples:

Let me add one more: Haskell already has guards on pattern matching,
with a well-defined syntax. All that Curry needs to do is extend their
semantics. Instead, Curry's guards are syntactically quite different (they
appear at the end of the equation rather than the beginning, which is bad
for large equations.

Simon
Received on Mi Dez 18 1996 - 17:13:42 CET