Argument order of = equivalence predicates
Olin Shivers 18 Feb 1999 20:02 UTC
From: John Stone <xxxxxx@cs.rice.edu>
Date: Thu, 18 Feb 1999 13:41:44 -0600 (CST)
* Argument order of = equivalence predicates
Terminological point: It isn't an equivalence predicate unless it
is reflexive, _symmetric_, and transitive. Notational point: It's
extremely misleading to use the identifier `=' for a predicate that is not
an equivalence predicate.
It's pleasant have a mathematician around to keep things on the straight
and narrow.
This is a good point. The = function not even slightly an equivalence relation
(although in practice, that is the common use). As I spell out in careful
detail in the topic's new text, you can *usefully* (not even pathologicaly)
use a function like <. So I should call it a "comparison" function. How's
that?
But I'm tempted to leave the parameter named "=" since it's short & convenient.
"CMP" or "COMPARE" seems more awkward.
-Olin