Path: utzoo!utgpu!water!watmath!clyde!bellcore!faline!ulysses!allegra!mit-eddie!ll-xn!ames!sdcsvax!ucsdhub!hp-sdd!hplabs!hpcea!hpfcdc!hpldola!kens
From: kens@hpldola.UUCP
Newsgroups: comp.lang.c++
Subject: Re: Operator overloading considered harmful
Message-ID: <11430004@hpldola.HP.COM>
Date: 11 Jan 88 19:40:50 GMT
References: <240@vsi1.UUCP>
Organization: HP Elec. Design Div. -ColoSpgs
Lines: 22
Posted: Mon Jan 11 14:40:50 1988

On the contrary, C++ operator overloading isn't powerful enough.  My
complaint about operator overloading in C++ is that it only allows
redefinition of the existing operators, not the declaration of new
operators (with potentially mixfix syntax).  (I'm quite aware of the
problems in parsing such languages;  the difficulty does not make my
point moot.)

In most disciplines notations have evolved to make the concepts common
to that discipline easy to express in an unambiguous manner.  Some
classic examples include vector algebra, the Einstein summation
convention, and bras and kets from quantum mechanics.  None of these
notations may be expressed easily in C++, and all of them are desirable
from an application standpoint.  The use of existing notations can
contribute not only to readability but to productivity through making
simple semantic errors more apparent.

I won't argue the point that the facility can be abused.  This merely
points out that software developers must agree on common notation for
an application and document thoroughly the notation used.

	Ken Shrum
	hplabs!hpldola!kens