Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!genrad!decvax!decwrl!sun!guy
From: guy@sun.UUCP
Newsgroups: net.lang.c
Subject: Re: Precedent for use of =
Message-ID: <4275@sun.uucp>
Date: Thu, 19-Jun-86 14:55:13 EDT
Article-I.D.: sun.4275
Posted: Thu Jun 19 14:55:13 1986
Date-Received: Sat, 21-Jun-86 10:44:39 EDT
References: <1331@brl-smoke.ARPA>
Organization: Sun Microsystems, Inc.
Lines: 26

> Is it really worth using a notation that is contrary to that of the more
> universal mathematics, and thus guaranteed to confuse novices and catch pros
> off-guard in order to type "=" instead of ":="?  I think adding a few
> characters to improve readability and understandability is worth the
> time and effort.

Is it really worth changing a well-established language syntax, in such a
way that would break the hell out of existing programs, in order to conform
with what is admittedly the dominant notation?  I think leaving C alone to
keep from breaking existing programs and programmers is worth the confusion
it may cause to some people.

> Consequently, I wrote a pre-processor for C called ac68 that uses := for
> assignment, = for equality, and has all the cumulative operators in the
> style of Algol 68: +:=, -:=, *:=, &:=, <<:=, etc.  Unfortunately, the C
> and dbx messages refer to the operators that get generated.  Still, the
> code is a lot more readable.

Unless you advertise the language accepted by this preprocessor as a
language which is similar to C, but is NOT C, I sincerely doubt that.  I
find that the fake ALGOL 68 crap that Steve Bourne used to write the Bourne
shell and "adb" to make the code a lot *less* readable.  Let C be C!
-- 
	Guy Harris
	{ihnp4, decvax, seismo, decwrl, ...}!sun!guy
	guy@sun.com (or guy@sun.arpa)