Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site kobold.UUCP
Path: utzoo!linus!decvax!genrad!grkermit!masscomp!kobold!tjt
From: tjt@kobold.UUCP (T.J.Teixeira)
Newsgroups: net.lang
Subject: Re: Strong Typing and Ignorance
Message-ID: <225@kobold.UUCP>
Date: Thu, 29-Dec-83 10:39:51 EST
Article-I.D.: kobold.225
Posted: Thu Dec 29 10:39:51 1983
Date-Received: Sat, 31-Dec-83 04:27:09 EST
References: <2892@utcsrgv.UUCP>, <137@csd1.UUCP> <718@cbosgd.UUCP>, <217@kobold.UUCP> <6421@unc.UUCP>
Organization: Masscomp, Westford, MA
Lines: 17

Using Tim's definition of strong typing, he should rest at ease.
According to the reference manual in K&R, section 7.13, the "?:"
operator is strongly typed.  The two expressions must be of an
arithmetic type (in which case the usual arithmetic conversions are
applied), both pointers of the same type, or one a pointer and the
other 0.

I concede that Tim's definition of "strong typing" makes sense, and I
won't submit an alternative definition.  However, I will point out that
sloppy usage substitutes "strong typing" for "strong type checking".
"strong type checking" means that the compiler checks for non-sensical
types of expressions in all contexts.  I think that we can agree that
using a floating point number as a pointer is non-sensical, as are the
results of most argument type mismatches.
-- 
	Tom Teixeira,  Massachusetts Computer Corporation.  Westford MA
	...!{ihnp4,harpo,decvax,ucbcad,tektronix}!masscomp!tjt   (617) 692-6200