Xref: utzoo comp.lang.c:6888 comp.lang.misc:987
Path: utzoo!utgpu!water!watmath!clyde!rutgers!rochester!crowl
From: crowl@cs.rochester.edu (Lawrence Crowl)
Newsgroups: comp.lang.c,comp.lang.misc
Subject: Re: Clarifying the C "int" Datatypes
Message-ID: <6377@sol.ARPA>
Date: 30 Jan 88 04:10:45 GMT
References: <6240@sol.ARPA> <10179@ccicpg.UUCP>
Reply-To: crowl@cs.rochester.edu (Lawrence Crowl)
Organization: U of Rochester, CS Dept, Rochester, NY
Lines: 23

In article <10179@ccicpg.UUCP> miket@ccicpg.UUCP (Mike Tracy) writes:
>In article <6240@sol.ARPA> crowl@cs.rochester.edu (Lawrence Crowl) writes:
>>Since I've noticed some confusion on the C "int" datatypes, I thought I would
>>post a clarification.  The "int" datatypes are not integers, they are machine
>>"words". 
>
>I'll have to disagree with you.  Many machines use a 16 bit word while an
>'int' is (many times) 32 bits.  It has more to do with the natural size
>of objects that the system manipulates.  ...

You have just confirmed my statement.  The fact that some C compilers use a word
size that is not identical to the bus size of the machine is irrelavent to the
point I was making.  The crucial difference is this: is the datatype supporting
the machine, or is it supporting an abstraction.  In C, "int" supports the
machine instead of supporting an abstraction.  To see this, note that the &
operator is meaningless for integers, but not for machine words.  The C support
for & indicates that it is oriented towards machine words and not towards the
integer abstraction.  Note that the second you start defining C semantics in
terms of bits, you have oriented yourself towards machine words.
-- 
  Lawrence Crowl		716-275-9499	University of Rochester
		      crowl@cs.rochester.edu	Computer Science Department
...!{allegra,decvax,rutgers}!rochester!crowl	Rochester, New York,  14627