signed/unsigned char/short/int/long

[Doug Gwyn]

In article <375 at aber-cs.UUCP> pcg at cs.aber.ac.uk (Piercarlo Grandi) writes:
>    [B] I have made it up.
>    [B] I have never read/understood the Reference Manual.
>    [B] My advisor (if I had one) must be on drugs.

The first and third of these refer to an inquiry I made about your
claims that "near" and "far" were in ANSI C and that "noalias" had
been in many drafts of the proposed ANSI C standard.  I don't know
why you're suggesting that they had been applied to your "char int"
notion.  I don't remember anyone other than you suggesting the
second.  At least we finally got you to admit that your notion
differs from the description in K&R 1st Ed.

"signed" was adopted by X3J11, from already-existing practice, in
order to remedy a deficiency in K&R 1st Ed. C regarding char types.
"char int" as you propose it is not common practice.  X3J11 is not
inventing a new language, but rather canonicalizing the one in use
(inventing only when necessary to repair perceived significant
problems).  You may not like C's lack of orthogonality in its type
scheme, but it happens to be the way the C language actually is.
Feel free to fix this when you invent a new programming language.