Path: utzoo!attcan!uunet!lll-winken!lll-tis!helios.ee.lbl.gov!pasteur!ucbvax!decwrl!labrea!sri-unix!garth!smryan
From: smryan@garth.UUCP (Steven Ryan)
Newsgroups: comp.lang.c
Subject: Re: Need C language Description
Message-ID: <1132@garth.UUCP>
Date: 1 Aug 88 20:15:25 GMT
References: <12707@agate.BERKELEY.EDU> <1104@garth.UUCP> <8270@brl-smoke.ARPA> <1112@garth.UUCP> <8279@brl-smoke.ARPA>
Reply-To: smryan@garth.UUCP (Steven Ryan)
Organization: INTERGRAPH (APD) -- Palo Alto, CA
Lines: 25

>>>The fellow can probably get what he's looking for, and more, by
>>>talking with the people at Metaware, Tom Penello for example.

As somebody else pointed out, their compiler is unusual and their definition
is thus unlikely to be widely accepted.

>>A compiler is not a definition. A compiler is an implementation of a
>>definition.
>
>Thanks for explaining something totally irrelevant.
>"Gee, I didn't know that."

That's alright, son, most people don't know either.

How people believe the formal definition of C is something other than `whatever
PCC does on a PDP-11?'

The only formal definition of Cyber Fortran 200 (FTN200) is the compiler.
That's what my project leader said (in essence) when I worked on the
compiler.

How languages do you know which are defined instead of having reference
manuals which are really tutorials in disguise? How often do people resolve
a fine point in the language, not by finding the precise production rules,
but by coding it up on their favorite box, possibly even going down to the
resultant assembly listing?