C newsgroups (was: Re: Rationale for posting draft V7 C description)

[Steve Summit]

In article <2713 at lupine.NCD.COM> rfg at NCD.COM (Ron Guilmette) writes:
>This group is for discussing the use of (and the rules of) ANSI C.

Which group?  Posting an article containing the words "this
group" under a Newsgroups: line containing two groups is bound to
be confusing (at best).

Comp.std.c is, to be sure, mostly about ANSI X3.159-1989.  But
even there, discussions about what isn't in the standard (and why
not), and what should be in a future version of the standard (and
why) would seem quite appropriate, so there is no implied
constraint that discussion stay entirely inside the current
"rules of" ANSI C.

Comp.lang.c is pretty much wide open for any discussion
pertaining to the language C (which is, as has been pointed out,
a much bigger animal than the ANSI C standard).  I am aware of
only two topics which are really discouraged on comp.lang.c:

     1.	Questions about specific compilers (e.g. PC compilers),
	and interfacing with specific operating systems, which
	enjoy a much more receptive audience (and are more likely
	to receive an answer) in a system-specific newsgroup such
	as comp.unix.programmer or comp.os.msdos.programmer .

    2.	Suggestions for "improvements" to the language which are
	so radical or far-reaching as to define another language,
	and are encouraged to move to comp.lang.c++, comp.lang.misc,
	or alt.cfutures (or occasionally alt.religion.computers).

Though many of the remaining questions and discussions which
arise can seem frivolous, they come with the territory; it would
be impossible to come up with any blanket prohibitions against
frivolous or inappropriate postings without destroying the
character of the newsgroup.

As ANSI C becomes more and more widespread, I expect comp.lang.c
to be an important support mechanism for people who (through no
fault of their own) are still using older compilers.

>...the people who read and write
>this group mostly want to see people starting using (and conforming to)
>the One and Only approved standard for the language C.
>Talking about non-conforming (decrepit and anachronistic) implementations
>of the C language does not help to promote that goal.

You said "mostly," so I will not take issue with this directly,
but I can say that I do not participate in these newsgroups to
promote any One True Way.  The ANSI standard is good (doubtless
the best that could have been done), and has been endorsed by
many people (including, I hear, dmr), but it is not perfect, and
it does not instantly render the compilers which it outdates as
"decrepit and anachronistic."  I am generally more productive
under some of those older but exhaustively shaken-down compilers,
even if they don't support prototypes or trigraphs, than I am
under brand-new, whiz-bang, feature-overladen modern compilers
which often tend to have the most maddening and unnecessary bugs.
(This is not to say that all old compilers are Good, nor all new
ones Bad, but just that the dichotomy between old and new is not
absolute and one-sided.)

                                            Steve Summit
                                            scs at adam.mit.edu