[snips]
I suggest that those who want to re-widen the scope of comp.lang.c do
so simply by posting, without bothering to continually respond to the
complaints of those who don't want them to.
Problem with this is, those who are going to complain will continue to do
so unless and until the "wideners" cease and desist. So you end up with a
newsgroup which is more noise than signal.
The problem seems to stem from the odd notion that "C" is, somehow, a
random concept, equally applied to anything convenient: if it has structs
and ints and a function called main, it's C.
This is not the case. C is a very well-defined entity; it has an
internationally ratified standard, one designed precisely to ensure that,
as much as possible, the language is the same across all systems, all
implementations, free from ties to particular hardware, operating systems
and the like. It is designed to be maximally usable, maximally portable.
If Joe Sixpack comes along, extolling the virtues of some new language
feature he's created in his compiler, that's all well and good, but it is
*not* something usable by others using different compilers, is it?
Whatever it might be, it is not _C_, as it is neither proper portable code
according to the standard, nor is it a proposed addition to the language,
something which other compilers might benefit from.
If it _were_ proper C code, it wouldn't require his compiler. If it was
being offered as a proposed enhancement to C, it would be in the group
discussing the language standard.
The same is true of all sorts of extensions. Compiler A supports "void
main()". Fine, great... but so what? That's _a compiler_. The language,
C, expressly forbids this, at least for hosted environments. Discussing
the use of void main() is *not* discussing C, it is discussing "Vendor X's
not-quite-C".
If you have a thousand different dialects of C come in here, and all of
them are "acceptable", you're going to quickly run into a problem: when
someone asks "What am I doing wrong" the only possible response is "We
can't tell, because we don't know what language you're using."
This is not helpful. By keeping the group focused on C - real C, C as
defined by the standards - it prevents such degeneration, it keeps the
group useful and vibrant. Proposed extensions to the language _already_
have a place to be discussed, they don't need to be discussed here.
Implementation-specific extensions also often have a place to be
discussed, they don't need to be discussed here.
What does someone working on a Vax care about some Windows-only extension
to C provided by a single implementation, one which he cannot run on his
system? Nothing. Yet if the issue were an actual _C_ question, he would
be just as able to offer answers, just as able to test the code as anyone
else. That benefits not just him, but everyone in the group; keeping the
group focused on C, not C-plus-random-additions, maximizes the likelihood
of getting - or being able to provide - help and guidance for all.
Why would anyone want to intentionally degrade such a useful resource,
simply to bring in things which have no business in here in the first
place? If they dislike the forum that much, they could simply stop
following it; there is no need to actively participate in hastening its
downfall.
CLC is what it is because it has always adhered to a simple rule: keep it
C. That's what it's here for. C is for _everybody_, not just those using
particular tools, particular OSen. This rule maximizes the benefits for
all; breaking it might make one or two people happy, but at the cost of
degrading the group as a whole.
Keep CLC what it is - helpful, useful, available to all... but remember
that it is only useful to all _because_ it adheres to the rule that it
only discusses a language available to all.
Just my pair of skunks.
