split comp.lang.c

[E. Robert Tisdale]

This article proposes splitting the comp.lang.c newsgroup.
The comp.lang.c newsgroup would be restricted to C99
and evolving C standards.
A new newsgroup (comp.lang.c89) would be devoted to
C89 and discourage any discussion of C99
or any enhancements to C89.

C89 and C99 are two completely different languages
and discussing them in a single forum
only leads to misunderstanding and confusion.

 

[Keith Thompson]

This article proposes splitting the comp.lang.c newsgroup.
The comp.lang.c newsgroup would be restricted to C99
and evolving C standards.
A new newsgroup (comp.lang.c89) would be devoted to
C89 and discourage any discussion of C99
or any enhancements to C89.

Since most valid C89 (or C90) code is also valid C99 code, it's hard
to know whether questions about such code belong in comp.lang.c89 or
comp.lang.c. Most readers would just have to read both newsgroups,
and cross-posting would be rampant.

If I were going to take this proposal seriously, I'd probably (a)
suggest that the new newsgroup be called comp.lang.c90 rather than
comp.lang.c89 (the 1990 ISO standard supersedes the 1989 ANSI
standard), and (b) ask whether the changes in Technical Corrigendum 1
(published in 1994, I believe) would be considered topical.

Assuming the division could be clearly defined, it's unlikely that we
would have any luck enforcing it.

The comp.lang.c newsgroup survived the transition from K&R to ANSI C
without the need for a split; I hardly think splitting it now would be
beneficial.

C89 and C99 are two completely different languages
and discussing them in a single forum
only leads to misunderstanding and confusion.

Though C89 and C99 are not identical, they are not two completely
different languages. In particular, they're much closer to each other
than either is to C++. (If you're trying to make a point about C
vs. C++, please make it elsewhere.)

 

[E. Robert Tisdale]

Since most valid C89 (or C90) code is also valid C99 code,
it's hard to know whether questions about such code
belong in comp.lang.c89 or comp.lang.c.
Most readers would just have to read both newsgroups,
and cross-posting would be rampant.

If I were going to take this proposal seriously, I'd probably (a)
suggest that the new newsgroup be called comp.lang.c90 rather than
comp.lang.c89 (the 1990 ISO standard supersedes the 1989 ANSI
standard), and (b) ask whether the changes in Technical Corrigendum 1
(published in 1994, I believe) would be considered topical.

Assuming the division could be clearly defined,
it's unlikely that we would have any luck enforcing it.

I suppose that you could make comp.lang.c89 a moderated newsgroup.

The comp.lang.c newsgroup survived the transition
from K&R to ANSI C without the need for a split;

But it does *not* appear that the comp.lang.c newsgroup
is ever going to make the transition from C89 to C99.
In order to make this transition, the subscribers
to the comp.lang.c newsgroup must encourage the use
of the new language features and discourage the use
of anachronisms such as C preprocessor macros, for example.
Any reference to the old standard should be deprecated.
Only references to the new standard should be allowed.

I hardly think splitting it now would be beneficial.


Though C89 and C99 are not identical,
they are not two completely different languages. In particular,
they're much closer to each other than either is to C++.
(If you're trying to make a point about C vs. C++,
please make it elsewhere.)

C89 and C99 certainly *are* two different languages.
C89 compilers will *not* accept all valid C99 programs.

 

[Martin Ambuhl]

This article proposes splitting the comp.lang.c newsgroup.
The comp.lang.c newsgroup would be restricted to C99
and evolving C standards.
A new newsgroup (comp.lang.c89) would be devoted to
C89 and discourage any discussion of C99
or any enhancements to C89.

C89 and C99 are two completely different languages
and discussing them in a single forum
only leads to misunderstanding and confusion.

I propose a newsgroup comp.lang.c.troll.e.robert.tisdale

 

[jacob navia]

Mmm Robert, you are in full regression mode.

I thought you would propose splitting it into
2009 and 1999 groups, and start a discussion
about how C could be improved.

But no. You propose to go back to 89... Ahh the nostalgia

Nostalgia isn't what it used to be...

jacob

 

[Keith Thompson]

Keith Thompson wrote: [...]

The comp.lang.c newsgroup survived the transition from K&R to ANSI C
without the need for a split;

But it does *not* appear that the comp.lang.c newsgroup
is ever going to make the transition from C89 to C99.

That requires the widespread availability of C99 compilers, which
hasn't happened yet -- and this newsgroup is unlikely to have any real
influence on that. We didn't really make the transition from K&R to
ANSI C until ANSI-compliant compilers were almost universally
available.

In order to make this transition, the subscribers
to the comp.lang.c newsgroup must encourage the use
of the new language features and discourage the use
of anachronisms such as C preprocessor macros, for example.

The preprocessor is described in section 6.10 of the C99 standard,
with no indication that it's deprecated. Its functionality is very
close to what's defined by the C90 standard. If you want to
discourage the use of macros, that's fine, but it has absolutely
nothing to do with the distinction between C89/C90 and C99. (This is
the kind of nonsense that prevents us from just ignoring you; we're
afraid that some inexperienced readers might think you know what
you're talking about.)

Any reference to the old standard should be deprecated.
Only references to the new standard should be allowed.

If you can find enough people who agree with you, it might happen.
That seems unlikely.

C89 and C99 certainly *are* two different languages.
C89 compilers will *not* accept all valid C99 programs.

Is that intended to contradict my statement? It doesn't. I
specifically said that C89 and C99 are not identical. I was disputing
your claim that they're "two completely different languages".

 

[Ben Pfaff]

Mmm Robert, you are in full regression mode.

I thought you would propose splitting it into
2009 and 1999 groups, and start a discussion
about how C could be improved.

That's a lot of groups! How would anyone ever decide which to
post to?

 

[Dan Pop]

This article proposes splitting the comp.lang.c newsgroup.

If you want to be taken seriously, post a formal CFD. Then, I'd gladly
refute your proposal.

Dan

 

[Default User]

That's a lot of groups! How would anyone ever decide which to
post to?

Whichever had the fewest appearances by E. Robert Trollsdale?




Brian Rodenborn

 

[Dave Vandervies]

That's a lot of groups! How would anyone ever decide which to
post to?

They'd just crosspost between all the ones that looked like they might
be remotely related to what they had to say.
Which would be wasted effort, since almost everybody would be reading
all of them anyways.


dave

 

[Jack Klein]

That's a lot of groups! How would anyone ever decide which to
post to?

They'd cross-post questions about programming the Windows API with MFC
to all of them, of course.