Should we broaden the topicality of this group?

[Richard Heathfield]

[Discussions of topicality are always topical. Nothing in this article
should be construed as suggesting otherwise - where I say something like
"only <foo> is topical here", this should be read as if it said "only
<foo> and topicality discussions are topical here"!]

One of the (several) bugbears that haunts this group is the issue of
topicality. There appear to be several topicality camps within clc at the
moment, which I'll attempt to categorise as neutrally as I can manage.
I'll assign a name and a letter to each group in an attempt to make it
easier for us all to discuss the various camps.

Group P (Purists): only code that can be written entirely in K&R C, C90, or
C99 is topical here; where functions or symbols that are not defined by
the Standard are mentioned, the source code should be available, at least
in principle, and that source code should be written entirely in one of
the above "standard dialects". Citations from K&R, from either of the
ISO/IEC 9899 documents, or from drafts are all acceptable.

Group N (Neopurists): only code that can be written entirely in the latest
version of C is topical here; where functions or symbols that are not
defined by the Standard are mentioned, the source code should be
available, at least in principle, and that source code should be written
entirely in C99. Drafts of the Standard don't count, and any citation from
a draft is off-topic.

Group C (Conservatives): whilst the principal focus should be on standard
C, discussions of particular implementations and extensions for
illustrative purposes are acceptable.

Group M (Moderates): As well as standard C, common extensions should be
considered topical (e.g. those that are common in Windows (GUI and
console) and in Unix-a-likes).

Group X (eXperimentalists): These folks would like discussions of possible
changes and enhancements to the language to be considered topical.

Group L (Liberals): Anything written in a vaguely C-like syntax or in a
language based on C is topical (including C++, Objective-C, Java, and C#).

Group A (Anarchists): Anything and everything is topical, including device
drivers, Shrek, CD marker pens, jokes about viola players, and the
Metropolitan District of South Humberside.

At present, the majority of regular contributors fall into group P, or so
it seems to me, but all the other groups are represented here to a greater
or lesser degree.

Some fall into more than one camp - for example, at least one person (who
will remain nameless!) appears to be both a Neopurist and a Liberal (or
possibly a Moderate), despite the apparent contradiction between these
positions.

When deciding what is topical and what is not, we should remember that
comp.lang.c is *not* the only newsgroup in town. There are newsgroups for
discussing the language definition, POSIX, Windows, and individual
implementations. On the other hand, it cannot be denied that some people
gravitate towards comp.lang.c even though their discussion subject bears
little or no relation to standard C. Should we be sending these folks on,
or should we be answering their questions here?

I have carefully refrained from putting forward my own opinion, partly
because I should imagine that just about everyone who reads this already
knows what it is, and partly because I don't want this thread to get
bogged down until people have had a chance to think about the issues for
themselves.

Any topicality stance is bound to have advantages *and* disadvantages. So
please think carefully before weighing in too heavily in favour of a
particular stance!

Thanks. Over to you.

 

[santosh]

[Discussions of topicality are always topical. Nothing in this article
should be construed as suggesting otherwise - where I say something like
"only <foo> is topical here", this should be read as if it said "only
<foo> and topicality discussions are topical here"!]

One of the (several) bugbears that haunts this group is the issue of
topicality. There appear to be several topicality camps within clc at the
moment, which I'll attempt to categorise as neutrally as I can manage.
I'll assign a name and a letter to each group in an attempt to make it
easier for us all to discuss the various camps.

Group P (Purists): only code that can be written entirely in K&R C, C90,
or C99 is topical here; where functions or symbols that are not defined by
the Standard are mentioned, the source code should be available, at least
in principle, and that source code should be written entirely in one of
the above "standard dialects". Citations from K&R, from either of the
ISO/IEC 9899 documents, or from drafts are all acceptable.

Group N (Neopurists): only code that can be written entirely in the latest
version of C is topical here; where functions or symbols that are not
defined by the Standard are mentioned, the source code should be
available, at least in principle, and that source code should be written
entirely in C99. Drafts of the Standard don't count, and any citation from
a draft is off-topic.

Group C (Conservatives): whilst the principal focus should be on standard
C, discussions of particular implementations and extensions for
illustrative purposes are acceptable.

Group M (Moderates): As well as standard C, common extensions should be
considered topical (e.g. those that are common in Windows (GUI and
console) and in Unix-a-likes).

Group X (eXperimentalists): These folks would like discussions of possible
changes and enhancements to the language to be considered topical.

Group L (Liberals): Anything written in a vaguely C-like syntax or in a
language based on C is topical (including C++, Objective-C, Java, and C#).

Group A (Anarchists): Anything and everything is topical, including device
drivers, Shrek, CD marker pens, jokes about viola players, and the
Metropolitan District of South Humberside.

At present, the majority of regular contributors fall into group P, or so
it seems to me, but all the other groups are represented here to a greater
or lesser degree.

Some fall into more than one camp - for example, at least one person (who
will remain nameless!) appears to be both a Neopurist and a Liberal (or
possibly a Moderate), despite the apparent contradiction between these
positions.

When deciding what is topical and what is not, we should remember that
comp.lang.c is *not* the only newsgroup in town. There are newsgroups for
discussing the language definition, POSIX, Windows, and individual
implementations. On the other hand, it cannot be denied that some people
gravitate towards comp.lang.c even though their discussion subject bears
little or no relation to standard C. Should we be sending these folks on,
or should we be answering their questions here?

Well if a reasonably suitable and active group, or groups, that are better
suited to the question exist, then, IMHO, a redirection is appropriate. The
tricky issue is what to do when no suitable group exists, or that such a
group appears to be dead. In this case, if someone knows something that
might possibly help the poster, then a suitably qualified answer with
possible redirections to other places on the Net might not be too out of
line.

Any topicality stance is bound to have advantages *and* disadvantages. So
please think carefully before weighing in too heavily in favour of a
particular stance!

Thanks. Over to you.

I might consider myself to be a member of group C and in my opinion, the
topics defined under groups P, N, and C should be topical to this group.

 

[Spiros Bousbouras]

One of the (several) bugbears that haunts this group is the issue of
topicality.

Personally I don't think that disagreements on topicality are
the real source of acrimony in this group. The reasons are probably
psychological. If this was a marriage the couple would visit a
marriage counsellor but since there's no such thing as a newsgroup
counsellor we'll just have to accept that the acrimony will
continue and all the metadiscussion in the world about topicality
isn't going to change that. I've seen situations in various
message boards where some of the regulars , perfectly respectable
and knowledgeable people in their own right , did not get along
with each other and metadiscussion didn't change anything. So I'll
probably stay out of the rest of this discussion since I cosnider
it pointless.

 

[Joachim Schmitz]

Personally I don't think that disagreements on topicality are
the real source of acrimony in this group.

It's not the only, but a major one.

Others are:

misunderstandings, mainly because english isn't everyone's native language.
rudeness
spam, and people quoting spam
trolls

in general: not knowing or caring about the netiquette
http://en.wikipedia.org/wiki/Netiquette, http://tools.ietf.org/html/rfc1855

Bye, Jojo

 

[Joachim Schmitz]

....

One of the (several) bugbears that haunts this group is the issue of
topicality. There appear to be several topicality camps within clc at the
moment, which I'll attempt to categorise as neutrally as I can manage.
I'll assign a name and a letter to each group in an attempt to make it
easier for us all to discuss the various camps. ....
Thanks. Over to you.

I was really unsure how much of your post to quote or snip, without risking
being flamed all over the planet.
Let's hope for the best...

I'd vote against N (not enough compliant compilers around, if need be get a
group comp.lang.c99 created), X (comp.std.c is the group for this), L (There
are active groups available that deal with these topics and these topics are
too different from C) and A (would significantly affect the readbility and
S/N ratio here).
For X and L a _polite_ pointer to the relevant groups should do. I don't
care about politeness to Anarchists .
Not sure about M (e.g. POSIX is at least a fully fledged standard, Windows
is not). I think a short (and polite) answer plus a redirect/pointer to an
appropriate group should be OK.
Leaves (for me) P and C.

Bye, Jojo

 

[Justin Spahr-Summers]

One of the (several) bugbears that haunts this group is the issue of
topicality. There appear to be several topicality camps within clc at the
moment, which I'll attempt to categorise as neutrally as I can manage.
I'll assign a name and a letter to each group in an attempt to make it
easier for us all to discuss the various camps.

Group P (Purists): only code that can be written entirely in K&R C, C90, or
C99 is topical here; where functions or symbols that are not defined by
the Standard are mentioned, the source code should be available, at least
in principle, and that source code should be written entirely in one of
the above "standard dialects". Citations from K&R, from either of the
ISO/IEC 9899 documents, or from drafts are all acceptable.

Group N (Neopurists): only code that can be written entirely in the latest
version of C is topical here; where functions or symbols that are not
defined by the Standard are mentioned, the source code should be
available, at least in principle, and that source code should be written
entirely in C99. Drafts of the Standard don't count, and any citation from
a draft is off-topic.

Group C (Conservatives): whilst the principal focus should be on standard
C, discussions of particular implementations and extensions for
illustrative purposes are acceptable.

Group M (Moderates): As well as standard C, common extensions should be
considered topical (e.g. those that are common in Windows (GUI and
console) and in Unix-a-likes).

Group X (eXperimentalists): These folks would like discussions of possible
changes and enhancements to the language to be considered topical.

Group L (Liberals): Anything written in a vaguely C-like syntax or in a
language based on C is topical (including C++, Objective-C, Java, and C#).

Group A (Anarchists): Anything and everything is topical, including device
drivers, Shrek, CD marker pens, jokes about viola players, and the
Metropolitan District of South Humberside.

I don't weigh in on this newsgroup often, but I do pay attention to as
much of it as is possible, so I'd like to think that I do hold some
reasonable views on the subject.

I wholeheartedly agree that there are other places for X, L, and A...
the newsgroup name may be perhaps misleading to some of those in group
X, but, as others have said, a pointer to comp.std.c should be
sufficient.

Group M is a bit trickier. I would say that, in general, those posters
should be redirected to the appropriate newsgroup (if possible and if
anyone knows it); however, it seems that sometimes the regulars here
can be overzealous in pointing out off-topic posts. If a code example
given in a question simply CONTAINS references to some non-standard
extensions, but the question applies to other (standard) things, I'd
say it's very relevant here.

Out of the remaining three, I'd place myself most in group C. I don't
find a problem with providing much of anything (including non-
standard or non-portable examples) for illustrative purposes, because
it may provide someone reading the topic with some insight for the
future.

As for group N... while I personally try to make my code as C99-
conformant as possible, and take advantages of the new features it
provides when possible, I agree with Joachim that it's not completely
implemented in a large enough number of compilers and standard
libraries to be the ONLY possible discussion topic.

Besides my above mention of why I prefer group C to group P, I have
one more (slight) objection to raise to group P. Namely, it's been 17
years since C90... why should anyone still be encouraging K&R C styles
or providing examples in K&R C? As demonstrated from C99 (with the
removal of implicit int), more and more of it may become obsolete in
the future. I feel it's our responsibility to point this out to
posters who mention the C Programming Language (ed. 1) or give
examples that reek of K&R C. I'd like to have the same objection to
C90, but C99 just hasn't seen a lot of implementation - certainly not
enough to make C90 anywhere near obsolescent.

 

[Joachim Schmitz]

Besides my above mention of why I prefer group C to group P, I have
one more (slight) objection to raise to group P. Namely, it's been 17
years since C90... why should anyone still be encouraging K&R C styles
or providing examples in K&R C?

It is OK to point out that K&R(pre ANSI) style is obsolecent, but it isn't
off topic IMHO.
Where else should the person having such code ask for
help/insight/wehatever?
There's quite a difference between encourage and call something off topic.

Bye, Jojo

 

[=?iso-2022-kr?q?=1B=24=29CHarald_van_D=0E=29=26=0F]

[Discussions of topicality are always topical. Nothing in this article
should be construed as suggesting otherwise - where I say something like
"only <foo> is topical here", this should be read as if it said "only
<foo> and topicality discussions are topical here"!]

I consider discussions of topicality as off-topic here -- certainly, they
have nothing to do whatsoever with C -- but acceptable. There's plenty
that I don't think is inappropriate for this group, and plenty that I
think is off-topic for this group, but neither is a strict subset of the
other. Just something that might be worth considering; I don't believe
this is covered by your groups.

 

[John Smith]

[Discussions of topicality are always topical. Nothing in this article
should be construed as suggesting otherwise - where I say something like
"only <foo> is topical here", this should be read as if it said "only
<foo> and topicality discussions are topical here"!]

Any topicality stance is bound to have advantages *and* disadvantages. So
please think carefully before weighing in too heavily in favour of a
particular stance!

Thanks. Over to you.

IMHO, setting strict limits on topicality is a futile exercise.
As a general principle why not just agree that any topic of
interest related to C programming is appropriate, otherwise
ignored? Marginally topical issues will inevitably be discussed
whatever formal guidelines are adopted.

JS

 

[Stephen Sprunk]

[Discussions of topicality are always topical. Nothing in this article
should be construed as suggesting otherwise - where I say something like
"only <foo> is topical here", this should be read as if it said "only
<foo> and topicality discussions are topical here"!]

I consider discussions of topicality as off-topic here -- certainly, they
have nothing to do whatsoever with C -- but acceptable.

Topicality is always on topic.

Personally, I'm in group C; references to specific implementations
(including the DS9k) should be considered acceptable for _illustrative_
purposes. I can see the point of the P folks because it's a slippery slope,
but the educational value of providing examples of _why_ something is
nonportable outweighs the risk, IMHO.

It'd be a lot easier for me to join group P if there were comp.lang.c.posix
and comp.lang.c.windows. Shuffling people off to platform-specific groups
that are not specific to C and do not have the C expertise available here,
even if they do discuss C a lot by default, doesn't seem friendly.

S

 

[Peter J. Holzer]

One of the (several) bugbears that haunts this group is the issue of
topicality. There appear to be several topicality camps within clc at the
moment, which I'll attempt to categorise as neutrally as I can manage.
I'll assign a name and a letter to each group in an attempt to make it
easier for us all to discuss the various camps. [...]
Group C (Conservatives): whilst the principal focus should be on standard
C, discussions of particular implementations and extensions for
illustrative purposes are acceptable. [...]
Group X (eXperimentalists): These folks would like discussions of possible
changes and enhancements to the language to be considered topical.

I guess I'm mostly in group C, then, with occasional forays into group X.

Basically, I think the topic of this group should be "the language C in
the real world", as opposed to "the language C as described in the
standard", which is on-topic in comp.std.c. In the real world, C is a
group of dialects, and while the principal focus should be on elements
of the standard language, I see nothing wrong in discussing the
idiosyncracies of a specific dialect.

hp

 

[Peter J. Holzer]

Besides my above mention of why I prefer group C to group P, I have
one more (slight) objection to raise to group P. Namely, it's been 17
years since C90... why should anyone still be encouraging K&R C styles
or providing examples in K&R C?

I don't think anybody should "encourage" K&R C, but I think it is
on-topic here, and if it's only to be able to discuss the history of C.

hp

 

[Flash Gordon]

Harald van Dijk wrote, On 29/09/07 17:08:

[Discussions of topicality are always topical. Nothing in this article
should be construed as suggesting otherwise - where I say something like
"only <foo> is topical here", this should be read as if it said "only
<foo> and topicality discussions are topical here"!]

I consider discussions of topicality as off-topic here -- certainly, they
have nothing to do whatsoever with C -- but acceptable.

There is a long standing convention that discussion of topicality is
always topical.

There's plenty
that I don't think is inappropriate for this group, and plenty that I
think is off-topic for this group, but neither is a strict subset of the
other. Just something that might be worth considering; I don't believe
this is covered by your groups.

I consider pre-standard C to be topical, and C as defined by all
versions of the standard, including drafts (some implementations may
have been updated to be in line with a draft before the next version is
published).

I see nothing wrong with using real (or imaginary) implementations to
make a point. Sometimes even posting a snippet of assembler to
illustrate what optimisations a compiler can do is appropriate.

Questions asking "can you do X" or "is Y part of standard C" are always
acceptable in my opinion. The answer might be no, or only by using
extensions, but that is not a problem.

Questions about the Windows API or Unix extensions or extensions for
embedded systems I would not consider topical, since there are
groups/mailing-lists/web-forums focused on these things. The same
applies to instructions on how to install/use specific toolsets,
although the odd suggestion of switches to get better diagnostics is fine.

Questions about which toolset to use are marginally topical but probably
better dealt with on groups about the relevant OS. I'm unlikely to
complain about such discussions.

I see nothing wrong with providing an answer to a question I consider to
be off topic *provided* it includes a redirection to somewhere more
appropriate for the discussion to be continued.

I also don't mind the odd digression, although I reserve the right to be
a grumpy git at times.

 

[pemo]

Group P (Purists): only code that can be written entirely in K&R C, C90, or
C99 is topical here; where functions or symbols that are not defined by
the Standard are mentioned, the source code should be available, at least
in principle, and that source code should be written entirely in one of
the above "standard dialects". Citations from K&R, from either of the
ISO/IEC 9899 documents, or from drafts are all acceptable.

Group N (Neopurists): only code that can be written entirely in the latest
version of C is topical here; where functions or symbols that are not
defined by the Standard are mentioned, the source code should be
available, at least in principle, and that source code should be written
entirely in C99. Drafts of the Standard don't count, and any citation from
a draft is off-topic.

Group C (Conservatives): whilst the principal focus should be on standard
C, discussions of particular implementations and extensions for
illustrative purposes are acceptable.

Group M (Moderates): As well as standard C, common extensions should be
considered topical (e.g. those that are common in Windows (GUI and
console) and in Unix-a-likes).

<snip>

My first thoughts are I'd like to go with P, N and M [does M include
C?]

However - isn't this purely academic? Nothing will change here IMHO.

 

[Harald van Dijk]

[Discussions of topicality are always topical. Nothing in this article
should be construed as suggesting otherwise - where I say something
like "only <foo> is topical here", this should be read as if it said
"only <foo> and topicality discussions are topical here"!]

I consider discussions of topicality as off-topic here -- certainly,
they have nothing to do whatsoever with C -- but acceptable.

Topicality is always on topic.

Topicality discussions are always appropriate on newsgroups, but they
rarely have anything to do with the topic of the groups. If you want to
call messages that do not relate to the topic of a group topical, go
ahead, but I don't think it's a good idea, so I won't.

 

[Eric Sosman]

[...]
I'll assign a name and a letter to each group in an attempt to make it
easier for us all to discuss the various camps.

Group P (Purists): [...]
Group N (Neopurists): [...]
Group C (Conservatives): [...]
Group M (Moderates): [...]
Group X (eXperimentalists): [...]
Group L (Liberals): [...]
Group A (Anarchists): [...]

Group EAS (figure it out): A group of one who feels no sympathy
for what he perceives as a false dichotomy. (Eptachotomy?)

 

[Mark McIntyre]

On Sat, 29 Sep 2007 13:16:42 +0000, in comp.lang.c , Richard

Thanks. Over to you.

IMHO, given that several dozen other groups exist specialising in
windows, linux, databases, GUI programming, networking, etc etc etc
etc, the scope of this group is just fine.
--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

 

[Malcolm McLean]

Group C (Conservatives): whilst the principal focus should be on standard
C, discussions of particular implementations and extensions for
illustrative purposes are acceptable.

I'll vote for C.

Firstly what irritates some people is not so much the fact of off-topic
redirections as the tone. Supercilious references to "your particular
platform" when the particular platform is Windows don't always help. The
reality is that for many people Windows is computing, and no one likes to be
made to feel inferior because he doesn't have a nice Unix box to play with.

Then I don't think there is enough in the standard itself to sustain a
newsgroup. There is a limit to the amount you can say about padding and
bitfields. Nor, and this is strictly a personal view, do I find
standards-based threads the most useful.

If we are move away from the standard that implies some relaxation of
topicality. This implies a certain amount of commonsense. A thread about
linked lists is appropriate, but there comes a point at which data
structures become so specialised as to belong in comp.programming. However
it is not the end of the world if the line isn't always drawn exactly where
it ought to be. Similarly indscussion of particular compilers or development
environments. There comes a point at which the discussion ceases to be of
any interest to those who don't use those tools, however knowing that xcc
has a flag to turn on bounds checking and that it is implemented with triple
pointers is not that point.

Also quantity is important. For instance one poster to comp.programming used
to write about left-wing politics, though with some relation to programming.
A single thread on Marxist analysis of software development is OK, if there
are constant atempts to derail every programming discussion into politics
then it become wearing. The same goes for people's projects. We are
interested in hearing what you are working on, but not ad nauseum.

 

[Ian Collins]

Group C (Conservatives): whilst the principal focus should be on standard
C, discussions of particular implementations and extensions for
illustrative purposes are acceptable.

This gets my vote..

Group X (eXperimentalists): These folks would like discussions of possible
changes and enhancements to the language to be considered topical.

With a spot of this from time to time to liven things up.

The definition of C would open up the scope to include POSIX, provided
the appropriate links to the OpenSolaris or BSD source was included

 

[Keith Thompson]

I see nothing wrong with providing an answer to a question I consider
to be off topic *provided* it includes a redirection to somewhere more
appropriate for the discussion to be continued.

[...]

A problem with that is that off-topic answers tend to lead to lengthy
off-topic discussions. I've seen too many cases where someone
provides an answer that is, for example, Unix-specific and wrong -- or
that somebody here incorrectly thinks is wrong. (Leaving bad
information uncorrected here just isn't going to happen, whether it's
topical or not.)

I suggest that a more effective strategy is to encourage the OP to
re-post the question in a more appropriate newsgroup. If I know the
answer, I can respond there, where my answer can be checked by (other)
experts.