NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part1/3461

Internet Message Format | 1996-08-05 | 8.4 KB

Path: newshost.lanl.gov!tanmoy
From: [e-mail] (Tanmoy Bhattacharya)
Newsgroups: comp.lang.c
Subject: Re: Off topic post
Date: 29 Jan 1996 06:56:29 GMT
Organization: Los Alamos National Laboratory
Message-ID: <[e-mail]>
References: <4dvrq8$[e-mail]> <[e-mail]>
    <[e-mail]> <4egtr8$[e-mail]>
NNTP-Posting-Host: qcd.lanl.gov
Mime-Version: 1.0
Content-Type: text
In-reply-to: [e-mail]'s message of 28 Jan 1996 22:37:28 GMT

In article <4egtr8$[e-mail]> [e-mail] (Ray Dunn) writes:
<snip>
   >I do not disagree that my responses are sometimes needlessly
   >unpleasant. What I object to is not use of keywords like `far', but
   >a lack of knowledge that the posted code is not valid C.

   If you agree that they're needlessly unpleasant, how can you justify
   them in this forum?

When they are indeed needlessly unpleasant, I do not justify them: I
apologize. The reason they sometimes happen is that irritation is not
something easy to keep in check. I should have stated that I do not
consider this as an instance of needless unpleasntness, or I would not
have responded in the first place.

On rereading I realize that I forgot to underline sometimes in my
statement, my fault.

   I thought part of the reason d'etre of this newsgroup was to *teach*
   people what is and isn't conforamnt code.  In addition, some extensions
   are required when writing on certain platforms.  Although it can be
   possible under certain circumstances to manipulate your code and the
   conditions under which it is compiling to avoid having to use some
   of them, it can be cumbersome, is not always possible, and ceratinly is
   often not productive.

   I wouldn't post such non-conformant code, but to make more than passing
   reference to it when someone does is unnecessary - so long as the
   problem being presented doesn't *rely* on the use of these extensions -
   and if they do, it can be pointed out to them without being nasty.

I had written `I cannot discuss the code as written because it uses
`far' and every conforming compiler has to diagnose this as a syntax
violation.' (words to that effect: I do not remember the exact words.)
Do you consider that nasty or rude? I do think that this could be
considered a `passing reference' strong enough to point out that it is
not conformant.

   >   A post with a keyword like far tells me that the user, probably,
   >does not know the difference between the language as defined, and a
   >compiler extension. You will notice that my response to a person who
   >used an incomplete array type in a struct used essentially similar
   >language.

   I tend to differentiate between extensions required for a platform and
   variations in syntax allowed by some compilers.  But even in that case,
   it's easy to be caught.  I have been myself - posted a query about an
   lvalue cast which is not legal ANSII but allowed by MSC.

Who hasn't made mistakes? But when I make one, I expect to be
corrected immediately: and am plainly disappointed if nobody notices
:-(

   Why should posters be castigated for this sort of error but not for a
   semantic error, like, for example using the wrong type for a variable,
   or being unable to cast correctly?  In all cases, if we could each
   spout *and* be understandingly familiar with the standard 100%, no-one
   would have to post *any* query.

Every question asked properly deserves an answer: no poster deserves
castigation for any behaviour, though it is easy to get irritated by
posters who do not read the FAQ. I hope I have only very rarely posted
anything rude in response to a question (except in exasparation: I
remember being especially unpleasant to someone who knew he/she was
asking a question about C++ in a C newsgroup). I am sometimes
unpleasant to people who _answer_ wrong (especially when they make
trivial mistakes showing they are absolute beginners, or when their
answer is wrong _and_ contrary to the FAQ), but, for those who ask, I
try my best to figure out what the misconception of the poster is:
instead of merely answering the question.

I do also point out glaring non-portabilities and other errors when I
notice them. I do not understand why you think that is unpleasant.

   >   In short, I believe that anyone writing in a language ought to know
   >the difference between portable and non-portable constructs.

   The language, as defined, is not just an intellectual curiosity, it is
   something that has been implemented in various real-world environments
   to achieve, in the main, commercial objectives.

   A *huge* number of programmers (probably the biggest percentage) are
   coding programs which by definition are not portable, because they are
   interfacing with operating system API's and dealing with difficult
   system environments, and *must* use language extensions.  I'd rather
   see their 'C' queries posted and responded to here when their problems
   are essentially language ones, than have them scared off.

   The language standard as defined is not just an intellectual curiosity,
   it is something that has been implemented in various real-world
   environments to achieve, in the main, commercial objectives. I think we
   are all here to help 'C' programmers learn and be more productive in
   the environment that they have to work, not just to have intellectual
   discussions.

Correct. But, they do need to distinguish the peculiarities of their
platform from questions which are problems of their understanding of
the basic language. If the question is really specific to their
platform, it probably is not a C question. (There are exceptions.) So,
how difficult is it to try to reproduce the problem in a portable
context before posting to the newsgroup?

   >Whether my replies are constructive, however, I really do not know.

   In the main, Tanmoy, I think most posters would agree that your
   responses are very constructive when they stick to technicalities.
   However it's often painful to observe your apparently deliberate
   attempts to humiliate and embarrass the authors of posts who don't fit
   into your idea of Politically Correct comp.lang.c questions.

I disagree that I want Politically Correct comp.lang.c questions. I
think that anyone asking questions needs to do a bit of homework to
isolate the problem from system dependencies. And, I do want people to
know what is and isn't portable.

As to your comment about humiliating and embarrassing authors, let me
state that I do not try either of those when I feel that somebody is
making an honest mistake not covered by the FAQ. It is difficult to
imagine why I should humiliate someone for not knowing something, if
they are already asking it. A certain amount of embarrassment arises
out of being wrong in the first place: that I cannot avoid. I myself
have made stupid errors here (as any who has read my post for a while
can testify), and sure, I am embarrassed at my own stupidity. That I
cannot avoid.

   It's unpleasant enough in the "chat" newsgroups.  Let's try and keep
   this place a pleasant and profitable place to browse.

   As far as this discussion being "off topic", Tanmoy is right of course,
   but in the absence of a comp.lang.c.d newsgroup where else would you
   suggest?

Such discussions are probably better by email if the number of
participants are few. The reason I changed the subject is not as any
reflection on your post, but rather to point out to the those that are
uninterested that they need not download this. If you feel that it is
relevant to this newsgroup, it should continue here. (In any case, I
do not have much else to say ... so this thread is going to die soon
anyway.)

   BTW, it's ironic that Tanmoy shows such sensitivity to "correctness" in
   posts to the newsgroup, while at the same time he uses a sig that's way
   out of line according to Usenet standards.

I agree that my sig is longer than that recommended. I have found it
useful to keep it this way though.

Cheers
Tanmoy
--
[e-mail](128.165.23.46) DECNET: BETA::"[e-mail]"(1.218=1242)
Tanmoy Bhattacharya O:T-8(MS B285)LANL,NM87545 H:#9,3000,Trinity Drive,NM87544
Others see <gopher://yaleinfo.yale.edu:7700/00/Internet-People/internet-mail>,
<http://alpha.acast.nova.edu/cgi-bin/inmgq.pl>or<ftp://csd4.csd.uwm.edu/pub/
internetwork-mail-guide>. -- <http://nqcd.lanl.gov/people/tanmoy/tanmoy.html>
fax: [phone]   voice: 1 [phone]    [ Home: 1 [phone] ]