home *** CD-ROM | disk | FTP | other *** search
- Path: solon.com!not-for-mail
- From: seebs@solutions.solon.com (Peter Seebach)
- Newsgroups: comp.lang.c
- Subject: Re: Off topic post
- Date: 1 Feb 1996 13:31:07 -0600
- Organization: Usenet Fact Police (Undercover)
- Message-ID: <4er4dr$eu7@solutions.solon.com>
- References: <TANMOY.96Jan27121202@qcd.lanl.gov> <DLyM02.9zn@emr1.emr.ca> <4ejfuc$i80@solutions.solon.com> <4eopgv$mub@crl13.crl.com>
- NNTP-Posting-Host: solutions.solon.com
-
- In article <4eopgv$mub@crl13.crl.com>, James Aragon <aragon@crl.com> wrote:
- >Peter Seebach (seebs@solutions.solon.com) wrote:
- >Then wouldn't it be to everyone's benefit if Tanmoy (and other comp.lang.c
- >regulars) were to assume that a person who posts a question that does not
- >follow those reasonable rules does NOT expect an answer from Tanmoy (or
- >other regular with the same "reasonable rules"). Then the best use of his
- >finite time is to SKIP that post, and leave it to those who DO want to
- >answer it. He need not waste his time commenting on posts that he doesn't
- >like.
-
- But if he has a newsreader of finite speed, it may take an hour or more to
- skip a hundred posts of no interest; an hour he could have spent answering
- questions. For that matter, what kind of fool would be explicitly looking for
- the sorts of incorrect or meaningless answers frequently posted by people other
- than those establihed regulars?
-
- Beyond any question of Tanmoy's time, there is the issue of the news
- hierarchy. If there is a better place for a question, ask it there, not here.
-
- >> 2 and 5 between them rule out the far keyword; either it is irrelevant
- >> to the question, or the question is not about C.
- > ^^^^^^^^^^^
-
- >Is anyone else as tired of this bit of false pedantry as
-
- You seem to have misunderstood the point completely.
-
- First, it's not "false" pedantry; it's pedantry explicitly covered by the
- relevant international standard. Definitionally, any program not following
- the rules in that sntandard is not a C program.
-
- However, in practice, we allow "programs with mistakes" for discussion.
-
- >I am? If we're
- >going to hear this claim repeatedly, could we please at least not omit the
- >word "standard" from in front of "C"? There was C long before there was a
- >C standard and there are many varieties of C other than standard C.
- >Although standard C is probably the most useful variety by virtue of
- >being, well, standardized, let's at least quit pretending that any
- >deviation from, or extension to, the standard, however slight, results in
- >some totally new, unnamed, hitherto unknown programming language with
- >completely mysterious and unknown semantics.
-
- Several points you seem to be missing:
- 1. It essentially does. A trivial extension like "exceptions are handled"
- can completely alter the semantics of a simple piece of code in nonobvious
- ways. Consider the effect of signal(SIGFPE, ...) on code.
- 2. This is the big one. When I say "this question is not about C", I'm not
- saying "it's actually about the distinct language borland-c-3.0-for-dos".
- I'm saying it is not about the *LANGUAGE* C. This newsgroup, comp.*LANG*.c,
- is for discussions of and about the C *language*.
-
- If your question concerns the semantics of an OS provided system call, *your
- question is about the OS*. Examples:
-
- C or not C: Explained!
-
- 1. "Unix provides a function execl() taking a sequence of char *
- arguments. I am wondering if I can just use "NULL" for the last
- one, or whether I need to cast it to (char *).
-
- 2. "I'm writing a translation program in C. I want a list of French
- words and their English Equivalents."
-
- 3. "I am looking for code in C to caculate Fibonacci numbers."
-
- Question 1 is the only one about C. The issue being discussed is an issue
- discussed in and defined by the C standard; the behavior of arguments to
- variadic functions. Question 2 has *nothing* to do with C. Question 3 is
- more sane for comp.sources,wanted. But it's not about the language; it's
- about an algorithm.
-
- Likely reasonable C topics might be things like "is this legal", "why does
- this code not work", or "how could I improve this".
-
- The central problem is, of course, that there is often no way for a novice
- user to determine whether or not the problem they have is dependant on OS
- specific semantics, or on the C language.
-
- The famed "DGROUP exceeds 64k" message is produced by a C compiler, but has
- nothing to do with the language. It's closer to a message like "/: write
- failed: file system full." which has nothing to do with C, even when a
- compiler says it.
-
- Generally, if your compiler doesn't seem to be working, you should talk to
- your compiler's tech support; it's their product. If your OS doesn't do what
- you want it to do, talk to the vendor. If your graphics library doesn't work,
- talk to the vendor. If you think it may be a usage problem, go to a group
- related to the source of the problem, not simply the first one you find that
- has a related word in the name.
-
- -s
- --
- Peter Seebach - seebs@solon.com - Copyright 1995 Peter Seebach.
- C/Unix wizard -- C/Unix questions? Send mail for help. No, really!
- Using trn? Weird new newsgroup problem? I know the fix! Email me!
- The *other* C FAQ - ftp taniemarie.solon.com /pub/c/afq - Not A Flying Toy
-