Strange Question

[REH]

OK, I was at a code review today and got asked a strange questions (at
least I thought it was). I pointed out that the naming convention they
were using for macros clashed with one of the reserved forms for
identifies in the standard. After some shock and disbelief, the lead
of the project agreed to change to convention. One of the engineers
piped up and asked, "Do we really want to accept the standard?" I
didn't know how to answer other than response, "If you are writing C
code, why would you not?"

REH

 

[Richard Heathfield]

REH said:

OK, I was at a code review today and got asked a strange questions (at
least I thought it was). I pointed out that the naming convention they
were using for macros clashed with one of the reserved forms for
identifies in the standard. After some shock and disbelief, the lead
of the project agreed to change to convention. One of the engineers
piped up and asked, "Do we really want to accept the standard?" I
didn't know how to answer other than response, "If you are writing C
code, why would you not?"

There are definite benefits to /not/ accepting the Standard. It means you
can write any old junk you like and expect it to work. If it doesn't, you
simply yell "there's a bug in the compiler!" and switch to another vendor.
And another. And another...

 

[Richard Tobin]

There are definite benefits to /not/ accepting the Standard. It means you
can write any old junk you like and expect it to work. If it doesn't, you
simply yell "there's a bug in the compiler!" and switch to another vendor.

It's much the same if you *do* follow the standard...

-- Richard

 

[bcaller]

Really.. you should follow conventions. In practise, who can be
bothered
____________
Compilr - the online C#, C, VB.NET, Fortran compiler -
http://www.caller.me.uk/Compilr

 

[Keith Thompson]

Really.. you should follow conventions. In practise, who can be
bothered

Conventions like quoting previous context when posting a followup?

There are reports that groups.google.com has updated their software so
followups include the previous article by default. Did you
deliberately delete the lines starting with "> "?

 

[spibou]

One of the engineers
piped up and asked, "Do we really want to accept the standard?" I
didn't know how to answer other than response, "If you are writing C
code, why would you not?"

Because your compiler supports some very useful extensions ?

 

[spibou]

Conventions like quoting previous context when posting a followup?

Similar remarks have been made to me and I must say I find them very
puzzling.

First , do people who post here really have trouble remembering
the context of a thread which only contains 5-6 posts ? I don't see how
anyone can participate in a thread if their memory is that bad.

Second , is it so hard for people to obtain a newsreader which can
display all the
messages of a thread together and make it easy to switch back and
forth between messages ? Pine can do that , is available for a large
variety of platforms and free. I'm sure emacs can as well. I'm sure the
same applies to many other newsreaders.

Third , if it is so hard why not access the group through Google ? Or
is
that considered sacrilegious ?

Quoting context when the context is obvious is distracting and wastes
the reader's time.

 

[kyle york]

Greetings,

Keith Thompson wrote:




Similar remarks have been made to me and I must say I find them very
puzzling.

First , do people who post here really have trouble remembering
the context of a thread which only contains 5-6 posts ? I don't see how
anyone can participate in a thread if their memory is that bad.

If I only read one thread/day, this wouldn't be a problem. When I read
several hundred it is. Also realize that my newsfeed may not get all of
the messages, or may get them out of order in which case I've no context
to determine to what the responses relate.

Third , if it is so hard why not access the group through Google ? Or
is
that considered sacrilegious ?

I download messages & read at my leasure, often times when not connected
to a network. Google doesn't help much here, does it?

Quoting context when the context is obvious is distracting and wastes
the reader's time.

Do explain the logic in that statement. Someone can read this message as
written, without any others around it, and follow it clearly. How
exactly is this distracting and wasting time? If I don't need the
context, it's easy enough to skip the quoted parts.

 

[spibou]

Do explain the logic in that statement. Someone can read this message as
written, without any others around it, and follow it clearly. How
exactly is this distracting and wasting time? If I don't need the
context, it's easy enough to skip the quoted parts.

But in order to realize that _you_could_have_skipped_them_
you need to read them first. If it turns out that you didn't need
to read them then you have wasted your time. Furthermore if
you are replying to a long post and you need to concentrate on
specific parts to comment on it makes it harder to locate those
parts on repeated readings if there is additional stuff.

Spiros Bousbouras

 

[Flash Gordon]

Because your compiler supports some very useful extensions ?

And when you then need, or would find it useful, to use a different
compiler?

There are times when extensions are required, but why limit yourself to
your current platform without good reason? See
http://clc-wiki.net/wiki/Portability_and_ANSI_C_Compliance for rather
more reasons for following the standard.

 

[Eric Sosman]

REH wrote:




Because your compiler supports some very useful extensions ?

It's one thing to make a conscious decision to stray
outside the Standard's bounds, but it's quite a different
matter to stumble over the line without knowing you've
done so. The former is the exercise of choice in the
never-ending balance between conflicting goals that is
engineering. The latter is just a display of ignorance.

In my country, the standards of automobile operation
tell me to drive on the right side of the road and allow
opposing traffic to use the left side (from my perspective,
obviously). Occasionally I will nonetheless steer my car
into the left lane in order to pass a slower vehicle or
make an awkward turn or something, but I do so only as
a conscious matter, having assured myself that it's the
proper maneuver for the moment and that it is safe; I still
"accept the standard" even though I cross the line. I put
it to you that you would not want to share a roadway with
someone who was flat-out unaware of the "keep right" rule,
much less with someone who rejected it out of hand ...

 

[CBFalconer]

Similar remarks have been made to me and I must say I find them very
puzzling.

First , do people who post here really have trouble remembering
the context of a thread which only contains 5-6 posts ? I don't see how
anyone can participate in a thread if their memory is that bad.

Why do you assume the other messages have ever arrived? Even if
they have, why do you assume they are still available?

Second , is it so hard for people to obtain a newsreader which can
display all the messages of a thread together and make it easy to
switch back and forth between messages ? Pine can do that , is
available for a large variety of platforms and free. I'm sure emacs
can as well. I'm sure the same applies to many other newsreaders.

See above.

Third , if it is so hard why not access the group through Google ? Or
is that considered sacrilegious ?

Google is an inefficient pain in the neck means of accessing
usenet. All the above caveats still apply, and in addition google
mungs articles.

Quoting context when the context is obvious is distracting and wastes
the reader's time.

On the contrary, it makes the article self-cohesive and coherent.

--
"Our enemies are innovative and resourceful, and so are we.
They never stop thinking about new ways to harm our country
and our people, and neither do we." -- G. W. Bush.
"The people can always be brought to the bidding of the
leaders. All you have to do is tell them they are being
attacked and denounce the pacifists for lack of patriotism
and exposing the country to danger. It works the same way
in any country." --Hermann Goering.

 

[spibou]

(e-mail address removed) wrote:

Why do you assume the other messages have ever arrived? Even if
they have, why do you assume they are still available?

Because I don't know enough about how usenet works no doubt. Where
could I find out more about how the data flows on usenet ie how the
information
travells from the moment one writes a message on one's computer either
using
a newsreader or Google until it finds its way into other people's
computers ?

Google is an inefficient pain in the neck means of accessing
usenet. All the above caveats still apply, and in addition google
mungs articles.

Mungs them in what way ?

One thing I like about Google is how it displays several messages of a
thread in one
page. I find it much more convenient to go back and forth between
messages
and compare what they're saying if they are on the same page rather
than
having only one message on a page. Is there a newsreader which does
that ?

Spiros Bousbouras

 

[Chris Torek]

But in order to realize that _you_could_have_skipped_them_
you need to read them first. [snippage] Furthermore if
you are replying to a long post and you need to concentrate on
specific parts to comment on it makes it harder to locate those
parts on repeated readings if there is additional stuff.

[Beeblebrox's remarks snipped, because Mr Bousbouras is certain he
will have already read them, and remember them.]

Mr Bousbouras, would you care to respond to Mr Beeblebrox's remarks?

 

[Mark McIntyre]

First , do people who post here really have trouble remembering
the context of a thread which only contains 5-6 posts ?

Today I can see ~130 messages in 30 different threads, in one group
alone. I follow around thirteen technical groups covering C, C++,
Samba, wireless networking and network security. Trying to remember
every single message in every single thread is pretty much impossible.
And this isn't my day-job, which is running a complex project
involving cabling up and kitting out a new building, testing and
deploying around 130 applications and shipping users into the space,
half way round the world from my own office. And I'm not getting any
younger either. Sometimes I get home and call my kids the wrong names.

Second , is it so hard for people to obtain a newsreader which can
display all the messages of a thread together and make it easy to switch back and
forth between messages ?

I'd counter by asking whether its really so hard to keep appropriate
context, as I've done here. Personally I find this much easier *when
making posts*, as it helps me keep my own thoughts in order.

Pine can do that , is available for a large
variety of platforms and free. I'm sure emacs can as well. I'm sure the
same applies to many other newsreaders.

Mine can too, but I choose not to fill my hard disk with thousands of
defunct messages, and my ISP chooses not to retain messages for more
than about a week.

Third , if it is so hard why not access the group through Google ? Or
is that considered sacrilegious ?

So you're saying I should have to fire up *another* application, to
find out what someone said a few days back? Why me?
--
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

 

[spibou]

And when you then need, or would find it useful, to use a different
compiler?

You rewrite the programme or you modify the compiler (if you have the
source) or you don't use it at all.

There are times when extensions are required, but why limit yourself to
your current platform without good reason? See
http://clc-wiki.net/wiki/Portability_and_ANSI_C_Compliance for rather
more reasons for following the standard.

I don't doubt that there are good reasons for following the standard.
You asked
for a reason for *not* following the standard which I provided.

 

[spibou]

But in order to realize that _you_could_have_skipped_them_
you need to read them first. [snippage] Furthermore if
you are replying to a long post and you need to concentrate on
specific parts to comment on it makes it harder to locate those
parts on repeated readings if there is additional stuff.

[Beeblebrox's remarks snipped, because Mr Bousbouras is certain he
will have already read them, and remember them.]

Mr Bousbouras, would you care to respond to Mr Beeblebrox's remarks?

Mr. Beeblebrox is an idiot and his comments are not worth replying to


And there's no need for that Mr. Bousbouras stuff ; Spiros will do.

Spiros Bousbouras

 

[Keith Thompson]

Similar remarks have been made to me and I must say I find them very
puzzling.

The issue has been discussed here at great length many times. One of
the *good* things about Google Groups is that it lets you go back and
read old discussions.

[snip]

Third , if it is so hard why not access the group through Google ?
Or is that considered sacrilegious ?

Jumping over to Google Groups to look up the context of a single
article is rarely worth the effort. If you post here, it's your
responsibility to make it easy for your readers to know what you're
talking about.

Quoting context when the context is obvious is distracting and wastes
the reader's time.

Contextless followups are distracting and waste the readers' time.

 

[Giannis Papadopoulos]

You rewrite the programme or you modify the compiler (if you have the
source) or you don't use it at all.

How logical does it sound to modify the compiler (and all its future
versions) just to comply with a non-conforming program?

Have you ever tried to alter even one line in your compiler and then
assert its correct operation?


--
one's freedom stops where others' begin

Giannis Papadopoulos
Computer and Communications Engineering dept. (CCED)
University of Thessaly
http://dop.freegr.net/

 

[Peter Nilsson]

OK, I was at a code review today and got asked a strange questions (at
least I thought it was). I pointed out that the naming convention they
were using for macros clashed with one of the reserved forms for
identifies in the standard.

Which ones, __XXXX or EXXXX?

The former is a definite no no, the latter is less so.

After some shock and disbelief, the lead
of the project agreed to change to convention. One of the engineers
piped up and asked, "Do we really want to accept the standard?" I
didn't know how to answer other than response, "If you are writing C
code, why would you not?"

Do they want to be able to demonstrate code correctness on the
basis of code semantics, or purely on the basis that it runs and
gives the right output on the current development suite and
target?

Do they want their code to be robust with respect to changes of
compiler,
standard library or platform?

Do they anticipate using portions of the same code in future projects?

The C standard is not the only standard with respect to C. For
instance,
POSIX has extensions, restrictions, and is even in contradiction with
ISO C in a few respects.

The issue comes down to whether the code being written is to be
semantically sound. If they are relying on assumptions and extensions,
that's fine so long as they have a rigorous basis for trusting those
assumptions and extensions.

If they are flying in the face of some of the C semantics, then
the real question is whether they 'need' to do so, and what do
they actually gain from doing that? If the answer is no and nothing
much respectively, then writing non standard code is what Larry Wall
calles False Laziness.

It may not hurt them in the short term, but it may well hurt them
in the long term.

The C language certainly has some problems, but ignoring
its core has the tendancy of introducing unnecessary future
complications.

But at the end of the day, some people actually enjoy having
to rewrite source code. ;-)