Do you have the C or C++ standard? (serious question)

[Michael Mair]

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

C90: I have the last public draft for searching and there is one
copy of the standard in our department (~50 people, half of them
involved with C) which mostly resides on my desk because I often
look something up (either for me or for colleagues). This is not
for day-to-day-programming but for conformance questions around
our product.
I have the German translation of K&R2 but it is so awful that I
do not work with it.
C95: We don't have Amendment 1 or one of the TC's.
C99: I have that (privately) as pdf and use it mostly if looking
something up for comp.lang.c or as reference for library functions
if I do not like the man page.
N1124.pdf: The same; in fact, this is my preferred version.

C++98: No standard; this is a language which I work with daily,
so this is probably the "interesting" part for you.
For my work, the BS book and the c.l.c++ FAQ are sufficient and,
to put it bluntly, I certainly am not ready to spend money or
time for standardese I am not interested in. If the need would
arise, then of course I'd read it (and expect the company to
provide a copy).


Cheers
Michael

 

[Chris Hills]

[..]
I have asked this question in several places and the majority have
said they do not have a copy of the language standard for c or C++

Many have the K&R2 or the BS books but as far as most are concerned
the compiler manuals are more important than the standard.

Well, good for you!

What is? I don't follow your logic

The knowledge of how many programmers have a copy
of the Standard is about as useless as how many drivers have a copy
of the statute on driving/vehicles for their country/state. What is
it you're after?

No. Some one else said that all programmers always had a copy of the
standard to hand. I thought they did not. So far it appears that most
don't

 

[Dave Steffen]

Hi,

It came up in a standards panel meeting the other day that "all c or
C++ programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

I've got a pdf of the '98 C++ standard. I'm the only one (of 6)
software guys who has it, but the others usually come to me for that
sort of question anyway.

I'd say it's more likely that "many" have the standards, and most or
all programming groups have at least one.

 

[red floyd]

I've got the PDF's for C90, C99, C++98 and C++03 on my laptop.
I have never found the need for a hard copy.

I own the dead tree for C++03. I can't stand E-books.

 

[Default User]

Chris Hills wrote:

Well, do most of you have a copy of the relevant ISO language
standard of your own or is there one on your desk at work?

I have a PDF of the C++ standard, which is easy because my division of
the company has a site license for it. For C, which I'm not actually
working in professional capacity at this time, I just use n869. On the
whole I prefer the text. It's not 100% the final standard, of course.





Brian

 

[Martin Ambuhl]

Hi,

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

Yes, I have copies of the standards ISO/IEC 9899:1999 and ISO/ANSII
9899:1990, along with various TCs, as well as standards documents for
C++, Pascal, F77, F77, F90, Algol, Lisp, and ALGOL.

 

[Walter Roberson]

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

C89 and POSIX.1 (1990) on my desk within arms reach; one or the other might
migrate to the bookshelf for awhile, but it's easier to leave them
within reach.

ISO C++ (first version): in the room across the hall; I haven't used it
for a number of years.

All 3 standards were purchased on my budget, and it is rare for anyone else here
to refer to them. I actively think about and test for portability when
I'm coding; it doesn't seem much of a priority for others here.

 

[dcorbit]

Hi,

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?


--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills Staffs England /\/\/\/\/
/\/\/ (e-mail address removed) www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

I have the C standard, the C++ standard, and the SQL standard from ANSI
in PDF format.

I couldn't live without them. I don't refer to them daily (and
sometimes will even go for several weeks without touching them). But
when you need them, you really need them.

I spend far more time in the SQL standard than in the other documents.

 

[dcorbit]

Hi,

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?


--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills Staffs England /\/\/\/\/
/\/\/ (e-mail address removed) www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

I have the C99 standard, the C++ standard and the SQL standard (all
parts). All are official ANSI documents, purchased in PDF format.

How else am I going to know for sure if anything is correct? I
literally can't imagine working without them.

Now, I don't refer to them every day, but when I need one of them, I
really need it.

I spend more time in the SQL standard (by far) but it is also the most
complex.

 

[pete]

Hi,

It came up in a standards panel meeting
the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

I would guess that the percentage of regulars here
that do have a copy,
is much higher than the percentage of all C programmers
that have a copy.

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

I have the following as pdf's:
ISO/IEC 9899:1990
ISO/IEC 9899:1999 (E)
ISO/IEC 2382-l
ISO 1087-1
ISO 1087-2

 

[pete]

Chris Hills wrote:

I have asked this question in several places
and the majority have said
they do not have a copy of the language standard for c or C++

Many have the K&R2

I have the K&R2, and also the 1978 K&R,
as well as a bunch of ISO standards.

 

[Jack Klein]

Hi,

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

Personally, I own my own PDF copies of C90, C99, and the TCs to C99,
and C++98 and 03.

But I have worked at and with a lot of organizations over the years,
and by my sampling, which I think is fairly accurate, the vast
majority of C and C++ programmers not only do not have a copy of a
language standard, they don't feel the need for one and haven't ever
read one.

Most C programmers have a copy of K&R2 on their shelf, I don't see
nearly as many copies of Stroustrup (any version) on the C++
programmers' shelves. Other books abound, including H&S for the C
programmers, but there are also a distressingly large number of
Schildt books.

When the C++ standard was first published by ANSI as an $18.00 PDF, I
emailed the project lead of a large C++ development in our
organization (dozens of programmers), suggesting they just spend the
$18.00 per developer, or investigate getting a source license.

The result was total apathy. If it wasn't MFC, it wasn't worth even
thinking about.

And this is an organization that, in the years since, has reached CMM
Level 4. Code inspections for new code are mandatory, development and
documentation follows a mature process.

This is not the case with our embedded development, where much of the
work is safety critical and we have to provide mitigations for
hardware and electronic faults. We use PC-Lint to check for MISRA C
conformance and require ISO conformance except for documented
extensions.

For far too many programmers, formal language study ended with college
graduation. For those without formal programming education, they
learned by "fooling around" with Microsoft or Borland compilers on a
PC over the years.

I think you will find that the vast majority of working C and C++
programmers today do not have and do not feel the need for an actually
copy, electronic or paper, of the language standards.

 

[Neil]

Hi,

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

None Here, I have never seen a hard copy.

 

[Greg Comeau]

IMO, whoever said this is probably almost 100% incorrect.

Personally, I own my own PDF copies of C90, C99, and the TCs to C99,
and C++98 and 03.

But I have worked at and with a lot of organizations over the years,
and by my sampling, which I think is fairly accurate, the vast
majority of C and C++ programmers not only do not have a copy of a
language standard, they don't feel the need for one and haven't ever
read one.

I agree completely with Jack. And for the most part, the don't
by software books or magazines either on a consistent basis.

The result was total apathy. ...
For far too many programmers, formal language study ended with college
graduation. For those without formal programming education, they
learned by "fooling around" ...
programmers today do not have and do not feel the need ...

Again, completely agree.

 

[jacob navia]

Hi,

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

Yes, I have bought the PDF from ANSI? the book, and many TR.
But I am not representative since as a compiler writer I need those
things for my daily work...

jacob

 

[Victor Bazarov]

[..]
I have asked this question in several places and the majority have
said they do not have a copy of the language standard for c or C++

Many have the K&R2 or the BS books but as far as most are concerned
the compiler manuals are more important than the standard.

Well, good for you!

What is? I don't follow your logic

That you're satisfied with the majority of answers in the places you
asked. No matter whether it's representative of "all programmers", or
whether it's acceptable as a proof of anything, if you are talking of
it, you must like it. I am glad for you. That's my logic. You don't
have to follow it. I am just commending on your achievement.

No. Some one else said that all programmers always had a copy of the
standard to hand. I thought they did not. So far it appears that most
don't

How many did you ask? How many programmers are there? What's your
expected margin of error when conducting a poll of that sort (it is
possible that all answer truthfully and never forget about it)? Is
the number of programmers you've asked larger than the margin of error?

Make of it whatever you like, but always supply a disclaimer about how
trustworthy your statistical information is.

V

 

[Ivan Vecerina]

: It came up in a standards panel meeting the other day that "all c
: or C++ programmers" have a copy of ISO C and/or C++ ...
:
: I challenged this and said most don't (outside those working
: on the standards).
:
: Well, do most of you have a copy of the relevant ISO language
: standard of your own or is there one on your desk at work?

I have a personal purchased .pdf copy of the 98 C++ standard.
For extensions since, and the C99 standard, I satisfy myself
with free pre-final drafts available on the internet.

I am just a user of the C++ language (albeit an "advanced" one).
Yet I studied the standard because I care for formal correctness
of my programs.
Unfortunately, if feel I am in a small minority (<5%?).
I take it as my responsibility to write code that actually works
according to the formal specification of the language.

It is sad, but most (C++) programmers seem to satisfy themselves
with knowing that their code passes casual tests, or at best,
available test cases. They don't even care about undefined
behavior and the invisible failures/leaks that may come with it.
They never go beyond reading recipe books (a la "Effective C++").

Thing is, many languages do not even have a formal definition.
Their users are accustomed to the fact of simply relying on
a vendor or solution provider to do their best not to break
previous behavior...


Regards,
Ivan

 

[Michael Mair]

: It came up in a standards panel meeting the other day that "all c
: or C++ programmers" have a copy of ISO C and/or C++ ...
:
: I challenged this and said most don't (outside those working
: on the standards).
:
: Well, do most of you have a copy of the relevant ISO language
: standard of your own or is there one on your desk at work?

I have a personal purchased .pdf copy of the 98 C++ standard.
For extensions since, and the C99 standard, I satisfy myself
with free pre-final drafts available on the internet.

I am just a user of the C++ language (albeit an "advanced" one).
Yet I studied the standard because I care for formal correctness
of my programs.
Unfortunately, if feel I am in a small minority (<5%?).
I take it as my responsibility to write code that actually works
according to the formal specification of the language.

It is sad, but most (C++) programmers seem to satisfy themselves
with knowing that their code passes casual tests, or at best,
available test cases. They don't even care about undefined
behavior and the invisible failures/leaks that may come with it.
They never go beyond reading recipe books (a la "Effective C++").

This heavily depends on the environment.
I have only use for the standard if the BS book cannot answer my
questions -- up to now, I did not have a use for the standard.
As the subset of C++ I work in for my day job is rather restricted
(nearly no templates, nearly no namespaces, ... "without training for
all developers, we will not introduce something that may be used
wrongly") and the application is guaranteed to be never, ever ported
to another platform, the actual need for knowledge about some of the
darker corners of the language is not so large that it must be
satisfied by the standard itself. Template classes and functions are
provided by third party libraries.
As for recipe books: If a book explains the background correctly
and concisely, then in fact this can lead to more intensive study
of language aspects. If people are not interested because they made
their way based on recipes instead of understanding, then this is
another problem...

Note that I do not advocate sloppy programming; learning the language
by testing what the compiler eats (ideally, on the lowest warning
level) or programming by changing the source until the compilation,
linking or run-time error disappears without real understanding of
the cause will not find any mercy from my side.

Thing is, many languages do not even have a formal definition.
Their users are accustomed to the fact of simply relying on
a vendor or solution provider to do their best not to break
previous behavior...

True.


Cheers
Michael

 

[kwikius]

Hi,

It came up in a standards panel meeting the other day that "all c or C++
programmers" have a copy of ISO C and/or C++ ...

I challenged this and said most don't (outside those working on the
standards).

Well, do most of you have a copy of the relevant ISO language standard
of your own or is there one on your desk at work?

If this is the full extent of the C++ committees concerns, I reckon C++
is dead meat.

regards
Andy Little

 

[Steve Pope]

Chris Hills wrote:

This will not work. Who will say in public

"I do not have the standard's copy" ???

(Raises hand.)

I have a draft, which is free. I have yet to get bitten
by something in the draft that would have been different
in the actual standard.

Yes, I'm cheap sometimes.

Steve