• Re: [meta] O tempora, o mores! (was Re: does an expression in alogical

    From DAVID BROWN@3:633/10 to COMP.LANG.C++,COMP.LANG.C on Thursday, November 19, 2026 11:02:00
    Subject: Re: [meta] O tempora, o mores! (was Re: does an expression in alogical statement

    To: COMP.LANG.C++,COMP.LANG.C
    From Newsgroup: comp.lang.c

    On 19/11/2026 09:25, Janis Papanagnou wrote:
    On 2026-11-19 08:39, David Brown wrote:
    On 18/11/2026 23:44, Keith Thompson wrote:
    Ross Finlayson <ross.a.finlayson@gmail.com> writes:
    [...]
    [...]

    C and C++ are distinct languages. They are treated by
    implementations
    as distinct languages. "C/C++" is a meaningless marketing term.

    To be fair, "C/C++" was not as meaningless in the very early days of
    C++

    "C/C++" is still a sensible shortcut formulation if one is referring
    to
    common and typical elements of these [two] languages.


    Sure. But I think you need to be careful that it is clear from the
    context whether you are talking about C /and/ C++, C /or/ C++, a common

    subset of the languages, or an imaginary single language. Shortcuts
    are
    fine when everyone knows what you mean by them!

    In fact, with C++ I don't think it always makes sense to talk about
    "The
    C++ language" any more, without additional context. C++11 is a very significantly different language from the original C++98, and C++23
    (and
    especially the up-and-coming C++26) is very different again. If and
    when C23 gains significant popularity, the same could perhaps be said
    of C.

    (That's all subjective opinion, and open to free debate!)

    when the language was an experimental "C with classes" extension to
    C.
    Thus there is the rather well-renowned "C/C++ Users Journal". But
    by
    the time Stroustrup's "The C++ Programming Language" was published
    in
    1985, I would say C++ was established as its own language
    independent
    of (but cooperating with) C. Other people may pick a different date

    or event for the division, but they have certainly been distinct
    languages for a very long time now.

    Janis


    --- Synchronet 3.19b-Linux NewsLink 1.113
    ? Synchronet ? Mythical Kingdom
    --- Synchronet 3.19b-Linux NewsLink 1.113
    * Mythical Kingdom SBBS - Dayton OH - telnet://bbs.mythicalkingdom.com

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From JANIS PAPANAGNOU@3:633/10 to COMP.LANG.C++,COMP.LANG.C on Thursday, November 19, 2026 11:17:00
    Subject: Re: [meta] O tempora, o mores! (was Re: does an expression in alogical statement

    To: COMP.LANG.C++,COMP.LANG.C
    From Newsgroup: comp.lang.c

    On 2026-11-19 11:02, David Brown wrote:
    On 19/11/2026 09:25, Janis Papanagnou wrote:
    On 2026-11-19 08:39, David Brown wrote:
    On 18/11/2026 23:44, Keith Thompson wrote:
    Ross Finlayson <ross.a.finlayson@gmail.com> writes:
    [...]
    [...]
    C and C++ are distinct languages. They are treated by
    implementations
    as distinct languages. "C/C++" is a meaningless marketing term.

    To be fair, "C/C++" was not as meaningless in the very early days
    of C++

    "C/C++" is still a sensible shortcut formulation if one is referring
    to
    common and typical elements of these [two] languages.

    Sure. But I think you need to be careful that it is clear from the
    context whether you are talking about C /and/ C++, C /or/ C++, a
    common
    subset of the languages, or an imaginary single language.

    All fair points. - Where the context requires it they should clearly
    be differentiated.

    (I had the impression that in Ross posts it was clear, but given that
    there's some confusion with his style and thinking it might not have
    been as obvious and unambiguous as it may have appeared to me.)

    Shortcuts are fine when everyone knows what you mean by them!

    In fact, with C++ I don't think it always makes sense to talk about
    "The
    C++ language" any more, without additional context. C++11 is a very significantly different language from the original C++98, and C++23
    (and
    especially the up-and-coming C++26) is very different again.

    All very true.

    If and
    when C23 gains significant popularity, the same could perhaps be said
    of C.

    Yes, in context cases where the difference matters.

    Janis

    [...]

    --- Synchronet 3.19b-Linux NewsLink 1.113
    ? Synchronet ? Mythical Kingdom
    --- Synchronet 3.19b-Linux NewsLink 1.113
    * Mythical Kingdom SBBS - Dayton OH - telnet://bbs.mythicalkingdom.com

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From DAVID BROWN@3:633/10 to COMP.LANG.C++,COMP.LANG.C on Thursday, November 19, 2026 13:33:00
    Subject: Re: [meta] O tempora, o mores! (was Re: does an expression in alogical statement

    To: COMP.LANG.C++,COMP.LANG.C
    From Newsgroup: comp.lang.c

    On 19/11/2026 11:17, Janis Papanagnou wrote:
    On 2026-11-19 11:02, David Brown wrote:
    On 19/11/2026 09:25, Janis Papanagnou wrote:
    On 2026-11-19 08:39, David Brown wrote:
    On 18/11/2026 23:44, Keith Thompson wrote:
    Ross Finlayson <ross.a.finlayson@gmail.com> writes:
    [...]
    [...]
    C and C++ are distinct languages. They are treated by
    implementations
    as distinct languages. "C/C++" is a meaningless marketing term.

    To be fair, "C/C++" was not as meaningless in the very early days
    of
    C++

    "C/C++" is still a sensible shortcut formulation if one is
    referring to
    common and typical elements of these [two] languages.

    Sure. But I think you need to be careful that it is clear from the
    context whether you are talking about C /and/ C++, C /or/ C++, a
    common subset of the languages, or an imaginary single language.

    All fair points. - Where the context requires it they should clearly
    be differentiated.

    (I had the impression that in Ross posts it was clear, but given that
    there's some confusion with his style and thinking it might not have
    been as obvious and unambiguous as it may have appeared to me.)


    To the limited extent that his posts seemed clear to me, he had been
    talking about the imaginary single language that compilers "treat as
    being the same". It does not appear that he understands that they are
    two different languages sharing a common subset, and that they are
    often
    used in very different ways. So if he was clear, he was clearly wrong.

    But maybe I (and others) have misunderstood him. It would be helpful
    if
    he clarified things, and said that he now understands what Keith and I
    wrote. Once we get on common ground, maybe we can move forward. But I

    am not holding my breath.

    --- Synchronet 3.19b-Linux NewsLink 1.113
    ? Synchronet ? Mythical Kingdom
    --- Synchronet 3.19b-Linux NewsLink 1.113
    * Mythical Kingdom SBBS - Dayton OH - telnet://bbs.mythicalkingdom.com

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From JANIS PAPANAGNOU@3:633/10 to COMP.LANG.C++,COMP.LANG.C on Thursday, November 19, 2026 14:36:00
    Subject: Re: [meta] O tempora, o mores! (was Re: does an expression in alogical statement

    To: COMP.LANG.C++,COMP.LANG.C
    From Newsgroup: comp.lang.c

    On 2026-11-19 13:33, David Brown wrote:

    To the limited extent that his posts seemed clear to me, he had been
    talking about the imaginary single language that compilers "treat as
    being the same". It does not appear that he understands that they
    are
    two different languages sharing a common subset, and that they are
    often
    used in very different ways. So if he was clear, he was clearly
    wrong.
    But maybe I (and others) have misunderstood him. It would be helpful
    if
    he clarified things, and said that he now understands what Keith and
    I
    wrote. Once we get on common ground, maybe we can move forward. But
    I
    am not holding my breath.

    It may be that I've read past some "formally wrong" statements he made.
    It's that I usually try to understand what one wants to express beyond
    how it's actually formulated. - Maybe it's what (I think) Dan has once formulated; that it might distract the discussion if one isn't focused
    on the topic but more concerned about "exact" representation. - I think
    that observation is worthwhile to think about! - But I also know that
    some folks - you know who they are! - are now tempted to point out the
    very importance of exact representation. That's also not wrong but it's
    missing the point made.[*]

    Janis

    [*] I tried anticipating that point to prevent unnecessary replies.

    --- Synchronet 3.19b-Linux NewsLink 1.113
    ? Synchronet ? Mythical Kingdom
    --- Synchronet 3.19b-Linux NewsLink 1.113
    * Mythical Kingdom SBBS - Dayton OH - telnet://bbs.mythicalkingdom.com

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From DAVID BROWN@3:633/10 to COMP.LANG.C++,COMP.LANG.C on Thursday, November 19, 2026 14:52:00
    Subject: Re: [meta] O tempora, o mores! (was Re: does an expression in alogical statement

    To: COMP.LANG.C++,COMP.LANG.C
    From Newsgroup: comp.lang.c

    On 19/11/2026 14:36, Janis Papanagnou wrote:
    On 2026-11-19 13:33, David Brown wrote:

    To the limited extent that his posts seemed clear to me, he had been

    talking about the imaginary single language that compilers "treat as

    being the same". It does not appear that he understands that they
    are
    two different languages sharing a common subset, and that they are
    often used in very different ways. So if he was clear, he was
    clearly
    wrong. But maybe I (and others) have misunderstood him. It would be

    helpful if he clarified things, and said that he now understands
    what
    Keith and I wrote. Once we get on common ground, maybe we can move
    forward. But I am not holding my breath.

    It may be that I've read past some "formally wrong" statements he
    made.
    It's that I usually try to understand what one wants to express
    beyond
    how it's actually formulated. - Maybe it's what (I think) Dan has
    once
    formulated; that it might distract the discussion if one isn't
    focused
    on the topic but more concerned about "exact" representation. - I
    think
    that observation is worthwhile to think about! - But I also know that
    some folks - you know who they are! - are now tempted to point out
    the
    very importance of exact representation. That's also not wrong but
    it's
    missing the point made.[*]

    Janis

    [*] I tried anticipating that point to prevent unnecessary replies.


    I think it is usually possible to tell what a poster means, even if
    they
    are, as you put it, "formally wrong". For example, it does not matter
    if they refer to a constant 1234 as an "integer literal" (the term in
    C23 and C++, IIRC), an "integer constant" (the term in C prior to C23),

    or non-standard but obvious terms like a "literal int" or "explicit
    integer in the source code". It can make sense to steer the poster
    towards the standard terms because it keeps things absolutely clear and

    is helpful if someone wants to look things up in the standards, but it
    it rarely useful to focus on it.

    However, if someone is mixing up different concepts - "integer
    literals"
    and "values of objects of type int" - then it is definitely worth
    correcting and clearing things up.

    As far as I can see, Ross' "C/C++" mixup here is more in that second
    category. But I can't be sure until he replies and clarifies (if he
    chooses to do so).

    --- Synchronet 3.19b-Linux NewsLink 1.113
    ? Synchronet ? Mythical Kingdom
    --- Synchronet 3.19b-Linux NewsLink 1.113
    * Mythical Kingdom SBBS - Dayton OH - telnet://bbs.mythicalkingdom.com

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)