PL/I Language Reference (IBM)

[Robin Vowels]

On Mar 16, 7:31 am, John W Kennedy <jwke...@attglobal.net> wrote:

> The IBM PL/I standard was last updated in the late
> 60s, but is often a good guide for why a modern compiler does something
> that seems strange.

The IBM standard is the current reference, not some manual that was
out-of-date
in the 1960s.

That manual was replaced with a new one back in the 1960s.

Since the original manual was issued, various new features were added
to the language,
including DO FOREVER, DO UNTIL, REORDER, not to mention the many new
features added in the 1990s including defined types, etc etc, new
built-in functions (perhaps 50). Some features were removed or
changed.

Forget the 1960s manuals. They are grossly obsolete and have no
bearing on
IBM's current compilers.

The PL/I "Language Reference" is the most appropriate reference for
IBM's
compilers.

[Peter Flass]

I have a pretty large collection of PL/I manuals for many systems.
Until IBM's "recent" changes the implementations were all pretty
consistent and followed either IBM's implementation, which is (mostly)
full ANSI with extensions, or "Subset G".

Some of what IBM has done is good, I wish someone could get a new
standards committee to look at it, but apparently it isn't going to happen.

--
Pete

[John W Kennedy]

On 2013-03-16 08:56:30 +0000, Robin Vowels said:

> On Mar 16, 7:31 am, John W Kennedy <jwke...@attglobal.net> wrote:
>
>> The IBM PL/I standard was last updated in the late
>> 60s, but is often a good guide for why a modern compiler does something
>> that seems strange.
>
> The IBM standard is the current reference, not some manual that was
> out-of-date
> in the 1960s.

No, the current Language Reference, like its earlier versions, the
Language References for the Optimizer and Checker and Language
Reference for (F), is hopelessly inadequate as a standard. Over and
over again here I have pointed out places where:

The compiler does something a user didn't expect,
The Language Reference doesn't explain why, and
The Language Specification does.

Whether you like it or not, whether I like it or not, huge parts of the
language are still implemented according to the Language Specification
where the Language Reference is silent or ambiguous.

--
John W Kennedy
"Only an idiot fights a war on two fronts. Only the heir to the throne
of the kingdom of idiots would fight a war on twelve fronts"
-- J. Michael Straczynski. "Babylon 5", "Ceremonies of Light and Dark"

[Robin Vowels]

On Mar 17, 2:36 am, John W Kennedy <jwke...@attglobal.net> wrote:
> On 2013-03-16 08:56:30 +0000, Robin Vowels said:
>
> > On Mar 16, 7:31 am, John W Kennedy <jwke...@attglobal.net> wrote:
>
> >> The IBM PL/I standard was last updated in the late
> >> 60s, but is often a good guide for why a modern compiler does something
> >> that seems strange.
>
> > The IBM standard is the current reference, not some manual that was
> > out-of-date
> > in the 1960s.
>
> No, the current Language Reference, like its earlier versions, the
> Language References for the Optimizer and Checker and Language
> Reference for (F), is hopelessly inadequate as a standard. Over and
> over again here I have pointed out places where:
>
> The compiler does something a user didn't expect,
> The Language Reference doesn't explain why, and
> The Language Specification does.
>
> Whether you like it or not, whether I like it or not, huge parts of the
> language are still implemented according to the Language Specification

Naturally, you'd expect that, as it's PL/I.
However, many features were changed, added, and deleted,
and that manual is obsolete and of no practical value.

> where the Language Reference is silent or ambiguous.

No it's not.
And certainly not "huge parts" are "silent" or "ambiguous".
The current PL/I Language Reference covers all of what's implemented
in a non-ambiguous way non-silent way.
To be sure, the odd clarification has been required over the years,
and probably still is, like any other reference manual.

If you think that "huge parts" are silent or ambiguous
you should describe them here.

[Peter Flass]

On 3/18/2013 10:07 PM, Robin Vowels wrote:
>
> If you think that "huge parts" are silent or ambiguous
> you should describe them here.
>

Maybe not "huge" parts, but I found several cases where the only way to
see what was happening was to compile and maybe run a program.

--
Pete

[John W Kennedy]

We've been through several. Your invariable reaction has been to stick
your fingers in your ears and go "Lalalalalalala! I'm not listening!"

--
John W Kennedy
"The pathetic hope that the White House will turn a Caligula into a
Marcus Aurelius is as naïve as the fear that ultimate power inevitably
corrupts."
-- James D. Barber (1930-2004)

[Robin Vowels]

On Mar 20, 1:49 am, John W Kennedy <jwke...@attglobal.net> wrote:

> We've been through several. Your invariable reaction has been to stick
> your fingers in your ears and go "Lalalalalalala! I'm not listening!"

You're still being a baby, I see.

[John W Kennedy]

That's it? That's really the best you can do? Playground taunts? Can't
you at least compare me to Hitler, or say that I'm a Communist? Or is
that that your long-term memory is so snookered that you really can't
remember the several times this same thing has come up in the last year
or so?

--
John W Kennedy
"Those in the seat of power oft forget their failings and seek only the
obeisance of others! Thus is bad government born! Hold in your heart
that you and the people are one, human beings all, and good government
shall arise of its own accord! Such is the path of virtue!"
-- Kazuo Koike. "Lone Wolf and Cub: Thirteen Strings" (tr. Dana Lewis)

[Robin Vowels]

On Mar 21, 4:30 am, John W Kennedy <jwke...@attglobal.net> wrote:
> On 2013-03-19 21:32:34 +0000, Robin Vowels said:
>
> > On Mar 20, 1:49 am, John W Kennedy <jwke...@attglobal.net> wrote:
>
> >> We've been through several. Your invariable reaction has been to stick
> >> your fingers in your ears and go "Lalalalalalala! I'm not listening!"
>
> > You're still being a baby, I see.
>
> That's it? That's really the best you can do? Playground taunts? Can't
> you at least compare me to Hitler, or say that I'm a Communist? Or is
> that that your long-term memory is so snookered that you really can't
> remember the several times this same thing has come up in the last year
> or so?

So, obviously, when it comes to the crunch, and you are
called to account, you have nothing to show to support your claims.

[John W Kennedy]

I already have, liar. The fact that you ignore anything you don't want
to see doesn't change that.

--
John W Kennedy
A proud member of the reality-based community.

[Robin Vowels]

You're still behaving like a baby.

And you still haven't provided anything to support your claims.

We're waiting.

[Peter Flass]

Okay you two, each go to your corners for a timeout.

--
Pete

[Robin Vowels]

On Mar 22, 12:47 am, John W Kennedy <jwke...@attglobal.net> wrote:

Still waiting on your response to your statement :

"Whether you like it or not, whether I like it or not, huge parts of
the
language are still implemented according to the Language Specification

where the Language Reference is silent or ambiguous."

Your statement in reference to the Language Reference is
unsupportable.

As for my experience, for the past 40 years,
I have not needed anything other than the
PL/I Language Reference.
This would also be the case for the many other
programmers who have relied on the Language reference
for writing PL/I programs.
To suggest that we (they and I) need to refer to a
45-year old out-of-date manual is patently absurd.

[Peter Flass]

The only unambiguous specification (IIRC) are the standards, which
aren't particularly useful for programmers attempting to write PL/I
programs. By nature everything else will have less information so there
will be a lot of cases that might be interpreted ambiguously.

--
Pete