On 4/9/23 9:56 AM, bill wrote:
> On 4/9/2023 10:39 AM, Craig A. Berry wrote:
>>
>> On 4/9/23 9:32 AM, Jan-Erik Söderholm wrote:
>>> Hi.
>>>
>>> So, got the files downloaded.
>>> Now, have anyone tried with VitualBox so far?
>>
>> Yes, and there have been reports that it's what at least some VSI
>> developers use.  If you happen to be on macOS Ventura, you'll need a
>> pre-release install of 7.0.7 because 7.0.6 (the current release) doesn't
>> work with EFI-enabled booting.
>
> I haven't downloaded it yet but I hope to get to it this week.
> I will probably try VirtualBox on Windows10.
> Can anyone tell me what compilers come with it as that is what I
> would be most interested in trying out.

Looks like there are field test versions of C (traditional front end
with LLVM translator), C++ (based on clang++), and FORTRAN. IIRC COBOL
is supposed to be next with BASIC "hopefully this year."

On 4/9/2023 11:27 AM, Craig A. Berry wrote:
>
> On 4/9/23 9:56 AM, bill wrote:
>> On 4/9/2023 10:39 AM, Craig A. Berry wrote:
>>>
>>> On 4/9/23 9:32 AM, Jan-Erik Söderholm wrote:
>>>> Hi.
>>>>
>>>> So, got the files downloaded.
>>>> Now, have anyone tried with VitualBox so far?
>>>
>>> Yes, and there have been reports that it's what at least some VSI
>>> developers use.  If you happen to be on macOS Ventura, you'll need a
>>> pre-release install of 7.0.7 because 7.0.6 (the current release) doesn't
>>> work with EFI-enabled booting.
>>
>> I haven't downloaded it yet but I hope to get to it this week.
>> I will probably try VirtualBox on Windows10.
>> Can anyone tell me what compilers come with it as that is what I
>> would be most interested in trying out.
>
> Looks like there are field test versions of C (traditional front end
> with LLVM translator), C++ (based on clang++), and FORTRAN.  IIRC COBOL
> is supposed to be next with BASIC "hopefully this year."
>

Well, that's a small bummer. My biggest interest is in COBOL but I
guess I could play with the Fortran a bit. But my interest in Fortran
runs out around Fortran-77. :-)

bill

On 2023-04-08 16:04, Zane H. Healy wrote:
> issinoho <iain@issinoho.com> wrote:
>> VSI TCP/IP Services for OpenVMS x86_64 Version X6.0
>> on a VMware, Inc. VMware7,1 running OpenVMS E9.2-1
>>
>> I've found the DHCP functionality pretty broken in this - I'm assuming
>> it's TCPIP 6 rather than anything to do with x86, but who knows...
>>
>> DHCP assigns the interface address and routing fine, but Bind doesn't get
>> Configured or Enabled. Doing this manually gets things working, however
>> as soon as a reboot kicks off the DHCP process again, Bind loses it's
>> settings. Strangely, this also means that the result of a "show network"
>> command displays an unconfigured IP stack even when it's working.
>>
>> Moving to static IP fixes all of the above.
>
> Last night, I brought up TCPIP with DHCP, and was able to connect to the
> system just fine. So at least very basic DHCP seemed to work. Having said
> that one of the first things I did this morning was to convert to a Static
> IP, as I have DNS setup for my home network.

Nothing in that prevents you from still using DHCP, though. I used DHCP
even for machines with fixed IP. I have it configured in my DHCP server
to give fixed IP addresses to specific MAC addresses, and DNS
configuration is also handed out via DHCP. I just like administrating it
all in one place. :-)

Johnny

Can anyone suggest what sort of cheap server/workstation I could get to
run this? All my modern systems are AMD

I have looked at various DL380p Gen 8, but the cpus seen too old

--
Chris

On 4/9/2023 11:35 AM, bill wrote:
> On 4/9/2023 11:27 AM, Craig A. Berry wrote:
>> On 4/9/23 9:56 AM, bill wrote:
>>> I haven't downloaded it yet but I hope to get to it this week.
>>> I will probably try VirtualBox on Windows10.
>>> Can anyone tell me what compilers come with it as that is what I
>>> would be most interested in trying out.
>>
>> Looks like there are field test versions of C (traditional front end
>> with LLVM translator), C++ (based on clang++), and FORTRAN.  IIRC COBOL
>> is supposed to be next with BASIC "hopefully this year."
>
> Well, that's a small bummer.  My biggest interest is in COBOL but I
> guess I could play with the Fortran a bit.  But my interest in Fortran
> runs out around Fortran-77.  :-)

It should compile Fortran 77 just fine.

Arne

On Monday, 10 April 2023 at 01:05:52 UTC+1, Arne Vajhøj wrote:
> On 4/9/2023 11:35 AM, bill wrote:
> > On 4/9/2023 11:27 AM, Craig A. Berry wrote:
> >> On 4/9/23 9:56 AM, bill wrote:
> >>> I haven't downloaded it yet but I hope to get to it this week.
> >>> I will probably try VirtualBox on Windows10.
> >>> Can anyone tell me what compilers come with it as that is what I
> >>> would be most interested in trying out.
> >>
> >> Looks like there are field test versions of C (traditional front end
> >> with LLVM translator), C++ (based on clang++), and FORTRAN. IIRC COBOL
> >> is supposed to be next with BASIC "hopefully this year."
> >
> > Well, that's a small bummer. My biggest interest is in COBOL but I
> > guess I could play with the Fortran a bit. But my interest in Fortran
> > runs out around Fortran-77. :-)
> It should compile Fortran 77 just fine.
>
> Arne

For anyone contemplating using VirtualBox, I couldn't work out (easily) how to get a serial connection to the terminal and found the entire VMware Player experience vastly better. It just works.

On 4/9/2023 8:05 PM, Arne Vajhøj wrote:
> On 4/9/2023 11:35 AM, bill wrote:
>> On 4/9/2023 11:27 AM, Craig A. Berry wrote:
>>> On 4/9/23 9:56 AM, bill wrote:
>>>> I haven't downloaded it yet but I hope to get to it this week.
>>>> I will probably try VirtualBox on Windows10.
>>>> Can anyone tell me what compilers come with it as that is what I
>>>> would be most interested in trying out.
>>>
>>> Looks like there are field test versions of C (traditional front end
>>> with LLVM translator), C++ (based on clang++), and FORTRAN.  IIRC COBOL
>>> is supposed to be next with BASIC "hopefully this year."
>>
>> Well, that's a small bummer.  My biggest interest is in COBOL but I
>> guess I could play with the Fortran a bit.  But my interest in Fortran
>> runs out around Fortran-77.  :-)
>
> It should compile Fortran 77 just fine.

I thought VMS Fortran was up at least to Fortran 99. I was just
hinting that, like COBOL, I have no interest in and see little
value in the more modern accretions.

bill

bill <bill.gunshannon@gmail.com> wrote:
>On 4/9/2023 8:05 PM, Arne Vajhøj wrote:
>> On 4/9/2023 11:35 AM, bill wrote:
>>> On 4/9/2023 11:27 AM, Craig A. Berry wrote:
>>>> On 4/9/23 9:56 AM, bill wrote:
>>>>> I haven't downloaded it yet but I hope to get to it this week.
>>>>> I will probably try VirtualBox on Windows10.
>>>>> Can anyone tell me what compilers come with it as that is what I
>>>>> would be most interested in trying out.
>>>>
>>>> Looks like there are field test versions of C (traditional front end
>>>> with LLVM translator), C++ (based on clang++), and FORTRAN.  IIRC COBOL
>>>> is supposed to be next with BASIC "hopefully this year."
>>>
>>> Well, that's a small bummer.  My biggest interest is in COBOL but I
>>> guess I could play with the Fortran a bit.  But my interest in Fortran
>>> runs out around Fortran-77.  :-)
>>
>> It should compile Fortran 77 just fine.
>
>I thought VMS Fortran was up at least to Fortran 99. I was just
>hinting that, like COBOL, I have no interest in and see little
>value in the more modern accretions.

It is, but if your brain is still at Fortran 77 or even 66, there are flags
to make the compiler behave that way too. Even computed if!
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On 4/10/2023 10:19 AM, bill wrote:
> On 4/9/2023 8:05 PM, Arne Vajhøj wrote:
>> On 4/9/2023 11:35 AM, bill wrote:
>>> On 4/9/2023 11:27 AM, Craig A. Berry wrote:
>>>> On 4/9/23 9:56 AM, bill wrote:
>>>>> I haven't downloaded it yet but I hope to get to it this week.
>>>>> I will probably try VirtualBox on Windows10.
>>>>> Can anyone tell me what compilers come with it as that is what I
>>>>> would be most interested in trying out.
>>>>
>>>> Looks like there are field test versions of C (traditional front end
>>>> with LLVM translator), C++ (based on clang++), and FORTRAN.  IIRC COBOL
>>>> is supposed to be next with BASIC "hopefully this year."
>>>
>>> Well, that's a small bummer.  My biggest interest is in COBOL but I
>>> guess I could play with the Fortran a bit.  But my interest in
>>> Fortran runs out around Fortran-77.  :-)
>>
>> It should compile Fortran 77 just fine.
>
> I thought VMS Fortran was up at least to Fortran 99.  I was just
> hinting that, like COBOL, I have no interest in and see little
> value in the more modern accretions.

Fortran 90/95.

So they added a ton of features. But they did not remove
the old stuff.

So your Fortran 77 code should compile fine.

(I believe the Fortran 90 standard explicit required Fortran
77 compatibility, but that the Fortran 95 standard removed some
of the 77 stuff, which does not matter as VSI Fortran kept
that stuff)

Arne

On 4/10/2023 4:16 PM, Arne Vajhøj wrote:
> On 4/10/2023 10:19 AM, bill wrote:
>> On 4/9/2023 8:05 PM, Arne Vajhøj wrote:
>>> On 4/9/2023 11:35 AM, bill wrote:
>>>> On 4/9/2023 11:27 AM, Craig A. Berry wrote:
>>>>> On 4/9/23 9:56 AM, bill wrote:
>>>>>> I haven't downloaded it yet but I hope to get to it this week.
>>>>>> I will probably try VirtualBox on Windows10.
>>>>>> Can anyone tell me what compilers come with it as that is what I
>>>>>> would be most interested in trying out.
>>>>>
>>>>> Looks like there are field test versions of C (traditional front end
>>>>> with LLVM translator), C++ (based on clang++), and FORTRAN.  IIRC
>>>>> COBOL
>>>>> is supposed to be next with BASIC "hopefully this year."
>>>>
>>>> Well, that's a small bummer.  My biggest interest is in COBOL but I
>>>> guess I could play with the Fortran a bit.  But my interest in
>>>> Fortran runs out around Fortran-77.  :-)
>>>
>>> It should compile Fortran 77 just fine.
>>
>> I thought VMS Fortran was up at least to Fortran 99.  I was just
>> hinting that, like COBOL, I have no interest in and see little
>> value in the more modern accretions.
>
> Fortran 90/95.
>
> So they added a ton of features. But they did not remove
> the old stuff.
>
> So your Fortran 77 code should compile fine.
>
> (I believe the Fortran 90 standard explicit required Fortran
> 77 compatibility, but that the Fortran 95 standard removed some
> of the 77 stuff, which does not matter as VSI Fortran kept
> that stuff)

As I understand it Fortran (much like COBOL) added a lot of junk
that had nothing to with the intended function of the language.
In the case of COBOL, thankfully, it was mostly ignored by true
practitioners of the art. Fortran did not fare as well.

bill

On 4/10/2023 6:14 PM, bill wrote:
> On 4/10/2023 4:16 PM, Arne Vajhøj wrote:
>> On 4/10/2023 10:19 AM, bill wrote:
>>> On 4/9/2023 8:05 PM, Arne Vajhøj wrote:
>>>> On 4/9/2023 11:35 AM, bill wrote:
>>>>> Well, that's a small bummer.  My biggest interest is in COBOL but I
>>>>> guess I could play with the Fortran a bit.  But my interest in
>>>>> Fortran runs out around Fortran-77.  :-)
>>>>
>>>> It should compile Fortran 77 just fine.
>>>
>>> I thought VMS Fortran was up at least to Fortran 99.  I was just
>>> hinting that, like COBOL, I have no interest in and see little
>>> value in the more modern accretions.
>>
>> Fortran 90/95.
>>
>> So they added a ton of features. But they did not remove
>> the old stuff.
>>
>> So your Fortran 77 code should compile fine.
>>
>> (I believe the Fortran 90 standard explicit required Fortran
>> 77 compatibility, but that the Fortran 95 standard removed some
>> of the 77 stuff, which does not matter as VSI Fortran kept
>> that stuff)
>
> As I understand it Fortran (much like COBOL) added a  lot of junk
> that had nothing to with the intended function of the language.
> In the case of COBOL, thankfully, it was mostly ignored by true
> practitioners of the art.  Fortran did not fare as well.

I think that a large part of Fortran developers actually like
the new features in 90/95 (or at least a significant part of them) and
use them.

Fortran 77 to 90/95 is not so much a paradigm change but
more practical oriented stuff. Examples:
- free format instead of fixed format
- "switch"
- dynamic memory allocation
- module concept

Huge change in how code is written and in many ways a totally
new language, but still in the Fortran spirit.

Fortran 90/95 to 2003/2008/2018 is a bit more mixed with
some paragdigm changing features (OOP) but also some
practical stuff (definition of interface with C, lots
of stuff to facilitate parallelization).

Arne

bill <bill.gunshannon@gmail.com> wrote:
>
>As I understand it Fortran (much like COBOL) added a lot of junk
>that had nothing to with the intended function of the language.
>In the case of COBOL, thankfully, it was mostly ignored by true
>practitioners of the art. Fortran did not fare as well.

This was the case after fortran90. But the stuff added to fortran90
was very much consistent with the intended function of scientific
programming. Stuff like vector and matrix operations makes a lot of
code using matrices easier to write and clearer to understand, as well
as easier for the compiler to vectorize.

Some of the stuff added in the layers after fortran90 have been
messy, like pointers. Engineers shouldn't be allowed to touch
pointers. Coroutines are kind of cool though.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On 4/10/2023 6:14 PM, bill wrote:
> On 4/10/2023 4:16 PM, Arne Vajhøj wrote:
>> On 4/10/2023 10:19 AM, bill wrote:
>>> On 4/9/2023 8:05 PM, Arne Vajhøj wrote:
>>>> On 4/9/2023 11:35 AM, bill wrote:
>>>>> Well, that's a small bummer.  My biggest interest is in COBOL but I
>>>>> guess I could play with the Fortran a bit.  But my interest in
>>>>> Fortran runs out around Fortran-77.  :-)
>>>>
>>>> It should compile Fortran 77 just fine.
>>>
>>> I thought VMS Fortran was up at least to Fortran 99.  I was just
>>> hinting that, like COBOL, I have no interest in and see little
>>> value in the more modern accretions.
>>
>> Fortran 90/95.
>>
>> So they added a ton of features. But they did not remove
>> the old stuff.
>>
>> So your Fortran 77 code should compile fine.
>>
>> (I believe the Fortran 90 standard explicit required Fortran
>> 77 compatibility, but that the Fortran 95 standard removed some
>> of the 77 stuff, which does not matter as VSI Fortran kept
>> that stuff)
>
> As I understand it Fortran (much like COBOL) added a  lot of junk
> that had nothing to with the intended function of the language.
> In the case of COBOL, thankfully, it was mostly ignored by true
> practitioners of the art.  Fortran did not fare as well.

Practical:

$ type bill.for
program bill
integer*4 hw(2), lbl
data hw/4hHi B,4hill!/
assign 100 to lbl
goto lbl
stop
100 write(6,200) hw
200 format(1x,2a4)
end
$ fortran bill
$ link bill
$ run bill
Hi Bill!
$ fortran/stand=f90 bill

integer*4 hw(2), lbl
...............^
%F90-W-WARNING, Fortran 90 does not allow this length specification. [4]
at line number 2 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
...............^
%F90-W-WARNING, In Fortran 90, the obsolescent Hollerith constant is not
permitted in this context. [4HHi B]
at line number 3 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
......................^
%F90-W-WARNING, In Fortran 90, the obsolescent Hollerith constant is not
permitted in this context. [4Hill!]
at line number 3 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
...............^
%F90-W-WARNING, A character string or Hollerith string is non-standard
in this context. [4HHi B]
at line number 3 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
......................^
%F90-W-WARNING, A character string or Hollerith string is non-standard
in this context. [4Hill!]
at line number 3 in file DISK2:[ARNE]bill.for;1

assign 100 to lbl
.......^
%F90-W-WARNING, The ASSIGN statement is obsolete.
at line number 4 in file DISK2:[ARNE]bill.for;1

goto lbl
.......^
%F90-W-WARNING, The assign GO TO statement is obsolete.
at line number 5 in file DISK2:[ARNE]bill.for;1
$ link bill
%LINK-W-WRNERS, compilation warnings
in module BILL file DISK2:[ARNE]bill.OBJ;3
$ run bill
Hi Bill!
$ fortran/stand=f95 bill

program bill
^ %F90-W-WARNING, Fixed form source is an obsolescent feature in Fortran 95.
at line number 1 in file DISK2:[ARNE]bill.for;1

integer*4 hw(2), lbl
...............^
%F90-W-WARNING, Fortran 95 does not allow this length specification. [4]
at line number 2 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
...............^
%F90-W-WARNING, The Hollerith constant has been deleted from Fortran 95.
[4HHi B]
at line number 3 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
......................^
%F90-W-WARNING, The Hollerith constant has been deleted from Fortran 95.
[4Hill!]
at line number 3 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
...............^
%F90-W-WARNING, A character string or Hollerith string is non-standard
in this context. [4HHi B]
at line number 3 in file DISK2:[ARNE]bill.for;1

data hw/4hHi B,4hill!/
......................^
%F90-W-WARNING, A character string or Hollerith string is non-standard
in this context. [4Hill!]
at line number 3 in file DISK2:[ARNE]bill.for;1

assign 100 to lbl
.......^
%F90-W-WARNING, The ASSIGN statement has been deleted in Fortran 95.
at line number 4 in file DISK2:[ARNE]bill.for;1

goto lbl
.......^
%F90-W-WARNING, The assign GO TO statement has been deleted in Fortran 95.
at line number 5 in file DISK2:[ARNE]bill.for;1
$ link bill
%LINK-W-WRNERS, compilation warnings
in module BILL file DISK2:[ARNE]bill.OBJ;4
$ run bill
Hi Bill!

Arne

On Friday, April 7, 2023 at 3:52:24 PM UTC+1, issinoho wrote:
> On Friday, 7 April 2023 at 12:31:26 UTC+1, Simon Clubley wrote:
> > On 2023-04-06, Andrew Back <and...@carrierdetect.com> wrote:
> > > Software and PAKs download and going to give it a go with Qemu.
> > >
> > I wonder what the criteria is for people being offered access ?
> >
> > I have also received nothing, but it's clear others have. (Not that it
> > matters at the moment anyway because I am currently way too busy with
> > other things. :-))
> >
> > I wonder if VSI are manually selecting a few people to offer it to
> > before making it generally available to hobbyists ?
> >
> > It's certainly not just being a member of the community program
> > because I've been registered there since HPE stopped offering
> > hobbyist licences.
> >
> > Simon.
> >
> > --
> > Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
> > Walking destinations on a map are further away than they appear.
> VSI TCP/IP Services for OpenVMS x86_64 Version X6.0
> on a VMware, Inc. VMware7,1 running OpenVMS E9.2-1
>
> I've found the DHCP functionality pretty broken in this - I'm assuming it's TCPIP 6 rather than anything to do with x86, but who knows...
>
> DHCP assigns the interface address and routing fine, but Bind doesn't get Configured or Enabled. Doing this manually gets things working, however as soon as a reboot kicks off the DHCP process again, Bind loses it's settings. Strangely, this also means that the result of a "show network" command displays an unconfigured IP stack even when it's working.
>
> Moving to static IP fixes all of the above.

DHCP client on OpenVMS is not used much so I'm not surprised. I've seen issues in the past with it on the rare occasion I've used it.

On 2023-04-07, Steven Schweda <sms.antinode@gmail.com> wrote:
>
>> I wonder what the criteria is for people being offered access ?
>
> A somewhat tardy and incomplete explanation of what they is was
> provided recently:
>
> https://forum.vmssoftware.com/viewtopic.php?f=37&t=8595#p18051

Thank you Steven, that was useful.

Of course, now I wonder why they need to do what is obviously a batched
manual process, given that early releases have been out for a while.

I did find the number of 1000+ accounts interesting. That was more than I was
expecting (unless many of them are the I-tried-it-once-and-didn't-like-it
accounts. :-))

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 2023-04-10, Arne Vajhøj <arne@vajhoej.dk> wrote:
> $ fortran/stand=f95 bill
>
> program bill
> ^
> %F90-W-WARNING, Fixed form source is an obsolescent feature in Fortran 95.
> at line number 1 in file DISK2:[ARNE]bill.for;1
>
> integer*4 hw(2), lbl
> ..............^
> %F90-W-WARNING, Fortran 95 does not allow this length specification. [4]
> at line number 2 in file DISK2:[ARNE]bill.for;1
>
> data hw/4hHi B,4hill!/
> ..............^
> %F90-W-WARNING, The Hollerith constant has been deleted from Fortran 95.
> [4HHi B]
> at line number 3 in file DISK2:[ARNE]bill.for;1
>
> data hw/4hHi B,4hill!/
> .....................^
> %F90-W-WARNING, The Hollerith constant has been deleted from Fortran 95.
> [4Hill!]
> at line number 3 in file DISK2:[ARNE]bill.for;1
>
> data hw/4hHi B,4hill!/
> ..............^
> %F90-W-WARNING, A character string or Hollerith string is non-standard
> in this context. [4HHi B]
> at line number 3 in file DISK2:[ARNE]bill.for;1
>
> data hw/4hHi B,4hill!/
> .....................^
> %F90-W-WARNING, A character string or Hollerith string is non-standard
> in this context. [4Hill!]
> at line number 3 in file DISK2:[ARNE]bill.for;1
>
> assign 100 to lbl
> ......^
> %F90-W-WARNING, The ASSIGN statement has been deleted in Fortran 95.
> at line number 4 in file DISK2:[ARNE]bill.for;1
>
> goto lbl
> ......^
> %F90-W-WARNING, The assign GO TO statement has been deleted in Fortran 95.
> at line number 5 in file DISK2:[ARNE]bill.for;1
> $ link bill
> %LINK-W-WRNERS, compilation warnings
> in module BILL file DISK2:[ARNE]bill.OBJ;4
> $ run bill
> Hi Bill!
>

These should all be -E- IMHO. If you want to write FORTRAN IV code, then
use a FORTRAN IV compiler, not a Fortran 95 compiler.

This is no different from people now using ANSI function definitions in C,
instead of those (horrible) K&R function definitions.

Another example is moving from VAX C to DEC C standards. You can't compile
half of a source file using /STANDARD=VAXC, and the other half of the same
file using a modern standard.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 4/11/2023 9:19 AM, Simon Clubley wrote:
> On 2023-04-10, Arne Vajhøj <arne@vajhoej.dk> wrote:
>> $ fortran/stand=f95 bill
>>
>> program bill
>> ^
>> %F90-W-WARNING, Fixed form source is an obsolescent feature in Fortran 95.
>> at line number 1 in file DISK2:[ARNE]bill.for;1
>>
>> integer*4 hw(2), lbl
>> ..............^
>> %F90-W-WARNING, Fortran 95 does not allow this length specification. [4]
>> at line number 2 in file DISK2:[ARNE]bill.for;1
>>
>> data hw/4hHi B,4hill!/
>> ..............^
>> %F90-W-WARNING, The Hollerith constant has been deleted from Fortran 95.
>> [4HHi B]
>> at line number 3 in file DISK2:[ARNE]bill.for;1

> These should all be -E- IMHO. If you want to write FORTRAN IV code, then
> use a FORTRAN IV compiler, not a Fortran 95 compiler.

That code and that qualifier combined does not make any sense.

But it was really my point.

VSI Fortran compiler supports 77, 90 and 95.

Bill wants to do 77.

And the only thing he need to do 77 with VSI Fortran to
avoid errors and warnings is to avoid to "accidentally"
type /stand=f90 or /stand=f95.

:-)

Re E instead of W then I would tend to agree that features
that has been removed not just deprecated/marked obsolete in
the standard should be E not W when explicit specifying that
standard.

Arne

On 4/11/2023 9:12 AM, Simon Clubley wrote:
> On 2023-04-07, Steven Schweda <sms.antinode@gmail.com> wrote:
>> https://forum.vmssoftware.com/viewtopic.php?f=37&t=8595#p18051

> I did find the number of 1000+ accounts interesting. That was more than I was
> expecting (unless many of them are the I-tried-it-once-and-didn't-like-it
> accounts. :-))

VSI once published this:

https://vmssoftware.com/images/blog/blog-article-chart.png

it is not at a 1000, but if the trend showed continued
after September 2021 then >1000 could easily be reached.

Arne

On 4/11/2023 9:29 AM, Arne Vajhøj wrote:
> On 4/11/2023 9:19 AM, Simon Clubley wrote:
>> On 2023-04-10, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>> $ fortran/stand=f95 bill
>>>
>>>         program bill
>>> ^
>>> %F90-W-WARNING, Fixed form source is an obsolescent feature in
>>> Fortran 95.
>>> at line number 1 in file DISK2:[ARNE]bill.for;1
>>>
>>>         integer*4 hw(2), lbl
>>> ..............^
>>> %F90-W-WARNING, Fortran 95 does not allow this length
>>> specification.   [4]
>>> at line number 2 in file DISK2:[ARNE]bill.for;1
>>>
>>>         data hw/4hHi B,4hill!/
>>> ..............^
>>> %F90-W-WARNING, The Hollerith constant has been deleted from Fortran 95.
>>>     [4HHi B]
>>> at line number 3 in file DISK2:[ARNE]bill.for;1
>
>> These should all be -E- IMHO. If you want to write FORTRAN IV code, then
>> use a FORTRAN IV compiler, not a Fortran 95 compiler.
>
> That code and that qualifier combined does not make any sense.
>
> But it was really my point.
>
> VSI Fortran compiler supports 77, 90 and 95.
>
> Bill wants to do 77.

Bill doesn't necessarily want to do 77. Bill is also perfectly
happy with Fortran IV. :-)

bill

bill <bill.gunshannon@gmail.com> wrote:
>On 4/10/2023 7:09 PM, Scott Dorsey wrote:
>>
>> Some of the stuff added in the layers after fortran90 have been
>> messy, like pointers. Engineers shouldn't be allowed to touch
>> pointers.
>
>There are many things engineers should not be allowed to touch.
>Another would be writing business applications in Fortran. But
>that's probably a story for another time. :-)

I can understand how folks like quants who want to do real math might
want to be using fortran for business applications. Unfortunately it
doesn't have money types, but you can always use integer and do some
fancy stuff to add the decimal point.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote:
>On 2023-04-11, Richard <legalize+jeeves@mail.xmission.com> wrote:
>>
>> Nit: Not according to the standard. F90/F95 removed no features and only
>> "obsoleted" some features, which is why the compiler issued warnings
>> and not errors.
>
>So which version of Fortran outright removed Hollerith constants ?

Wait, hollerith fields were removed?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On 4/10/2023 7:09 PM, Scott Dorsey wrote:
>
> Some of the stuff added in the layers after fortran90 have been
> messy, like pointers. Engineers shouldn't be allowed to touch
> pointers.

There are many things engineers should not be allowed to touch.
Another would be writing business applications in Fortran. But
that's probably a story for another time. :-)

bill

On 4/11/2023 5:47 PM, Scott Dorsey wrote:
> Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote:
>> On 2023-04-11, Richard <legalize+jeeves@mail.xmission.com> wrote:
>>> Nit: Not according to the standard. F90/F95 removed no features and only
>>> "obsoleted" some features, which is why the compiler issued warnings
>>> and not errors.
>>
>> So which version of Fortran outright removed Hollerith constants ?
>
> Wait, hollerith fields were removed?

Hollerith constants as in:

iv = 4hABCD

were removed but still described
in Fortran 77. And supported in VMS Fortran.

Hollerith descriptor formats as in:

100 format(4HABCD)

were removed in Fortran 95. But still supported
in VMS Fortran.

Arne

[Please do not mail me a copy of your followup]

Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> spake the secret code
<u13mpq$2kv9o$2@dont-email.me> thusly:

>On 2023-04-10, Arne Vajh�j <arne@vajhoej.dk> wrote:
>> $ fortran/stand=f95 bill
>>
>> program bill
>> ^
>> %F90-W-WARNING, Fixed form source is an obsolescent feature in Fortran 95.
>> at line number 1 in file DISK2:[ARNE]bill.for;1
>
>These should all be -E- IMHO.

Nit: Not according to the standard. F90/F95 removed no features and only
"obsoleted" some features, which is why the compiler issued warnings
and not errors.
--
"The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline>
The Terminals Wiki <http://terminals-wiki.org>
The Computer Graphics Museum <http://computergraphicsmuseum.org>
Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

On 2023-04-11, Richard <legalize+jeeves@mail.xmission.com> wrote:
>
> Nit: Not according to the standard. F90/F95 removed no features and only
> "obsoleted" some features, which is why the compiler issued warnings
> and not errors.

So which version of Fortran outright removed Hollerith constants ?

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.