Which (if any) Windows based X server are people using to run CDE from their OpenVMS instance? Paid or free.
I've had mixed experiences with MobaXterm which does work for discrete apps however getting a reliable full screen session doesn't really cut it.
TIA.

On 17/09/2023 18:41, issinoho wrote:
> Which (if any) Windows based X server are people using to run CDE from their OpenVMS instance? Paid or free.
> I've had mixed experiences with MobaXterm which does work for discrete apps however getting a reliable full screen session doesn't really cut it.
> TIA.

XMing - <http://www.straightrunning.com/XmingNotes/> - I pay for it
again every now and then.

Cheers,
Gary B-)

issinoho <iain@issinoho.com> wrote:
>Which (if any) Windows based X server are people using to run CDE from their OpenVMS instance? Paid or free.

We have used StarNine's Xwin32.

>I've had mixed experiences with MobaXterm which does work for discrete apps however getting a reliable full screen session doesn't really cut it.
>TIA.

What happens when you try? I have not run the full CDE desktop remotely just
because it's so damn slow.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On 9/17/2023 4:41 AM, issinoho wrote:
> Which (if any) Windows based X server are people using to run CDE
> from their OpenVMS instance? Paid or free. I've had mixed experiences
> with MobaXterm which does work for discrete apps however getting a
> reliable full screen session doesn't really cut it.

When (approx. once per year) I want to run something X on
VMS then I use MobaXTerm.

20 years ago ReflectionX worked great.

Arne

Den 2023-09-17 kl. 18:46, skrev Arne Vajhøj:
> On 9/17/2023 4:41 AM, issinoho wrote:
>> Which (if any) Windows based X server are people using to run CDE
>> from their OpenVMS instance? Paid or free. I've had mixed experiences
>> with MobaXterm which does work for discrete apps however getting a
>> reliable full screen session doesn't really cut it.
>
> When (approx. once per year) I want to run something X on
> VMS then I use MobaXTerm.

X is dead, Web/HTML is what is used for applications today.

On 9/17/2023 6:03 PM, Jan-Erik Söderholm wrote:
> Den 2023-09-17 kl. 18:46, skrev Arne Vajhøj:
>> On 9/17/2023 4:41 AM, issinoho wrote:
>>> Which (if any) Windows based X server are people using to run CDE
>>> from their OpenVMS instance? Paid or free. I've had mixed experiences
>>> with MobaXterm which does work for discrete apps however getting a
>>> reliable full screen session doesn't really cut it.
>>
>> When (approx. once per year) I want to run something X on
>> VMS then I use MobaXTerm.
>
> X is dead, Web/HTML is what is used for applications today.
>
>

Tell that to the millions of Linux and BSD desktop users.
Just because VMS users don;t want a desktop doesn't mean the
rest of the industry agrees.

Personally, I really miss my VMS desktop.

bill

On 9/17/2023 6:03 PM, Jan-Erik Söderholm wrote:
> Den 2023-09-17 kl. 18:46, skrev Arne Vajhøj:
>> On 9/17/2023 4:41 AM, issinoho wrote:
>>> Which (if any) Windows based X server are people using to run CDE
>>> from their OpenVMS instance? Paid or free. I've had mixed experiences
>>> with MobaXterm which does work for discrete apps however getting a
>>> reliable full screen session doesn't really cut it.
>>
>> When (approx. once per year) I want to run something X on
>> VMS then I use MobaXTerm.
>
> X is dead, Web/HTML is what is used for applications today.

Web is the default GUI today.

The GUI for the typical business application (the stuff that
35 years ago would run on VMS TDMS/FMS/DECForms or similar for
other platforms and 25 years ago would run as VB6 or Delphi
Windows applications) is web based today and have been
for many years.

But client applications are still used on non-server OS
(which does not include VMS!).

Among other needs there is the little detail that
you can't do the web browser as a web application!
For all those web applications to work there need to be
a browser application running on the client system.

Windows, Android, iOS, macOS, Linux etc. all need
a GUI framework for client usage.

For Linux (and other *nix) that means either X or
Wayland at the bottom of the GUI stack - typical with
Gnome & GTK or KDE & Qt at the upper level.

So X or Wayland is still relevant for some platforms.
But IMHO not important for VMS even though it is nice to
have.

Personally I have never liked X. I failed miserable
learning X 30 years ago. It was probably a mistake
to pick Macro-32 as the language for X. :-)

Arne

On Sun, 2023-09-17 at 18:25 -0400, bill wrote:
> Tell that to the millions of Linux and BSD desktop users.
> Just because VMS users don;t want a desktop doesn't mean the
> rest of the industry agrees.
>
> Personally, I really miss my VMS desktop.

It should be straightforward to port the existing X11 driver for
Virtualbox or VMware to OpenVMS. The x86 driver already exists for
Linux.
--
Tactical Nuclear Kittens

On 2023-09-17, Jan-Erik Söderholm <jan-erik.soderholm@telia.com> wrote:
>
> X is dead, Web/HTML is what is used for applications today.
>

The applications you are thinking of are only a small subset of the
usage cases for X.

IOW, Web/HTML is dominant in certain classes of problems, but not all
of them.

If there was no longer any need for X, then why is there such a massive
negative pushback against Wayland by system admins/etc who use X on a
daily basis to carry out their jobs ?

Simon.

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

On 2023-09-17, Arne Vajhøj <arne@vajhoej.dk> wrote:
>
> Personally I have never liked X. I failed miserable
> learning X 30 years ago. It was probably a mistake
> to pick Macro-32 as the language for X. :-)
>

Picking Macro-32 to write any application is _always_ a mistake. :-)

Traditional X is both very good for enabling remote GUI operations and
for being able to operate a GUI over a low-bandwidth network connection.

Simon.

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

Den 2023-09-18 kl. 14:13, skrev Simon Clubley:
> On 2023-09-17, Jan-Erik Söderholm <jan-erik.soderholm@telia.com> wrote:
>>
>> X is dead, Web/HTML is what is used for applications today.
>>
>
> The applications you are thinking of are only a small subset of the
> usage cases for X.
>
> IOW, Web/HTML is dominant in certain classes of problems, but not all
> of them.
>
> If there was no longer any need for X, then why is there such a massive
> negative pushback against Wayland by system admins/etc who use X on a
> daily basis to carry out their jobs ?
>
> Simon.
>

OK. And yes, I saw the other posts...

Right, I might have been thinking mostly on application development.
And yes, there are sys admin tool that you might want to run over X.
And of course, since this is cov, I was mainly thinking about OpenVMS.

On 9/18/2023 8:17 AM, Simon Clubley wrote:
> On 2023-09-17, Arne Vajhøj <arne@vajhoej.dk> wrote:
>> Personally I have never liked X. I failed miserable
>> learning X 30 years ago. It was probably a mistake
>> to pick Macro-32 as the language for X. :-)
>
> Picking Macro-32 to write any application is _always_ a mistake. :-)

The logic was that I had Macro-32 code for VT (SYS$QIOW not SMG$),
disk (SYS$QIOW not RMS) and tape (SYS$QIOW again), so for this new
type of IO then I decided to use Macro-32 as well. Total failure.

Arne

On 2023-09-18 08:51, Single Stage to Orbit wrote:
> On Sun, 2023-09-17 at 18:25 -0400, bill wrote:
>> Tell that to the millions of Linux and BSD desktop users.
>> Just because VMS users don;t want a desktop doesn't mean the
>> rest of the industry agrees.
>>
>> Personally, I really miss my VMS desktop.
>
> It should be straightforward to port the existing X11 driver for
> Virtualbox or VMware to OpenVMS. The x86 driver already exists for
> Linux.

Uh? How did you come to the conclusion that some driver from Linux would
be straight forward to port to VMS? There aren't exactly much
commonality between the internals in Linux and VMS...

Johnny

On 2023-09-18 12:17:41 +0000, Simon Clubley said:

> On 2023-09-17, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>
>> Personally I have never liked X. I failed miserable
>> learning X 30 years ago. It was probably a mistake
>> to pick Macro-32 as the language for X. :-)
>>
>
> Picking Macro-32 to write any application is _always_ a mistake. :-)

If you write it, sure.

On Mon, 2023-09-18 at 17:04 +0200, Johnny Billquist wrote:
> > It should be straightforward to port the existing X11 driver for
> > Virtualbox or VMware to OpenVMS. The x86 driver already exists for
> > Linux.
>
> Uh? How did you come to the conclusion that some driver from Linux
> would be straight forward to port to VMS? There aren't exactly much
> commonality between the internals in Linux and VMS...

The driver model in VMS may be very different to what is expected in
Linux but the basics of controlling the device itself does't change.
Likewise for X11.

Hopefully VSI has documentation on implementing drivers in x86_64 on
their website, I would definitely take a look.
--
Tactical Nuclear Kittens

On 9/18/2023 10:41 AM, Arne Vajhøj wrote:
> On 9/18/2023 8:17 AM, Simon Clubley wrote:
>> On 2023-09-17, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>> Personally I have never liked X. I failed miserable
>>> learning X 30 years ago. It was probably a mistake
>>> to pick Macro-32 as the language for X. :-)
>>
>> Picking Macro-32 to write any application is _always_ a mistake. :-)

The use of "always" is your mistake.

That mentioned, Macro-32 should not be used unless there is a valid reason to do
so. Usually there is no such reason.

> The logic was that I had Macro-32 code for VT (SYS$QIOW not SMG$),
> disk (SYS$QIOW not RMS) and tape (SYS$QIOW again), so for this new
> type of IO then I decided to use Macro-32 as well. Total failure.

Well, that figures. When using Macro-32 for I/O and other things, there is much
required detail work that the VMS compilers are happy to do for you. Also note
that QIO is usable from any of the VMS languages that use the VMS common calling
standard.

Dealing with FAB, RAB, and such is far from my favorite activities.

--
David Froble Tel: 724-529-0450
Dave Froble Enterprises, Inc. E-Mail: davef@tsoft-inc.com
DFE Ultralights, Inc.
170 Grimplin Road
Vanderbilt, PA 15486

On 2023-09-18, Brian Schenkenberger <mail@SendSpamHere.ORG> wrote:
> On 2023-09-18 12:17:41 +0000, Simon Clubley said:
>
>> On 2023-09-17, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>>
>>> Personally I have never liked X. I failed miserable
>>> learning X 30 years ago. It was probably a mistake
>>> to pick Macro-32 as the language for X. :-)
>>>
>>
>> Picking Macro-32 to write any application is _always_ a mistake. :-)
>
> If you write it, sure.
>

Yeah, some of us prefer a real language. :-) :-)

Simon.

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

On 9/17/2023 4:41 PM, issinoho wrote:
> Which (if any) Windows based X server are people using to run CDE from their OpenVMS instance? Paid or free.
> I've had mixed experiences with MobaXterm which does work for discrete apps however getting a reliable full screen session doesn't really cut it.
> TIA.

The few times I need to use X on Windows, I just run VcXsrv.

https://sourceforge.net/projects/vcxsrv/

Has worked for me for the stuff I use it for. YMMV.

--
Johann | email: invalid -> com | www.myrkraverk.com/blog/
I'm not from the Internet, I just work there. | twitter: @myrkraverk

On 9/18/2023 10:41 AM, Arne Vajhøj wrote:
> On 9/18/2023 8:17 AM, Simon Clubley wrote:
>> On 2023-09-17, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>> Personally I have never liked X. I failed miserable
>>> learning X 30 years ago. It was probably a mistake
>>> to pick Macro-32 as the language for X. :-)
>>
>> Picking Macro-32 to write any application is _always_ a mistake. :-)
>
> The logic was that I had Macro-32 code for VT (SYS$QIOW not SMG$),
> disk (SYS$QIOW not RMS) and tape (SYS$QIOW again), so for this new
> type of IO then I decided to use Macro-32 as well. Total failure.

I packaged the VT code up.

So if someone want to do some advanced VT stuff, doesn't
like SMG$ or curses and can live with the code being in
Macro-32:

https://www.vajhoej.dk/arne/info-vax/scr/scr.zip

It is not well documented, but I created a text file
listing all functions and 3 Fortran examples using it.
The API is traditional VMS high level aka strings
by descriptor and integers by reference (it was
intended to be called from Fortran).

It is somewhat questionable whether anyone would
develop for VT today and even more questionable
using Macro-32, but despite the code being written
more than 30 years ago then I still think it is
sort of elegant and worth preserving.

Arne

On 30/09/2023 17:45, Arne Vajhøj wrote:
> On 9/18/2023 10:41 AM, Arne Vajhøj wrote:
>> On 9/18/2023 8:17 AM, Simon Clubley wrote:
>>> On 2023-09-17, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>>> Personally I have never liked X. I failed miserable
>>>> learning X 30 years ago. It was probably a mistake
>>>> to pick Macro-32 as the language for X. :-)
>>>
>>> Picking Macro-32 to write any application is _always_ a mistake. :-)
>>
>> The logic was that I had Macro-32 code for VT (SYS$QIOW not SMG$),
>> disk (SYS$QIOW not RMS) and tape (SYS$QIOW again), so for this new
>> type of IO then I decided to use Macro-32 as well. Total failure.
>
> I packaged the VT code up.
>
> So if someone want to do some advanced VT stuff, doesn't
> like SMG$ or curses and can live with the code being in
> Macro-32:
>
> https://www.vajhoej.dk/arne/info-vax/scr/scr.zip
>
> It is not well documented, but I created a text file
> listing all functions and 3 Fortran examples using it.
> The API is traditional VMS high level aka strings
> by descriptor and integers by reference (it was
> intended to be called from Fortran).
>
> It is somewhat questionable whether anyone would
> develop for VT today and even more questionable
> using Macro-32, but despite the code being written
> more than 30 years ago then I still think it is
> sort of elegant and worth preserving.
>
> Arne

I rather liked SMG$ - I created a few functions when writing code for
use on HHTs using terminal emulation - saved rewriting the screen by
using pop up windows

Chris
--
Chris

On 9/17/23 08:41, issinoho wrote:
> Which (if any) Windows based X server are people using to run CDE from their OpenVMS instance? Paid or free.
> I've had mixed experiences with MobaXterm which does work for discrete apps however getting a reliable full screen session doesn't really cut it.
> TIA.

This might be heresy, but have used cygwin X under windows for some
years. Currently, cygwin, with xfce4 desktop, nedit full screen
editor, gcc and any number of shell windows. Bash shell and all the
usual 'nix utilities work as expected. A bit fiddly to set up
initially, but rock solid once working. Example here under server
2012, long term test bed code for an instrumentation project on
freebsd / linux, running for weeks nonstop without issue...

Chris