OpenVMS graphics - once more

[Dirk Munk]

In the threat about OpenVMS Future Directions there has been a lot of
discussion about the question if OpenVMS should get graphics
capabilities. With that I mean if OpenVMS should support graphics
hardware, not if OpenVMS should have a GUI. A Gui could be a X-Windows
interface, or a browser interface, and of course both can be supported.
Graphics hardware is something very different.

The last graphics hardware that OpenVMS supported (to my knowledge) was
a very simple ATI Radeon 7200 card. That was a low-end PCI device, not
really the kind of hardware you need on a for instance a CAD/CAM
workstation.

A typical server doesn't have graphics hardware, with one exception.
These days servers have an Integrated Lights Out (ILO) port with its own
network interface. It has a browser interface, and it replaces the
traditional Bios interface/SRM console.

On the other hand a typical desktop/laptop CPU had an integrated GPU
these days. I'm sure there is a market for OpenVMS desktop and even
laptop systems. Suppose a customer with a OpenVMS production server has
some kind of software problem, then a software engineer with the sources
and compilers on his laptop can visit the company, determine what is
wrong, and fix the software on his laptop. Very simple example, but you
get the picture. These days laptops are more than capable for these tasks.

VSI can not support any kind of desktop or laptop hardware, so _*if*_
OpenVMS is to run directly on such hardware, then it is save to assume
it will be Intel based HP hardware. Those CPUs have an embedded Intel
HD4000 series GPU, more than sufficient for this kind of work. VSI would
only have to write graphics drivers for these GPUs _*if*_ OpenVMS is to
run on the bare desktop/laptop metal.

Another option would be to run OpenVMS as a guest operating system on
top of a host operating system. You could run it on top of VMware,
Windows Virtual PC, or what ever VSI thinks is most suitable. No need
for specific hardware drivers, it's all under control of the host
operating system. Suppose you would run Windows and OpenVMS on the same
hardware, then you could have OpenVMS set up a Samba server for file
exchange between OpenVMS and Windows, and use a browser on Windows for
your all important window to web based services. Simple, cheap and very
efficient for VSI and us users. And you can run it on any PC hardware of
your choice.

And what if we want to use OpenVMS as a high end workstation with high
performance graphics? In that case forget about consumer graphics, and
go for professional hardware. The AMD FirePro Workstation Graphics for
instance would be far more suitable.

[Phillip Helbig (undress to reply)]

In article <5512f$54db40ab$5ed4324a$54...@news.ziggo.nl>, Dirk Munk

<mu...@home.nl> writes:

> The last graphics hardware that OpenVMS supported (to my knowledge) was
> a very simple ATI Radeon 7200 card. That was a low-end PCI device, not
> really the kind of hardware you need on a for instance a CAD/CAM
> workstation.

There was a time, though, when VMS workstations were marketed for CAD/
CAM.

> A typical server doesn't have graphics hardware, with one exception.
> These days servers have an Integrated Lights Out (ILO) port with its own
> network interface. It has a browser interface, and it replaces the
> traditional Bios interface/SRM console.

Right, but that is also low-end.

> VSI can not support any kind of desktop or laptop hardware,

Why not?

[Dirk Munk]

hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply)) wrote:
> In article <5512f$54db40ab$5ed4324a$54...@news.ziggo.nl>, Dirk Munk
> <mu...@home.nl> writes:
>
>> The last graphics hardware that OpenVMS supported (to my knowledge) was
>> a very simple ATI Radeon 7200 card. That was a low-end PCI device, not
>> really the kind of hardware you need on a for instance a CAD/CAM
>> workstation.
>
> There was a time, though, when VMS workstations were marketed for CAD/
> CAM.

I know, and *if* there is a market for it and *if* VSI has the money,
time, and staff for it, you could do service that market again. But at
the moment VSI has very different priorities.

>
>> A typical server doesn't have graphics hardware, with one exception.
>> These days servers have an Integrated Lights Out (ILO) port with its own
>> network interface. It has a browser interface, and it replaces the
>> traditional Bios interface/SRM console.
>
> Right, but that is also low-end.
>
>> VSI can not support any kind of desktop or laptop hardware,
>
> Why not?
>

Very simple, there are many pieces of hardware in a PC that all need
drivers. VSI simply doesn't have the capability to produce all of these
drivers of all the suppliers, even if the producers of the hardware
would supply them with all the technical details. Normally the producer
of the hardware will also deliver Windows and Linux drivers, but surely
not OpenVMS drivers.

[Bill Gunshannon]

In article <mbfhp3$2bmm$3...@news.kjsl.com>,

hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply)) writes:
> In article <5512f$54db40ab$5ed4324a$54...@news.ziggo.nl>, Dirk Munk
> <mu...@home.nl> writes:
>
>> The last graphics hardware that OpenVMS supported (to my knowledge) was
>> a very simple ATI Radeon 7200 card. That was a low-end PCI device, not
>> really the kind of hardware you need on a for instance a CAD/CAM
>> workstation.
>
> There was a time, though, when VMS workstations were marketed for CAD/
> CAM.

Yes, but I expect that was mostly using Tektronix terminals rather than
built in graphics. And I don't mean 4010's. I worked with a number of
(huge!!) terminals with pens and pads for graphics entry.

>
>> A typical server doesn't have graphics hardware, with one exception.
>> These days servers have an Integrated Lights Out (ILO) port with its own
>> network interface. It has a browser interface, and it replaces the
>> traditional Bios interface/SRM console.
>
> Right, but that is also low-end.
>
>> VSI can not support any kind of desktop or laptop hardware,
>
> Why not?

I would ask the same. As I stated in another message, all the reasons why
VMS faded from the graphics/GUI/desktop market are going to be gone once
VMS is avaiable on x86-64 systems.

bill


--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

[Phillip Helbig (undress to reply)]

In article <ck11f7...@mid.individual.net>,

bi...@server3.cs.scranton.edu (Bill Gunshannon) writes:

> > There was a time, though, when VMS workstations were marketed for CAD/
> > CAM.
>
> Yes, but I expect that was mostly using Tektronix terminals rather than
> built in graphics. And I don't mean 4010's. I worked with a number of
> (huge!!) terminals with pens and pads for graphics entry.

As late as the EV45/EV5 workstations (Alphastation 255, 500, etc), they
were being market for CAD/CAM.

[Simon Clubley]

On 2015-02-11, Bill Gunshannon <bi...@server3.cs.scranton.edu> wrote:
>
> I would ask the same. As I stated in another message, all the reasons why
> VMS faded from the graphics/GUI/desktop market are going to be gone once
> VMS is avaiable on x86-64 systems.
>

Sorry Bill, but no they are not because there's one real big one left - the
software to drive this hardware.

Have you ever written any device drivers Bill ?

Have you ever considered the effort involved in tracking the continual
hardware changes in the x86 world, including the continual device driver
changes or outright new drivers required ?

In addition, have you also considered that much of the required
programming information might not simply be available to VSI (given
it's relative small size) ?

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world

[Jan-Erik Soderholm]

Bill Gunshannon skrev den 2015-02-11 13:49:
> In article <mbfhp3$2bmm$3...@news.kjsl.com>,
> hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply)) writes:
>> In article <5512f$54db40ab$5ed4324a$54...@news.ziggo.nl>, Dirk Munk
>> <mu...@home.nl> writes:
>>
>>> The last graphics hardware that OpenVMS supported (to my knowledge) was
>>> a very simple ATI Radeon 7200 card. That was a low-end PCI device, not
>>> really the kind of hardware you need on a for instance a CAD/CAM
>>> workstation.
>>
>> There was a time, though, when VMS workstations were marketed for CAD/
>> CAM.
>
> Yes, but I expect that was mostly using Tektronix terminals rather than
> built in graphics. And I don't mean 4010's. I worked with a number of
> (huge!!) terminals with pens and pads for graphics entry.
>

I one (early 80s) looked at a job as sysmgr at a place that
run Computervision system. VAX/VMS servers and Tec terminals,
as I remember it...

[Dirk Munk]

Yes, but in those days graphics hardware was simply there to show what
the CPU had calculated. These days an enormous amount of computing is
done in graphics cards, that sometimes have gigabytes of internal
memory. You simply can't compare those two generations.

[Bill Gunshannon]

In article <cda2d$54db552d$5ed4324a$41...@news.ziggo.nl>,

The Tektronix Terminals were complete computers, themselves. They had
8086 CPUs and could run CP/M-86 as well as functioning as CAD terminals
just on their firmware. Kinda like graphics cards with in an external
box. :-)

[gérard Calliet]

Le 11/02/2015 12:44, Dirk Munk a écrit :
> On the other hand a typical desktop/laptop CPU had an integrated GPU
> these days. I'm sure there is a market for OpenVMS desktop and even
> laptop systems. Suppose a customer with a OpenVMS production server has
> some kind of software problem, then a software engineer with the sources
> and compilers on his laptop can visit the company, determine what is
> wrong, and fix the software on his laptop. Very simple example, but you
> get the picture. These days laptops are more than capable for these tasks.

Good story.

Now what we do is quite the same using on our laptop some alpha emulator
(there are a lot of solutions). You develop a large software for
OpenVMS, you have Itanium and Alpha versions, and you are able to
explore bugs or problems on your laptop during the journey to your
customer, and there you connect on the site with your laptop.
A little bit complicate because of the differents hardware (itanium and
emalted alpha).

The idea behind this story is about contagion. If OpenVMS is good
somewhere (for example big sustainable software, or best maintained data
center) you can develop some neighbourg markets. You have not to conquer
The deskstop market, but if OpenVMS get some deskstop facilities, or
support 3 ot 4 of the best internet frameworks, you will get by
contagion parts in these markets. You can do it now, and when OpenVMS is
on x86, you will have already some idea of where OpenVMS should extend
its market.

The OpenVMS parts of market on the other worlds : deskstop, internet
appliances, embedded devices would be found (also, not only) using this
type of story. In the advertising domain the name of this strategy is
viral advertising (I don't no the exact words in english). If sometimes
there exist a special embeddable OpenVMS, it could be introduced knowing
qualities of OpenVMS in other domains.

Also, OpenVMS should conquer specific parts of the known markets.
Sometimes in economics offer is before demand. An example.

We know a lot of small or medium companies, or departments of bigger
companies that are bored to have to maintain their 20 or 30 deskstops,
with the little network, and in the middle some medium servers to the
bigger tasks. Everything has to reboot, reinstall, you have a lot of
thing to know. A lot of time lost to nothing. You could propose an
OpenVMS oriented and centered offer where OpenVMS stability is a big
gain of time. In this schema the customer has some central OpenVMS
center, X11 or other solution to the 20 or 30 employees, some smart
appliance for their IoT, and all that maintainable, with a reboot a year
maximum. In this example OpenVMS is worth a need which is not yet trully
identified.

---
L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast.
http://www.avast.com

[Bill Gunshannon]

In article <mbfjtp$5nf$1...@dont-email.me>,

Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
> On 2015-02-11, Bill Gunshannon <bi...@server3.cs.scranton.edu> wrote:
>>
>> I would ask the same. As I stated in another message, all the reasons why
>> VMS faded from the graphics/GUI/desktop market are going to be gone once
>> VMS is avaiable on x86-64 systems.
>>
>
> Sorry Bill, but no they are not because there's one real big one left - the
> software to drive this hardware.

Assuming that they are going to stay with X rather than write a VMS specific
proprietary protocol, most of it already exists and is readily available.
Only device drivers for specific graphics cards would be needed and there
is a lot of source available for those, too. (Not VMS source, but enough
to reveal the needed programming information.)

>
> Have you ever written any device drivers Bill ?

Yes, but not for a graphics card. :-)

>
> Have you ever considered the effort involved in tracking the continual
> hardware changes in the x86 world, including the continual device driver
> changes or outright new drivers required ?

It's called business. They have to do the same for storage devices.
Or are you suggesting VMS stick with MSCP to avoid "hardware changes"
and "new drivers"? Or, just ignore another market segment that VMS
abandoned years ago.

>
> In addition, have you also considered that much of the required
> programming information might not simply be available to VSI (given
> it's relative small size) ?

If they are planning to enter the PC Gaming world, probably true, but
to provide desktop capabilities desired by the people most likely to
use VMS, I disagree.

I think this is a good time to get back into the desktop market (well,
x86-64 VMS is still a few years away, but when it becomes a reality...)
and even the laptop market. Macs are a niche market that really offered
nothing that wasn't already avaialable and yet they attracted a large
and growing following. I (and I am sure others here) think VMS can do
the same.

And people call me a naysayer.....

[MG]

Dirk Munk schreef op 11-feb-2015 om 12:44:

> The last graphics hardware that OpenVMS supported (to my knowledge) was
> a very simple ATI Radeon 7200 card. That was a low-end PCI device, not
> really the kind of hardware you need on a for instance a CAD/CAM
> workstation.

You mean the Radeon 7500? Mostly the PCI variant of it. It wasn't
actually such a bad card, when it originally came out... But it
was already quite old by the time it became somewhat 'common' for
VMS (and Tru64 UNIX) usage. It would actually work fine for CAD/CAM,
with its lower memory (fine for wireframes and shaded solids).

But what 'CAD/CAM' applications are there for under VMS...?

> And what if we want to use OpenVMS as a high end workstation with high
> performance graphics? In that case forget about consumer graphics, and
> go for professional hardware. The AMD FirePro Workstation Graphics for
> instance would be far more suitable.

Everybody gave up with that, it seems. It's mostly Windows and, I
guess, OS X workstations nowadays. (Also a number of Linux x86/-64
ones, but mostly for post-production from what I understand, to a
lesser degree technical graphics like CAD and CAM.)

- MG

[MG]

Phillip Helbig (undress to reply) schreef op 11-feb-2015 om 13:26:

> There was a time, though, when VMS workstations were marketed for
> CAD/CAM.

I keep hearing that, from time to time, but where is the proof?
Nobody could ever cite me specific software packages. Even old
product brochures don't mention too terribly much... (The only
traces of software somewhat belonging into this category, that I
could ever find documented traces of, was for ULTRIX.)

- MG

[David Froble]

Dirk Munk wrote:
> In the threat about OpenVMS Future Directions there has been a lot of
> discussion about the question if OpenVMS should get graphics
> capabilities. With that I mean if OpenVMS should support graphics
> hardware, not if OpenVMS should have a GUI. A Gui could be a X-Windows
> interface, or a browser interface, and of course both can be supported.
> Graphics hardware is something very different.

Let's talk about the "real world" for a bit, huh?

If a company was serious about attracting new users to it's GUI capable
systems, they are going to have to either cater to what the users
expect, or, they are going to have to come up with something so great
that the users of the world will want to switch to it.

First, let's rule out the "something great" for now, as I don't have a
clue as to what such might be, and if I did, I would not be
participating on c.o.v, I'd be busy making enough money to have my own
tropical island ....

What we got left in the "real world" is what the vast majority of users
think is a GUI based computer. It even has a name. It's called
weendoze. Oh, Ok, Windows.

I cannot count the times customers have asked for Codis to be a weendoze
based product. One of their arguments is that since "everybody" knows
weendoze, they would not have to train new hires and temporary hires.

They are of course wrong in that supposition, but don't know it.

The point is, there are people in management, and elsewhere, that have
this concept, and won't understand that they'd have just as much
training on the "application" regardless of the user interface.

Now, I were to go to the great effort to provide them with an x-weendoze
user interface, they'd give me a wierd look and ask, "what's this", and
then once again start talking about weendoze.

People believe their perceptions. People make decisions based on their
perceptions. The vast majority of users have the perception that GUI
and weendoze are one in the same. Trying to fight this perception is
an uphill (more like up the side of a sheer cliff) battle, which you
just won't win. People will look at you like you're nuts. Good way to
not win customers.

What we've done to handle the problem is to nightly build a MySQL
database which mirrors the data files in the application. Thee office
types who use weendoze can now use their favorite tools to access the
data, and even better, if they do manage to change anything, it's only a
throw away copy. This has done two things.

1) It has silenced the calls for a weendoze based system

2) It has short of shown that the original arguments were bogus, and
that the real call for weendoze was from the office and management types
who use weendoze in their work and who didn't want to learn anything
else. Still no or few calls for weendoze from the users of the application.

So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
going to be serious about a VMS GUI for the masses, continuing with
x-windows isn't going to cut the mustard. They would need to come up
with a new GUI that had the "look and feel" of MS weendoze. Not saying
that would be enough, but anything less would just be a waste of time
and money.

Me, I'd be happy right now for what we got on x86.

[pcov...@gmail.com]

Unigraphics was the CAD/CAM package available. and is now called NX and still one of the best cad packages there is.

PC

[Dirk Munk]

MG wrote:
> Dirk Munk schreef op 11-feb-2015 om 12:44:
>> The last graphics hardware that OpenVMS supported (to my knowledge) was
>> a very simple ATI Radeon 7200 card. That was a low-end PCI device, not
>> really the kind of hardware you need on a for instance a CAD/CAM
>> workstation.
>
> You mean the Radeon 7500? Mostly the PCI variant of it.

May have been the 7500, and indeed it was the PCI version.

> It wasn't
> actually such a bad card, when it originally came out... But it
> was already quite old by the time it became somewhat 'common' for
> VMS (and Tru64 UNIX) usage.

I know, when I bought one there were far more powerful Radeon cards.

> It would actually work fine for CAD/CAM,
> with its lower memory (fine for wireframes and shaded solids).

But these days many of the calculations are done by the graphics card.

>
> But what 'CAD/CAM' applications are there for under VMS...?

Very few, perhaps none. It's just hypothetical.

>
>
>> And what if we want to use OpenVMS as a high end workstation with high
>> performance graphics? In that case forget about consumer graphics, and
>> go for professional hardware. The AMD FirePro Workstation Graphics for
>> instance would be far more suitable.
>
> Everybody gave up with that, it seems. It's mostly Windows and, I
> guess, OS X workstations nowadays. (Also a number of Linux x86/-64
> ones, but mostly for post-production from what I understand, to a
> lesser degree technical graphics like CAD and CAM.)

Again, my suggestion is just hypothetical. *If* there is a demand for a
VMS workstation with high performance graphics, then look at these AMD
cards.

[Phillip Helbig (undress to reply)]

In article <54db6a0a$0$3853$e4fe...@dreader34.news.xs4all.nl>, MG

I have some old brochures. :-)

[Dirk Munk]

David Froble wrote:
> Dirk Munk wrote:
>> In the threat about OpenVMS Future Directions there has been a lot of
>> discussion about the question if OpenVMS should get graphics
>> capabilities. With that I mean if OpenVMS should support graphics
>> hardware, not if OpenVMS should have a GUI. A Gui could be a X-Windows
>> interface, or a browser interface, and of course both can be
>> supported. Graphics hardware is something very different.
>
> Let's talk about the "real world" for a bit, huh?
>
> If a company was serious about attracting new users to it's GUI capable
> systems, they are going to have to either cater to what the users
> expect, or, they are going to have to come up with something so great
> that the users of the world will want to switch to it.
>
> First, let's rule out the "something great" for now, as I don't have a
> clue as to what such might be, and if I did, I would not be
> participating on c.o.v, I'd be busy making enough money to have my own
> tropical island ....
>
> What we got left in the "real world" is what the vast majority of users
> think is a GUI based computer. It even has a name. It's called
> weendoze. Oh, Ok, Windows.
>
> I cannot count the times customers have asked for Codis to be a weendoze
> based product. One of their arguments is that since "everybody" knows
> weendoze, they would not have to train new hires and temporary hires.

What kind of user interface does Codis have? VT100 style? Of course you
could try to give it a GUI interface, there is plenty of information on
how to do that. For instance you could use Java, and no one using a PC
would ever guess he or she is in fact using a OpenVMS application.

>
> They are of course wrong in that supposition, but don't know it.
>
> The point is, there are people in management, and elsewhere, that have
> this concept, and won't understand that they'd have just as much
> training on the "application" regardless of the user interface.
>
> Now, I were to go to the great effort to provide them with an x-weendoze
> user interface, they'd give me a wierd look and ask, "what's this", and
> then once again start talking about weendoze.
>
> People believe their perceptions. People make decisions based on their
> perceptions. The vast majority of users have the perception that GUI
> and weendoze are one in the same. Trying to fight this perception is
> an uphill (more like up the side of a sheer cliff) battle, which you
> just won't win. People will look at you like you're nuts. Good way to
> not win customers.
>
> What we've done to handle the problem is to nightly build a MySQL
> database which mirrors the data files in the application.

That's a way to do it, but there tools that make it possible to use SQL
calls to read RMS datafiles. So any Windows application that can access
a remote database can be used.

> Thee office
> types who use weendoze can now use their favorite tools to access the
> data, and even better, if they do manage to change anything, it's only a
> throw away copy. This has done two things.
>
> 1) It has silenced the calls for a weendoze based system
>
> 2) It has short of shown that the original arguments were bogus, and
> that the real call for weendoze was from the office and management types
> who use weendoze in their work and who didn't want to learn anything
> else. Still no or few calls for weendoze from the users of the
> application.
>
> So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
> going to be serious about a VMS GUI for the masses, continuing with
> x-windows isn't going to cut the mustard. They would need to come up
> with a new GUI that had the "look and feel" of MS weendoze. Not saying
> that would be enough, but anything less would just be a waste of time
> and money.

For what kind of system? Who would be using those Windows lookalike
systems? Office staff?

[Bill Gunshannon]

In article <mbftgs$2fsn$1...@news.kjsl.com>,

And some of it wold not have mentioned The VAX or VMS specifically.
PLOT-10 was a library usually in source that supported Tektronix.
Same frop Westinghouse GCS. Maybe I need to dig out my Software
Source Books and see what's listed.

[JF Mezei]

On 15-02-11 06:44, Dirk Munk wrote:

> A typical server doesn't have graphics hardware, with one exception.

Actually they do because they still have a console and plug for monitor
and keyboard. And that graphics card also maintains the display so it
can be used by the remote desktop software. (aka: access the server's
GUI from another desktop). As such, the server still needs to be
graphics aware.

> These days servers have an Integrated Lights Out (ILO) port with its own
> network interface. It has a browser interface, and it replaces the
> traditional Bios interface/SRM console.

ILO is generally IPMI and the actual protocol is command driven, not
HTTP driven. This is what is used to get status, power up/down and
monitor hardware irrespective of software running.

The other part of it would be the EFI interface that allows one to setup
the system. I believe that on Xserves, when you boot into EFI (not easy
task), it runs on the display and expects a keyboard.

> On the other hand a typical desktop/laptop CPU had an integrated GPU
> these days. I'm sure there is a market for OpenVMS desktop and even
> laptop systems.

It is my opinion that one should not see the desktop as a "market" for
VMS, but more like a necessary step to get developpers to produce
software for VMS.

Running VMS on a Laptop has interesting potential when managing a data
centre with the laptop usable to manage VMS with SYSMAN etc. VSI could
even produce a GUI version of SYSMAN with the laptop running as member
of cluster.

It should also produce tooks to monitor servers via IPMI (as the Server
Monitor app that Apple produced back when it was in the server business)


Of course, all of this greatly depends on what sort of hardware VSI will
initially be supporting. If it will be targeting BIOS low end servers
without IPMI/EFI, then it needs to have BIOS support (puke, barf).

[Simon Clubley]

On 2015-02-11, Bill Gunshannon <bi...@server3.cs.scranton.edu> wrote:
> In article <mbfjtp$5nf$1...@dont-email.me>,
> Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>>
>> Sorry Bill, but no they are not because there's one real big one left - the
>> software to drive this hardware.
>
> Assuming that they are going to stay with X rather than write a VMS specific
> proprietary protocol, most of it already exists and is readily available.
> Only device drivers for specific graphics cards would be needed and there
> is a lot of source available for those, too. (Not VMS source, but enough
> to reveal the needed programming information.)
>

The source tells you what needs to be done in a specific set of
circumstances; it does not tell you _why_ it needs to be done.

The source also does not give you a full model of operation for the
device in question but any decent specification will.

Knowledge obtained from the source code only is fragile and is prone
to break when the circumstances change. For example, what about things
like error recovery which VMS is strong on but may not be such a
high priority in another OS's driver ?

>>
>> Have you ever considered the effort involved in tracking the continual
>> hardware changes in the x86 world, including the continual device driver
>> changes or outright new drivers required ?
>
> It's called business. They have to do the same for storage devices.
> Or are you suggesting VMS stick with MSCP to avoid "hardware changes"
> and "new drivers"? Or, just ignore another market segment that VMS
> abandoned years ago.
>

Storage devices are a core part of a server operating system; high
performance graphics cards are not.

There also tends to be much more of a open standards culture for those
storage devices and those storage interfaces have a design lifetime
which is a lot longer than the graphics card interfaces have.

And yes, I know there are still device specific bugs to handle; however
this is an issue which exists on top of the interface specs and is true
regardless of whatever class of device you are writing a driver for.

[JF Mezei]

On 15-02-11 13:14, Simon Clubley wrote:

> Storage devices are a core part of a server operating system; high
> performance graphics cards are not.

Is that truly the case anymore ? Wit various libraries that allow one
to use the graphic card(s) as compute engines, is this something a
server operating system can afford to ignore for much longer ?

Sure, if you are running COBOL or BASIC, chances are your computations
involve decimal monetary amounts and not floating points.

But if you are wanting to create waves around the Titanic, then that
extra CPU power comes in really handy (I mention that because those
waves had been done on Alphas as I recall).

Remember that VSI need to not only look at yesterday (what current
customers have told them what they are doing) but also tomorrow.

[Dirk Munk]

JF Mezei wrote:
> On 15-02-11 06:44, Dirk Munk wrote:
>
>> A typical server doesn't have graphics hardware, with one exception.
>
> Actually they do because they still have a console and plug for monitor
> and keyboard. And that graphics card also maintains the display so it
> can be used by the remote desktop software. (aka: access the server's
> GUI from another desktop). As such, the server still needs to be
> graphics aware. >
>
>> These days servers have an Integrated Lights Out (ILO) port with its own
>> network interface. It has a browser interface, and it replaces the
>> traditional Bios interface/SRM console.
>
> ILO is generally IPMI and the actual protocol is command driven, not
> HTTP driven. This is what is used to get status, power up/down and
> monitor hardware irrespective of software running.

Well, even on my simple several years old Asus server motherboard there
is a one chip web server ILO management port. You can use it to power-up
and power-down the server, do bios settings, read voltages and so on.
These interfaces are quite common these days.

>
> The other part of it would be the EFI interface that allows one to setup
> the system. I believe that on Xserves, when you boot into EFI (not easy
> task), it runs on the display and expects a keyboard.
>
>
>> On the other hand a typical desktop/laptop CPU had an integrated GPU
>> these days. I'm sure there is a market for OpenVMS desktop and even
>> laptop systems.
>
> It is my opinion that one should not see the desktop as a "market" for
> VMS, but more like a necessary step to get developpers to produce
> software for VMS.

A platform for a developer can also be regarded as a market, and even
more if it enables the developer to work at any place he likes. These
days you don't have to sit in an office to do your work, and with a
simple NAS and iSCSI (VMS must support ISCSI!) it's even possible to
support a multi-terrabyte database at home.

>
> Running VMS on a Laptop has interesting potential when managing a data
> centre with the laptop usable to manage VMS with SYSMAN etc. VSI could
> even produce a GUI version of SYSMAN with the laptop running as member
> of cluster.
>
> It should also produce tooks to monitor servers via IPMI (as the Server
> Monitor app that Apple produced back when it was in the server business)
>
>
> Of course, all of this greatly depends on what sort of hardware VSI will
> initially be supporting. If it will be targeting BIOS low end servers
> without IPMI/EFI, then it needs to have BIOS support (puke, barf).
>

Any modern x64 server supports EFI, and most certainly when the x64
version of VMS is ready.

[johnwa...@yahoo.co.uk]

Quite.

For reference, from a time when graphics cards were simpler, there's
the time and effort (and pain) that went into VGA emulation in the
SRM console on Alphas. Different graphics cards have different VGA
BIOSes built in; they contain x86 code which the system BIOS executes
to set up the graphics card's basic operation. Later versions of the
SRM console had a basic (not BASIC :)) x86 emulator built in, simply
so that the SRM console didn't have to have detailed knowledge of how
to initialise and drive a generic VGA card.
VGA card.

Want something more sophisticated than simple VGA frame buffer? Then
expect more complexity and more device-specificness.

More info, try the Alpha Architecture Reference Manual. Probably not
in the first edition, seems to be in 4th edition downloadable from
https://archive.org/details/dec-alpha_arch_ref (~8MB, ~1000 pages, not
all of which are about VGA BIOS emulation).

I'm not sure where the EFI vs legacy-BIOS road leads in this respect,
but it would be a very pleasant surprise if things had improved.

Have a lot of fun

[JF Mezei]

On 15-02-11 13:49, johnwa...@yahoo.co.uk wrote:

> I'm not sure where the EFI vs legacy-BIOS road leads in this respect,
> but it would be a very pleasant surprise if things had improved.

Wouldn't it be the same , where EFI provides extremely basic console
support for initial output of early boot sequence, and eventually the OS
has to take control of the graphics device with its own drivers ?

or could an OS continue to rely on the EFI provided access to the
console screen/keyboard throughoput the lifetime of the instance ?

[mcle...@gmail.com]

A radical thought or two ...

Could VMS have an X86 emulator to handle graphics (or for that matter any Windows or Chrome stuff at all) ?

When VMS is available on X86 having an emulator to run X86 seems silly so would it be possible and useful to be able to activate native X86 assembler directly and then return into VMS?

I have no idea about the technical issues involved; I simply ask the questions.

[Jan-Erik Soderholm]

Dirk Munk skrev den 2015-02-11 19:44:
> JF Mezei wrote:
>> On 15-02-11 06:44, Dirk Munk wrote:
>>
>>> A typical server doesn't have graphics hardware, with one exception.
>>
>> Actually they do because they still have a console and plug for monitor
>> and keyboard. And that graphics card also maintains the display so it
>> can be used by the remote desktop software. (aka: access the server's
>> GUI from another desktop). As such, the server still needs to be
>> graphics aware. >
>>
>>> These days servers have an Integrated Lights Out (ILO) port with its own
>>> network interface. It has a browser interface, and it replaces the
>>> traditional Bios interface/SRM console.
>>
>> ILO is generally IPMI and the actual protocol is command driven, not
>> HTTP driven. This is what is used to get status, power up/down and
>> monitor hardware irrespective of software running.
>
> Well, even on my simple several years old Asus server motherboard there is
> a one chip web server ILO management port. You can use it to power-up and
> power-down the server, do bios settings, read voltages and so on. These
> interfaces are quite common these days.
>

OK, sure. But what has that to do with VMS? Even if that "server
motherboard" might run VMS, you'd probably use your standard
laptop to access the ILO. VMS would never know...

>>
>> The other part of it would be the EFI interface that allows one to setup
>> the system. I believe that on Xserves, when you boot into EFI (not easy
>> task), it runs on the display and expects a keyboard.
>>
>>
>>> On the other hand a typical desktop/laptop CPU had an integrated GPU
>>> these days. I'm sure there is a market for OpenVMS desktop and even
>>> laptop systems.
>>
>> It is my opinion that one should not see the desktop as a "market" for
>> VMS, but more like a necessary step to get developpers to produce
>> software for VMS.
>
> A platform for a developer can also be regarded as a market, and even more
> if it enables the developer to work at any place he likes.

I have a Win7 laptop and I work at any place I like.
I must be missing something here...

[Dirk Munk]

It was all about the discussion that a server doesn't need to have a
graphics card these days. It can be replaced by an ILO port with
integrated web server.

>
>>>
>>> The other part of it would be the EFI interface that allows one to setup
>>> the system. I believe that on Xserves, when you boot into EFI (not easy
>>> task), it runs on the display and expects a keyboard.
>>>
>>>
>>>> On the other hand a typical desktop/laptop CPU had an integrated GPU
>>>> these days. I'm sure there is a market for OpenVMS desktop and even
>>>> laptop systems.
>>>
>>> It is my opinion that one should not see the desktop as a "market" for
>>> VMS, but more like a necessary step to get developpers to produce
>>> software for VMS.
>>
>> A platform for a developer can also be regarded as a market, and even
>> more
>> if it enables the developer to work at any place he likes.
>
> I have a Win7 laptop and I work at any place I like.
> I must be missing something here...
>

Yes you can, but can you compile and link VMS programs on your laptop?
The issue was if there can be demand for VMS on a laptop.

[JF Mezei]

On 15-02-11 18:43, Dirk Munk wrote:

> It was all about the discussion that a server doesn't need to have a
> graphics card these days. It can be replaced by an ILO port with
> integrated web server.

ILO port does not manage the OS instance, it manages the hardware. And
depending on system, the equivalent of SRM.

At best, you can use the ILO to order a "soft" power-off (where it sends
a signal to OS which has a couple of seconds before power is shut).


This is quite different from logging into the server to manage it, its
parameners, networking (IP address etc), user authentication etc.

And in cases where networking is out, you really need a way to login to
the server's console to fix it. Can't be done remotely.

[John E. Malmberg]

On 2/11/2015 8:52 AM, David Froble wrote:
>
> So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
> going to be serious about a VMS GUI for the masses, continuing with
> x-windows isn't going to cut the mustard. They would need to come up
> with a new GUI that had the "look and feel" of MS weendoze. Not saying
> that would be enough, but anything less would just be a waste of time
> and money.

X-11 look and feel is not the problem. It should not be hard to port
any of the various Linux GUI Shells to VMS. If the GTK+ 2.x+ port ever
gets done, you might be surprised at what will simply compile and work
with out any source changes.

VMS graphics are based on Motif and CDE, neither of which are common in
the Linux/Windows/OS-x and and now Android.

X-11 is a problem though for mobile and portable devices. VNC can
survive, but there can be a lot of latency in updating. RDP attempts to
deal with that, but is not a public standard protocol AFAIK.


As others have pointed out, medium to high end video drivers are an issue.

Modern video cards have effectively two APIs.

One is a public one that has the video card run in the lower performance
modes. Essentially, your $50+ video card is emulating a less than $20
video card. Open source drivers are available for these modes. In some
cases some of the high performance modes have been figured out or
released, but not the highest performance modes.

The other is private one that needs a binary driver from from the video
card vendor, or the chip set vendor. This is for the high performance
modes.

When you load a linux distribution that only contains GPL components,
many of them will provide an installer that will go to the chip set
vendor to download the closed source drivers when they detect the
advanced video chip-sets.

Other Linux distributions will just include the closed source drivers as
they are not so strict about what they include, so you do not realize
that the source is not available.

The specifications for writing the high performance drivers for many
video chip sets are secret, and you can not even get them with an NDA
agreement.

If you are an x86 based OS, you probably want to find a way to use
unmodified linux or Windows binary drivers, especially for graphics devices.

Regards,
-John

[Dirk Munk]

John E. Malmberg wrote:
> On 2/11/2015 8:52 AM, David Froble wrote:
>>
>> So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
>> going to be serious about a VMS GUI for the masses, continuing with
>> x-windows isn't going to cut the mustard. They would need to come up
>> with a new GUI that had the "look and feel" of MS weendoze. Not saying
>> that would be enough, but anything less would just be a waste of time
>> and money.
>
> X-11 look and feel is not the problem. It should not be hard to port
> any of the various Linux GUI Shells to VMS. If the GTK+ 2.x+ port ever
> gets done, you might be surprised at what will simply compile and work
> with out any source changes.
>
> VMS graphics are based on Motif and CDE, neither of which are common in
> the Linux/Windows/OS-x and and now Android.
>
> X-11 is a problem though for mobile and portable devices. VNC can
> survive, but there can be a lot of latency in updating. RDP attempts to
> deal with that, but is not a public standard protocol AFAIK.
>
>
> As others have pointed out, medium to high end video drivers are an issue.
>
> Modern video cards have effectively two APIs.
>
> One is a public one that has the video card run in the lower performance
> modes. Essentially, your $50+ video card is emulating a less than $20
> video card.

I suppose you are referring to the VESA Bios Extensions?

> Open source drivers are available for these modes. In some
> cases some of the high performance modes have been figured out or
> released, but not the highest performance modes.
>
> The other is private one that needs a binary driver from from the video
> card vendor, or the chip set vendor. This is for the high performance
> modes.
>
> When you load a linux distribution that only contains GPL components,
> many of them will provide an installer that will go to the chip set
> vendor to download the closed source drivers when they detect the
> advanced video chip-sets.
>
> Other Linux distributions will just include the closed source drivers as
> they are not so strict about what they include, so you do not realize
> that the source is not available.
>
> The specifications for writing the high performance drivers for many
> video chip sets are secret, and you can not even get them with an NDA
> agreement.
>
> If you are an x86 based OS, you probably want to find a way to use
> unmodified linux or Windows binary drivers, especially for graphics
> devices.

My laptop has two graphics devices, the Intel HD4000 GPU embedded in the
CPU, and a Nvidia card. I the laptop doesn't need high performance
graphics, the Intel GPU is used. If it does need high performance, it
can switch to the Nvidia card (it never does).

The Intel drivers don't receive many updates, the Nvidia drivers almost
every month so it seems.

I suppose VSI could get a lot of information on the Intel GPU, and as I
wrote before, the Intel graphics are more than sufficient for typical
VMS work.

>
> Regards,
> -John

[Jan-Erik Soderholm]

Not "on" but "from".

> The issue was if there can be demand for VMS on a laptop.
>

Right.

[Paul Sture]

An idea that refuses to go away, so I'll spit it out:

A Raspberry Pi, Arduino or similar probably has more than enough
graphics capability to handle console work, and in those we have large
runs of the same graphics components, and those are Someone Else's
Problem.

;-)

P.S. The Raspberry Pi 2, just out, doesn't like camera flashes,
apparently, but a bit of Blu-Tack will fix this.
<http://www.theregister.co.uk/2015/02/08/raspberry_pi_2_camera_flash_glitch/>

--
1972 - IBM begins development on its last tape drive (3480) ever because
of the declining cost of disk drives.

[Phillip Helbig (undress to reply)]

In article <ck1f9t...@mid.individual.net>,

bi...@server3.cs.scranton.edu (Bill Gunshannon) writes:

> >> I keep hearing that, from time to time, but where is the proof?
> >> Nobody could ever cite me specific software packages. Even old
> >> product brochures don't mention too terribly much... (The only
> >> traces of software somewhat belonging into this category, that I
> >> could ever find documented traces of, was for ULTRIX.)
> >
> > I have some old brochures. :-)
>
> And some of it wold not have mentioned The VAX or VMS specifically.
> PLOT-10 was a library usually in source that supported Tektronix.
> Same frop Westinghouse GCS. Maybe I need to dig out my Software
> Source Books and see what's listed.

The ones I have are specifically marketing the ALPHAstation running VMS
for CAD/CAM stuff.

[Phillip Helbig (undress to reply)]

In article <54db9397$0$4485$b1db1813$2411...@news.astraweb.com>, JF

Mezei <jfmezei...@vaxination.ca> writes:

> > A typical server doesn't have graphics hardware, with one exception.
>
> Actually they do because they still have a console and plug for monitor
> and keyboard.

Certainly not all do. Many of mine don't. Connections for keyboard and
terminal, yes, but not necessarily a graphics monitor.

[Bill Gunshannon]

In article <c3243$54dbfbfd$5ed4324a$31...@news.ziggo.nl>,

I think this pretty much sums it up. It is not the low level graphics
capabilities that are the problem. It is the VMS X implementation.
Given something more modern than the DECWindows we have I could do (and
actually usually do) most desktop tasks on VMS. No one is going to use
VMS to play World of Warcraft (although I might try Minecraft!! :-).
But things like Open Office and better IDE's can all be don on VMS. All
it would take is a version of X-11 modern enough to let these Open Source
programs be built on VMS. I know it is a major undertaking, but all the
sources to current X-11 distributions (there are 2 but one has one out in
the marketplace) are available. And that includes X-servers for numerous
video cards.

[Bob Koehler]

In article <f056991c-4b5e-40cc...@googlegroups.com>, mcle...@gmail.com writes:
> A radical thought or two ...
>
> Could VMS have an X86 emulator to handle graphics (or for that matter any Windows or Chrome stuff at all) ?

Folks ahave run x86 emulators under VMS for a variety of
applications.

Some VMS device drivers have a bit of x86 emulation to operate code
stored in ROM on device controllers.

Not new.

[Bob Koehler]

In article <f056991c-4b5e-40cc...@googlegroups.com>, mcle...@gmail.com writes:

> A radical thought or two ...
>
> Could VMS have an X86 emulator to handle graphics (or for that matter any Windows or Chrome stuff at all) ?

Missed this on my last post (you need to wrap your lines at a
reasonable length, the above is going off the right side of my
screen).

Running Chrome et. al. does'nt just need an x86 emulator. Chrome
runs on tops of an OS, be it Windows, Mac OS, Android, or other.
Which means you have to deal with it's calls to that OS, or be
running a copy of that OS in the emulation.

[Bob Koehler]

In article <3d151$54dbe922$5ed4324a$56...@news.ziggo.nl>, Dirk Munk <mu...@home.nl> writes:
>
> Yes you can, but can you compile and link VMS programs on your laptop?
> The issue was if there can be demand for VMS on a laptop.

I don't know about his laptop, but yes you can. I've compiled,
linked, and run VMS programs on my laptop via SIMH emulation.
And Tadpole made an Alpha laptop which ran VMS native.

But I do think there is a real call for VMS on the desktop to
attract developers. A lot of developers will be put off by the
notion that they must log into a server to do their work, even if
you sell a really cheap server.

[David Froble]

Yes, a lot of VT terminal stuff. Doesn't matter.

>> They are of course wrong in that supposition, but don't know it.
>>
>> The point is, there are people in management, and elsewhere, that have
>> this concept, and won't understand that they'd have just as much
>> training on the "application" regardless of the user interface.
>>
>> Now, I were to go to the great effort to provide them with an x-weendoze
>> user interface, they'd give me a wierd look and ask, "what's this", and
>> then once again start talking about weendoze.
>>
>> People believe their perceptions. People make decisions based on their
>> perceptions. The vast majority of users have the perception that GUI
>> and weendoze are one in the same. Trying to fight this perception is
>> an uphill (more like up the side of a sheer cliff) battle, which you
>> just won't win. People will look at you like you're nuts. Good way to
>> not win customers.
>>
>> What we've done to handle the problem is to nightly build a MySQL
>> database which mirrors the data files in the application.
>
> That's a way to do it, but there tools that make it possible to use SQL
> calls to read RMS datafiles. So any Windows application that can access
> a remote database can be used.

Maybe we don't want the "office types" in the data files. Why write
code to keep things "correct" just to let someone go and change anything
they want?

>> Thee office
>> types who use weendoze can now use their favorite tools to access the
>> data, and even better, if they do manage to change anything, it's only a
>> throw away copy. This has done two things.
>>
>> 1) It has silenced the calls for a weendoze based system
>>
>> 2) It has short of shown that the original arguments were bogus, and
>> that the real call for weendoze was from the office and management types
>> who use weendoze in their work and who didn't want to learn anything
>> else. Still no or few calls for weendoze from the users of the
>> application.
>>
>> So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
>> going to be serious about a VMS GUI for the masses, continuing with
>> x-windows isn't going to cut the mustard. They would need to come up
>> with a new GUI that had the "look and feel" of MS weendoze. Not saying
>> that would be enough, but anything less would just be a waste of time
>> and money.
>
> For what kind of system? Who would be using those Windows lookalike
> systems? Office staff?

Anybody who thinks (has the perception) that all computers look like
weendoze ....

[David Froble]

I'm going to disagree. Why? those "billion weendoze users" Steve has
mentioned. They have perceptions and expectations. Ignore their
perceptions and expectations, and you won't attract them.

Note, I'm not advocating going after the GUI desktop market. All I'm
saying is that if you're going to do so, you need to meet those
perceptions and expectations. Anything else is a waste of time and
effort (and money) which would be better spent on some of the things VMS
needs.

[David Froble]

Dirk Munk wrote:

> Yes you can, but can you compile and link VMS programs on your laptop?
> The issue was if there can be demand for VMS on a laptop.

If the issue doesn't involve needing performance, then an emulator on a
notebook would serve just fine. No matter the difference in performance
between native and emulation, a compile and link just won't be an issue.

Perhaps better places to spend the limited time and money ????

[John E. Malmberg]

On 2/12/2015 8:34 AM, David Froble wrote:

> Bill Gunshannon wrote:
>>
>> I think this pretty much sums it up. It is not the low level graphics
>> capabilities that are the problem. It is the VMS X implementation.
>> Given something more modern than the DECWindows we have I could do (and
>> actually usually do) most desktop tasks on VMS. No one is going to use
>> VMS to play World of Warcraft (although I might try Minecraft!! :-).
>> But things like Open Office and better IDE's can all be don on VMS. All
>> it would take is a version of X-11 modern enough to let these Open Source
>> programs be built on VMS. I know it is a major undertaking, but all the
>> sources to current X-11 distributions (there are 2 but one has one out in
>> the marketplace) are available. And that includes X-servers for numerous
>> video cards.

There is nothing that is needed for the X-11 on VMS that is preventing
the ports of the above programs. While it would be nice to get to more
current, the existing X-11 is good enough.

The issue is that the GTK+ 2.x was not ported to VMS or maintained, and
that is the API that all the GUI programs expect.

I ported most of GTK+ 2.x to VMS over 10 years ago using shared images
and PCSI install kits, and it seems to have been mostly ignored.

I did not have time to put the finishing touches on the final component,
the actual GTK+ library.

As I remember it, the missing steps for the final kit were:
* Creating the transfer vectors.
* Fixing an issue with handling the VMS X-11 delete key.
GTK+ 2.x only accepted Control-H for delete.
* Building the PCSI kit for distribution.

That is what has been blocking getting most Linux GUI applications on
VMS. And it has been sitting in the FTP directory on
Eisner.Encompasserve.org

Of course now the code is 10 years out of date, and everything needs a
refresh.

The GTK+ lower level rendering libraries can not deal with 8 bit color,
so use on the older VMS displays is not possible. That is not an X-11
limitation.

> I'm going to disagree. Why? those "billion weendoze users" Steve has
> mentioned. They have perceptions and expectations. Ignore their
> perceptions and expectations, and you won't attract them.

> Note, I'm not advocating going after the GUI desktop market. All I'm
> saying is that if you're going to do so, you need to meet those
> perceptions and expectations. Anything else is a waste of time and
> effort (and money) which would be better spent on some of the things VMS
> needs.

A current GTK+ port essentially gives you a cross platform GUI. It also
handles that VMS does not have a fork().

The demo GTK+ programs just compiled and worked. Only issue was that
the X11 Delete key from VMS workstations was not recognized, and that is
something that can easily be fixed.

Regards,
-John

[David Froble]

But .... what am I going to do with all these server type system I got
laying around?

:-)

Really, I got no problem running EDT on a VT emulator on my Weendoze
2000 system ....

[Dirk Munk]

Depends on what you want to achieve. If you can make the office people
happy with a GUI, well then go for it. Lots of people in the thread are
complaining VMS has no GUI, and here it is possible to replace the VT
interface by a GUI interface.

>
>>> They are of course wrong in that supposition, but don't know it.
>>>
>>> The point is, there are people in management, and elsewhere, that have
>>> this concept, and won't understand that they'd have just as much
>>> training on the "application" regardless of the user interface.
>>>
>>> Now, I were to go to the great effort to provide them with an x-weendoze
>>> user interface, they'd give me a wierd look and ask, "what's this", and
>>> then once again start talking about weendoze.
>>>
>>> People believe their perceptions. People make decisions based on their
>>> perceptions. The vast majority of users have the perception that GUI
>>> and weendoze are one in the same. Trying to fight this perception is
>>> an uphill (more like up the side of a sheer cliff) battle, which you
>>> just won't win. People will look at you like you're nuts. Good way to
>>> not win customers.
>>>
>>> What we've done to handle the problem is to nightly build a MySQL
>>> database which mirrors the data files in the application.
>>
>> That's a way to do it, but there tools that make it possible to use
>> SQL calls to read RMS datafiles. So any Windows application that can
>> access a remote database can be used.
>
> Maybe we don't want the "office types" in the data files. Why write
> code to keep things "correct" just to let someone go and change anything
> they want?
>

That depends on what the office people want to do. Normally the data
files should be accessed read-only of course, and that should be possible.

>>> Thee office
>>> types who use weendoze can now use their favorite tools to access the
>>> data, and even better, if they do manage to change anything, it's only a
>>> throw away copy. This has done two things.
>>>
>>> 1) It has silenced the calls for a weendoze based system
>>>
>>> 2) It has short of shown that the original arguments were bogus, and
>>> that the real call for weendoze was from the office and management types
>>> who use weendoze in their work and who didn't want to learn anything
>>> else. Still no or few calls for weendoze from the users of the
>>> application.
>>>
>>> So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
>>> going to be serious about a VMS GUI for the masses, continuing with
>>> x-windows isn't going to cut the mustard. They would need to come up
>>> with a new GUI that had the "look and feel" of MS weendoze. Not saying
>>> that would be enough, but anything less would just be a waste of time
>>> and money.
>>
>> For what kind of system? Who would be using those Windows lookalike
>> systems? Office staff?
>
> Anybody who thinks (has the perception) that all computers look like
> weendoze ....

So just to proof a point, not because someone actually needs it?

[Dirk Munk]

David Froble wrote:
> Dirk Munk wrote:
>
>> Yes you can, but can you compile and link VMS programs on your laptop?
>> The issue was if there can be demand for VMS on a laptop.
>
> If the issue doesn't involve needing performance, then an emulator on a
> notebook would serve just fine.

You want to emulate a x64 CPU on a x64 system?????

[David Froble]

The office people are not the users of the application.

While there isn't so much manual order entry any more, we showed
multiple times that a GUI would just slow down the users.

Note, it's never been the users who complained.

>>
>>>> They are of course wrong in that supposition, but don't know it.
>>>>
>>>> The point is, there are people in management, and elsewhere, that have
>>>> this concept, and won't understand that they'd have just as much
>>>> training on the "application" regardless of the user interface.
>>>>
>>>> Now, I were to go to the great effort to provide them with an
>>>> x-weendoze
>>>> user interface, they'd give me a wierd look and ask, "what's this", and
>>>> then once again start talking about weendoze.
>>>>
>>>> People believe their perceptions. People make decisions based on their
>>>> perceptions. The vast majority of users have the perception that GUI
>>>> and weendoze are one in the same. Trying to fight this perception is
>>>> an uphill (more like up the side of a sheer cliff) battle, which you
>>>> just won't win. People will look at you like you're nuts. Good way to
>>>> not win customers.
>>>>
>>>> What we've done to handle the problem is to nightly build a MySQL
>>>> database which mirrors the data files in the application.
>>>
>>> That's a way to do it, but there tools that make it possible to use
>>> SQL calls to read RMS datafiles. So any Windows application that can
>>> access a remote database can be used.
>>
>> Maybe we don't want the "office types" in the data files. Why write
>> code to keep things "correct" just to let someone go and change anything
>> they want?
>>
>
> That depends on what the office people want to do. Normally the data
> files should be accessed read-only of course, and that should be possible.

We fixed it to our satisfaction with the relational database. They get
to use the tools they know. We stop hearing the complaints.

>>>> Thee office
>>>> types who use weendoze can now use their favorite tools to access the
>>>> data, and even better, if they do manage to change anything, it's
>>>> only a
>>>> throw away copy. This has done two things.
>>>>
>>>> 1) It has silenced the calls for a weendoze based system
>>>>
>>>> 2) It has short of shown that the original arguments were bogus, and
>>>> that the real call for weendoze was from the office and management
>>>> types
>>>> who use weendoze in their work and who didn't want to learn anything
>>>> else. Still no or few calls for weendoze from the users of the
>>>> application.
>>>>
>>>> So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
>>>> going to be serious about a VMS GUI for the masses, continuing with
>>>> x-windows isn't going to cut the mustard. They would need to come up
>>>> with a new GUI that had the "look and feel" of MS weendoze. Not saying
>>>> that would be enough, but anything less would just be a waste of time
>>>> and money.
>>>
>>> For what kind of system? Who would be using those Windows lookalike
>>> systems? Office staff?
>>
>> Anybody who thinks (has the perception) that all computers look like
>> weendoze ....
>
> So just to proof a point, not because someone actually needs it?

Indeed! Nobody actually NEEDS weendoze. But for many, it's all they know.

Now, keep in mind the billion weendoze users Steve has claimed. If, and
I say IF, one would want to attract some of those billion, don't you
have to go to them, instead of expecting them to come to you?

That is my whole point. If one were to target the billion weendoze
users, the only reasonable way is to give them what they expect. They
won't go for anything else.

[johnwa...@yahoo.co.uk]

But which "what they expect from Windows" do you give them?

W98? NT3.51? NT4? Win2K? [not Vista] WinXP? Win 7? [not Win8] Win 10?

Despite the tales of "Windows familiarity", those flavours of Windows
have as many differences as they have similarities.

Same goes for Office 9x vs Office 200x-without-Ribbon vs Office-with-Ribbon.

The talk of "Windows familiarity" is at odds with reality outside the IT
world, and is likely irrelevant anyway when what the end user needs to see
is a specific line of business application set rather than a Windows
desktop and everything that goes with it.

Anyway, you found a solution that works for your customers, which is grand.

[David Froble]

A very good point, which makes things even worse.

> W98? NT3.51? NT4? Win2K? [not Vista] WinXP? Win 7? [not Win8] Win 10?

Not even MS got away with all of those. I cannot imagine an unknown VSI
even thinking of getting away with all of those.

> Despite the tales of "Windows familiarity", those flavours of Windows
> have as many differences as they have similarities.
>
> Same goes for Office 9x vs Office 200x-without-Ribbon vs Office-with-Ribbon.
>
> The talk of "Windows familiarity" is at odds with reality outside the IT
> world, and is likely irrelevant anyway when what the end user needs to see
> is a specific line of business application set rather than a Windows
> desktop and everything that goes with it.

Maybe, but, if you're going after the entire market, then it's not just
special appications. If it is a special application, then for at least
the application, it doesn't matter.

The only reason we had problems was that some non-users of the
application wanted to pull some data from the data files, and they only
wanted to do it in the manner they wanted. It really didn't have
anything to do with the application software. Or data.

[johnwa...@yahoo.co.uk]

But why would anyone other than MS think it sensible to go after the
entire market (from embedded to datacentre?) with one (allegedly)
'familiar'/'compatible' product?

One size does not fit all. Mainframes still exist. Tandem still exists
(new name, same ideas). UNIX-like OSes still exist. And for now at
least, VMS still exists, and hopefully is looking better than it has
for a while.

Oh well. Interesting times.

[Phillip Helbig (undress to reply)]

Dirk Munk wrote:

> In the threat about OpenVMS Future Directions there has been a lot of
> discussion about the question if OpenVMS should get graphics
> capabilities. With that I mean if OpenVMS should support graphics
> hardware, not if OpenVMS should have a GUI. A Gui could be a X-Windows
> interface, or a browser interface, and of course both can be
> supported. Graphics hardware is something very different.
> Let's talk about the "real world" for a bit, huh?

It seems there are two camps here, the larger(?) touting VMS as a
"server operating system", having given up the idea that VMS can be used
for any graphics stuff, and the other wondering whether it could or
should compete with high-end graphics on other systems.

However, I think the goal, at least for now, is to have the graphics
capabilities of, say, a $200 laptop, so that 99% of people (whether
previously familiar with VMS or not) can do 99% of what they need to do
on it.

To those who run a web browser on one platform and use it to access
stuff they use on VMS: you are wasting WAY too much time.

A related issue is a VMS laptop. Again, target the $200 PC, not the
$5000 MacBook Pro. Is emulation an alternative? Not in all cases.
While it might be fast enough to emulate a VAX, it is probably not fast
enough to usefully emulate VMS running natively on the same hardware.
Also, there are perhaps license costs and certainly overhead associated
with the host operating system, not to mention security.

Maybe at my funeral I'll get someone to say "ashes to ashes, desktop to
datacenter". :-)

[johnwa...@yahoo.co.uk]

One way to address running VMS on random x86 boxes (laptops etc) is to
make nuVMS run reasonably well in a VMware environment (or similar). Out
of the box with no engineering effort necessary, the VMware layer in
theory isolates nuVMS from the obscurities of the underlying hardware,
by providing pseudodevices (disk, graphics, etc) which are (allegedly)
the same interfaces and behaviours regardless of the real hardware
underneath. It seems good enough to keep the Windows world happy.

The additional complexity and the performance impact means it's not
necessarily something for every occasion, and there may be costs
associated in some circumstances, but it has its applications.

If you've never tried it, you may want to give it a go. You may get a
surprise.

[Jan-Erik Soderholm]

Phillip Helbig (undress to reply) skrev den 2015-02-13 11:12:
> Dirk Munk wrote:
>
>> In the threat about OpenVMS Future Directions there has been a lot of
>> discussion about the question if OpenVMS should get graphics
>> capabilities. With that I mean if OpenVMS should support graphics
>> hardware, not if OpenVMS should have a GUI. A Gui could be a X-Windows
>> interface, or a browser interface, and of course both can be
>> supported. Graphics hardware is something very different.
>> Let's talk about the "real world" for a bit, huh?
>
> It seems there are two camps here, the larger(?) touting VMS as a
> "server operating system", having given up the idea that VMS can be used
> for any graphics stuff, and the other wondering whether it could or
> should compete with high-end graphics on other systems.
>
> However, I think the goal, at least for now, is to have the graphics
> capabilities of, say, a $200 laptop, so that 99% of people (whether
> previously familiar with VMS or not) can do 99% of what they need to do
> on it.

Why in earth would someone not "previously familiar with VMS"
even get the thought of running VMS on a laptop?

Most people needing a laptop get a laptop with whatever
is preinstalled from the local IT-shop.

Over here in Sweden a few years ago some distributors tried to
ship pre-configured Linux laptops side-by-side with the Windows
laptops. It totaly went away a year later and today those
laptops doesn't exist any longer. The market was non-existent.

If I look at the largest Swedish online distributor and look
what OS'es are available, they are like this (no of laptop
offerings with each OS):

Win 8/8.1: 514
Win 7: 380
Chrome: 23
MacOS X: 29 (Apples, of course)
Ubuntu: 1
None: 4

And the Linux/Ubuntu one is a high-end modell at aprox $1.800.

So how can anyone even think that there would be a market for
*VMS* based laptops?

Say that I can do 99% of what I need to do on my VMS laptop.
But that last 1% happens to be that my internet banking appl
doesn't exist for VMS. Now that will be enought for that
solution beeing totaly nonworking for "99% of people".

>
> To those who run a web browser on one platform and use it to access
> stuff they use on VMS: you are wasting WAY too much time.

Trying to do everything on VMS would be a total wast of time...

[VAXman-]

In article <mbkp6r$p1n$1...@news.albasani.net>, Jan-Erik Soderholm <jan-erik....@telia.com> writes:
>Phillip Helbig (undress to reply) skrev den 2015-02-13 11:12:

>> Dirk Munk wrote:
>>
>>> In the threat about OpenVMS Future Directions there has been a lot of
>>> discussion about the question if OpenVMS should get graphics
>>> capabilities. With that I mean if OpenVMS should support graphics
>>> hardware, not if OpenVMS should have a GUI. A Gui could be a X-Windows
>>> interface, or a browser interface, and of course both can be
>>> supported. Graphics hardware is something very different.
>>> Let's talk about the "real world" for a bit, huh?
>>
>> It seems there are two camps here, the larger(?) touting VMS as a
>> "server operating system", having given up the idea that VMS can be used
>> for any graphics stuff, and the other wondering whether it could or
>> should compete with high-end graphics on other systems.
>>
>> However, I think the goal, at least for now, is to have the graphics
>> capabilities of, say, a $200 laptop, so that 99% of people (whether
>> previously familiar with VMS or not) can do 99% of what they need to do
>> on it.
>

>Why in earth would someone not "previously familiar with VMS"
>even get the thought of running VMS on a laptop?

Perhaps, those same people who tossed out WEENDOZE and installed a Linux
distribution.

>Most people needing a laptop get a laptop with whatever
>is preinstalled from the local IT-shop.

Yeah, one size fits all, eh? Just like prophylactics. Don't confuse the
freedom of choices with the favorability of choices!

I've only ever purchased one non-Apple laptop brand new because I did NOT
want to pay any Billy-tax. That laptop NEVER started up with/on WEENDOZE.
There's just too many hurdles to leap to make a WEENDOZE box even remotely
usable out of the box. Then, there's its virus collection affinity that's
another reason not to play.

>Over here in Sweden a few years ago some distributors tried to
>ship pre-configured Linux laptops side-by-side with the Windows
>laptops. It totaly went away a year later and today those
>laptops doesn't exist any longer. The market was non-existent.
>
>If I look at the largest Swedish online distributor and look
>what OS'es are available, they are like this (no of laptop
>offerings with each OS):
>
>Win 8/8.1: 514
>Win 7: 380
>Chrome: 23
>MacOS X: 29 (Apples, of course)
>Ubuntu: 1
>None: 4
>
>And the Linux/Ubuntu one is a high-end modell at aprox $1.800.
>
>So how can anyone even think that there would be a market for
>*VMS* based laptops?

I'd buy one. ;)

>Say that I can do 99% of what I need to do on my VMS laptop.
>But that last 1% happens to be that my internet banking appl
>doesn't exist for VMS. Now that will be enought for that
>solution beeing totaly nonworking for "99% of people".

So, what do you use to access your internet banking? WEENDOZE? Secure
banking... NOT!

>> To those who run a web browser on one platform and use it to access
>> stuff they use on VMS: you are wasting WAY too much time.
>

>Trying to do everything on VMS would be a total wast of time...

'Tis true. I once tried grilling a steak on VMS. It's a good think I'm
a fan of very rare. ;)


--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.

[Jan-Erik Soderholm]

You know, you aren't even funny. And certainly not worth
any serious comments, of course. With that kind of VMS
"friends", VMS doesn't need any enemies...

But a few comments...

The largest supplier of internet banking solutions in Sweden
stopped their support for Linux last year (or if it was
earlier). The user base was to small and the cost was
to high. Most banks supports Windows and Apple today.

>
> I'd buy one. ;)
>

Yes, that is definitely what is called a "market"... :-)

The rest was mostly childish rants.

[Phillip Helbig (undress to reply)]

In article <mbkp6r$p1n$1...@news.albasani.net>, Jan-Erik Soderholm
<jan-erik....@telia.com> writes:

> Why in earth would someone not "previously familiar with VMS"
> even get the thought of running VMS on a laptop?

Even some people younger than 20 occasionally get enough of modern
assembly-line pop and start listening to the Beatles. :-)

If ONLY people previously familiar with VMS use it in the future, then
its days are truly numbered. Why should someone not previously familiar
with VMS avoid it?

> Most people needing a laptop get a laptop with whatever
> is preinstalled from the local IT-shop.

True, but that doesn't mean that no-one would be interested in a VMS
laptop.

> So how can anyone even think that there would be a market for
> *VMS* based laptops?

I don't see the masses moving to VMS laptops. However, many people
previously familiar with VMS would be interested in a VMS laptop, so
there is a market. How big, how viable? I don't know. I know people
half my age who recently learned VMS and like it. Some would like to
have a VMS laptop.

> Say that I can do 99% of what I need to do on my VMS laptop.
> But that last 1% happens to be that my internet banking appl
> doesn't exist for VMS. Now that will be enought for that
> solution beeing totaly nonworking for "99% of people".

I have done essentially all my banking electronically for years. I
don't have any apps. The bank has a web server, I access it with a web
browser. Yes, even my old VMS web browser works fine.

> Trying to do everything on VMS would be a total wast of time...

Not everything, just all computing. :-)

[Dirk Munk]

I totally agree with you, I think this is the way to go. You can use any
type of x64 desktop or laptop system, no need for VSI to produce drivers
for all kind of motherboards etc. It will give us a replacement for the
former VAX and Alpah workstations, and believe me once you've worked
with one, you have no desire to return to just a VT interface.

I'm sure the VMS GUI needs a lot of work, but these days a GUI is
standard for developers. I'm not familiar with all the differences
between X-Windows and Motif, or similar interfaces, so I will just call
them X-Windows. This is the standard GUI interface for Unix, so it will
do just fine for VMS as well.

Running Windows and VMS at the same time on one platform (desktop or
laptop) is great. Given the fact that the demand for web based
applications is enormous, you can develop your web based application for
VMS on the same hardware as you can use to test the application with a
browser on Windows.

[Phillip Helbig (undress to reply)]

In article <00AF2AE6...@SendSpamHere.ORG>, VAXman-

@SendSpamHere.ORG writes:

> >Trying to do everything on VMS would be a total wast of time...

It's a shame that the newsgroup alt.sex.fetish.dec-hardware has become
inactive. :-)

> 'Tis true. I once tried grilling a steak on VMS. It's a good think I'm
> a fan of very rare. ;)

Actually, VMS is quite well done! :-)

[Phillip Helbig (undress to reply)]

In article <mbkrpr$u15$1...@news.albasani.net>, Jan-Erik Soderholm

<jan-erik....@telia.com> writes:

> The largest supplier of internet banking solutions in Sweden
> stopped their support for Linux last year (or if it was
> earlier). The user base was to small and the cost was
> to high. Most banks supports Windows and Apple today.

What does "support" mean? Fire up a web browser on the appropriate
platform and see if it works? Or do you have to actually install some
application to access your account?

I certainly want my account accessible to me from ANY web browser. This
has greater flexibility.

HTTPs itself is of course pretty secure. In addition, I get a TAN sent
to my mobile phone (a Nokia 3310, so no danger of it being hacked) which
I have to enter to confirm the transaction.

If a bank insisted I have to install some software to do my banking, and
only supported a couple of platforms, then I would change the bank.

[Jan-Erik Soderholm]

Phillip Helbig (undress to reply) skrev den 2015-02-13 14:02:
> In article <mbkrpr$u15$1...@news.albasani.net>, Jan-Erik Soderholm
> <jan-erik....@telia.com> writes:
>
>> The largest supplier of internet banking solutions in Sweden
>> stopped their support for Linux last year (or if it was
>> earlier). The user base was to small and the cost was
>> to high. Most banks supports Windows and Apple today.
>
> What does "support" mean?

A small box with a PIC card that connects over USB. (Can also
be used USB-less but with an extra manual chalenge/response
between the internat bank and the box. Like:

[Jan-Erik Soderholm]

Sent to fast...

Phillip Helbig (undress to reply) skrev den 2015-02-13 14:02:>> In
article <mbkrpr$u15$1...@news.albasani.net>, Jan-Erik Soderholm
> <jan-erik....@telia.com> writes:
>
>> The largest supplier of internet banking solutions in Sweden
>> stopped their support for Linux last year (or if it was
>> earlier). The user base was to small and the cost was
>> to high. Most banks supports Windows and Apple today.
>
> What does "support" mean?

A small box with a PIN card that connects over USB. (Can also
be used USB-less but with an extra manual challenge/response
between the internet bank and the box. Like:

http://secure.pensionsmyndigheten.se/images/18.70e56c27145a7fdc95824ca7/1387278996576/bankdosa.png

>
> Fire up a web browser on the appropriate
>> platform and see if it works? Or do you have to actually install some
>> application to access your account?

Not if using the box without the USB cable. Then it doesn't
matter what platform is used (apart from the web page itself).
To use the USB cable (easier and less keyboarding) you need
a small "security aplication".

There is also something called "Mobile BankID" that is recomended
for thos without a supported platform.

[Bob Koehler]

In article <mbihve$i3c$1...@dont-email.me>, David Froble <da...@tsoft-inc.com> writes:
>
> But .... what am I going to do with all these server type system I got
> laying around?

Put graphjics cards in them! :-)

> Really, I got no problem running EDT on a VT emulator on my Weendoze
> 2000 system ....

I do, but I do it anyhow if I have to.

But on my hobbyist cluster I have a real VT320 plugged into the serial
ports on my Alpha and my VAX, so I don't have to combine pain and
suffering with pleasure.

[Bob Koehler]

In article <f8bf12f4-6e95-468b...@googlegroups.com>, johnwa...@yahoo.co.uk writes:
>
> The talk of "Windows familiarity" is at odds with reality outside the IT
> world, and is likely irrelevant anyway when what the end user needs to see
> is a specific line of business application set rather than a Windows
> desktop and everything that goes with it.
>

It's hype, just like "open systems" was hype and "all UNIX are the
same" was hype.

Unfortunatley the software industry is driven by dogma and hype.

(Just saw another article disputing the "no goto" dogma.)

[David Froble]

VAXman- @SendSpamHere.ORG wrote:

> So, what do you use to access your internet banking? WEENDOZE? Secure
> banking... NOT!

<Sarcasm>

But, everyone else does it ....

</Sarcasm>

:-)

Actually, that is the exact words of one previous customer when I
objected to putting customer credit card and checking account data on an
IIS server without any encryption or other protection.

The whole "security" attitude is wrong. Instead of going after the
hackers, law enforcement should be going after those who sit their money
on the curb, and then complain when someone picks it up.

[Jan-Erik Soderholm]

Exactly. That is why I (and the banking business at large, at least
over here) use a tested and secure solution using bank card reader
with a chip-card. There is no known breaking in for these.

There are very few bank customers left (probably 2-3%) that doesn't
do their bank business over "the net" today.

[Phillip Helbig (undress to reply)]

In article <ZQqVH$RnJ...@eisner.encompasserve.org>,

koe...@eisner.nospam.decuserve.org (Bob Koehler) writes:

> It's hype, just like "open systems" was hype and "all UNIX are the
> same" was hype.

There are lies, damn lies, and open systems.

[Richard Maher]

On 2/13/2015 8:03 PM, Jan-Erik Soderholm wrote:
>
> Why in earth would someone not "previously familiar with VMS"
> even get the thought of running VMS on a laptop?
>

Because they like to wait 1/2 hr watching the console before it's ready
to do anything at boot-time.

Honestly Jan-Erik I wish these guys were trolls instead of seriously
asking for this VMS-killing crap :-( But don't feed them anyway eh?

[JF Mezei]

On 15-02-13 05:12, Phillip Helbig (undress to reply) wrote:

> However, I think the goal, at least for now, is to have the graphics
> capabilities of, say, a $200 laptop, so that 99% of people (whether
> previously familiar with VMS or not) can do 99% of what they need to do
> on it.

I wouldn't go as far as a $200 laptop, but would state that VMS should
be able to run on as wide a range of hardware as possible. If a number
of popular servers have graphic cards and support/expect a console, then
VMS should support that. (even if that console is via VMware).

Yes, eventually, leveraging graphics capabilities may be fun, but
initially it would really be to enable basic system management on
hardware that expects a GUI console of some basic kind.

[Dirk Munk]

Jan-Erik Soderholm wrote:
> Phillip Helbig (undress to reply) skrev den 2015-02-13 11:12:
>> Dirk Munk wrote:
>>
>>> In the threat about OpenVMS Future Directions there has been a lot of
>>> discussion about the question if OpenVMS should get graphics
>>> capabilities. With that I mean if OpenVMS should support graphics
>>> hardware, not if OpenVMS should have a GUI. A Gui could be a X-Windows
>>> interface, or a browser interface, and of course both can be
>>> supported. Graphics hardware is something very different.
>>> Let's talk about the "real world" for a bit, huh?
>>
>> It seems there are two camps here, the larger(?) touting VMS as a
>> "server operating system", having given up the idea that VMS can be used
>> for any graphics stuff, and the other wondering whether it could or
>> should compete with high-end graphics on other systems.
>>
>> However, I think the goal, at least for now, is to have the graphics
>> capabilities of, say, a $200 laptop, so that 99% of people (whether
>> previously familiar with VMS or not) can do 99% of what they need to do
>> on it.
>
> Why in earth would someone not "previously familiar with VMS"
> even get the thought of running VMS on a laptop?

Perhaps because when x64 VMS is out, it is a very logical idea that you
can run it on a x64 laptop? Because he wants to develop programs on VMS
and use his laptop for that purpose?

I can run VMS on my laptop, in fact I can run a cluster on my laptop. I
have an Alpha emulator that will let me emulate an ES40 class sytem.

[Phillip Helbig (undress to reply)]

In article <mblot7$vmt$1...@speranza.aioe.org>, Richard Maher

<maher_rj...@hotmail.com> writes:

> > Why in earth would someone not "previously familiar with VMS"
> > even get the thought of running VMS on a laptop?
>

> Because they like to wait 1/2 hr watching the console before it's ready
> to do anything at boot-time.

Confusing an early-1980s VAX with a more modern VMS machine? 1/2 an
hour to boot? Really?

> Honestly Jan-Erik I wish these guys were trolls instead of seriously
> asking for this VMS-killing crap :-( But don't feed them anyway eh?

Right. VMS is SO thriving right now that one has to be really, really
worried about a few newsgroup posts killing it off. :-)

[Dirk Munk]

Like I wrote before, you can support any graphics card by using the VESA
BIOS Extensions. It will not give you high performance, but you can use
any graphics card.

[Phillip Helbig (undress to reply)]

In article <54de6acc$0$16690$c3e8da3$76a7...@news.astraweb.com>, JF

Mezei <jfmezei...@vaxination.ca> writes:

> > However, I think the goal, at least for now, is to have the graphics
> > capabilities of, say, a $200 laptop, so that 99% of people (whether
> > previously familiar with VMS or not) can do 99% of what they need to do
> > on it.
>

> I wouldn't go as far as a $200 laptop, but would state that VMS should
> be able to run on as wide a range of hardware as possible. If a number
> of popular servers have graphic cards and support/expect a console, then
> VMS should support that. (even if that console is via VMware).
>
> Yes, eventually, leveraging graphics capabilities may be fun, but
> initially it would really be to enable basic system management on
> hardware that expects a GUI console of some basic kind.

I agree.

I spent a couple of hours today with Mozilla on an XP1000. It worked
fine. No, it's not good enough for YouPorn, but a slight improvement
might be good enough to enable it to do much useful stuff.

[Phillip Helbig (undress to reply)]

In article <651d0$54de6f32$5ed4324a$24...@news.ziggo.nl>, Dirk Munk

<mu...@home.nl> writes:

> > Why in earth would someone not "previously familiar with VMS"
> > even get the thought of running VMS on a laptop?
>

> Perhaps because when x64 VMS is out, it is a very logical idea that you
> can run it on a x64 laptop? Because he wants to develop programs on VMS
> and use his laptop for that purpose?

Indeed. What will be very, very, very bad publicity for VMS on x86
would be that if, when it is available, it can't do things which today
are considered basic requirements and are literally available in
bargain-basement PCs.

[John Reagan]

On Friday, February 13, 2015 at 5:20:42 PM UTC-5, Phillip Helbig (undress to reply) wrote:

> Indeed. What will be very, very, very bad publicity for VMS on x86
> would be that if, when it is available, it can't do things which today
> are considered basic requirements and are literally available in
> bargain-basement PCs.

So you make the same comment for NonStop on x86 that won't run on a laptop?

There is software development vs deployment.

I will point out however that NonStop does provide supported Windows-hosted cross-compilers for "laptop-based" development. OpenVMS does not.

[Phillip Helbig (undress to reply)]

In article <37e8b511-a180-4f72...@googlegroups.com>, John

Reagan <xyzz...@gmail.com> writes:

> So you make the same comment for NonStop on x86 that won't run on a laptop?

NonStop never was, and never claimed to be, a desktop-to-datacenter OS.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of JF
> Mezei
> Sent: 11-Feb-15 1:39 PM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] OpenVMS graphics - once more
>
> On 15-02-11 13:14, Simon Clubley wrote:
>
> > Storage devices are a core part of a server operating system; high
> > performance graphics cards are not.
>
> Is that truly the case anymore ? Wit various libraries that allow one
> to use the graphic card(s) as compute engines, is this something a
> server operating system can afford to ignore for much longer ?
>
> Sure, if you are running COBOL or BASIC, chances are your computations
> involve decimal monetary amounts and not floating points.
>
> But if you are wanting to create waves around the Titanic, then that
> extra CPU power comes in really handy (I mention that because those
> waves had been done on Alphas as I recall).
>
> Remember that VSI need to not only look at yesterday (what current
> customers have told them what they are doing) but also tomorrow.
>

For some environments like E-Sports (online gaming), there is a definite
trend to move most of the compute to the server as a means to reduce
all of the hacking on the client that is such a big scourge in the online
gaming world today. Most businesses would likely agree.

In fact, many of the games are moving away from intensive 3D displays
on the client because of the delays in lag time that these type of displays
cause. It also reduces their target market as only a small % of the gaming
base are able to upgrade their graphics card every year.

With internet connections exponentially increasing (my son just moved
Into new Apt in Irvine and has 750Mbps internet connection down AND up),
the future is going to be much more thin client / big server type envs.

Singapore announced plans to begin deployment of 2Gbps fibre to homes
in the next year.
http://news360.com/article/265876689#

Custs (Corp and personal) are tied of fat client backups, virus checking,
hacking, complex config mgmt etc.

Imho, in the future, OpenVMS needs to have:

- server features that make it the preferred back end platform from a
scalability, security, availability and rock solid perspective. With that
means getting big name ISV's back on platform. Also why I would love
to see Galaxy features come back as Windows/Linux are far away from
offering these and imho, would be a huge server differentiator on the
back end - especially with larger numbers of cores coming. Galaxy
provides capability to dynamically balance cores across different OS's
based on business rules, manually, or via loads.

- Server features that simplify issues that all IT shops have with things
like centralized configuration mgmt. Huge challenges today with most
commodity AND enterprise OS's today.

- thin client with browser that can be used as part of an overall secure
and Mgmt solutions. Browser does not need to be leading edge, but
simply competitive that supports industry web standards for things
like HTML5 and especially similar web security stds.

- hugely simplified X86-64 pricing model. Does not need to be cheaper
than Windows/Linux, but needs to be competitive and much more
simple from a Cust perspective. Perhaps host based model combined
with tier level like Workstation, Department, Enterprise? This alone
would raise eyebrows (in good way) as the per core/socket model
used by Linux/Windows/Enterprise OS's today is way too complex
and restrictive. Custs want simplicity and flexibility.

Regards,

Kerry Main
Back to the Future IT Inc.
.. Learning from the past to plan the future

Kerry dot main at backtothefutureit dot com

[David Froble]

Got to say that there is much in the above to consider.

Just recently there has been some guy claiming that we're creating a
"data dark age", because data is stored using current formats, which may
not be known in the future. Talking many years, not just a couple.
While DVD drives can stile read and write CDs, and BlueRay drives still
support DVD, and probably CD, this may not continue long term.

For example, the Magna Carta (spelling) and US Declaration of
Independance and such are on paper, which if it doesn't degrade are
readable. However, not so with with encoding on magnetic and optical media.

If there was places for people to store things, Ok, even Cloud, and such
were long term, then as storage was upgraded, data could be converted
and maintained.

Now, as for the current "cloud storage", it is for hire, and without any
care from the providers for the data. Ok, my opinion. What would be
needed would be storage that did care about the data, and perhaps not
dependent on individual customers who's data might disappear the day
they stopped paying.

Perhaps much more longetivity than people putting stuff on their PC and
losing it every time a disk dies or they get a new PC. Yes, I know of
people who do not carry data forward. People need to learn better
habits with their weendoze toys.

So yes, perpetual storage, sort of like Google, only much better, might
be desirable. Thin clients that are enough to access the "world data
store" might be best for many people. That's how many are now using
PCs, basically to run a browser and email. Many even use email services
with a browser, no local client.

Then there is the issue of sorting through all the "dross" ....

[Dirk Munk]

I've been aware of this problem for decades now. Not only do we need
devices to read the media, but also interfaces to connect the devices,
software (drivers etc.) to read the media, and software to read the
actual data (word processing software, database software and so on).

The Royal Library in the Netherlands has a special department that is
specialized in transferring data from all kind of data form floppy
disks, casettes, and whatever kind of media to modern media and data
formats. Ministries, municipalities, provinces, and other authorities
have these old media in their archives, and no computer systems to read
them. The documents on these media have to be present in the archives,
but quite often the paper versions got lost.

There used to be a law that laid down what kind of paper, pens etc. to
use for official documents, thus making sure that documents are still
readable in a few hundred years. It wouldn't surprise when in the near
future someone fetches a document from an archive, just to find out the
pages are almost blank because the inkjet ink has faded beyond readability.

Do you know the story about the library of congress? They had some old
tapeunits scrapped, because they didn't use them any more. Then they
found out they had racks full of old 7-track tapes they couldn't read,
because they had scrapped the tapeunits, so IBM had to build new ones.

Didn't the same thing happen with NASA too? Luckily they found an old
tape drive in a museum in Australia if I remember correctly.

In my view microfilm is a very good way to store documents. you only
need a magnifying glass to read a microfilm, and that is basic technology.

Combine it with digital technology (overlapping), and you have it all.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> David Froble
> Sent: 15-Feb-15 1:37 PM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] OpenVMS graphics - once more
>
> Kerry Main wrote:
> >> -----Original Message-----
> >> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> JF
> >> Mezei
> >> Sent: 11-Feb-15 1:39 PM
> >> To: info...@info-vax.com
> >> Subject: Re: [New Info-vax] OpenVMS graphics - once more
> >>

[snip..]

>
> Got to say that there is much in the above to consider.
>
> Just recently there has been some guy claiming that we're creating a
> "data dark age", because data is stored using current formats, which may
> not be known in the future. Talking many years, not just a couple.
> While DVD drives can stile read and write CDs, and BlueRay drives still
> support DVD, and probably CD, this may not continue long term.
>

Not just "some guy" ... pretty well respected guy.

http://www.bbc.com/news/science-environment-31450389

Vint Cerf, a "father of the internet", says he is worried that all the
images and documents we have been saving on computers will
eventually be lost.

Currently a Google vice-president, he believes this could occur as
hardware and software become obsolete....[see link]

[snip..]

[Craig A. Berry]

On 2/15/15 12:36 PM, David Froble wrote:

> Just recently there has been some guy claiming that we're creating a
> "data dark age", because data is stored using current formats, which may
> not be known in the future. Talking many years, not just a couple.
> While DVD drives can stile read and write CDs, and BlueRay drives still
> support DVD, and probably CD, this may not continue long term.

Those are media formats, not data formats. Both can be a problem for
long-term preservation.

> For example, the Magna Carta (spelling)

This you spelled correctly, though I believe all extant copies were
written on vellum (parchment made from calf skin), not paper.

> and US Declaration of Independance and such are on paper,

The word is "independence" and it was written on parchment, not paper.
Good paper can last a long time, too, though probably not as long as
parchment. Bad paper, such as what was used for most of the cheap
printing in the 19th century, has a high acid content and crumbles to
dust in a few decades.

> which if it doesn't degrade are readable. However, not so with with
> encoding on magnetic and optical media.

I think the jury is still out on optical media, and of course there are
and have been quite a few different optical media technologies to date.
None of them was designed for preservation that I'm aware of, but some
of them might well last a while. Preserving devices that can read the
media formats is of course an independent problem from the degradation
of the media itself.

Memristors may well solve the media degradation problem but not the
media format preservation (or continual re-conversion) problem.

> If there was places for people to store things, Ok, even Cloud, and such
> were long term, then as storage was upgraded, data could be converted
> and maintained.

> So yes, perpetual storage, sort of like Google, only much better, might
> be desirable.

Professional librarians have been wrestling with this ever since first
presented with digital materials to preserve. The technical challenges
are by no means solved, but they pale in comparison to the institutional
challenges. The only things likely to work for data preservation are
long-term institutional stability and a foundational commitment to
preservation, but these values are the antithesis of those prevalent in
the current internet economy. Google probably thinks it will last
forever, and maybe it will, but unless copies of its vast digital
holdings are held elsewhere (as some of the lawsuits against Google
Books have tried to ensure), a change in management or priorities could
trigger enormous losses.

It's no guarantee, but something like the Hathi Trust,[1] which was
specifically founded to solve this problem, is probably the best thing
currently going.

[1] http://www.hathitrust.org

[Phillip Helbig (undress to reply)]

In article <mbr2eu$1jp$1...@dont-email.me>, "Craig A. Berry"

<craig...@nospam.mac.com> writes:

> This you spelled correctly, though I believe all extant copies were
> written on vellum (parchment made from calf skin), not paper.

Certainly not ALL copies. I could write one right now, on paper. Oh,
you mean all ORIGINALS? But surely there is just one original. Or do
you mean all copies made at the time? :-)

[Craig A. Berry]

As with many charters and other official documents, there was more than
one "original" copy, or to be pedantically correct, "exemplified" copy.

[JF Mezei]

Digital data has the advantage of being loss-less copiable from one
media format to another as technology progresses and older media no
longer in vogue.

This is not the case for paper copies or analogue film, both of which
not only degrade over time, but a copy is never as good as the original.

Apple has traditionally led the way with media evolution. Made the 5"
floppies redundant with the newer 3.5" "hard" floppies. Gave up the
diskettes in favour of CDs and short period of ZIP drives and then DVDs
and now gave up optical media in favour of digital downloads and USB sticks.

I recently fired up my old G3 iMac which has a ZIP drive to scann
thorugh the hadful of ZIp disks I had in case there was something of
value to preserve.

I keep my Vaxstation 3100 as my last device capable of reading 3.5"
floppies.


My guess is that at this time, SSD drives with USB attachement are the
way to archive off-line.

I am in a multi year process of scanning slides. They are being stored
as .TIFF files with LZW compression. This appears to be the safest long
term format. I had tried the Kodak PhotoCD in the 1990s, and to rescue
those, I had to fire up an old version of Photoshop running on MacOS 8.6
(emulated with sheepshaver) to rescue the photos because current
Photoshop no longer supports this.

Similarly, I had some Freehand documents and Adobe Illustrator no longer
has the ability to read them, so I have to fire up the emulator and use
Freehand to convert those files to .eps.

One need to have good awareness of what formats one has, and when
applications lose the ability to read such formats. One doesn't delete
an app until you are sure the new one has the ability to read all thsoe
formats.

[Bill Gunshannon]

In article <mbqokh$ovu$1...@dont-email.me>,

> Got to say that there is much in the above to consider.
>
> Just recently there has been some guy claiming that we're creating a
> "data dark age", because data is stored using current formats, which may
> not be known in the future. Talking many years, not just a couple.
> While DVD drives can stile read and write CDs, and BlueRay drives still
> support DVD, and probably CD, this may not continue long term.

That "guy" was Vint Cerf. And hopefully people will actuall listen to
him. Cause those of us who have been saying this for at least 20 years
have been pretty much ignored.

>
> For example, the Magna Carta (spelling) and US Declaration of
> Independance and such are on paper, which if it doesn't degrade are
> readable. However, not so with with encoding on magnetic and optical media.

And thus the reason many of us still have the capability to read 3.5",
5.25" and yes, even 8" disks in myriad formats. Not to mention that
I even still have a 9-track tape drive. But finding people who have
not let their media deteriorate too far is definitely getting rarer.

>
> If there was places for people to store things, Ok, even Cloud, and such
> were long term, then as storage was upgraded, data could be converted
> and maintained.

Long term? I wonder.

>
> Now, as for the current "cloud storage", it is for hire, and without any
> care from the providers for the data. Ok, my opinion. What would be
> needed would be storage that did care about the data, and perhaps not
> dependent on individual customers who's data might disappear the day
> they stopped paying.

Or when the "cloud" company goes belly-up. or did you miss that article
in the news?

>
> Perhaps much more longetivity than people putting stuff on their PC and
> losing it every time a disk dies or they get a new PC. Yes, I know of
> people who do not carry data forward. People need to learn better
> habits with their weendoze toys.
>
> So yes, perpetual storage, sort of like Google,

Google? Perpetual? Hahahahahahaha........
Do people even realize how much has been lost? Do I have to trot out
my favorite list again? Starting with The Software Tools Virtual
Operating System? How about The Graphics Compatability System?
WATBOL? The Firmware sourcecode for the DR-100 and DR-200 digital
repeater systems. And the list goes on and on and on.

> only much better, might
> be desirable. Thin clients that are enough to access the "world data
> store" might be best for many people. That's how many are now using
> PCs, basically to run a browser and email. Many even use email services
> with a browser, no local client.
>
> Then there is the issue of sorting through all the "dross" ....

bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

[Bill Gunshannon]

In article <mbr2eu$1jp$1...@dont-email.me>,
"Craig A. Berry" <craig...@nospam.mac.com> writes:

> On 2/15/15 12:36 PM, David Froble wrote:
>
>> Just recently there has been some guy claiming that we're creating a
>> "data dark age", because data is stored using current formats, which may
>> not be known in the future. Talking many years, not just a couple.
>> While DVD drives can stile read and write CDs, and BlueRay drives still
>> support DVD, and probably CD, this may not continue long term.
>
> Those are media formats, not data formats. Both can be a problem for
> long-term preservation.
>
>> For example, the Magna Carta (spelling)
>
> This you spelled correctly, though I believe all extant copies were
> written on vellum (parchment made from calf skin), not paper.
>
>> and US Declaration of Independance and such are on paper,
>
> The word is "independence" and it was written on parchment, not paper.
> Good paper can last a long time, too, though probably not as long as
> parchment. Bad paper, such as what was used for most of the cheap
> printing in the 19th century, has a high acid content and crumbles to
> dust in a few decades.

Think celuloid. Anybody here old enough to remember The Jackie Gleason
Show? (No, not The Honeymooners!) Care to guess why you can't buy copies
of all those old shows when so much nostalgia TV is available?

[William Pechter]

In article <4c2d9f33-09a1-4dcf...@googlegroups.com>,
pcov...@gmail.com <pcov...@gmail.com> wrote:
>On Wednesday, February 11, 2015 at 9:41:44 AM UTC-5, MG wrote:
>> Phillip Helbig (undress to reply) schreef op 11-feb-2015 om 13:26:
>> > There was a time, though, when VMS workstations were marketed for
>> > CAD/CAM.
>>
>> I keep hearing that, from time to time, but where is the proof?
>> Nobody could ever cite me specific software packages. Even old
>> product brochures don't mention too terribly much... (The only
>> traces of software somewhat belonging into this category, that I
>> could ever find documented traces of, was for ULTRIX.)
>>
>> - MG
>
>Unigraphics was the CAD/CAM package available. and is now called NX and
>still one of the best cad packages there is.
>
>PC

I saw it on an 11/780 or was it an 11/785 back in '85 or so at Eatontown
NJ doing CAD/CAM -- customer was Bendix Aerospace IIRC.

Slick 3D CAD part design work... They supported a number of some type of
smart terminal off the one VAX.

Bill

Bill
--
--
Digital had it then. Don't you wish you could buy it now!
pechter-at-gmail.com http://xkcd.com/705/

[William Pechter]

In article <mbkimn$tqq$1...@news.kjsl.com>,
Phillip Helbig (undress to reply) <hel...@asclothestro.multivax.de> wrote:

>Maybe at my funeral I'll get someone to say "ashes to ashes, desktop to
>datacenter". :-)


Damn. Saving this for my kid to use at my eulogy.

[William Pechter]

In article <mbltbo$19ia$4...@news.kjsl.com>,

Phillip Helbig (undress to reply) <hel...@asclothestro.multivax.de> wrote:

Remember: These laptops and VM's (if both are available) will be tools
for SALES DEMOS...and for Trade Shows...

There's something amazing about being able to see the product demonstrated
live with no GoTo Meeting, MyPC or other software and internet connection being
required.

[William Pechter]

In article <00AF2AE6...@SendSpamHere.ORG>,
<VAXman- @SendSpamHere.ORG> wrote:
>In article <mbkp6r$p1n$1...@news.albasani.net>, Jan-Erik Soderholm

><jan-erik....@telia.com> writes:
>>Phillip Helbig (undress to reply) skrev den 2015-02-13 11:12:
>>Trying to do everything on VMS would be a total wast of time...
>

>'Tis true. I once tried grilling a steak on VMS. It's a good think I'm
>a fan of very rare. ;)
>--
>VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
>
>I speak to machines with the voice of humanity.

You need an 8650. Just pull the board and sear on each side.
I wish I had the 8650 they put in at Fort Monmouth as a souvenir.

[William Pechter]

In article <ck3p70...@mid.individual.net>,
Bill Gunshannon <bill...@cs.uofs.edu> wrote:
>In article <c3243$54dbfbfd$5ed4324a$31...@news.ziggo.nl>,
> Dirk Munk <mu...@home.nl> writes:
>> John E. Malmberg wrote:
>>> On 2/11/2015 8:52 AM, David Froble wrote:
>>>>
>>>> So, back to VMS. In my opinion, (we're all entitled to one), if VSI is
>>>> going to be serious about a VMS GUI for the masses, continuing with
>>>> x-windows isn't going to cut the mustard. They would need to come up
>>>> with a new GUI that had the "look and feel" of MS weendoze. Not saying
>>>> that would be enough, but anything less would just be a waste of time
>>>> and money.
>>>
>>> X-11 look and feel is not the problem. It should not be hard to port
>>> any of the various Linux GUI Shells to VMS. If the GTK+ 2.x+ port ever
>>> gets done, you might be surprised at what will simply compile and work
>>> with out any source changes.
>>>
>>> VMS graphics are based on Motif and CDE, neither of which are common in
>>> the Linux/Windows/OS-x and and now Android.
>>>
>>> X-11 is a problem though for mobile and portable devices. VNC can
>>> survive, but there can be a lot of latency in updating. RDP attempts to
>>> deal with that, but is not a public standard protocol AFAIK.
>>>
>>>
>>> As others have pointed out, medium to high end video drivers are an issue.
>>>
>>> Modern video cards have effectively two APIs.
>>>
>>> One is a public one that has the video card run in the lower performance
>>> modes. Essentially, your $50+ video card is emulating a less than $20
>>> video card.
>>
>> I suppose you are referring to the VESA Bios Extensions?
>>
>>> Open source drivers are available for these modes. In some
>>> cases some of the high performance modes have been figured out or
>>> released, but not the highest performance modes.
>>>
>>> The other is private one that needs a binary driver from from the video
>>> card vendor, or the chip set vendor. This is for the high performance
>>> modes.
>>>
>>> When you load a linux distribution that only contains GPL components,
>>> many of them will provide an installer that will go to the chip set
>>> vendor to download the closed source drivers when they detect the
>>> advanced video chip-sets.
>>>
>>> Other Linux distributions will just include the closed source drivers as
>>> they are not so strict about what they include, so you do not realize
>>> that the source is not available.
>>>
>>> The specifications for writing the high performance drivers for many
>>> video chip sets are secret, and you can not even get them with an NDA
>>> agreement.
>>>
>>> If you are an x86 based OS, you probably want to find a way to use
>>> unmodified linux or Windows binary drivers, especially for graphics
>>> devices.
>>
>> My laptop has two graphics devices, the Intel HD4000 GPU embedded in the
>> CPU, and a Nvidia card. I the laptop doesn't need high performance
>> graphics, the Intel GPU is used. If it does need high performance, it
>> can switch to the Nvidia card (it never does).
>>
>> The Intel drivers don't receive many updates, the Nvidia drivers almost
>> every month so it seems.
>>
>> I suppose VSI could get a lot of information on the Intel GPU, and as I
>> wrote before, the Intel graphics are more than sufficient for typical
>> VMS work.
>
>I think this pretty much sums it up. It is not the low level graphics
>capabilities that are the problem. It is the VMS X implementation.
>Given something more modern than the DECWindows we have I could do (and
>actually usually do) most desktop tasks on VMS. No one is going to use
>VMS to play World of Warcraft (although I might try Minecraft!! :-).
>But things like Open Office and better IDE's can all be don on VMS. All
>it would take is a version of X-11 modern enough to let these Open Source
>programs be built on VMS. I know it is a major undertaking, but all the
>sources to current X-11 distributions (there are 2 but one has one out in
>the marketplace) are available. And that includes X-servers for numerous
>video cards.

>
>bill
>
>--
>Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
>bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
>University of Scranton |
>Scranton, Pennsylvania | #include <std.disclaimer.h>

There's even another possibility. If there was an available Linux emulation
layer under VMS it's possible you could even run MS-Office under it.

Since you will have x86-64 architecture you don't have to emulate the
CPU or instruction set fully. Porting a Virtual Machine Layer under
VMS might be interesting as well.

Crossover Office https://www.codeweavers.com/ will run Office 2010
under Linux with X11. It also will do Quicken 2015 as well.

Current version is 14.0.3

[William Pechter]

In article <54dbeb3e$0$29561$c3e8da3$e074...@news.astraweb.com>,
JF Mezei <jfmezei...@vaxination.ca> wrote:
>On 15-02-11 18:43, Dirk Munk wrote:
>
>> It was all about the discussion that a server doesn't need to have a
>> graphics card these days. It can be replaced by an ILO port with
>> integrated web server.
>
>ILO port does not manage the OS instance, it manages the hardware. And
>depending on system, the equivalent of SRM.
>
>At best, you can use the ILO to order a "soft" power-off (where it sends
>a signal to OS which has a couple of seconds before power is shut).
>
>
>This is quite different from logging into the server to manage it, its
>parameners, networking (IP address etc), user authentication etc.
>
>And in cases where networking is out, you really need a way to login to
>the server's console to fix it. Can't be done remotely.
>
>

The IBM x86-64 boxes I worked with had a web based app that would allow you to
login to the OS on the server console and do anything including reconfig bios.
The low end minimum IPMI manages the hardware through the BMC controller, but
a lot of server vendors have the more enhanced features on an optional add-in
board to reduce costs on boxes that don't need the functionality.

DELL had the DRAC board. IBM had their RSA (Remote Supervisor Adapter).
HP had ILO varients (IIRC that was a Compaq name). Sun had ILOM.

Some run java or other embedded software giving full console emulation.
Some add CD/DVD/Floppy emulation attaching the remote box storage over LAN.
Some have on-board storage for media upload that the board makes available.

These technologies on PC servers are about 15 years old IIRC.

These boards either have their own ethernet interface (which can be on a
non-server private switch/vlan) or they share one port with the host.

They can (should) be on different networks than the production interfaces.

The emulation lets you do full bring up from dead remotely if the hardware
is functioning. They also monitor voltages, ac line input and power supply
status, fan speed/temp.

[William Pechter]

In article <mbihve$i3c$1...@dont-email.me>,
David Froble <da...@tsoft-inc.com> wrote:
>Bob Koehler wrote:
>> In article <3d151$54dbe922$5ed4324a$56...@news.ziggo.nl>, Dirk Munk
><mu...@home.nl> writes:
>>> Yes you can, but can you compile and link VMS programs on your laptop?
>>> The issue was if there can be demand for VMS on a laptop.
>>
>> I don't know about his laptop, but yes you can. I've compiled,
>> linked, and run VMS programs on my laptop via SIMH emulation.
>> And Tadpole made an Alpha laptop which ran VMS native.
>>
>> But I do think there is a real call for VMS on the desktop to
>> attract developers. A lot of developers will be put off by the
>> notion that they must log into a server to do their work, even if
>> you sell a really cheap server.
>>
>
>But .... what am I going to do with all these server type system I got
>laying around?
>
>:-)
>
>Really, I got no problem running EDT on a VT emulator on my Weendoze
>2000 system ....


OK -- what's a good VT emulator. Most of'em I've used suck.

[David Froble]

SmarTerm

Don't know if it's still marketed. I got a real old copy. Works great.

Well, except for the F14 or F12, whatever it is, for switching between
overstrike and insert. Stuck with ^A for that.

[David Froble]

There is an evaluation copy available at:

http://www.esker.com/host-access/smarterm.asp

[Simon Clubley]

On 2015-03-05, William Pechter <pec...@S20.pechter.dyndns.org> wrote:
>
> OK -- what's a good VT emulator. Most of'em I've used suck.
>

Currently logged into Eisner using a xterm on Linux with a custom
keyboard mapping script I picked up from comp.os.vms about a decade ago.
EVE works just fine in this setup (I'm typing this response using it)
although the keyboard is always in application keypad mode with this
script.

I also use PuTTY on Linux which works ok apart from a really annoying
issue where PF1 through PF4 doesn't work in the keyboard mapping I
selected (the PC keyboard friendly layout instead of the strict DEC
keyboard mode) unless the keypad is in application keypad mode.

On Windows PCs, I've installed Teraterm which works ok apart from not
been able to get it to do double height/double width characters.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world

[Bill Gunshannon]

In article <mdacm9$r0s$1...@pechter.eternal-september.org>,

pec...@S20.pechter.dyndns.org (William Pechter) writes:
> In article <4c2d9f33-09a1-4dcf...@googlegroups.com>,
> pcov...@gmail.com <pcov...@gmail.com> wrote:
>>On Wednesday, February 11, 2015 at 9:41:44 AM UTC-5, MG wrote:
>>> Phillip Helbig (undress to reply) schreef op 11-feb-2015 om 13:26:
>>> > There was a time, though, when VMS workstations were marketed for
>>> > CAD/CAM.
>>>
>>> I keep hearing that, from time to time, but where is the proof?
>>> Nobody could ever cite me specific software packages. Even old
>>> product brochures don't mention too terribly much... (The only
>>> traces of software somewhat belonging into this category, that I
>>> could ever find documented traces of, was for ULTRIX.)
>>>
>>> - MG
>>
>>Unigraphics was the CAD/CAM package available. and is now called NX and
>>still one of the best cad packages there is.
>>
>>PC
>
> I saw it on an 11/780 or was it an 11/785 back in '85 or so at Eatontown
> NJ doing CAD/CAM -- customer was Bendix Aerospace IIRC.
>
> Slick 3D CAD part design work... They supported a number of some type of
> smart terminal off the one VAX.
>

The VAX at West Point in the (very) early 1980's had no problem
supporting multiple Textronix CAD terminals with digitizers when
I was there. Don't remember what they ran for software but the
system was in The Geography and Computer Science Department so
it was likely mapping systems.

[Bill Gunshannon]

In article <mdae47$4jp$1...@pechter.eternal-september.org>,

Maybe it's just me, :-) but the link above gets nothing and if you use
just HTTP: you get a black page with three useless, unanswered questions.

[Bill Gunshannon]

In article <mdae47$4jp$1...@pechter.eternal-september.org>,
pec...@S20.pechter.dyndns.org (William Pechter) writes:

Oh wait. I missed the "s". Makes a big difference. I wonder who owns
the other page? :-)

[Bill Gunshannon]

In article <mdaeus$7li$2...@pechter.eternal-september.org>,

Putty? MSKermit?

Those are my two favorites with MSKermit making the better console
terminal for a VAX or PDP-11 (I have even used it to load PDP bootstraps!)

[Hunter Goatley]

On 3/6/2015 9:04 AM, Bill Gunshannon wrote:
> In article <mdaeus$7li$2...@pechter.eternal-september.org>,
> pec...@S20.pechter.dyndns.org (William Pechter) writes:
>
>> OK -- what's a good VT emulator. Most of'em I've used suck.
>>
>
> Putty? MSKermit?
>
> Those are my two favorites with MSKermit making the better console
> terminal for a VAX or PDP-11 (I have even used it to load PDP bootstraps!)
>

For the past 15 years or so, my favorite has been TeraTerm Pro (which is
also my favorite SSH client for Windows).

http://ttssh2.sourceforge.jp/index.html.en

--
Hunter
------
Hunter Goatley, Process Software, http://www.process.com/
goath...@goatley.com http://hunter.goatley.com/