VSI licencing policy (again), was: Re: VSI has a new CEO

[Simon Clubley]

On 2021-08-03, Dave Froble <da...@tsoft-inc.com> wrote:
> On 8/3/2021 1:41 PM, Simon Clubley wrote:
>>
>> Unfortunately, VSI do not seem to show any interest in addressing this.
>>
>> I wonder how much business it has cost them and how much it's going to
>> cost them simply because a customer cannot allow this situation to occur.
>>
>> A business may love VMS and want to stay with it, but they are not
>> going to allow the collapse of a vendor to be the collapse of their
>> own business, even if that means moving away from VMS.
>
> On this we agree 100%.
>

Even when people disagree with me on the other things I say, everyone
still appears to agree with me on this.

Why can't VSI see just how strongly customers feel about this and why
are they not doing anything to address this showstopping problem for
keeping many people as VSI customers ?

Do VSI simply not understand the sheer strength of customer feeling
about this ?

If you have your application source code, there are now solutions for
porting your VMS applications, including its VMS-specific code, over
to another operating system such as Linux.

How many people are now exploring this option as a direct result of
VSI's new time-limited licences policy ?

Simon.

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

[Phillip Helbig (undress to reply)]

In article <see0k9$km$1...@dont-email.me>, Simon Clubley

<clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> >> Unfortunately, VSI do not seem to show any interest in addressing this.
> >>
> >> I wonder how much business it has cost them and how much it's going to
> >> cost them simply because a customer cannot allow this situation to occur.
> >>
> >> A business may love VMS and want to stay with it, but they are not
> >> going to allow the collapse of a vendor to be the collapse of their
> >> own business, even if that means moving away from VMS.
> >
> > On this we agree 100%.
>
> Even when people disagree with me on the other things I say, everyone
> still appears to agree with me on this.

Right. If Simon and I both agree, then it must be right! :-)

> Why can't VSI see just how strongly customers feel about this and why
> are they not doing anything to address this showstopping problem for
> keeping many people as VSI customers ?
>
> Do VSI simply not understand the sheer strength of customer feeling
> about this ?
>
> If you have your application source code, there are now solutions for
> porting your VMS applications, including its VMS-specific code, over
> to another operating system such as Linux.
>
> How many people are now exploring this option as a direct result of
> VSI's new time-limited licences policy ?

Maybe people should write to VSI, but also publicly state, that they
would stick with VMS if they could get a non-expiring license (even if
it costs more, as long as the additional cost is not absurd), but if not
then they will move off of VMS and never become a VSI customer or, if
they are, will stop being one as soon as they can port stuff.

[Arne Vajhøj]

On 8/4/2021 8:19 AM, Simon Clubley wrote:
> On 2021-08-03, Dave Froble <da...@tsoft-inc.com> wrote:
>> On 8/3/2021 1:41 PM, Simon Clubley wrote:
>>> Unfortunately, VSI do not seem to show any interest in addressing this.
>>>
>>> I wonder how much business it has cost them and how much it's going to
>>> cost them simply because a customer cannot allow this situation to occur.
>>>
>>> A business may love VMS and want to stay with it, but they are not
>>> going to allow the collapse of a vendor to be the collapse of their
>>> own business, even if that means moving away from VMS.
>>
>> On this we agree 100%.
>
> Even when people disagree with me on the other things I say, everyone
> still appears to agree with me on this.
>
> Why can't VSI see just how strongly customers feel about this and why
> are they not doing anything to address this showstopping problem for
> keeping many people as VSI customers ?
>
> Do VSI simply not understand the sheer strength of customer feeling
> about this ?

Good questions.

Maybe too many random people complain in comp.os.vms and too few
paying customers complain directly?

Maybe VSI has not seen a financially attractive alternative?

Maybe VSI is working on something that is just mot ready yet.

Guesswork.

> If you have your application source code, there are now solutions for
> porting your VMS applications, including its VMS-specific code, over
> to another operating system such as Linux.

Such solutions has existed for decades.

> How many people are now exploring this option as a direct result of
> VSI's new time-limited licences policy ?

Probably very few. The remaining VMS world just don't jump so fast.

But if not eventually handled then it will yet another problem for VMS.

And VMS need fewer problems not more problems.

Arne

[Phillip Helbig (undress to reply)]

In article <see2c7$5q6$1...@gioia.aioe.org>, =?UTF-8?Q?Arne_Vajh=c3=b8j?=

<ar...@vajhoej.dk> writes:

> > Do VSI simply not understand the sheer strength of customer feeling
> > about this ?
>
> Good questions.
>
> Maybe too many random people complain in comp.os.vms and too few
> paying customers complain directly?

I resemble that remark! :-)

Seriously, yes, I agree; see my previous post on the topic.

> But if not eventually handled then it will yet another problem for VMS.
>
> And VMS need fewer problems not more problems.

So Arne agrees as well. We are on to the truth!

[John Dallman]

In article <see0k9$km$1...@dont-email.me>,

clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) wrote:

> Why can't VSI see just how strongly customers feel about this and
> why are they not doing anything to address this showstopping problem
for
> keeping many people as VSI customers ?

Has anyone raised this in one of their webinars? That seems like a way to
talk to them directly.

John

[Jan-Erik Söderholm]

Den 2021-08-04 kl. 14:19, skrev Simon Clubley:
> On 2021-08-03, Dave Froble <da...@tsoft-inc.com> wrote:
>> On 8/3/2021 1:41 PM, Simon Clubley wrote:
>>>
>>> Unfortunately, VSI do not seem to show any interest in addressing this.
>>>
>>> I wonder how much business it has cost them and how much it's going to
>>> cost them simply because a customer cannot allow this situation to occur.
>>>
>>> A business may love VMS and want to stay with it, but they are not
>>> going to allow the collapse of a vendor to be the collapse of their
>>> own business, even if that means moving away from VMS.
>>
>> On this we agree 100%.
>>
>
> Even when people disagree with me on the other things I say, everyone
> still appears to agree with me on this.

Please define "everyone".

>
> Why can't VSI see just how strongly customers...

"Customers"? How do you know what (real) customers think?
Maybe you are mixing up c.o.v with the real customers?

> feel about this and why are they not doing anything...

How do you know that they (the real customers) are doing nothing?
Do you really think that everything happening between VSI and its
customers would be copied to c.o.v? Don't think so...

> to address this showstopping problem for
> keeping many people as VSI customers ?
>
> Do VSI simply not understand the sheer strength of customer feeling
> about this ?

You obviously do not know much about how VSI and customres are
communicating. And it is *not* through c.o.v!

When was the last time you had any direct communication with VSI?

>
> If you have your application source code, there are now solutions for
> porting your VMS applications, including its VMS-specific code, over
> to another operating system such as Linux.
>
> How many people are now exploring this option as a direct result of
> VSI's new time-limited licences policy?

If you are one of those customers, do talk to VSI, just as all
other customers are doing that have some thoughts around this.

>
> Simon.
>

[John Vottero]

On Wednesday, August 4, 2021 at 8:19:24 AM UTC-4, Simon Clubley wrote:
> On 2021-08-03, Dave Froble <da...@tsoft-inc.com> wrote:
> > On 8/3/2021 1:41 PM, Simon Clubley wrote:
> >>
> >> Unfortunately, VSI do not seem to show any interest in addressing this.
> >>
> >> I wonder how much business it has cost them and how much it's going to
> >> cost them simply because a customer cannot allow this situation to occur.
> >>
> >> A business may love VMS and want to stay with it, but they are not
> >> going to allow the collapse of a vendor to be the collapse of their
> >> own business, even if that means moving away from VMS.
> >
> > On this we agree 100%.
> >
>
> Even when people disagree with me on the other things I say, everyone
> still appears to agree with me on this.
>

Not everyone.

VSI is smart enough to know that they cannot expect much revenue from new sales of VMS.

VSI must charge a yearly fee to ensure a revenue stream. If they didn't, many customers would get to a stable x86 version and then stop paying. VSI also knows that if they try to do a yearly fee with just language in the license, they will spend more money trying to collect the fees than the fees themselves. What happens when the VMS administrator leaves a company? Have you ever tried calling a billion dollar company and asking who the new VMS admin is? When the license key drops dead, the new VMS admin will call VSI.

But, IT DOESN'T MATTER because VSI will never just drop dead. If VSI decides they just can't make it work, they will lay off everyone except the one person that sends out invoices and generates new license keys. That would be a multi-million dollar business that only has one employee.

[Arne Vajhøj]

On 8/4/2021 9:58 AM, John Vottero wrote:
> On Wednesday, August 4, 2021 at 8:19:24 AM UTC-4, Simon Clubley
> wrote:
>> On 2021-08-03, Dave Froble <da...@tsoft-inc.com> wrote:
>>> On 8/3/2021 1:41 PM, Simon Clubley wrote:
>>>> A business may love VMS and want to stay with it, but they are
>>>> not going to allow the collapse of a vendor to be the collapse
>>>> of their own business, even if that means moving away from
>>>> VMS.
>>>
>>> On this we agree 100%.
>>
>> Even when people disagree with me on the other things I say,
>> everyone still appears to agree with me on this.
>>
>
> Not everyone.

I think practically everyone.

> VSI is smart enough to know that they cannot expect much revenue from
> new sales of VMS.
>
> VSI must charge a yearly fee to ensure a revenue stream. If they
> didn't, many customers would get to a stable x86 version and then
> stop paying.

Yes. That is probably VSI's problem with the forever license.

But that does not make the customers problem go away.

It just means that asking for a forever license is unlikely to
happen as it would not solve VSI's problem.

The key is to find a solution that works for both VSI and
the customers.

I gave my suggestion a few months back: an option for annual
extension of a 5 year license. That would give VSI annual
revenue. It would give customer that need assurance 4-5
years to migrate if VSI went under.

> But, IT DOESN'T MATTER because VSI will never just drop dead. If VSI
> decides they just can't make it work, they will lay off everyone
> except the one person that sends out invoices and generates new
> license keys. That would be a multi-million dollar business that only
> has one employee.

Or someone will takeover the remains and do that model.

The problem is if they do not switch to that model in time
and has to file for bankruptcy and the court appoints a lawyer
to run the business and that lawyer will need to spend time
figuring things out - including the contract with HPE.

The basic law of capitalism/greed says that it will be sorted
out - if money is to be made by issuing licenses then licenses
will eventually be issued.

Problem is that such legal matters can take a long time to sort out.

And customers with only a month left of their license could
be having a big problem.

Arne

[Phillip Helbig (undress to reply)]

In article <see4nb$tqq$1...@dont-email.me>,
=?UTF-8?Q?Jan-Erik_S=c3=b6derholm?= <jan-erik....@telia.com>
writes:

> > Even when people disagree with me on the other things I say, everyone
> > still appears to agree with me on this.
>
> Please define "everyone".
>
> > Why can't VSI see just how strongly customers...
>
> "Customers"? How do you know what (real) customers think?
> Maybe you are mixing up c.o.v with the real customers?

There is at least some overlap.

> > feel about this and why are they not doing anything...
>
> How do you know that they (the real customers) are doing nothing?
> Do you really think that everything happening between VSI and its
> customers would be copied to c.o.v? Don't think so...
>
> > to address this showstopping problem for
> > keeping many people as VSI customers ?
> >
> > Do VSI simply not understand the sheer strength of customer feeling
> > about this ?
>
> You obviously do not know much about how VSI and customres are
> communicating. And it is *not* through c.o.v!
>
> When was the last time you had any direct communication with VSI?
>
> >
> > If you have your application source code, there are now solutions for
> > porting your VMS applications, including its VMS-specific code, over
> > to another operating system such as Linux.
> >
> > How many people are now exploring this option as a direct result of
> > VSI's new time-limited licences policy?
>
> If you are one of those customers, do talk to VSI, just as all
> other customers are doing that have some thoughts around this.

I don't think that current customers of VSI are the problem. They
obviously already have a relationship with VSI and have already made
some sort of commitment. The problem is more with those who are waiting
to see what happens before making a commitment to VSI. VMS folks are a
conservative lot; if they were willing to change things on a whim, they
would no longer be on VMS. There are also potential new customers, but
that is probably not an issue for them. What fraction of VMS customers
are now VSI customers? Why are the others not already VSI customers?

[calliet gérard]

Le 04/08/2021 à 15:29, Jan-Erik Söderholm a écrit :
>
> "Customers"? How do you know what (real) customers think?
> Maybe you are mixing up c.o.v with the real customers?

There are been a meeting in France last June. The last point was about
licensing.

There is a report
here:https://www.vmsgenerations.fr/wp-content/uploads/2021/07/compte-rendu-24-juin-2021.pdf
(in french, I hope they will do a translation sometime).

I give an (automatic made) translation of the part of the report that
interests us here:

"""""NEW LICENSING METHOD

The next topic was the discussion with VSI about the marketing of VMS
and the new licensing mode.

After the March 2021 meeting we started a dialogue with VSI to express
the users' reactions on the reservations and blockages introduced by the
by the new VMS licensing mode since May 2020.

This subscription-based licensing mode combining the right to use and
support is based on a time-limited PAK that must be renewed with VSI
otherwise the system will stop working.

The new risks introduced by this model (the need for VSI to be still in
business to renew the subscription, and lack of visibility on the prices
that will be charged at renewal) are considered unacceptable to a
majority of users. The takeover of the future of applications by the
publisher is a profound imbalance to the sole benefit of
VSI.

VSI finally informed us that for certain critical market segments
an exception would be possible, with licenses not limited in time.

But we learned that this was accompanied by a contract that obliges
the customer to renew his subscription with VSI and otherwise shut down
his systems. This "exception" does not change the imbalance introduced
by VSI and the takeover of the applications.

Our action having been visible on comp.os.vms, several contributions
have proposed other licensing modes that would be more acceptable and
would encourage investment in VMS. We have consolidated these proposals
in a message to VSI asking them if these options had been evaluated and
to tell us why they had not been selected.

Indeed, many roundtable participants confirmed that
the current licensing model is widely rejected, to the point of
triggering plans to leave VMS. All those who expressed themselves
pointed out the difficulty in the current conditions introduced by VSI
to get a message of sustainability across to the management in order to
obtain the investments."""""

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

[Phillip Helbig (undress to reply)]

In article <imvsv9...@mid.individual.net>,

calliet_gérard?= <gerard....@pia-sofer.fr> writes:

> VSI finally informed us that for certain critical market segments
> an exception would be possible, with licenses not limited in time.

That is good.

> But we learned that this was accompanied by a contract that obliges
> the customer to renew his subscription with VSI and otherwise shut down
> his systems. This "exception" does not change the imbalance introduced
> by VSI and the takeover of the applications.

Just so I understand: if one has a non-renewable license, then one has
to keep a subscription with VSI? Subscription for what? Licenses?
Support? Both?

I see the imbalance argument (VSI could, theoretically, drastically
increase prices, knowing that there is a captive market), but it does
address the license issue, assuming that most or all of those who really
want such a non-expiring license would purchase support anyway.

Of course, the whole point is that one can continue to use the system
even if VSI is no longer around, so presumably that is allowed.

[Simon Clubley]

On 2021-08-04, Arne Vajhøj <ar...@vajhoej.dk> wrote:
> On 8/4/2021 8:19 AM, Simon Clubley wrote:
>>
>> Why can't VSI see just how strongly customers feel about this and why
>> are they not doing anything to address this showstopping problem for
>> keeping many people as VSI customers ?
>>
>> Do VSI simply not understand the sheer strength of customer feeling
>> about this ?
>
> Good questions.
>
> Maybe too many random people complain in comp.os.vms and too few
> paying customers complain directly?
>

Gerard and the French users had a go recently and we all know the
results of that so far. :-(

> Maybe VSI has not seen a financially attractive alternative?
>
> Maybe VSI is working on something that is just mot ready yet.
>
> Guesswork.
>
>> If you have your application source code, there are now solutions for
>> porting your VMS applications, including its VMS-specific code, over
>> to another operating system such as Linux.
>
> Such solutions has existed for decades.
>

The point that VSI are acting as if users do not have alternatives
and will eventually end up doing business with VSI on VSI's terms.

If VSI do actually believe that, I suspect they may be in for a
nasty surprise unless they become more flexible.

[Simon Clubley]

A group of French users formally discussed this with VSI recently.

They got a very tone deaf response from VSI that basically said
nothing significant was going to change.

[Simon Clubley]

On 2021-08-04, Phillip Helbig (undress to reply) <hel...@asclothestro.multivax.de> wrote:
> In article <imvsv9...@mid.individual.net>,
> calliet_gérard?= <gerard....@pia-sofer.fr> writes:
>
>> VSI finally informed us that for certain critical market segments
>> an exception would be possible, with licenses not limited in time.
>
> That is good.
>

Not good unfortunately. There are only a very few of these segments and
"critical" is defined by VSI and not the user. Your normal commercial
business critical system that keeps your company running does not qualify
based on the last information available.

Based on the last information available, it's more for some safety
critical systems (for example) only.

[Simon Clubley]

On 2021-08-04, calliet gérard <gerard....@pia-sofer.fr> wrote:
>
> Our action having been visible on comp.os.vms, several contributions
> have proposed other licensing modes that would be more acceptable and
> would encourage investment in VMS. We have consolidated these proposals
> in a message to VSI asking them if these options had been evaluated and
> to tell us why they had not been selected.
>

Did VSI respond to any of the alternative suggestions you put to them ?

> Indeed, many roundtable participants confirmed that
> the current licensing model is widely rejected, to the point of
> triggering plans to leave VMS. All those who expressed themselves
> pointed out the difficulty in the current conditions introduced by VSI
> to get a message of sustainability across to the management in order to
> obtain the investments."""""
>

To Jan-Erik: you have now heard directly from users trying to talk
to VSI about this. Do you still think there isn't a problem ?

[Jan-Erik Söderholm]

Den 2021-08-04 kl. 20:10, skrev Simon Clubley:
> On 2021-08-04, calliet gérard <gerard....@pia-sofer.fr> wrote:
>>
>> Our action having been visible on comp.os.vms, several contributions
>> have proposed other licensing modes that would be more acceptable and
>> would encourage investment in VMS. We have consolidated these proposals
>> in a message to VSI asking them if these options had been evaluated and
>> to tell us why they had not been selected.
>>
>
> Did VSI respond to any of the alternative suggestions you put to them ?
>
>> Indeed, many roundtable participants confirmed that
>> the current licensing model is widely rejected, to the point of
>> triggering plans to leave VMS. All those who expressed themselves
>> pointed out the difficulty in the current conditions introduced by VSI
>> to get a message of sustainability across to the management in order to
>> obtain the investments."""""
>>
>
> To Jan-Erik: you have now heard directly from users trying to talk
> to VSI about this. Do you still think there isn't a problem ?
>
> Simon.
>

Yes, I know about the french activities, and I have participated in the
on-line discussions about that. Have you?

My point is that does not happen on c.o.v.

Of course there is a problem. France has some issues with there nuclear
power plants and the metro systems. But I'm sure that it will be sorted...

But, that will be sorted outside of c.o.v.

[Arne Vajhøj]

On 8/4/2021 1:55 PM, Simon Clubley wrote:
> On 2021-08-04, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>> Maybe too many random people complain in comp.os.vms and too few
>> paying customers complain directly?
>
> Gerard and the French users had a go recently and we all know the
> results of that so far. :-(

If VSI hear it from more customers especially large customers
around the globe then it will help get it recognized as a problem.

Arne

[Phillip Helbig (undress to reply)]

In article <seekuj$9pd$3...@dont-email.me>, Simon Clubley

<clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> On 2021-08-04, Phillip Helbig (undress to reply) <hel...@asclothestro.multivax.de> wrote:
> > In article <imvsv9...@mid.individual.net>,

> > calliet_gérard <gerard....@pia-sofer.fr> writes:
> >
> >> VSI finally informed us that for certain critical market segments
> >> an exception would be possible, with licenses not limited in time.
> >
> > That is good.
> >
>
> Not good unfortunately. There are only a very few of these segments and
> "critical" is defined by VSI and not the user. Your normal commercial
> business critical system that keeps your company running does not qualify
> based on the last information available.
>
> Based on the last information available, it's more for some safety
> critical systems (for example) only.

I think that many here are critical of VSI's license policies. Does
that count? :-)

[calliet gérard]

Jan-Erik it has always a pleasure to meet you, in Sweeden, c.o.v. and
elsewhere on the net.

My question: if users club like in France speak on only little issues
which will be sorted anyway, if c.o.v. is not a place where serious
issues can be sorted, imagine there are serious issues, where can they
be sorted?

My opinion - and it not just from today -: VMS ecosystem is lacking real
places where, if there exists some issue, it can be seriously discuted
and evaluated. I know all of us have white hair, but it seems we are a
little bit childish about how the decisions can be made in an ecosystem.
Even Big Blue has ears.

Typically 80s? :)

[Dave Froble]

On 8/4/2021 8:19 AM, Simon Clubley wrote:

This is not going to be decided on c.o.v. This will be decided by
actual customers going to VSI and voicing their concerns, if any. I do
look for this to happen. Whether we port to x86 and take our customers
there might depend on such things.

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

[Dave Froble]

Keep in mind, there is an option. Permanent license. Hey, good enough
for nuke plants, good enough for others.

I really don't want to see this route chosen, because customers could
then go off support. I think all VMS users need to be on support, if
for no other reason, to financially support VSI.

[Dave Froble]

Part of that problem is that those customers won't be there waiting when
matters get sorted out. So, that's basically suicide for whoever takes
too long to sort matters.

I'll just give one opinion. But before I do, I want to state that I
believe that the only path forward for VSI is to have all VMS users
paying periodically for support. VMS users need VSI to be successful.

Should VSI cease to exist, and there is not a viable path forward for
VMS users, then existing methods, and possibly new methods, of bypassing
the licensing will be used.

Who could justify shutting down all businesses depending on VMS, and all
the employees of those businesses. Surely not interested governments.
I could see a government passing laws/regulations to enable such. I
sure would go to court to stop the destruction of businesses and jobs.

That's what we're talking about here. Destruction of businesses and
jobs. It ain't gonna happen. Yell and scream all you want about
copyright. It ain't gonna happen.

Perhaps part of the support agreement with VSI might be a stipulation
that if VSI is no longer able to support customers, then customers are
allowed to do whatever is required to continue in business using VMS.

This mess needs a resolution. It will happen.

[calliet gérard]

Indeed.

But very few are doing that. Because they react deciding to quit VMS.
And because it is one by one they quit, VSI cannot understand it's a
general issue they have to cope with.

For decades there has been hemorrhage out of VMS. We were able in 2014
to fight against that with the VSI creation. We didn't stop it, but we
were able to make some of the customers wait. The only message from VSI
had been since that "you'll have x86"... and the x86 horizon walked away
each year. It has been increadable difficult to make wait the customers
who didn't yet abandoned. You add the licensing issue, and everyone I
know has decided to go away. I think it is the real reason why the
customers don't speak with VSI to renegotiate.They are already out.

[Phillip Helbig (undress to reply)]

In article <sef80i$i74$2...@dont-email.me>, Dave Froble

<da...@tsoft-inc.com> writes:

> This is not going to be decided on c.o.v. This will be decided by
> actual customers going to VSI and voicing their concerns, if any. I do
> look for this to happen. Whether we port to x86 and take our customers
> there might depend on such things.

Right. But some folks might choose not to become VSI customers at all
if the only publicly available information is that there are no
non-expiring licenses.

[Phillip Helbig (undress to reply)]

In article <sef9de$q8k$1...@dont-email.me>, Dave Froble

<da...@tsoft-inc.com> writes:

> Keep in mind, there is an option. Permanent license. Hey, good enough
> for nuke plants, good enough for others.

Yes, but available to others?

> I really don't want to see this route chosen, because customers could
> then go off support. I think all VMS users need to be on support, if
> for no other reason, to financially support VSI.

My understanding was that the permanent license required a
"subscription" to VSI.

[Phillip Helbig (undress to reply)]

In article <sefamv$us$1...@dont-email.me>, Dave Froble

<da...@tsoft-inc.com> writes:

> Part of that problem is that those customers won't be there waiting when
> matters get sorted out. So, that's basically suicide for whoever takes
> too long to sort matters.
>
> I'll just give one opinion. But before I do, I want to state that I
> believe that the only path forward for VSI is to have all VMS users
> paying periodically for support. VMS users need VSI to be successful.
>
> Should VSI cease to exist, and there is not a viable path forward for
> VMS users, then existing methods, and possibly new methods, of bypassing
> the licensing will be used.
>
> Who could justify shutting down all businesses depending on VMS, and all
> the employees of those businesses. Surely not interested governments.
> I could see a government passing laws/regulations to enable such. I
> sure would go to court to stop the destruction of businesses and jobs.
>
> That's what we're talking about here. Destruction of businesses and
> jobs. It ain't gonna happen. Yell and scream all you want about
> copyright. It ain't gonna happen.
>
> Perhaps part of the support agreement with VSI might be a stipulation
> that if VSI is no longer able to support customers, then customers are
> allowed to do whatever is required to continue in business using VMS.
>
> This mess needs a resolution. It will happen.

The resolution will not be televised.

[Simon Clubley]

On 2021-08-04, calliet gérard <gerard....@pia-sofer.fr> wrote:
>
> My opinion - and it not just from today -: VMS ecosystem is lacking real
> places where, if there exists some issue, it can be seriously discuted
> and evaluated. I know all of us have white hair, but it seems we are a

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

> little bit childish about how the decisions can be made in an ecosystem.
> Even Big Blue has ears.
>

$ set response/mode=good_natured

Hey, speak for yourself. :-)

Some of us are still a good number of years away from that. :-)

(I hope. :-))

[Phillip Helbig (undress to reply)]

In article <segk95$jd4$1...@dont-email.me>, Simon Clubley

<clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> On 2021-08-04, calliet gérard <gerard....@pia-sofer.fr> wrote:
> >
> > My opinion - and it not just from today -: VMS ecosystem is lacking real
> > places where, if there exists some issue, it can be seriously discuted
> > and evaluated. I know all of us have white hair, but it seems we are a
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Those who still have hair can be thankful. :-|

[Simon Clubley]

On 2021-08-05, calliet gérard <gerard....@pia-sofer.fr> wrote:
>
> But very few are doing that. Because they react deciding to quit VMS.
> And because it is one by one they quit, VSI cannot understand it's a
> general issue they have to cope with.
>

Once an avalanche of users start to move away from VSI, then it's going
to be too late for VSI to do something about it.

"The avalanche has already started. It is too late for the pebbles to vote."
That's as true in real life for this as it was in the fictional context.

Going back to something I said yesterday, I wonder if VSI really do think
that the only choice for users is to deal with VSI, and hence VSI think
they can dictate to the users the terms on which VSI will deal with the
users.

I do suspect VSI are in for a nasty shock if they do believe that.

> For decades there has been hemorrhage out of VMS. We were able in 2014
> to fight against that with the VSI creation. We didn't stop it, but we
> were able to make some of the customers wait. The only message from VSI
> had been since that "you'll have x86"... and the x86 horizon walked away
> each year. It has been increadable difficult to make wait the customers
> who didn't yet abandoned. You add the licensing issue, and everyone I
> know has decided to go away. I think it is the real reason why the
> customers don't speak with VSI to renegotiate.They are already out.
>

I know VMS isn't really portable (at least not in the modern sense of
the word), but 7 years (and counting) is still a lot of time for a port.

[Simon Clubley]

On 2021-08-04, Dave Froble <da...@tsoft-inc.com> wrote:
>
> Should VSI cease to exist, and there is not a viable path forward for
> VMS users, then existing methods, and possibly new methods, of bypassing
> the licensing will be used.
>

What if VSI start releasing signed images for parts of VMS so that you
cannot do that ? (IOW, if you patch the binary, the signature will no
longer match so the image will not load or run).

[Arne Vajhøj]

On 8/4/2021 8:17 PM, Dave Froble wrote:
> Should VSI cease to exist, and there is not a viable path forward for
> VMS users, then existing methods, and possibly new methods, of bypassing
> the licensing will be used.
>
> Who could justify shutting down all businesses depending on VMS, and all
> the employees of those businesses.  Surely not interested governments. I
> could see a government passing laws/regulations to enable such.  I sure
> would go to court to stop the destruction of businesses and jobs.
>
> That's what we're talking about here.  Destruction of businesses and
> jobs.  It ain't gonna happen.  Yell and scream all you want about
> copyright.  It ain't gonna happen.

You can be pretty sure that federal copyright law and WTO rules
will not be changed to save VMS customers.

I will not rule out that some companies may decide to
break copyright - that happens frequently. But it is not a
general solution to the problem - legal, auditors, senior management,
board etc. will say no.

> This mess needs a resolution.

Yes.

> It will happen.

Hopefully VSI will not go under.

:-)

I hope that VSI comes up with a solution that will
provide some assurance.

Arne

[Bill Gunshannon]

On 8/5/21 8:07 AM, Simon Clubley wrote:
> On 2021-08-04, calliet gérard <gerard....@pia-sofer.fr> wrote:
>>
>> My opinion - and it not just from today -: VMS ecosystem is lacking real
>> places where, if there exists some issue, it can be seriously discuted
>> and evaluated. I know all of us have white hair, but it seems we are a
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>> little bit childish about how the decisions can be made in an ecosystem.
>> Even Big Blue has ears.
>>
>
> $ set response/mode=good_natured
>
> Hey, speak for yourself. :-)
>
> Some of us are still a good number of years away from that. :-)
>
> (I hope. :-))
>
> Simon.
>

I have had white hair since my early 30's.

bill

[Chris Townley]

I have had white hair since soon after I was born

--
Chris

[Bill Gunshannon]

And then you have Edgar and Johnny Winter. :-)

bill

[chris]

Definately must have been the life you have been living :-)...

[Bill Gunshannon]

I had an aunt who was white by 21. Probably genetics. Not the life
I was living. I was in the Army and that life agreed with me more than
anything else I have done in my life.

bill

[calliet gérard]

Le 05/08/2021 à 14:07, Simon Clubley a écrit :
> $ set response/mode=good_natured

My bad english is perhaps a cause of a misunderstanding.

I do like children with white hairs - oh, bad response -.

More seriously, I couldn't stand Jan-Erik speaking about little sortable
issues in France, and c.o.v. "being negative as usual".

In France, as an independent consultant for VMS, it seems there is no
future. Every customer we know have decided to go out. I know One
exception, yes.

Perhaps France is an exception, and for sure some big companies are VSI
best friends - or they say that -. And the VSI offer is good for them.
What will be the VMS future if only a small kernel of users continues?
Will not this happy fews being sometimes worried about their loneliness?

Anyway the situation in France is very bad, so it is a little bit
difficult to hear we have in France just some tractable issues.

And what about c.o.v.? A little bit sterile discutions? Not this one,
perhaps. More than that: is it so evident that criticizing VSI is just
empty words?

My summary is we have a cultural problem. VSI redoes the Digital way of
not hearing the customers, we redo the adoration way we had in the 80s.

I recognize in the way we justify everything done by The Company the
very precious and kind loyalty I appreciate in the community. My opinion
is we can be sometimes more loyal committing lèse majesté crimes, when
we think there is a danger in the kingdom :)

[calliet gérard]

Le 05/08/2021 à 14:21, Simon Clubley a écrit :
> On 2021-08-05, calliet gérard <gerard....@pia-sofer.fr> wrote:
>>
>> But very few are doing that. Because they react deciding to quit VMS.
>> And because it is one by one they quit, VSI cannot understand it's a
>> general issue they have to cope with.
>>
>
> Once an avalanche of users start to move away from VSI, then it's going
> to be too late for VSI to do something about it.
>
> "The avalanche has already started. It is too late for the pebbles to vote."

There is still a work for the Saint Bernard dogs (montain rescue dogs) :)

> That's as true in real life for this as it was in the fictional context.
>
> Going back to something I said yesterday, I wonder if VSI really do think
> that the only choice for users is to deal with VSI, and hence VSI think
> they can dictate to the users the terms on which VSI will deal with the
> users.
>
> I do suspect VSI are in for a nasty shock if they do believe that.

It seems VSI is like us: condemned to interpret a chrystal ball. We
think they think we think. *If* they organized strong ways of
consultation, we could abandone the chrystal ball.

>
>> For decades there has been hemorrhage out of VMS. We were able in 2014
>> to fight against that with the VSI creation. We didn't stop it, but we
>> were able to make some of the customers wait. The only message from VSI
>> had been since that "you'll have x86"... and the x86 horizon walked away
>> each year. It has been increadable difficult to make wait the customers
>> who didn't yet abandoned. You add the licensing issue, and everyone I
>> know has decided to go away. I think it is the real reason why the
>> customers don't speak with VSI to renegotiate.They are already out.
>>
>
> I know VMS isn't really portable (at least not in the modern sense of
> the word), but 7 years (and counting) is still a lot of time for a port.
>
> Simon.
>


--

[Jan-Erik Söderholm]

Den 2021-08-06 kl. 12:39, skrev calliet gérard:
> Le 05/08/2021 à 14:07, Simon Clubley a écrit :
>> $ set response/mode=good_natured
>
> My bad english is perhaps a cause of a misunderstanding.
>
> I do like children with white hairs - oh, bad response -.
>
> More seriously, I couldn't stand Jan-Erik speaking about little sortable
> issues in France, and c.o.v. "being negative as usual".

Hm... I'm sorry if it sounded like I thought that there wasn't some
serious issues in the France VMS community. I fully understand that.

Sorry if I expressed myself in an unclear way. Hard to tell since
I do not know what post from me you are refering to.

Regards,

Jan-Erik.

>
> In France, as an independent consultant for VMS, it seems there is no
> future. Every customer we know have decided to go out. I know One
> exception, yes.

At our site, the expected lifetime for our VMS system is 5-10 years.
After that it is very hard to tell what happens. Have we been able to
continue supporting the factory in an acceptable way? Who knows...

And I have my retirement in 2-3 years, if not sooner, anyway.

>
> Perhaps France is an exception, and for sure some big companies are VSI
> best friends - or they say that -. And the VSI offer is good for them. What
> will be the VMS future if only a small kernel of users continues? Will not
> this happy fews being sometimes worried about their loneliness?
>
> Anyway the situation in France is very bad, so it is a little bit difficult
> to hear we have in France just some tractable issues.
>
> And what about c.o.v.? A little bit sterile discutions? Not this one,
> perhaps. More than that: is it so evident that criticizing VSI is just
> empty words?

Not fully empty, but talking directly to VSI is probably better, IMHO.

[ultr...@gmail.com]

On Friday, August 6, 2021 at 7:01:02 AM UTC-4, Jan-Erik Söderholm wrote:
> Den 2021-08-06 kl. 12:39, skrev calliet gérard:
> > Le 05/08/2021 à 14:07, Simon Clubley a écrit :
> >> $ set response/mode=good_natured
> >
>

> Jan-Erik.
> >
> > In France, as an independent consultant for VMS, it seems there is no
> > future. Every customer we know have decided to go out. I know One
> > exception, yes.
> At our site, the expected lifetime for our VMS system is 5-10 years.
> After that it is very hard to tell what happens. Have we been able to
> continue supporting the factory in an acceptable way? Who knows...
>
> And I have my retirement in 2-3 years, if not sooner, anyway.
> >

Don't count on it our binary star wormwood in the bible is coming. You may not get to retire.

[John Wallace]

On 04/08/2021 13:19, Simon Clubley wrote:
> On 2021-08-03, Dave Froble <da...@tsoft-inc.com> wrote:
>> On 8/3/2021 1:41 PM, Simon Clubley wrote:
>>>
>>> Unfortunately, VSI do not seem to show any interest in addressing this.
>>>
>>> I wonder how much business it has cost them and how much it's going to
>>> cost them simply because a customer cannot allow this situation to occur.
>>>
>>> A business may love VMS and want to stay with it, but they are not
>>> going to allow the collapse of a vendor to be the collapse of their
>>> own business, even if that means moving away from VMS.
>>
>> On this we agree 100%.
>>
>
> Even when people disagree with me on the other things I say, everyone
> still appears to agree with me on this.
>
> Why can't VSI see just how strongly customers feel about this and why
> are they not doing anything to address this showstopping problem for
> keeping many people as VSI customers ?
>
> Do VSI simply not understand the sheer strength of customer feeling
> about this ?
>
> If you have your application source code, there are now solutions for
> porting your VMS applications, including its VMS-specific code, over
> to another operating system such as Linux.
>
> How many people are now exploring this option as a direct result of
> VSI's new time-limited licences policy ?
>
> Simon.
>

Subscription-based licences for VMS date back to at least VMS 4.6 (in
1987?) and V4.7, as can be seen from the VAX VMS Software Product
Description of the day which has the part numbers.

Back then they were called Periodic Payment Licences and seem to have
been associated with the then-new VAXBI series of machines.

I don't remember when they stopped being of interest to paying
customers. In fact I don't remember when DEC's introduction of
subscription based licencing *started* being of interest to VMS
customers I knew, but presumably someone somewhere liked the idea, at
least for a while.

Interesting to see that VSIVMS will not only have software licenced by
subscription, there will be "enforcement mechanisms" as part of the OS.

It's a good job such enforcement mechanisms (including the likes of what
used to be called DRM) have never ever in the history of digital
restrictions management or production IT been vulnerable to fail in new
and mysterious ways at the least convenient time possible and across the
whole user base. Well, not very often anyway.

[Simon Clubley]

On 2021-08-07, ultr...@gmail.com <ultr...@gmail.com> wrote:
>
> Don't count on it our binary star wormwood in the bible is coming. You may not get to retire.

On what date does it arrive ? It would be nice to be properly dressed
for the end of the world.

On a more serious note, how can anyone actually believe this stuff ?

This is not about believing in some god. (And while I don't believe
in a god, I strongly support the right of others to believe.)

This is actually full-blown doomsday cult stuff and given the casual
way Bob has mentioned this and similar things in the past, he clearly
believes the end of the world is just around the corner and has
integrated that belief into his way of life.

How can someone actually believe something that utterly extreme without
any proof ?

[Dave Froble]

On 8/7/2021 7:58 PM, Simon Clubley wrote:

Oh my Simon, you need to look at the world around you.

> On 2021-08-07, ultr...@gmail.com <ultr...@gmail.com> wrote:
>>
>> Don't count on it our binary star wormwood in the bible is coming. You may not get to retire.
>
> On what date does it arrive ? It would be nice to be properly dressed
> for the end of the world.

Why bother, no one will notice.

> On a more serious note, how can anyone actually believe this stuff ?

Just look at how many people voted for Trump. How many believed that
q-anon junk. How many believed trump won the election.

Belief - fiction with no requirement of proof, and no bearing on reality.

But it still happens.

Just look at those who won't get a Covid-19 vaccination.

Human intelligence is a myth.

> This is not about believing in some god. (And while I don't believe
> in a god, I strongly support the right of others to believe.)

Since nobody knows, or can know, agnostic is the best bet.

> This is actually full-blown doomsday cult stuff and given the casual
> way Bob has mentioned this and similar things in the past, he clearly
> believes the end of the world is just around the corner and has
> integrated that belief into his way of life.

If he truly believes that, why is he still doing anything? Just kick
back and wait.

Just like all those who just can't wait to get to heaven. What's
holding you back I'd ask them.

> How can someone actually believe something that utterly extreme without
> any proof ?

Many do.

It's called religion.

Better to call it superstition.

As mentioned by Jaba the Hut, "weak minded fools".

[Phillip Helbig (undress to reply)]

In article <sen6mn$6k5$1...@dont-email.me>, Simon Clubley

<clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> > Don't count on it our binary star wormwood in the bible is coming.
> You may not get to retire.
>

> On what date does it arrive ? It would be nice to be properly dressed
> for the end of the world.
>

> On a more serious note, how can anyone actually believe this stuff ?

Tens of millions in the USA; essentially all born-again Christians.

> How can someone actually believe something that utterly extreme without
> any proof ?

It happens. Bill Gates and George Soros invented COVID so that they
could get rich on the vaccine and inject chips into people. Hillary
Clinton runs a pedophile ring out of a pizza place, and she, Obama,
Gates, and Soros drink the blood of children to stay young. Yes,
millions of people do actually believe it, and believe it strongly
enough that it moves them to commit acts of terrorism.

[Henry Crun]

REMEMBER!! 50% of any population are below average!!
(And the avarege isn't all that high, either)

--
Mike R.
Home: http://alpha.mike-r.com/
QOTD: http://alpha.mike-r.com/qotd.php
No Micro$oft products were used in the URLs above, or in preparing this message.
Recommended reading: http://www.catb.org/~esr/faqs/smart-questions.html#before
and: http://alpha.mike-r.com/jargon/T/top-post.html
Missile address: N31.7624/E34.9691

[Arne Vajhøj]

On 8/8/2021 11:03 AM, Henry Crun wrote:
> REMEMBER!! 50% of any population are below average!!
> (And the avarege isn't all that high, either)

50% are below median.

Whether 50% are below average depends on the distribution.

Arne

[Simon Clubley]

On 2021-08-07, Dave Froble <da...@tsoft-inc.com> wrote:
>
> If he truly believes that, why is he still doing anything? Just kick
> back and wait.
>
> Just like all those who just can't wait to get to heaven. What's
> holding you back I'd ask them.
>

Maybe their god isn't all that keen on having them join him. :-)

[Dave Froble]

On 8/8/2021 8:51 PM, Simon Clubley wrote:
> On 2021-08-07, Dave Froble <da...@tsoft-inc.com> wrote:
>>
>> If he truly believes that, why is he still doing anything? Just kick
>> back and wait.
>>
>> Just like all those who just can't wait to get to heaven. What's
>> holding you back I'd ask them.
>>
>
> Maybe their god isn't all that keen on having them join him. :-)
>
> Simon.
>

Maybe they are not so sure of what they claim to believe.

[Lawrence D’Oliveiro]

On Friday, August 6, 2021 at 12:21:32 AM UTC+12, Simon Clubley wrote:
> On 2021-08-05, calliet gérard <gerard....@pia-sofer.fr> wrote:
>
>> The only message from VSI
>> had been since that "you'll have x86"... and the x86 horizon walked away
>> each year.
> >

> I know VMS isn't really portable (at least not in the modern sense of
> the word), but 7 years (and counting) is still a lot of time for a port.

This just reinforces my impression that proprietary software is very hard to make portable. Windows NT was supposedly designed from the beginning to be portable across more than just x86, and look at what a failure that was.

I suggested on one of VSI’s YouTube videos that they rearchitect VMS on top of a Linux kernel. That should simplify the job immensely, since Linux already runs on essentially every major processor architecture still in existence.

[John Dallman]

In article <c4f936d9-317d-4ce3...@googlegroups.com>,

lawren...@gmail.com (Lawrence D_Oliveiro) wrote:

> Windows NT was supposedly designed from the beginning to be portable
> across more than just x86, and look at what a failure that was.

It was initially shipped on MIPS, PowerPC and Alpha, and the dropping of
those platforms was because they weren't much used, rather than because
they didn't work. It then shipped on Itanium, which was dropped because
the market preferred x86-64, rather than because it didn't work. x86-64
is now dominant in the market, but ARM64 is shipping and working.

My employers have shipped Windows NT software on x86, Alpha, Itanium,
x86-64 and ARM64. I did the Itanium, x86-64 and ARM64 ports. Other
engineers did porting work for MIPS and PowerPC, but both of those
platforms were abandoned before being shipped, because their prospective
customers lost interest.

> I suggested on one of VSI's YouTube videos that they rearchitect
> VMS on top of a Linux kernel. That should simplify the job
> immensely, since Linux already runs on essentially every major
> processor architecture still in existence.

That might have been a good idea several years ago, but given they have
it working now, changing would be foolish. The unusual difficulties with
porting VMS seem to have been the need to create BLISS and MACRO-32
compilers, the out-of-date C compiler, and a much smaller team than DEC
had for the Alpha port, or HP for the Itanium port.

John

[Dave Froble]

The Macro-32 and Bliss compilers would have been needed regardless,
unless VSI was going to give up some customers. Don't know how many,
but, it could be significant.

I seem to recall that some of the real work was different memory
management and a few other things.

[Lawrence D’Oliveiro]

On Monday, August 9, 2021 at 4:27:34 PM UTC+12, John Dallman wrote:

> [Windows NT] was initially shipped on MIPS, PowerPC and Alpha, and the dropping of

> those platforms was because they weren't much used, rather than because
> they didn't work. It then shipped on Itanium, which was dropped because
> the market preferred x86-64, rather than because it didn't work.

All of which were supported on Linux, and continued to be supported on Linux long after Microsoft had abandoned them. So you see, it wasn’t just a matter of the popularity (or not) of those architectures.

Alpha is an interesting case. In spite of it being a 64-bit architecture, Windows NT only ever ran on it in 32-bit “TASO” mode. OpenVMS got as far as a hybrid 32/64-bit port, but I don’t think it ever managed to go full 64-bit.

The only two fully-64-bit OSes to run on Alpha were DEC’s Tru64 Unix ... and Linux.

[Jan-Erik Söderholm]

Den 2021-08-09 kl. 13:06, skrev Lawrence D’Oliveiro:
> On Monday, August 9, 2021 at 4:27:34 PM UTC+12, John Dallman wrote:
>
>> [Windows NT] was initially shipped on MIPS, PowerPC and Alpha, and the dropping of
>> those platforms was because they weren't much used, rather than because
>> they didn't work. It then shipped on Itanium, which was dropped because
>> the market preferred x86-64, rather than because it didn't work.
>
> All of which were supported on Linux, and continued to be supported on Linux long after Microsoft had abandoned them. So you see, it wasn’t just a matter of the popularity (or not) of those architectures.

I read Johns comment such as "they weren't much used [*by Windows users*].

And also "the [*Windows*] market preferred x86-64..."

Not the possible popularity of the platforms as such...

[John Dallman]

In article <ser2qf$bjk$1...@dont-email.me>, jan-erik....@telia.com
(Jan-Erik Söderholm) wrote:

> Den 2021-08-09 kl. 13:06, skrev Lawrence D_Oliveiro:
> > On Monday, August 9, 2021 at 4:27:34 PM UTC+12, John Dallman
> > wrote:
> > All of which were supported on Linux, and continued to be
> > supported on Linux long after Microsoft had abandoned them. So

> you see, it wasn_t just a matter of the popularity (or not) of

> those architectures.
>
> I read Johns comment such as "they weren't much used [*by Windows
> users*].
>
> And also "the [*Windows*] market preferred x86-64..."
>
> Not the possible popularity of the platforms as such...

Indeed. The Windows market popularity of those architectures is only
vaguely connected to their popularity in other markets.

Windows on PowerPC was a Microsoft/IBM co-production, and the only
supported machines outside Japan were rs/60000 models that were extremely
expensive. There was a plan to support PowerMac machines via the Common
Hardware Reference Platform, but since that would have opened the MacOS
market to hardware competition, Apple were less enthusiastic in practice
than they claimed in public, and it never went anywhere.

Windows on MIPS suffered because there were disagreements within SGI
about supporting it, or shunning it. Then the Intel Pentium Pro made the
contemporary MIPS processors look really slow, and NetPower, the company
that ex-SGI people had set up to do Windows MIPS high-end machines
switched to Intel.

Windows on Alpha lasted until the end of NT4. There was a 64-bit port,
used at Microsoft for development, but never released, because Itanium
was going to be the thing.

Itanium's only lasting markets were for HP-UX, VMS and NonStop, all of
which were HP proprietary operating systems at the time. A fair number of
customers did abandon them to get away from Itanium.

Windows on ARM64 has real potential, and I hope Microsoft are willing to
let it grow out of the niche they've allocated for it in their plans.

John

[Simon Clubley]

On 2021-08-09, Lawrence D?Oliveiro <lawren...@gmail.com> wrote:
> On Monday, August 9, 2021 at 4:27:34 PM UTC+12, John Dallman wrote:
>
>> [Windows NT] was initially shipped on MIPS, PowerPC and Alpha, and the dropping of
>> those platforms was because they weren't much used, rather than because
>> they didn't work. It then shipped on Itanium, which was dropped because
>> the market preferred x86-64, rather than because it didn't work.
>

> All of which were supported on Linux, and continued to be supported on Linux long after Microsoft had abandoned them. So you see, it wasn?t just a matter of the popularity (or not) of those architectures.
>
> Alpha is an interesting case. In spite of it being a 64-bit architecture, Windows NT only ever ran on it in 32-bit ?TASO? mode. OpenVMS got as far as a hybrid 32/64-bit port, but I don?t think it ever managed to go full 64-bit.
>

The lack of pure 64-bit mode for VMS on Alpha was nothing to do with
Alpha, but was to do with the VMS architecture and choice of implementation
languages.

Unlike with other operating systems where the lowest supported language
is C (with a bit of assembly thrown in for architecture-specific things),
_way_ too much would have broken in VMS code if they had tried to make
it a pure 64-bit environment.

[Dan Cross]

In article <608dc068-5afe-4f04...@googlegroups.com>,

Lawrence Dâ Oliveiro <lawren...@gmail.com> wrote:
>On Monday, August 9, 2021 at 4:27:34 PM UTC+12, John Dallman wrote:
>> [Windows NT] was initially shipped on MIPS, PowerPC and Alpha, and the dropping of
>> those platforms was because they weren't much used, rather than because
>> they didn't work. It then shipped on Itanium, which was dropped because
>> the market preferred x86-64, rather than because it didn't work.
>
>All of which were supported on Linux, and continued to be supported on Linux
>long after Microsoft had abandoned them. So you see, it wasn't
>just a matter of the popularity (or not) of those architectures.

FTR, NT first booted on an in-house i860 workstation built by
microsoft.

But porting VMS to run "on top of" Linux would likely be far more
work than just running directly on bare hardware. Linux's internal
kernel architecture is very different from that of VMS; a non-trivial
translation layer would have to be implemented.

>Alpha is an interesting case. In spite of it being a 64-bit architecture,
>Windows NT only ever ran on it in 32-bit 'TASO' mode. OpenVMS
>got as far as a hybrid 32/64-bit port, but I don't think it ever
>managed to go full 64-bit.

Running on the bare hardware, in a lot of respects, is not the
hard part. I imagine drivers and compilers are a much bigger
issue than booting, page tables and context switching. As others
pointed out, you need the compilers anyway to support customers.
Given the system call differences between VMS and Linux, it's
unclear how much you could really leverage Unix drivers, but
there's no reason those couldn't be ported if they can be used.

>The only two fully-64-bit OSes to run on Alpha were DEC's Tru64
>Unix ... and Linux.

I found that doubtful. NetBSD claims full 64-bit support on Alpha.
I imagine so does OpenBSD. I don't know what the status of
FreeBSD was, but I imagine they were full 64-bit. The plan 9
kernel on Alpha was a 64-bit kernel, though admittedly pretty
much a 32-bit API.

- Dan C.

[Arne Vajhøj]

On 8/8/2021 11:41 PM, Lawrence D’Oliveiro wrote:
> I suggested on one of VSI’s YouTube videos that they rearchitect VMS
> on top of a Linux kernel. That should simplify the job immensely,
> since Linux already runs on essentially every major processor
> architecture still in existence.

But would VMS userland on top of a Linux kernel
be VMS?

And is the porting effort really mostly kernel?

I suspect NO and NO.

Arne

[Bill Gunshannon]

On 8/9/21 11:36 AM, Arne Vajhøj wrote:
> On 8/8/2021 11:41 PM, Lawrence D’Oliveiro wrote:
>> I suggested on one of VSI’s YouTube videos that they rearchitect VMS
>> on top of a Linux kernel. That should simplify the job immensely,
>> since Linux already runs on essentially every major processor
>> architecture still in existence.
> But would VMS userland on top of a Linux kernel
> be VMS?
>

I agree. If you're going to run on VMS run on VMS. If you're
going to run on linux, run on linux.

> And is the porting effort really mostly kernel?

Mostly is a matter of opinion. Without the kernel none of the
layered products really matter. Linux already has a userland
that does all the stuff necessary to run an IS. The advantage
VMS is in VMS.

>
> I suspect NO and NO.

bill

[Arne Vajhøj]

On 8/9/2021 8:14 AM, Simon Clubley wrote:
> On 2021-08-09, Lawrence D?Oliveiro <lawren...@gmail.com> wrote:
>> Alpha is an interesting case. In spite of it being a 64-bit architecture, Windows NT only ever ran on it in 32-bit ?TASO? mode. OpenVMS got as far as a hybrid 32/64-bit port, but I don?t think it ever managed to go full 64-bit.
>
> The lack of pure 64-bit mode for VMS on Alpha

As states before there is no 32 bit mode in VMS.

But VMS compilers and API's support both 32 and 64 bit pointers.

> was nothing to do with
> Alpha, but was to do with the VMS architecture and choice of implementation
> languages.
>
> Unlike with other operating systems where the lowest supported language
> is C (with a bit of assembly thrown in for architecture-specific things),
> _way_ too much would have broken in VMS code if they had tried to make
> it a pure 64-bit environment.

I suspect that DEC could have rewritten VMS itself to all
64 bit pointers.

VMS applications out at customers would be a a bigger problem.

Arne

[Simon Clubley]

On 2021-08-09, Arne Vajhøj <ar...@vajhoej.dk> wrote:
> On 8/8/2021 11:41 PM, Lawrence D?Oliveiro wrote:
>> I suggested on one of VSI?s YouTube videos that they rearchitect VMS

>> on top of a Linux kernel. That should simplify the job immensely,
>> since Linux already runs on essentially every major processor
>> architecture still in existence.
> But would VMS userland on top of a Linux kernel
> be VMS?
>

The third party porting kits that allow you to port VMS applications
to run on top of Linux already do this to some extent.

So the question becomes: What _is_ VMS ? Is it the APIs ? Is it the
underlying design ? Is it something else ?

I think the APIs are part of it, but not sufficient by itself.

So while you can run some VMS applications on Linux with the help
of these tools, I don't think of that as being VMS on Linux.

[Lawrence D’Oliveiro]

On Monday, August 9, 2021 at 11:50:13 PM UTC+12, John Dallman wrote:

> Windows on ARM64 has real potential ...

Just like all the previous attempts at porting Windows to ARM, no doubt. Here is a processor architecture that ships more units per year than the entire population of the Earth, and this is, what, Microsoft’s third or fourth attempt at doing Windows on ARM? (Windows RT, anybody? And what about the even more laughable “Windows 10 IoT Edition” for the Raspberry π?) And so far it has been sputtering along about as nicely as the previous ones.

Is it any wonder that Microsoft is also trying desperately to turn Windows into Linux?

And if a juggernaut with resources on the scale of Microsoft has realized that Linux has become an irresistible force, how is a minnow like VSI, which by all accounts has already permanently lost much of the potential customer base for its products with all the development delays, supposed to survive?

[David Goodwin]

Windows has run on i860, MIPS, x86, Alpha, Clipper, PowerPC, Itanium and ARM64. Possibly SPARC too (it was announced but I don't know how far Intergraph got or if it was ever demonstrated).

I don't think it is reasonable to claim Microsoft failed to develop a portable operating system - the evidence clearly shows its unusually portable as far as proprietary operating systems go. Whether there is a *market* for non-x86 ports is an entirely different matter and not something you can really blame Microsoft for. They made it available, it worked, people weren't interested. I'm sure Microsoft could easily enough port Windows to RISC-V but there is little point in doing so if there is no one interested in buying it.

Additionally, current Windows on ARM isn't their third or fourth attempt - its their first. They ported Windows NT to ARM once and have made it available in various editions ever since as they've tried to find a niche for it. As I understand it there is nothing in particular wrong with the current variant of desktop windows on ARM - its windows, it runs windows apps (including the x86 variety via a mechanism similar to Windows 2000 on Alpha as I understand it). The bigger problem, again, is an apparent lack of demand and lack of worthwhile affordable hardware. Are people actually asking to run windows on ARM computers? Are there actually worthwhile ARM computers with good (comparable to x86) performance outside of Apple?

[Dave Froble]

That pretty much covers the issue.

WEENDOZE users are interested in cheap HW, and there sure is enough
cheap x86 HW, and many really don't know or care about anything else.

There are some uses where WEENDOZE on x86 seems to be the only option.
Look at what the gamers are using.

It might be a bit hard for ARM, or anything else to replace much of the
WEENDOZE world. Just how much cheaper can systems get? The CPU isn't
the only cost. Intel and AMD are in a tooth and nail fight for
performance, and users. Neither has much of a chance to inflate
pricing. Big benefit for the users.

[Michael S]

On Monday, August 16, 2021 at 7:53:08 AM UTC+3, dgso...@gmail.com wrote:
> On Tuesday, August 10, 2021 at 12:14:37 PM UTC+12, Lawrence D’Oliveiro wrote:
> > On Monday, August 9, 2021 at 11:50:13 PM UTC+12, John Dallman wrote:
> >
> > > Windows on ARM64 has real potential ...
> >
> > Just like all the previous attempts at porting Windows to ARM, no doubt. Here is a processor architecture that ships more units per year than the entire population of the Earth, and this is, what, Microsoft’s third or fourth attempt at doing Windows on ARM? (Windows RT, anybody? And what about the even more laughable “Windows 10 IoT Edition” for the Raspberry π?) And so far it has been sputtering along about as nicely as the previous ones.
> >
> > Is it any wonder that Microsoft is also trying desperately to turn Windows into Linux?
> >
> > And if a juggernaut with resources on the scale of Microsoft has realized that Linux has become an irresistible force, how is a minnow like VSI, which by all accounts has already permanently lost much of the potential customer base for its products with all the development delays, supposed to survive?
> Windows has run on i860, MIPS, x86, Alpha, Clipper, PowerPC, Itanium and ARM64. Possibly SPARC too (it was announced but I don't know how far Intergraph got or if it was ever demonstrated).

I'd like to see a proof w.r.t. i860 and Clipper.

Also, completely different kernel (WinCE) that exposed almost identical user-level API was available on Hitachi SH.

>
> I don't think it is reasonable to claim Microsoft failed to develop a portable operating system - the evidence clearly shows its unusually portable as far as proprietary operating systems go. Whether there is a *market* for non-x86 ports is an entirely different matter and not something you can really blame Microsoft for. They made it available, it worked, people weren't interested. I'm sure Microsoft could easily enough port Windows to RISC-V but there is little point in doing so if there is no one interested in buying it.
>
> Additionally, current Windows on ARM isn't their third or fourth attempt - its their first. They ported Windows NT to ARM once and have made it available in various editions ever since as they've tried to find a niche for it. As I understand it there is nothing in particular wrong with the current variant of desktop windows on ARM - its windows, it runs windows apps (including the x86 variety via a mechanism similar to Windows 2000 on Alpha as I understand it).

According to my understanding, only Microsoft itself and OEMs are allowed to develop kernel-mode modules, including hardware drivers, for Windows on ARM64.
That's far more restrictive than any previous port.

> The bigger problem, again, is an apparent lack of demand and lack of worthwhile affordable hardware.

WinARM64 laptops are not particularly expensive relative to other "premium" Windows laptops
The problem is - right now it seems that only "premium" is on market.

> Are people actually asking to run windows on ARM computers?

Why would they?
Why would "normal" people at all care about CPU architecture they are running?

> Are there actually worthwhile ARM computers with good (comparable to x86) performance outside of Apple?

Performance wise Samsung Galaxy Book S is not on par with the latest and greatest x86 laptops you can buy today, but it is faster than average of installed base.
Probably, non-trivially faster, too.

[Arne Vajhøj]

Unless emulation is extremely good and extremely fast then they care
about getting something that can run the software they want to use
and have used previously.

Remember that at one time the definition of "IBM compatible PC"
was "being able to run Lotus 1-2-3".

Arne

[Michael S]

Right now that's a reason to ask to for x86 rather than for Arm.

According to my understanding, not many apps are readily available on WinArm in native form
apart from Microsoft Office suite and Firefox.
But in theory, most of in-house stuff nowadays is written in .Net, so don't even need recompile
and most of Open Source stuff is, in theory, is not hard to recompile, assuming that Windows
version is properly maintained.
I never was interested enough to test to which degree a practice matches a theory.

For Win11 for Arm, Microsoft will try to push ability to run Android apps as an advantage.
I have no idea how many people really want to run Android apps on their laptops.
I certainly don't, but I don't pretend to be typical.

[David Goodwin]

On Monday, August 16, 2021 at 11:10:13 PM UTC+12, Michael S wrote:
> On Monday, August 16, 2021 at 7:53:08 AM UTC+3, dgso...@gmail.com wrote:
> > On Tuesday, August 10, 2021 at 12:14:37 PM UTC+12, Lawrence D’Oliveiro wrote:
> > > On Monday, August 9, 2021 at 11:50:13 PM UTC+12, John Dallman wrote:
> > >
> > > > Windows on ARM64 has real potential ...
> > >
> > > Just like all the previous attempts at porting Windows to ARM, no doubt. Here is a processor architecture that ships more units per year than the entire population of the Earth, and this is, what, Microsoft’s third or fourth attempt at doing Windows on ARM? (Windows RT, anybody? And what about the even more laughable “Windows 10 IoT Edition” for the Raspberry π?) And so far it has been sputtering along about as nicely as the previous ones.
> > >
> > > Is it any wonder that Microsoft is also trying desperately to turn Windows into Linux?
> > >
> > > And if a juggernaut with resources on the scale of Microsoft has realized that Linux has become an irresistible force, how is a minnow like VSI, which by all accounts has already permanently lost much of the potential customer base for its products with all the development delays, supposed to survive?
> > Windows has run on i860, MIPS, x86, Alpha, Clipper, PowerPC, Itanium and ARM64. Possibly SPARC too (it was announced but I don't know how far Intergraph got or if it was ever demonstrated).
> I'd like to see a proof w.r.t. i860 and Clipper.

Early Windows NT development was done on custom i860XR workstations before switching to custom MIPS workstations. The Intel codename for the i860XR was apparently "N-Ten", a possible source for the "NT" in "Windows NT".

The Clipper port was done by Intergraph and apparently demonstrated at a trade show in 1993 but never publicly released. IIRC Integraph switched to building SPARC machines shortly after and announced an effort to port Windows NT to SPARC but I suspect this port didn't get very far before it was abandoned.

[Arne Vajhøj]

On 8/16/2021 5:22 PM, Michael S wrote:
> On Monday, August 16, 2021 at 11:18:57 PM UTC+3, Arne Vajhøj wrote:
>> On 8/16/2021 7:10 AM, Michael S wrote:
>>> On Monday, August 16, 2021 at 7:53:08 AM UTC+3, dgso...@gmail.com wrote:
>>>> Are people actually asking to run windows on ARM computers?
>>>
>>> Why would they?
>>> Why would "normal" people at all care about CPU architecture they are running?
>> Unless emulation is extremely good and extremely fast then they care
>> about getting something that can run the software they want to use
>> and have used previously.
>>
>> Remember that at one time the definition of "IBM compatible PC"
>> was "being able to run Lotus 1-2-3".
>

> Right now that's a reason to ask to for x86 rather than for Arm.
>
> According to my understanding, not many apps are readily available on WinArm in native form
> apart from Microsoft Office suite and Firefox.
> But in theory, most of in-house stuff nowadays is written in .Net, so don't even need recompile
> and most of Open Source stuff is, in theory, is not hard to recompile, assuming that Windows
> version is properly maintained.
> I never was interested enough to test to which degree a practice matches a theory.

The in house business app GUI's that are not web GUI's are likely
written in .NET today and should run fine on Windows ARM.

> For Win11 for Arm, Microsoft will try to push ability to run Android apps as an advantage.
> I have no idea how many people really want to run Android apps on their laptops.
> I certainly don't, but I don't pretend to be typical.

I don't see it either.

If Android tablets was super popular then having a single app run on
both Windows ARM and Android tablet could be nice, but iOS tablets seems
way more popular than Android tablets.

Arne

[John Dallman]

In article <sfero7$1o3a$1...@gioia.aioe.org>, ar...@vajhoej.dk (Arne Vajhøj)
wrote:

> On 8/16/2021 5:22 PM, Michael S wrote:
> > I have no idea how many people really want to run Android apps on
> > their laptops.

> If Android tablets were super popular then having a single app run on

> both Windows ARM and Android tablet could be nice, but iOS tablets

> seem way more popular than Android tablets.

A lot of Android tablets are sold, but they're mostly "family devices"
for use by kids, very cheap, and not very powerful. Business use of
tablets is almost exclusively iOS, and the high-end ones are as fast as
good laptops.

Microsoft have said that Windows 11 will run Android apps on x86 Windows,
as well as ARM Windows. This is possible for a combination of reasons:

- Many Android apps are entirely written in Java or Kotlin (which runs
on Java infrastructure), and JIT-compiled for the device, on the
device, at install time. Such apps don't ship any ARM native code.

- Some Android apps that do ship native code ship both x86 and ARM,
because the usual app building tools do that by default.

- For apps that ship ARM native code but not x86, Intel have produced
a binary translator for running ARM code on their processors, which
will be bundled with Windows 11.

John

[Simon Clubley]

On 2021-08-16, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>
> Remember that at one time the definition of "IBM compatible PC"
> was "being able to run Lotus 1-2-3".
>

By that definition, VMS was at one time an "IBM compatible PC". :-)

Going the other way, I wonder if you could get DOSBox or Bochs to
run under VMS ? :-)

[Simon Clubley]

On 2021-08-16, David Goodwin <dgso...@gmail.com> wrote:
>
> Early Windows NT development was done on custom i860XR workstations before switching to custom MIPS workstations. The Intel codename for the i860XR was apparently "N-Ten", a possible source for the "NT" in "Windows NT".
>

Increase each of the letters in VMS by one and see what you get.

That's the other common source of the reported origin of WNT, ie: "Windows NT".

Not the first time something like that has apparently been done.
Decrease each of the letters in IBM and see what you get.

I'm afraid I can't do that. :-)

[Phillip Helbig (undress to reply)]

In article <sfgsa8$pr9$1...@dont-email.me>, Simon Clubley

<clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> On 2021-08-16, Arne Vajhøj <ar...@vajhoej.dk> wrote:
> >
> > Remember that at one time the definition of "IBM compatible PC"
> > was "being able to run Lotus 1-2-3".
>
> By that definition, VMS was at one time an "IBM compatible PC". :-)

Yes, and VMS met the Posix standard. :-|

[Phillip Helbig (undress to reply)]

In article <sfgsog$pr9$2...@dont-email.me>, Simon Clubley
<clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> On 2021-08-16, David Goodwin <dgso...@gmail.com> wrote:
> >
> > Early Windows NT development was done on custom i860XR workstations before switching to custom MIPS workstations. The Intel codename for the i860XR was apparently "N-Ten", a possible source for the "NT" in "Windows NT".
> >
>

> Increase each of the letters in VMS by one and see what you get.
>
> That's the other common source of the reported origin of WNT, ie: "Windows NT".
>
> Not the first time something like that has apparently been done.
> Decrease each of the letters in IBM and see what you get.
>
> I'm afraid I can't do that. :-)

For what it's worth, Clarke has denied that.

[Chris Townley]

NT was promulgated as Windows New Technology


--
Chris

[Jay E. Morris]

On 8/17/2021 4:05 AM, John Dallman wrote:
> In article <sfero7$1o3a$1...@gioia.aioe.org>, ar...@vajhoej.dk (Arne Vajhøj)
> wrote:
>> On 8/16/2021 5:22 PM, Michael S wrote:
>>> I have no idea how many people really want to run Android apps on
>>> their laptops.
>> If Android tablets were super popular then having a single app run on
>> both Windows ARM and Android tablet could be nice, but iOS tablets
>> seem way more popular than Android tablets.
>
> A lot of Android tablets are sold, but they're mostly "family devices"
> for use by kids, very cheap, and not very powerful. Business use of
> tablets is almost exclusively iOS, and the high-end ones are as fast as
> good laptops.
>

Don't tell the US Air Force that. Android tablets have been the standard
for at least 10 years. The just selected the Samsung S2 ruggedized
tablet for flight line use and logistical operations. And I don't think
buying the kids Android tables would lead to them having over 50% market
share verses iOS 36%.

https://www.statista.com/statistics/273268/worldwide-tablet-sales-by-operating-system-since-2nd-quarter-2010/

[Lawrence D’Oliveiro]

On Tuesday, August 17, 2021 at 9:23:00 AM UTC+12, Michael S wrote:
>
> According to my understanding, not many apps are readily available on WinArm in native form
> apart from Microsoft Office suite and Firefox.

On top of which, Microsoft’s x86 emulator only handles 32-bit code, not AMD64. The only thing that makes that palatable is the fact that much of Windows code never made the transition to 64-bit anyway.

> But in theory, most of in-house stuff nowadays is written in .Net ...

What, not Dotnet Core? What happened to Win64? Silverlight? WinRT? UWP? How many versions of “Project Reunion” have there been?

I don’t think Windows developers can say with any certainty right now which of Microsoft’s many platform APIs is the “core” one going forward...

[Arne Vajhøj]

The applications Michael is talking about are doing fine.

C# or VB.NET winform or WPF app acting as GUI for some business
application. "VB6 replacement"

Code should work unchanged from 1.0 (winform) / 3.0 (WPF) to 6.0.

So no problem for them.

The problems are in the more "exotic" sides of .NET.

.NET code running in the browser (SL) got canned. Just like
Java applets, Adobe Flash and other similar technologies. SL
never got much traction anyway.

The entire Metro/UWP and Xamarin/MAUI story is a big mess. But their
adoptation has so far been rather limited.

Arne

[John Dallman]

In article <184ecfb7-7efa-45e4...@googlegroups.com>,

lawren...@gmail.com (Lawrence D_Oliveiro) wrote:

> On top of which, Microsoft's x86 emulator only handles 32-bit code,
> not AMD64. The only thing that makes that palatable is the fact
> that much of Windows code never made the transition to 64-bit
> anyway.

That is due to change, but it hasn't happened yet.

> I don_t think Windows developers can say with any certainty right
> now which of Microsoft's many platform APIs is the _core_ one going
> forward...

Indeed. I've never regretted sticking to portable C and C++. Fortunately,
my management are almost as suspicious of marketing-driven programming
innovations as I am.


John

[John Dallman]

In article <sfhs0n$4qq$1...@dont-email.me>, mor...@epsilon3.comcon (Jay E.

Morris) wrote:

> > A lot of Android tablets are sold, but they're mostly "family
> > devices" for use by kids, very cheap, and not very powerful.

> Don't tell the US Air Force that. Android tablets have been the
> standard for at least 10 years. The just selected the Samsung S2
> ruggedized tablet for flight line use and logistical operations.

I said "mostly". The USAF has only about 675,000 people, and Android
tablet sales are measured in tens of millions per quarter.

Got a link to the ruggedized tablet? The nearest I can find is a 2016
model that's been discontinued:
<https://www.gsmarena.com/samsung_galaxy_tab_s2_9_7-7438.php>

John

[Jay E. Morris]

I shouldn't have said just as it was just before I retired three years
that they did that. Don't know how that fell out of my head. Couldn't
tell you what it's currently running on but could be part of the Air
Force Research Laboratory's Android Team Awareness Kit.

https://en.wikipedia.org/wiki/Android_Team_Awareness_Kit

[David Goodwin]

On Wednesday, August 18, 2021 at 8:32:37 PM UTC+12, Lawrence D’Oliveiro wrote:
> On Tuesday, August 17, 2021 at 9:23:00 AM UTC+12, Michael S wrote:
> >
> > According to my understanding, not many apps are readily available on WinArm in native form
> > apart from Microsoft Office suite and Firefox.
> On top of which, Microsoft’s x86 emulator only handles 32-bit code, not AMD64. The only thing that makes that palatable is the fact that much of Windows code never made the transition to 64-bit anyway.

That's because for most applications unless you need to allocate more than 2GB of RAM there is little benefit in being 64bit - all it really does is increase memory consumption a bit and probably CPU cache misses as well.

> > But in theory, most of in-house stuff nowadays is written in .Net ...
>
> What, not Dotnet Core? What happened to Win64? Silverlight? WinRT? UWP? How many versions of “Project Reunion” have there been?
>
> I don’t think Windows developers can say with any certainty right now which of Microsoft’s many platform APIs is the “core” one going forward...

.net core is just a newer version of .net same as Java 7 is a newer version of Java. Microsoft has actually dropped the 'core' branding now - the most recent version is simply called .net 5. Win64 is still around and still enhanced - lots of stuff is built against the C API and it will never go away. Silverlight died for the same reason Flash did. WinRT is still around. As is UWP.

And, importantly, all of these things are still supported on current windows. Even the VB6 runtime is still supported on current windows. This is a substantially better situation than on Linux where you'd have a hard time building a Qt 3 application from source on a recent Ubuntu release and running an existing Qt 3 binary is impossible unless it happened to be statically linked (and even then there is a good chance it won't work properly)

It would sure be nice if Microsoft would just pick a UI toolkit and stick with it but Microsoft is by no means alone when it comes to this problem.

[Arne Vajhøj]

On 8/18/2021 8:45 PM, David Goodwin wrote:
> On Wednesday, August 18, 2021 at 8:32:37 PM UTC+12, Lawrence
> D’Oliveiro wrote:
>> On Tuesday, August 17, 2021 at 9:23:00 AM UTC+12, Michael S wrote:
>>>
>>> According to my understanding, not many apps are readily
>>> available on WinArm in native form apart from Microsoft Office
>>> suite and Firefox.
>> On top of which, Microsoft’s x86 emulator only handles 32-bit code,
>> not AMD64. The only thing that makes that palatable is the fact
>> that much of Windows code never made the transition to 64-bit
>> anyway.
>
> That's because for most applications unless you need to allocate more
> than 2GB of RAM there is little benefit in being 64bit - all it
> really does is increase memory consumption a bit and probably CPU
> cache misses as well.

Floating Point is also very different between 32 bit and 64 bit.

And virtual memory space does not only apply to RAM but can also
be mapped to a file.

But yes all 3 cases are probably exceptions.

>>> But in theory, most of in-house stuff nowadays is written in .Net
>>> ...
>>
>> What, not Dotnet Core? What happened to Win64? Silverlight? WinRT?
>> UWP? How many versions of “Project Reunion” have there been?
>>
>> I don’t think Windows developers can say with any certainty right
>> now which of Microsoft’s many platform APIs is the “core” one going
>> forward...
>
> .net core is just a newer version of .net same as Java 7 is a newer
> version of Java. Microsoft has actually dropped the 'core' branding
> now - the most recent version is simply called .net 5.

.NET Core and .NET FX were two different .NET flavors.

.NET FX got discontinued and .NET Core 3 evolved into .NET 5.

.NET Core 3 to .NET 5 is a normal version upgrade like Java 6 to 7.

But .NET FX 4 to .NET 5 is a somewhat bigger upgrade - some things
are different and some features were dropped.

So some .NET FX 4 users may need to stay or rewrite.

But not the typical .NET winform or WPF GUI.

> It would sure be nice if Microsoft would just pick a UI toolkit and
> stick with it but Microsoft is by no means alone when it comes to
> this problem.

Java went: AWT -> Swing -> JavaFX.

Arne

[Lawrence D’Oliveiro]

On Monday, August 16, 2021 at 4:53:08 PM UTC+12, dgso...@gmail.com wrote:
> Additionally, current Windows on ARM isn't their third or fourth attempt - its their first.
> They ported Windows NT to ARM once and have made it available in various editions
> ever since as they've tried to find a niche for it.

Fitting in with Einstein’s definition of insanity: trying the same thing over and over, hoping for a different outcome each time.

[David Goodwin]

Good thing they've not been trying the same thing over and over then!

The initial ARM offering (Windows RT) was a variant of Windows 8 that could only run apps from the windows store. There was a version of Visual C++ that could target this variant of windows but I don't know if applications built with it could be distributed via the Windows Store or if Microsoft required "modern"-style apps written in .net. This product ultimately failed for reasons that should have been obvious to everyone within Microsoft before the first device running it was sold.

Later there was Windows 10 IoT Edition. This is just a rebrand of Windows Embedded which has been around in various forms since the 90s. The difference here is its now available on ARM in addition to whatever platforms it was previously available for. The IoT Core variant was made available for free on Raspberry Pis for hobbyist use. I've no idea if Windows IoT for ARM could be considered a failure as I doubt Microsoft publishes sales data for something like this. Given its been around since the 90s I assume if it wasn't making money they would have given up on it by now.

The most recent consumer-oriented effort is just Windows 10 (or 11) but on an ARM processor. Apparently it works fine. Its got some sort of binary translation/emulation thing (like Apples Rosetta 2 I guess) that lets it run unmodified x86 software. If an ARM Windows laptop appears to be just another laptop running Windows to end users then I'd say Microsoft has done all they can. Whether it would then be successful is down to the companies making the hardware it runs on. Seems unknown whether Qualcomm can be bothered trying to compete with Apple in making a high performance low power ARM CPU.

[John Dallman]

In article <6851ade5-b1ad-4062...@googlegroups.com>,

dgso...@gmail.com (David Goodwin) wrote:

> The most recent consumer-oriented effort is just Windows 10 (or 11)
> but on an ARM processor. Apparently it works fine.

It does, within the limits of its CPU power. The currently available
devices do not have leading-edge ARM processors.

> Its got some sort of binary translation/emulation thing (like
> Apples Rosetta 2 I guess) that lets it run unmodified x86 software.

That works too, although it's a bit slow. It is a binary translator, as
are all these things nowadays: instruction-by-instruction emulation is
Too Slow.

> If an ARM Windows laptop appears to be just another laptop running
> Windows to end users then I'd say Microsoft has done all they can.
> Whether it would then be successful is down to the companies making
> the hardware it runs on. Seems unknown whether Qualcomm can be bothered

> trying to compete with Apple in making a high performance low power
> ARM CPU.

Qualcomm's CEO has /claimed/ he'll deliver performance equivalent to
Apple's M1 next year. I have no idea if Qualcomm will manage this, but
there is no magic in the M1, and the same approach could be used by any
manufacturer prepared to go to the necessary effort.

John

[Lawrence D’Oliveiro]

On Wednesday, September 22, 2021 at 6:54:24 PM UTC+12, John Dallman wrote:
> In article <6851ade5-b1ad-4062...@googlegroups.com>,
> dgso...@gmail.com (David Goodwin) wrote:
>
>> The most recent consumer-oriented effort is just Windows 10 (or 11)
>> but on an ARM processor. Apparently it works fine.
>
> It does, within the limits of its CPU power. The currently available
> devices do not have leading-edge ARM processors.

It’s worth noting that these are not “ARM processors” in any general sense, but only a specific family of Qualcomm Snapdragon chips. Remember, the ARM ecosystem is huge, with a range of companies and volume of unit shipments that completely dwarf x86. For example, the Raspberry π runs a Broadcom processor. Samsung also make their own ARM chips. And of course we know about Apple’s ones.

The problem is that ARM lacks a standardized BIOS layer like has been a traditional part of x86. Without such a system, Windows is lost. Linux has its “device tree” system for coping with that, which does mean compiling custom kernels for different families of SoCs. But that’s a lot easier than a Windows compile.

>> Its got some sort of binary translation/emulation thing (like
>> Apples Rosetta 2 I guess) that lets it run unmodified x86 software.

32-bit only, as I pointed out.

[John Dallman]

In article <cd94fd57-441f-4af0...@googlegroups.com>,
lawren...@gmail.com (Lawrence D_Oliveiro) wrote:

> It_s worth noting that these are not _ARM processors_ in any

> general sense, but only a specific family of Qualcomm Snapdragon
> chips. Remember, the ARM ecosystem is huge, with a range of
> companies and volume of unit shipments that completely dwarf x86.

> For example, the Raspberry Pi runs a Broadcom processor. Samsung

> also make their own ARM chips. And of course we know about Apple's
> ones.

My view of ARM processors is concentrated on the fast 64-bit ones for
applications, mobile and otherwise, rather than embedded or other areas.
Within that:

Qualcomm's cores have their own branding, but are based on ARM Holding
core designs. Qualcomm make a fuss about their custom work but observing
the changes in performance over a few generations, it tracks closely with
ARM's Cortex-A7x series core designs. I've recently got to use Android
devices with ARM Cortex-X1 cores, in Snapdragon 888 SoCs, and they are
surprisingly close to Apple M1 performance.

Broadcom's chips use ARM Cortex-A cores, and they don't conceal that.

Samsung used to design their own cores, in the Exynos family, but they
used ARM Cortex-A cores for the American and South Korean markets, and
have recently given up on Exynos.

Apple's designs are entirely their own, of course. But lots of other SoC
manufacturers just use ARM Cortex-Ax cores.

> The problem is that ARM lacks a standardized BIOS layer like has
> been a traditional part of x86. Without such a system, Windows is
> lost.

That is, indeed, a big problem. Microsoft could produce a standard, but
are rather tentative about ARM in some ways.

John

[David Goodwin]

I've never got my hands on any Windows-compatible ARM hardware but as
they're not giving out the source code I assume they're currently requiring a
standard pre-boot environment (UEFI?) similar to how they required ARC
firmware on MIPS, PowerPC and Alpha hardware capable of running
Windows NT 3.x and 4.0.

For other hardware differences Windows NT has a Hardware Abstraction
Layer which was used in the past for handling handling odd or not quite
standard hardware. I guess if an ARM board doesn't quite meet the specs
a custom HAL implementation may solve the problem.

I guess Microsoft has the power to dictate requirements here given Windows
on ARM isn't a retail product. The only way to get it is with hardware so
its on the manufacturer to design their widget to be compatible with Windows
in one way or another if they want to sell it with Windows on it.

[Lawrence D’Oliveiro]

On Wednesday, September 22, 2021 at 11:55:56 PM UTC+12, dgso...@gmail.com wrote:
> I guess Microsoft has the power to dictate requirements here ...

To whom? Nobody else seems to be bothering with Windows-on-ARM.

[Scott Dorsey]

Just like they didn't bother with Windows on Alpha.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Bill Gunshannon]

On 9/23/21 7:41 AM, Scott Dorsey wrote:
> =?UTF-8?Q?Lawrence_D=E2=80=99Oliveiro?= <lawren...@gmail.com> wrote:
>> On Wednesday, September 22, 2021 at 11:55:56 PM UTC+12, dgso...@gmail.com wrote:
>>> I guess Microsoft has the power to dictate requirements here ...
>>
>> To whom? Nobody else seems to be bothering with Windows-on-ARM.
>
> Just like they didn't bother with Windows on Alpha.

When Windows on ARM comes in a turnkey, out-of-the-box package
with MS Office pre-loaded and costs less than x86/64 businesses
will take it up.

The majority of users don't care about processors. They only
care about getting the job done. Alpha Windows was released
before any applications were ready. Most people I knew who
actually bought one ended up dumping Windows and running Linux.
Why? Because Linux had applications.

I honestly believe VMS never had that problem because VMS users
run locally developed applications and not COTS.

bill

[Craig A. Berry]

Perhaps, but Microsoft makes their own hardware and is still releasing
new ARM-based devices:

<https://arstechnica.com/gadgets/2021/09/arm-based-surface-pro-x-gets-an-899-wi-fi-only-model-but-few-other-upgrades/>

<https://www.microsoft.com/en-us/surface/business/surface-pro-x/processor>

I think someone mentioned up-thread that the x86 emulation is 32-bit
only, but that is changing and will be in Windows 11:

<https://blogs.windows.com/windows-insider/2020/12/10/introducing-x64-emulation-in-preview-for-windows-10-on-arm-pcs-to-the-windows-insider-program/>

Maybe this will all fizzle and end up like Windows on Alpha or Itanium.
But Microsoft is big enough that a small percentage of Windows devices
running on ARM is still a big market, so it's not at all obvious this
will follow the same pattern.

[chris]

On 09/23/21 12:41, Scott Dorsey wrote:
> =?UTF-8?Q?Lawrence_D=E2=80=99Oliveiro?=<lawren...@gmail.com> wrote:
>> On Wednesday, September 22, 2021 at 11:55:56 PM UTC+12, dgso...@gmail.com wrote:
>>> I guess Microsoft has the power to dictate requirements here ...
>>
>> To whom? Nobody else seems to be bothering with Windows-on-ARM.
>
> Just like they didn't bother with Windows on Alpha.
> --scott

Perhaps, but Arm has far more traction and installed base than Alpha
ever had. It does need Microsoft to get serious about Arm on the desktop
and in servers. The processors are already there, but mass adoption of
arm and windows would upset the probably very cosy relationship with Intel.

Can't see any reason for the delay otherwise. Arm is everywhere else...

Chris

[Dave Froble]

My question is, "why"?

x86 is cheap.
x86 is everywhere.

What reason would Microsoft have to look at anything else?

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

[John Dallman]

In article <ir39oe...@mid.individual.net>, bill.gu...@gmail.com

(Bill Gunshannon) wrote:

> Alpha Windows was released before any applications were ready.
> Most people I knew who actually bought one ended up dumping
> Windows and running Linux. Why? Because Linux had applications.

It got used in some specialised fields, because it was faster than Intel
at the time. That advantage disappeared when Intel and AMD got into a
clockspeed battle in 1997, and when Compaq took over DEC, they weren't
interested in it. Microsoft continued to use Alpha-based machines to
develop 64-bit Windows until they had a reasonable number of Itanium
machines.

John

[John Dallman]

In article <sihsf2$2mo$1...@dont-email.me>, craig...@nospam.mac.com (Craig
A. Berry) wrote:

> On 9/22/21 8:37 PM, Lawrence D_Oliveiro wrote:
> > Nobody else seems to be bothering with Windows-on-ARM.
>
> Perhaps, but Microsoft makes their own hardware and is still
> releasing new ARM-based devices

They seem to have decided that Windows on ARM is /strictly/ for mobile
devices. It does help considerably with the battery life of small and
thin laptops.

John

[Simon Clubley]

On 2021-09-23, Dave Froble <da...@tsoft-inc.com> wrote:
> On 9/23/2021 10:27 AM, chris wrote:
>>
>> Perhaps, but Arm has far more traction and installed base than Alpha
>> ever had. It does need Microsoft to get serious about Arm on the desktop
>> and in servers. The processors are already there, but mass adoption of
>> arm and windows would upset the probably very cosy relationship with Intel.
>>

I wonder if the availability of RISC-V is likely to upset anything ? :-)

>> Can't see any reason for the delay otherwise. Arm is everywhere else...
>>
>

> My question is, "why"?
>
> x86 is cheap.
> x86 is everywhere.
>
> What reason would Microsoft have to look at anything else?
>

Why do phones use ARM processors instead of x86 processors ?

Simon.

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

[Stephen Hoffman]

On 2021-09-23 15:17:50 +0000, Dave Froble said:

> My question is, "why"?
>
> x86 is cheap.
> x86 is everywhere.
>
> What reason would Microsoft have to look at anything else?

Price and power efficiency, same as usual. Arm can be cheaper, more
power-efficient, and fast.

https://www.anandtech.com/show/15578/cloud-clash-amazon-graviton2-arm-against-intel-and-amd/9


Arm designs can also be juggernaut-scale, with 15 billion transistors
in one recent design; with fast big.LITTLE multiprocessor, a fast GPU,
statistics-math acceleration; that's a full-on SoC. And
power-efficient. For comparison, Itanium Poulson and Kittson are ~3
billion. And Alpha and Itanium processors and servers never really saw
appreciable work on power efficiency.

As for being "everywhere", the Arm installed base dwarfs those of Intel
and AMD and x86-64. And I'd suspect that Arm-related investments dwarf
Intel, too.

Intel has spectacular processor design and processor fabrication
abilities, but they're also necessarily working within a massive
software installed base, and with a complex and accreted architecture.
And their fabrication efforts have been falling short. TSMC and others
have massive investments in fabrication, as well. Intel has discussed
using TMSC to fab parts of some Intel-designed components.

https://www.reuters.com/business/intel-details-mixed-source-chip-strategy-tsmc-partnerships-2021-08-19/


Microsoft has been selling Arm clients for a while, and publicly
prototyping Arm servers for several years now, as have others. How far
Microsoft might get with Windows 11 for ARM64? There are a number of
folks working with the Windows ARM64 insiders' preview, including
having gotten that working on Apple M1.

https://blogs.windows.com/windowsdeveloper/2021/06/28/announcing-arm64ec-building-native-and-interoperable-apps-for-windows-11-on-arm/


How? If? When? Unknown. Architectural and product transitions tend to
be boring and slow and happening only around the periphery of other
markets, then the platforms and tools are ready, and then the changes
can then accelerate through the market.




--
Pure Personal Opinion | HoffmanLabs LLC

[Dave Froble]

I'm aware of many of the things ARM is used for. Yes, they are quite
useful.

But, are they much better for desktop and notebook PCs? I really can't
see them being much better in that environment.

Now, talking Microsoft, how successful have they been outside the
desktop and notebook PCs? Not very. So, for them, x86 does the job.

I have no idea of cost to get a decent WEENDOZE on ARM. But whatever it
is, would doing so be of adequate benefit to Microsoft? Maybe sometime,
but right now I don't see it.

Then again, I don't get out much ....

:-)

[John Dallman]

In article <siik2d$grk$1...@dont-email.me>, seao...@hoffmanlabs.invalid

(Stephen Hoffman) wrote:

> There are a number of folks working with the Windows ARM64 insiders'
> preview, including having gotten that working on Apple M1.

Just in time for Microsoft to announce that running ARM Windows on M1 is
"not a supported scenario."

https://www.theregister.com/2021/09/10/windows_11_m1/

John