Den 2021-06-02 kl. 13:17, skrev Andrew Brehm:
> On 02/06/2021 10:08, Jan-Erik Söderholm wrote:
>> Den 2021-06-02 kl. 09:23, skrev Andrew Brehm:
>>> On 31/05/2021 21:54, Phillip Helbig (undress to reply) wrote:
>>>> In article <s92qfa$7o3$1@dont-email.me>, Dave Froble
>>>> <davef@tsoft-inc.com> writes:
>>>>
>>>>> My idea is the same as it's been for years.  Do away with license PAKs,
>>>>> allow anyone to run VMS, require support for any commercial use of VMS.
>>>>> This would avoid all the issues about drop dead dates.
>>>>
>>>> How would you actually check whether commercial users had support?  Big
>>>> commercial users?  Sure.  Commercial users with one VMS system left?
>>>> Probably not.  And what about other people offering support, openly or
>>>> not, in return for money?  Could VSI prevent that?  Yes, someone who
>>>> needs important patches will pay for support.  But if you are relying on
>>>> that, then you will have unpatched VMS support in the wild at least
>>>> among non-commercial users (or, rather, all who don't want to pay for
>>>> support, whether commercial or not).  But old systems which haven't been
>>>> touched for years or decades probably won't be patched anyway.
>>>>
>>> I think all of that is too complicated.
>>>
>>> Perhaps the easiest distinction between commercial and non-commercial
>>> use is system specs.
>>>
>>> Make OpenVMS freely available to everyone and let it use up to 4 cores
>>> and up to 8 GB of RAM for free, then demand payment for more.
>>>
>>> This will allow everyone to use VMS for development and testing and will
>>> make serious customers pay. Likewise, if VSI goes away or someone
>>> forgets to renew support, VMS would simply collapse to using 4 cores and
>>> 8 GB only, keeping production system running but very slowly...
>>
>> We have 1 core and 4 GB RAM, and the system is quite fast.
>> And I do not consider a system supporting production in the main
>> factory of a leading producer of forrest/garden equipment, as "small".
>>
>> So no, I do not beleive in that solution.
>
> I don't think there will be too many systems like that in the near future.
> As cores and memory become cheaper and cheaper, systems with 1 core and 4
> GB of RAM will become a faint memory.

You select the config that is "good enough". And why wouldn't you if
you get it for free. Is it not that all 1 or 2 CPU Alphas suddenly
need a 4+ CPU/core system that is also way faster per core.

No, I do not beleive in your suggestion.

Den 2021-06-02 kl. 14:19, skrev Simon Clubley:
> On 2021-06-02, Jan-Erik Söderholm <jan-erik.soderholm@telia.com> wrote:
>>
>> So no, I do not beleive in that solution.
>
> I don't think it's going to be viable either.
>
> Do you like or dislike any of the other proposed solutions Jan-Erik ?
>
> Do you have any ideas of your own ?
>
> This is something that should have been sorted out at the start 5 years
> ago instead of just sprung on the community last year. VSI were seriously
> wrong to suddenly change the ground rules in that way IMHO.
>
> Simon.
>

You pay support to get new versions and patches.
If you don't pay, you stay where you are.

Simple as that.

In no way should the system suddenly just stop working.
Not in any circumstances.

On 6/2/2021 8:19 AM, Simon Clubley wrote:
> On 2021-06-02, Jan-Erik Söderholm <jan-erik.soderholm@telia.com> wrote:
>>
>> So no, I do not beleive in that solution.
>
> I don't think it's going to be viable either.
>
> Do you like or dislike any of the other proposed solutions Jan-Erik ?
>
> Do you have any ideas of your own ?
>
> This is something that should have been sorted out at the start 5 years
> ago instead of just sprung on the community last year. VSI were seriously
> wrong to suddenly change the ground rules in that way IMHO.
>
> Simon.
>

VSI mainly, I think, comes from the old way of doing things, large
license fee up front. I can see their confusion on how to move forward.

There are two issues here, and they are not very compatible.

1) Recurring support fees.

I think this is the way forward. There can be multiple fees, such as an
additional fee for something like cluster. I don't agree with that,
since it will inhibit the use of clusters. Still, support should depend
on usage.

2) VSI no longer around to issue new licenses.

No entity can take a route that might cause them to fail. No ISVs can
gamble on their work being for naught. There must be a way forward for
customers/users, regardless of whether VSI succeeds or fails.

Before license PAKs, there was still a license to use software. It
specified allowed usage. This can still happen.

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

On 6/2/2021 7:26 AM, Phillip Helbig (undress to reply) wrote:
> In article <ihp7s1Fjmf6U1@mid.individual.net>, Andrew Brehm
> <andrew@netneurotic.net> writes:
>
>
>>> I can think of many, many commercial applications which could get by
>>> with far fewer resources, say a webserver running a webshop. And I can
>>> think of non-commercial use which needs more resources, such as number
>>> crunching in academia.
>
>> And academic number crunching IS a commercial application and can
>> justify buying a licence. Why not? VSI could sell such licences for free
>> if this is required.
>
> No, it is not commercial.

You're being a bit too narrow with the term "commercial". Think instead
of a resource that normally costs money.

Academic or otherwise, number crunching for some purpose will cost,
regardless of the vendor.

> (It is not hobbyist use, at least in most
> cases, but definitely not commercial.) DEC and VMS used to be big in
> the academic market. It is a big mistake to think that there are only
> hobbyists and huge commercial users. In-between there are academic
> users, non-profit-organization users, small businesses, self-employed
> people, etc.

There are two broad uses.

1) For some purpose which has value. Call this "commercial".

2) A purpose with no value, other than entertainment(sometimes),
education, and such.

It should be up to VSI, IBM, and such to determine whether a usage
should require support fees. Don't get hung up on the word
"commercial", which has different meanings to different folks.

>>> Yes. But many commercial customers wouldn't have to pay anything and
>>> some non-commercial ones would.
>>
>> The first group will likely become smaller and smaller as time passes
>
> So with that the possibility of running VMS for free, e.g. for
> hobbyists, vanishes as well.

Your pet "ox" getting gored, huh?

>> and cores become cheaper. And the second group can always get a licence.
>
> Sure, but the whole point is that non-commercial customers shouldn't
> have to pay.

Why not? They are using someone else's labor, resources, and such, just
as "commercial" users do. The "free" usage has much more to do with the
benefits to the vendor, ie; exposure of their products so as to induce
more "commercial" usage.

>>>> Likewise, if VSI goes away or someone
>>>> forgets to renew support, VMS would simply collapse to using 4 cores and
>>>> 8 GB only, keeping production system running.
>>>
>>> Certainly not all production systems.
>>
>> Perhaps not, but the majority or at least some. It would still be
>> better than a complete halt as dictated by the current process.
>
> You can't define a production system as "more powerful than X" then,
> when the license no longer works, limit the functionality to "less
> powerful than X".
>
>> The actual comparison is the actual comparison when the project is
>> started? Do we use VMS or do we use Linux? What does each cost?
>
> For new products. What about moving to a different platform?
>

This whole discussion is a problem for VSI, and it is up to VSI to
resolve it to customer's satisfaction. Not something for c.o.v to
remedy. If they do not satisfy the customers, long term, all they will
have is short term customers.

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

On 2021-06-02, Dave Froble <davef@tsoft-inc.com> wrote:
> On 6/2/2021 7:26 AM, Phillip Helbig (undress to reply) wrote:
>>
>> No, it is not commercial.
>
> You're being a bit too narrow with the term "commercial". Think instead
> of a resource that normally costs money.
>
> Academic or otherwise, number crunching for some purpose will cost,
> regardless of the vendor.
>

Instead of commercial use when talking about licences, think in terms
of production use.

Academic number crunching is production use.

>
> This whole discussion is a problem for VSI, and it is up to VSI to
> resolve it to customer's satisfaction. Not something for c.o.v to
> remedy. If they do not satisfy the customers, long term, all they will
> have is short term customers.
>

Since VSI management seem to be burying their heads in the sand,
then at least the comp.os.vms ideas might start _some_ discussion
in VSI about options the VSI management may not have considered.

Simon.

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

On 6/2/2021 2:12 PM, Simon Clubley wrote:
> On 2021-06-02, Dave Froble <davef@tsoft-inc.com> wrote:
>> This whole discussion is a problem for VSI, and it is up to VSI to
>> resolve it to customer's satisfaction. Not something for c.o.v to
>> remedy. If they do not satisfy the customers, long term, all they will
>> have is short term customers.
>
> Since VSI management seem to be burying their heads in the sand,
> then at least the comp.os.vms ideas might start _some_ discussion
> in VSI about options the VSI management may not have considered.

There are many opinions about how to fix the problem.

But practically everyone agrees that it is a problem.

We know that VSI follow comp.os.vms so it is very likely
that at least someone in VSI got the message.

Whether the message get from the VSI tech guys to the VSI
biz guys is another question.

But with enough discussion here + direct letters + informal
conversations + various other channels then I am optimistic
that the message will eventually come through.

Arne

In article <ihpdcdFkn8kU1@mid.individual.net>, Bill Gunshannon
<bill.gunshannon@gmail.com> writes:

> >>> I can think of many, many commercial applications which could get by
> >>> with far fewer resources, say a webserver running a webshop. And I can
> >>> think of non-commercial use which needs more resources, such as number
> >>> crunching in academia.
> >
> >> And academic number crunching IS a commercial application and can
> >> justify buying a licence. Why not? VSI could sell such licences for free
> >> if this is required.
> >
> > No, it is not commercial. (It is not hobbyist use, at least in most
> > cases, but definitely not commercial.) DEC and VMS used to be big in
> > the academic market.
>
> The key here is "used to be". Does anyone know of any academic use of
> VMS today?

Not many. But most people here think that VSI should try to get new
customers. Academia might not bring in that much profit, but it trains
a new generation who want to use VMS. DEC's biggest mistake was losing
the academic market.

> > It is a big mistake to think that there are only
> > hobbyists and huge commercial users. In-between there are academic
> > users, non-profit-organization users, small businesses, self-employed
> > people, etc.
> >
> >>> Yes. But many commercial customers wouldn't have to pay anything and
> >>> some non-commercial ones would.

Right; that's why I think that basic payment on computing power is not a
good idea.

> >> The first group will likely become smaller and smaller as time passes
> >
> > So with that the possibility of running VMS for free, e.g. for
> > hobbyists, vanishes as well.
>
> You mean like it did for the VAX? Welcome to reality.

Right. The whole discussion here is to try to avoid what happened to
VAX, so a proposed solution which leads to a similar end result is a
non-starter.

> The only true non-commercial use is hobbyists. Non-profit,
> Government use, Academic use are all just as commercial as a
> bank, store or factory.

Depends on the definition, local laws, etc.

On 6/2/21 2:41 PM, Phillip Helbig (undress to reply) wrote:
> In article <ihpdcdFkn8kU1@mid.individual.net>, Bill Gunshannon
> <bill.gunshannon@gmail.com> writes:
>
>>>>> I can think of many, many commercial applications which could get by
>>>>> with far fewer resources, say a webserver running a webshop. And I can
>>>>> think of non-commercial use which needs more resources, such as number
>>>>> crunching in academia.
>>>
>>>> And academic number crunching IS a commercial application and can
>>>> justify buying a licence. Why not? VSI could sell such licences for free
>>>> if this is required.
>>>
>>> No, it is not commercial. (It is not hobbyist use, at least in most
>>> cases, but definitely not commercial.) DEC and VMS used to be big in
>>> the academic market.
>>
>> The key here is "used to be". Does anyone know of any academic use of
>> VMS today?
>
> Not many. But most people here think that VSI should try to get new
> customers. Academia might not bring in that much profit, but it trains
> a new generation who want to use VMS. DEC's biggest mistake was losing
> the academic market.

Don't blame DEC for that one. VMS was firmly entrenched in
academia during DEC's era. It was COMAPQ and HP that dropped
that ball. Neither of them had a clue what went on in
academia or how it worked. And, HP at least, was unwilling
to listen when suggestions were offered.

>
>>> It is a big mistake to think that there are only
>>> hobbyists and huge commercial users. In-between there are academic
>>> users, non-profit-organization users, small businesses, self-employed
>>> people, etc.
>>>
>>>>> Yes. But many commercial customers wouldn't have to pay anything and
>>>>> some non-commercial ones would.
>
> Right; that's why I think that basic payment on computing power is not a
> good idea.
>
>>>> The first group will likely become smaller and smaller as time passes
>>>
>>> So with that the possibility of running VMS for free, e.g. for
>>> hobbyists, vanishes as well.
>>
>> You mean like it did for the VAX? Welcome to reality.
>
> Right. The whole discussion here is to try to avoid what happened to
> VAX, so a proposed solution which leads to a similar end result is a
> non-starter.
>
>> The only true non-commercial use is hobbyists. Non-profit,
>> Government use, Academic use are all just as commercial as a
>> bank, store or factory.
>
> Depends on the definition, local laws, etc.

That's true. By redefining the word anything can become non-commercial.

bill

On 02/06/2021 13:21, Phillip Helbig (undress to reply) wrote:
> In article <ihp7m7FjjlnU1@mid.individual.net>, Andrew Brehm
> <andrew@netneurotic.net> writes:
>
>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>> And I do not consider a system supporting production in the main
>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>
>>> So no, I do not beleive in that solution.
>>
>> I don't think there will be too many systems like that in the near
>> future. As cores and memory become cheaper and cheaper, systems with 1
>> core and 4 GB of RAM will become a faint memory.
>
> Right, so according to your plan no system could run VMS without paying
> for support.
>

Except for very small non-productive systems, yes.

And that was the idea.

On 02/06/2021 15:56, Jan-Erik Söderholm wrote:
>>>>
>>>> Perhaps the easiest distinction between commercial and non-commercial use is system specs.
>>>>
>>>> Make OpenVMS freely available to everyone and let it use up to 4 cores and up to 8 GB of RAM for free, then demand payment for more.
>>>>
>>>> This will allow everyone to use VMS for development and testing and will make serious customers pay. Likewise, if VSI goes away or someone forgets to renew support, VMS would simply collapse to using 4 cores and 8 GB only, keeping production system running but very slowly...
>>>
>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>> And I do not consider a system supporting production in the main
>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>
>>> So no, I do not beleive in that solution.
>>
>> I don't think there will be too many systems like that in the near future. As cores and memory become cheaper and cheaper, systems with 1 core and 4 GB of RAM will become a faint memory.
>
> You select the config that is "good enough". And why wouldn't you if
> you get it for free. Is it not that all 1 or 2 CPU Alphas suddenly
> need a 4+ CPU/core system that is also way faster per core.

1 or 2 CPU Alphas are probably not in scope for new OpenVMS versions.

> No, I do not beleive in your suggestion.

It's not a question of belief. I just think it is a better solution than VSI's current idea.

On 02/06/2021 13:26, Phillip Helbig (undress to reply) wrote:
> In article <ihp7s1Fjmf6U1@mid.individual.net>, Andrew Brehm
> <andrew@netneurotic.net> writes:
>
>
>>> I can think of many, many commercial applications which could get by
>>> with far fewer resources, say a webserver running a webshop. And I can
>>> think of non-commercial use which needs more resources, such as number
>>> crunching in academia.
>
>> And academic number crunching IS a commercial application and can
>> justify buying a licence. Why not? VSI could sell such licences for free
>> if this is required.
>
> No, it is not commercial. (It is not hobbyist use, at least in most
> cases, but definitely not commercial.) DEC and VMS used to be big in
> the academic market. It is a big mistake to think that there are only
> hobbyists and huge commercial users. In-between there are academic
> users, non-profit-organization users, small businesses, self-employed
> people, etc.

I think academic use for non-educational purposes qualifies as "commercial". It can certainly qualify for the requirement for a licence.

As for the big mistake, I made no such assumption.

>>> Yes. But many commercial customers wouldn't have to pay anything and
>>> some non-commercial ones would.
>>
>> The first group will likely become smaller and smaller as time passes
>
> So with that the possibility of running VMS for free, e.g. for
> hobbyists, vanishes as well.

Not if hobbyists use small system. Also, hobbyist licences are up to VSI. They can grant licences for larger hobbyist systems.

>> and cores become cheaper. And the second group can always get a licence.
>
> Sure, but the whole point is that non-commercial customers shouldn't
> have to pay.

VSI can grant licences for free.

The idea is that very small systems would be free while larger systems need a licence. Whether that licence would be free or not free depending on use is independent of the basic idea of having automatic free licences for very small systems.

>>>> Likewise, if VSI goes away or someone
>>>> forgets to renew support, VMS would simply collapse to using 4 cores and
>>>> 8 GB only, keeping production system running.
>>>
>>> Certainly not all production systems.
>>
>> Perhaps not, but the majority or at least some. It would still be
>> better than a complete halt as dictated by the current process.
>
> You can't define a production system as "more powerful than X" then,
> when the license no longer works, limit the functionality to "less
> powerful than X".

Why not?

Many OS are licenced per core. VMware ESXi is, for example. They do define a number of cores as the limit up to which a certain licence works.

>> The actual comparison is the actual comparison when the project is
>> started? Do we use VMS or do we use Linux? What does each cost?
>
> For new products. What about moving to a different platform?
>

I wasn't talking about products. I am seeing this from the perspective of in-house IT. A project can be anything, a new product, an old product continue, a new version of some software, a reconfiguration, a move to a new platform or new hardware.

Den 2021-06-03 kl. 12:24, skrev Andrew Brehm:
> On 02/06/2021 15:56, Jan-Erik Söderholm wrote:
>>>>>
>>>>> Perhaps the easiest distinction between commercial and non-commercial
>>>>> use is system specs.
>>>>>
>>>>> Make OpenVMS freely available to everyone and let it use up to 4 cores
>>>>> and up to 8 GB of RAM for free, then demand payment for more.
>>>>>
>>>>> This will allow everyone to use VMS for development and testing and
>>>>> will make serious customers pay. Likewise, if VSI goes away or someone
>>>>> forgets to renew support, VMS would simply collapse to using 4 cores
>>>>> and 8 GB only, keeping production system running but very slowly...
>>>>
>>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>>> And I do not consider a system supporting production in the main
>>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>>
>>>> So no, I do not beleive in that solution.
>>>
>>> I don't think there will be too many systems like that in the near
>>> future. As cores and memory become cheaper and cheaper, systems with 1
>>> core and 4 GB of RAM will become a faint memory.
>>
>> You select the config that is "good enough". And why wouldn't you if
>> you get it for free. Is it not that all 1 or 2 CPU Alphas suddenly
>> need a 4+ CPU/core system that is also way faster per core.
>
> 1 or 2 CPU Alphas are probably not in scope for new OpenVMS versions.
>

The current roadmap has Alpha "Standard Support" into 2030 (as long
as the current roadmap reaches). But it also depends on what you
mean with a "new version". An Alpha 8.4-2L3? We don't know.

Anyway, for those a 1-2 core x86 VM will probably be "enough".
Should those production systems be free?

On 03/06/2021 12:44, Jan-Erik Söderholm wrote:
> Den 2021-06-03 kl. 12:24, skrev Andrew Brehm:
>
> The current roadmap has Alpha "Standard Support" into 2030 (as long
> as the current roadmap reaches). But it also depends on what you
> mean with a "new version". An Alpha 8.4-2L3? We don't know.
>
> Anyway, for those a 1-2 core x86 VM will probably be "enough".
> Should those production systems be free?
>

No. But I think they ought to keep running if unlicensed.

You are confusing enforcement mechanism with price policy.

I have not commented on how much OpenVMS should cost and what uses should be free. I just thought collapsing a running system to using at most 4 cores and 8 GB of memory would be an acceptable way to deal with forcing licence and support renewals and would allow for easier adoption among hobbyists, developers and sysadmins.

In article <ihs071F5l8iU1@mid.individual.net>, Andrew Brehm
<andrew@netneurotic.net> writes:

> > Den 2021-06-03 kl. 12:24, skrev Andrew Brehm:
> >
> > The current roadmap has Alpha "Standard Support" into 2030 (as long
> > as the current roadmap reaches). But it also depends on what you
> > mean with a "new version". An Alpha 8.4-2L3? We don't know.
> >
> > Anyway, for those a 1-2 core x86 VM will probably be "enough".
> > Should those production systems be free?
>
> No. But I think they ought to keep running if unlicensed.
>
> You are confusing enforcement mechanism with price policy.
>
> I have not commented on how much OpenVMS should cost and what uses
> should be free. I just thought collapsing a running system to using at
> most 4 cores and 8 GB of memory would be an acceptable way to deal with
> forcing licence and support renewals and would allow for easier adoption
> among hobbyists, developers and sysadmins.

I see your point, but it is not a good idea in practice, since it misses
the smaller production systems. Think of it this way: x86 systems are
powerful (because more modern) and even a minimal system is probably
more powerful than most Alphas (many still in use) and even some or most
Itaniums. If there is a collapse and the system needs more than 8 GB,
then your idea doesn't work. Also, more powerful systems can be had for
free (if not now then soon). Should hobbyists not be allowed to use
them?

OK, your idea is orthogonal to hobbyist licenses, which could exist for
any system. However, in practice, such licenses for low-powered systems
could be used by hobbyists if there are no license costs.

Things are complicated enough; we shouldn't make them more complicated,
especially not by defining something which is not future proof. I
remember when a 1-GB disk and 24 MB memory was a big system with a big
price.

On 6/3/2021 6:23 AM, Andrew Brehm wrote:
> On 02/06/2021 13:21, Phillip Helbig (undress to reply) wrote:
>> In article <ihp7m7FjjlnU1@mid.individual.net>, Andrew Brehm
>> <andrew@netneurotic.net> writes:
>>
>>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>>> And I do not consider a system supporting production in the main
>>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>>
>>>> So no, I do not beleive in that solution.
>>>
>>> I don't think there will be too many systems like that in the near
>>> future. As cores and memory become cheaper and cheaper, systems with 1
>>> core and 4 GB of RAM will become a faint memory.
>>
>> Right, so according to your plan no system could run VMS without paying
>> for support.
>>
>
> Except for very small non-productive systems, yes.
>
> And that was the idea.

What about large non-productive systems?

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

On 6/3/2021 6:24 AM, Andrew Brehm wrote:
> On 02/06/2021 15:56, Jan-Erik Söderholm wrote:
>>>>>
>>>>> Perhaps the easiest distinction between commercial and
>>>>> non-commercial use is system specs.
>>>>>
>>>>> Make OpenVMS freely available to everyone and let it use up to 4
>>>>> cores and up to 8 GB of RAM for free, then demand payment for more.
>>>>>
>>>>> This will allow everyone to use VMS for development and testing and
>>>>> will make serious customers pay. Likewise, if VSI goes away or
>>>>> someone forgets to renew support, VMS would simply collapse to
>>>>> using 4 cores and 8 GB only, keeping production system running but
>>>>> very slowly...
>>>>
>>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>>> And I do not consider a system supporting production in the main
>>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>>
>>>> So no, I do not beleive in that solution.
>>>
>>> I don't think there will be too many systems like that in the near
>>> future. As cores and memory become cheaper and cheaper, systems with
>>> 1 core and 4 GB of RAM will become a faint memory.
>>
>> You select the config that is "good enough". And why wouldn't you if
>> you get it for free. Is it not that all 1 or 2 CPU Alphas suddenly
>> need a 4+ CPU/core system that is also way faster per core.
>
> 1 or 2 CPU Alphas are probably not in scope for new OpenVMS versions.
>
>> No, I do not beleive in your suggestion.
>
> It's not a question of belief. I just think it is a better solution than
> VSI's current idea.

Well, just about anything is better than the current VSI direction.

I also do not think much of your suggestion. Too complex. Made for
lawyers.

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

On 6/3/2021 6:44 AM, Jan-Erik Söderholm wrote:
> Den 2021-06-03 kl. 12:24, skrev Andrew Brehm:
>> On 02/06/2021 15:56, Jan-Erik Söderholm wrote:
>>>>>>
>>>>>> Perhaps the easiest distinction between commercial and
>>>>>> non-commercial use is system specs.
>>>>>>
>>>>>> Make OpenVMS freely available to everyone and let it use up to 4
>>>>>> cores and up to 8 GB of RAM for free, then demand payment for more.
>>>>>>
>>>>>> This will allow everyone to use VMS for development and testing
>>>>>> and will make serious customers pay. Likewise, if VSI goes away or
>>>>>> someone forgets to renew support, VMS would simply collapse to
>>>>>> using 4 cores and 8 GB only, keeping production system running but
>>>>>> very slowly...
>>>>>
>>>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>>>> And I do not consider a system supporting production in the main
>>>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>>>
>>>>> So no, I do not beleive in that solution.
>>>>
>>>> I don't think there will be too many systems like that in the near
>>>> future. As cores and memory become cheaper and cheaper, systems with
>>>> 1 core and 4 GB of RAM will become a faint memory.
>>>
>>> You select the config that is "good enough". And why wouldn't you if
>>> you get it for free. Is it not that all 1 or 2 CPU Alphas suddenly
>>> need a 4+ CPU/core system that is also way faster per core.
>>
>> 1 or 2 CPU Alphas are probably not in scope for new OpenVMS versions.
>>
>
> The current roadmap has Alpha "Standard Support" into 2030 (as long
> as the current roadmap reaches). But it also depends on what you
> mean with a "new version". An Alpha 8.4-2L3? We don't know.
>
> Anyway, for those a 1-2 core x86 VM will probably be "enough".
> Should those production systems be free?
>

If they are, VSI will be missing out on much revenue, and that could
break them.

An old friend that sold services once told me, "I don't want to make a
killing on a single transaction, I just want a little nick. But I want
a LOT of nicks."

Not too many free WEENDOZE systems out there, right? Lots of little
"nicks" can add up nicely.

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

On 6/2/2021 7:26 AM, Phillip Helbig (undress to reply) wrote:

> Sure, but the whole point is that non-commercial customers shouldn't
> have to pay.

Why not? Should all those games and such written for WEENDOZE systems
be free? If so, they would not exist.

This whole topic isn't about cost, it's about drop dead dates.

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

On 6/3/21 9:43 AM, Dave Froble wrote:
> On 6/3/2021 6:24 AM, Andrew Brehm wrote:
>> On 02/06/2021 15:56, Jan-Erik Söderholm wrote:
>>>>>>
>>>>>> Perhaps the easiest distinction between commercial and
>>>>>> non-commercial use is system specs.
>>>>>>
>>>>>> Make OpenVMS freely available to everyone and let it use up to 4
>>>>>> cores and up to 8 GB of RAM for free, then demand payment for more.
>>>>>>
>>>>>> This will allow everyone to use VMS for development and testing and
>>>>>> will make serious customers pay. Likewise, if VSI goes away or
>>>>>> someone forgets to renew support, VMS would simply collapse to
>>>>>> using 4 cores and 8 GB only, keeping production system running but
>>>>>> very slowly...
>>>>>
>>>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>>>> And I do not consider a system supporting production in the main
>>>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>>>
>>>>> So no, I do not beleive in that solution.
>>>>
>>>> I don't think there will be too many systems like that in the near
>>>> future. As cores and memory become cheaper and cheaper, systems with
>>>> 1 core and 4 GB of RAM will become a faint memory.
>>>
>>> You select the config that is "good enough". And why wouldn't you if
>>> you get it for free. Is it not that all 1 or 2 CPU Alphas suddenly
>>> need a 4+ CPU/core system that is also way faster per core.
>>
>> 1 or 2 CPU Alphas are probably not in scope for new OpenVMS versions.
>>
>>> No, I do not beleive in your suggestion.
>>
>> It's not a question of belief. I just think it is a better solution than
>> VSI's current idea.
>
> Well, just about anything is better than the current VSI direction.
>
> I also do not think much of your suggestion.  Too complex.  Made for
> lawyers.
>

I think beancounters are a much bigger threat than lawyers. Whatever
finally becomes the business model for VMS the most important feature
it needs is to not be another argument for moving off of VMS.

bill

On 2021-06-03, Bill Gunshannon <bill.gunshannon@gmail.com> wrote:
>
> I think beancounters are a much bigger threat than lawyers. Whatever
> finally becomes the business model for VMS the most important feature
> it needs is to not be another argument for moving off of VMS.
>

As I've mentioned previously, there's also the "Will I get fired by
my boss if I make the decision to go with VSI ?" worry the customer's
manager faces.

Going with VSI or porting to Linux (for example) is just another
decision a manager has to make. They are unlikely to have some
emotional bond that drives them to VSI regardless.

The manager is unlikely to make the more risky choice without a
_very_ good reason so it is the job of VSI to show how going with
VSI is safer for the customer's manager than it is for the manager
to order a port to Linux (for example).

So far, VSI seem to be making a _very_ poor job of convincing those
customer managers that going with VSI is the safer option for them.

Having time-limited licences that will stop working if VSI goes bust
will not exactly convince the customer's manager that going with VSI
is the safer option for the manager when it comes to keeping their job.

Simon.

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

On 03/06/2021 15:48, Dave Froble wrote:
>> The current roadmap has Alpha "Standard Support" into 2030 (as long
>> as the current roadmap reaches). But it also depends on what you
>> mean with a "new version". An Alpha 8.4-2L3? We don't know.
>>
>> Anyway, for those a 1-2 core x86 VM will probably be "enough".
>> Should those production systems be free?
>>
>
> If they are, VSI will be missing out on much revenue, and that could break them.
>
> An old friend that sold services once told me, "I don't want to make a killing on a single transaction, I just want a little nick.  But I want a LOT of nicks."
>
> Not too many free WEENDOZE systems out there, right?  Lots of little "nicks" can add up nicely.
>

Actually, MSFT did exactly that: offer Windows for free for small devices.

https://www.pcworld.com/article/2139080/microsoft-makes-windows-free-on-iot-and-small-mobile-devices-but-not-pcs.html

Makers of tablets and phones with a screen size of less than 9" could get a free Windows licence. I think I owned at least one of those free-Windows devices. They were dirt cheap for a few years when Microsoft was trying to establish Windows Mobile and Windows on tablets.

I fear VSI will be missing out much more revenue if they try to sell a product that stops working when the subscription isn't renewed. That's not the sort of thing that convinces new customers. If they just want to milk existing customers, that is fine. But to get people to switch from VMS on Alpha or Itanium to VMS on x64 instead of to Linux or from non-VMS to VMS, it will be difficult outside the cloud.

On 03/06/2021 15:41, Dave Froble wrote:
> On 6/3/2021 6:23 AM, Andrew Brehm wrote:
>> On 02/06/2021 13:21, Phillip Helbig (undress to reply) wrote:
>>> In article <ihp7m7FjjlnU1@mid.individual.net>, Andrew Brehm
>>> <andrew@netneurotic.net> writes:
>>>
>>>>> We have 1 core and 4 GB RAM, and the system is quite fast.
>>>>> And I do not consider a system supporting production in the main
>>>>> factory of a leading producer of forrest/garden equipment, as "small".
>>>>>
>>>>> So no, I do not beleive in that solution.
>>>>
>>>> I don't think there will be too many systems like that in the near
>>>> future. As cores and memory become cheaper and cheaper, systems with 1
>>>> core and 4 GB of RAM will become a faint memory.
>>>
>>> Right, so according to your plan no system could run VMS without paying
>>> for support.
>>>
>>
>> Except for very small non-productive systems, yes.
>>
>> And that was the idea.
>
> What about large non-productive systems?
>

Get a hobbyist licence? I'm not sure I understand the question.

On 6/4/2021 2:32 AM, Andrew Brehm wrote:
> On 03/06/2021 15:48, Dave Froble wrote:
>> An old friend that sold services once told me, "I don't want to make a
>> killing on a single transaction, I just want a little nick.  But I
>> want a LOT of nicks."
>>
>> Not too many free WEENDOZE systems out there, right?  Lots of little
>> "nicks" can add up nicely.
>
> Actually, MSFT did exactly that: offer Windows for free for small devices.
>
> https://www.pcworld.com/article/2139080/microsoft-makes-windows-free-on-iot-and-small-mobile-devices-but-not-pcs.html
>
>
> Makers of tablets and phones with a screen size of less than 9" could
> get a free Windows licence. I think I owned at least one of those
> free-Windows devices. They were dirt cheap for a few years when
> Microsoft was trying to establish Windows Mobile and Windows on tablets.

I am not sure what we can conclude from that.

MS could not sell Windows Phone (which even though it shares
kernel with desktop Windows is not the same OS).

So they tried to give it away to get customers. But that did
not work either - still no demand.

So in the end they killed the product.

Arne