What Will Drive More OpenVMS Adoption?

[Ken Farmer]

Developers and community members are debating what’s needed to keep interest growing in OpenVMS. On the VMS newsgroup, a discussion is underway, more than a month of it, surrounding openings for the OS. Answers to the question about momentum cover well-traveled ground. Some of the solutions stand out as repeats of the HP 3000’s struggles.

Read more...
https://legacyos.org/what-will-drive-more-openvms-adoption/

[Mike K.]

Honestly, I think that any new users will probably be coming out of the hobbyist/community license program. Basically people who try it out and use it for personal projects, and decide that they like it so much compared to Linux or Windows that when the time comes, they push to use VMS as a server platform for a work project. For this to happen, though, VMS needs to actually have all the pieces in place to serve a modern web app. This means not just a web server, but things like PHP, Python, node.js, etc. And a DBMS that folks can actually use. While Oracle may be planning to port Rdb to x86 eventually, it's not there yet, and even when it comes, will not have any hobbyist option available to allow new users to get to know it. This needs to be fixed, either by beating Oracle with a clue stick until something happens (unlikely) or by updating the woefully outdated version of MariaDB currently available for VMS.

Documentation specifically aimed at new users who are primarily familiar with Linux of Window would be a help, as would a more reasonable set of default system parameter values that don't require an AUTOGEN cycle every time a significant software package is installed.

Mike

[Jan-Erik Söderholm]

As far as I know, and I have not seen that change lately, Rdb is free
to use for development or non-commercial use. That includes development
for commercial use. So if you are a buniess and have one development
and one production box, one of them is free.

Now, I'm no sure I see that making new VMS users queing up.

[Phillip Helbig (undress to reply)]

In article <sochlr$v4t$1...@dont-email.me>,
=?UTF-8?Q?Jan-Erik_S=c3=b6derholm?= <jan-erik....@telia.com>
writes:

> As far as I know, and I have not seen that change lately, Rdb is free
> to use for development

Yes.

> or non-commercial use.

No.

> That includes development
> for commercial use.

Actually, ONLY development for commercial use.

One is not allowed to use Rdb as a hobbyist, or for a non-profit entity
(which is by definition non-commercial), or whatever.

[Jan-Erik Söderholm]

OK. I do not know of any good source for it (in either way), but I seam
to remember from 6-7 years ago that it was OK for "hobbyist" use.

[Arne Vajhøj]

The license wording is:

<quote>
Oracle grants You a nonexclusive, nontransferable, limited license to
internally use the Programs, subject to the restrictions stated in this
Agreement, only for the purpose of developing, testing, prototyping, and
demonstrating Your application and only as long as Your application has
not been used for any data processing, business, commercial, or
production purposes, and not for any other purpose.
</quote>

I am not a lawyer but it seems pretty straight forward to me:
- development and test is OK no matter whether it is a commercial
software, open source or just for fun
- any usage to store real data whether it is commercial
or non-commercial (incl. catalog of your books) is not OK

Arne

[Phillip Helbig (undress to reply)]

In article <61aa18a0$0$700$1472...@news.sunsite.dk>,
=?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:

I seem to remember that "application" had been defined to be commercial.
Maybe my memory is wrong or maybe the wording has changed.

[Dave Froble]

Aren't all hobbyists working on a "commercial application" ?

They may never come up with something that sells, but they could still be
"working" on something.

:-)

--
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

[Arne Vajhøj]

On 12/3/2021 10:53 AM, Dave Froble wrote:
> Aren't all hobbyists working on a "commercial application" ?
>
> They may never come up with something that sells, but they could still
> be "working" on something.
>
> :-)

If they are working on something that they think would become
a commercial application then they should join ISV program instead
of CL program.

Arne

[Phillip Helbig (undress to reply)]

In article <sodehc$tf6$1...@dont-email.me>, Dave Froble

Not sure what the smiley means. Many, probably most or all, hobbyists
work on hobbies. :-) Some might also have day jobs involving VMS, of
course. They can work on many things, which is fine if they are not
commercial. Of course, not everything which is not commercial is a
hobby, but IIRC the Rdb license allowed only development of a commercial
application.

[Phillip Helbig (undress to reply)]

In article <61aa4030$0$699$1472...@news.sunsite.dk>,

=?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:

> On 12/3/2021 10:53 AM, Dave Froble wrote:

> > Aren't all hobbyists working on a "commercial application" ?
> >
> > They may never come up with something that sells, but they could still
> > be "working" on something.
> >
> > :-)
>

> If they are working on something that they think would become
> a commercial application then they should join ISV program instead
> of CL program.

For VMS, yes, of course, but what about for Rdb?

[Arne Vajhøj]

My understanding:

VMS Rdb
production commercial license commercial license
development ISV free developer license
fun CL free developer license

Arne

[Dave Froble]

For VMS I totally agree, and have advocated that in the past. But Arne, the
discussion was about Rdb.

[Simon Clubley]

On 2021-12-02, Mike K. <madcrow...@gmail.com> wrote:
> Honestly, I think that any new users will probably be coming out of the hobbyist/community license program. Basically people who try it out and use it for personal projects, and decide that they like it so much compared to Linux or Windows that when the time comes, they push to use VMS as a server platform for a work project. For this to happen, though, VMS needs to actually have all the pieces in place to serve a modern web app. This means not just a web server, but things like PHP, Python, node.js, etc. And a DBMS that folks can actually use. While Oracle may be planning to port Rdb to x86 eventually, it's not there yet, and even when it comes, will not have any hobbyist option available to allow new users to get to know it. This needs to be fixed, either by beating Oracle with a clue stick until something happens (unlikely) or by updating the woefully outdated version of MariaDB currently available for VMS.
>
> Documentation specifically aimed at new users who are primarily familiar with Linux of Window would be a help, as would a more reasonable set of default system parameter values that don't require an AUTOGEN cycle every time a significant software package is installed.
>

I think VSI have done some work with parameter defaults so at least
some of them will not be an issue on x86-64 VMS.

Is there any option in VMS where, if a system/process goes past some
percentage (say 80% to 90%) of any system parameter designed to limit
use of a resource, VMS will issue an OPCOM warning about that system
parameter (and maybe log it elsewhere as well) ? If not, would this
be a good option to add to VMS ?

BTW, I do think that at this point, the type of people wanting to
explore VMS without a specific reason are also the same type of
people who just want to explore z/OS because its there.

I wonder if there's any benefit to VSI (and VMS growth/retention in
general) if VSI construct and run a z/OS style Master the Mainframe program ?

IBM's Master the Mainframe program is here:

https://www.ibm.com/it-infrastructure/z/education/zxplore

Simon.

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

[Arne Vajhøj]

On 12/3/2021 12:52 PM, Dave Froble wrote:
> On 12/3/2021 11:05 AM, Arne Vajhøj wrote:
>> On 12/3/2021 10:53 AM, Dave Froble wrote:
>>> Aren't all hobbyists working on a "commercial application" ?
>>>
>>> They may never come up with something that sells, but they could
>>> still be
>>> "working" on something.
>>>
>>> :-)
>>
>> If they are working on something that they think would become
>> a commercial application then they should join ISV program instead
>> of CL program.
>

> For VMS I totally agree, and have advocated that in the past.  But Arne,
> the discussion was about Rdb.

They would not be hobbyists if they are working on something
that they think would become a commercial application.

Arne

[Arne Vajhøj]

On 12/3/2021 1:55 PM, Simon Clubley wrote:
> I think VSI have done some work with parameter defaults so at least
> some of them will not be an issue on x86-64 VMS.
>
> Is there any option in VMS where, if a system/process goes past some
> percentage (say 80% to 90%) of any system parameter designed to limit
> use of a resource, VMS will issue an OPCOM warning about that system
> parameter (and maybe log it elsewhere as well) ? If not, would this
> be a good option to add to VMS ?

I think VSI should get rid of >95% of SYSGEN and SYSUAF limits.

They made sense with a 256 KB VAX but not so much on a 256 GB x86-64.

Arne

[Bill Gunshannon]

No.... :-)

> They may never come up with something that sells, but they could still
> be "working" on something.
>
> :-)
>

bill

[DeanW]

On Thu, Dec 2, 2021 at 8:50 PM Mike K. via Info-vax <info...@rbnsn.com>
wrote:

> While Oracle may be planning to port Rdb to x86 eventually, it's not there
> yet, and even when it comes, will not have any hobbyist option available to
> allow new users to get to know it. This needs to be fixed, either by
> beating Oracle with a clue stick until something happens (unlikely) or by
> updating the woefully outdated version of MariaDB currently available for
> VMS.

If you want adoption- and this will be unpopular here- skip RDB and port
the current generation mainstream Oracle DB. There's already a huge
installed base for that, having VMS under it could be a selling point.

Don't hold your breath waiting for that; Oracle is pushing appliance-like
setups with favorable licensing / usage conditions over BYOPlatform
solutions these days to further set the vendor lock-in hooks into their own
products.

My current preference would be:
* VMS (implying modern networking, firewalling, logging to go with
traditional clustering strengths)
* WASD
* OpenJDK.current-ish, Python, Ruby?
* Postgres

Cluster that across a couple separate data centers (pick one or more of
Tier 3, Google, AMS, or Azure) and... what's the hook to make people learn
VMS?

Maybe the killer VMS platform is a rock-solid K8S / VM hosting platform.

But where I am currently doesn't know how to do a rolling upgrade, and
hours of downtime doesn't seem to bother any of TPTB so...

Missing being able to swap things around on the fly just by changing
logical names though.

[Phillip Helbig (undress to reply)]

In article <mailman.0.1638653703.1...@rbnsn.com>,

DeanW <dean.w...@gmail.com> writes:

> On Thu, Dec 2, 2021 at 8:50 PM Mike K. via Info-vax <info...@rbnsn.com>
> wrote:
>
> > While Oracle may be planning to port Rdb to x86 eventually, it's not there
> > yet, and even when it comes, will not have any hobbyist option available to
> > allow new users to get to know it.

Presumably it will have similar licensing to what it has now.

> > This needs to be fixed, either by
> > beating Oracle with a clue stick until something happens (unlikely) or by
> > updating the woefully outdated version of MariaDB currently available for
> > VMS.
>
> If you want adoption- and this will be unpopular here- skip RDB and port
> the current generation mainstream Oracle DB. There's already a huge
> installed base for that, having VMS under it could be a selling point.

Although I knew of some, there weren't that many Oracle Classic
installations under VMS even when VMS looked to have a long lifetime.
And VMS people are much more comfortable with Rdb (which is more VMS
like). At the application level, using standard SQL, does anyone even
have to know what database is running?

[John Dallman]

In article <sogpfo$18sn$1...@gioia.aioe.org>,
hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply))
wrote:

> At the application level, using standard SQL, does anyone even have to
> know what database is running?

Sadly, yes. SQL databases are not perfectly compatible:

<https://en.wikipedia.org/wiki/SQL#Interoperability_and_standardization>

The big database product my employer produces has a non-trivial
abstraction layer for SQL databases. It supports Oracle and MS SQL Server,
and has supported DB/2 in the past. The abstraction layer needs regular
maintenance to cope with changes in databases.

John

[DeanW]

On Sat, Dec 4, 2021 at 3:05 PM John Dallman via Info-vax <info...@rbnsn.com>
wrote:

> In article <sogpfo$18sn$1...@gioia.aioe.org>,
> hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply))
> wrote:
>
> > At the application level, using standard SQL, does anyone even have to
> > know what database is running?
>
> Sadly, yes. SQL databases are not perfectly compatible:
>

That's the wonderful thing about standards(sic)- you have so many to choose
from.

My management wants off Oracle, badly (hint, I work for Oregon gov't, see
"Oracle vs. Oregon") - I have to constantly explain to management that the
vendor specific features that previous generations of developers have taken
advantage of over the past 20+ years will not be cheap or easy to unravel.
Not just SQL dialect but how the security model is set up, which would take
a while to unravel and replicate; it'd be easier to start from scratch and
do a data migration. at the end, but that takes selling stakeholders on
more work than anyone thinks they want to engage in.

[Arne Vajhøj]

If the application use standard SQL-92 then the database does
not really matter )as long you can get a driver for it).

But sometimes vendor specific SQL is used.

And in the Oracle DB world stored procedures are very popular
and stored procedures are not portable among databases.

Having all business logic in a few thousand Oracle stored
procedures is a vendor lock-in that matters.

Arne

[Arne Vajhøj]

On 12/4/2021 4:34 PM, DeanW wrote:
> On Thu, Dec 2, 2021 at 8:50 PM Mike K. via Info-vax <info...@rbnsn.com>
> wrote:
>> While Oracle may be planning to port Rdb to x86 eventually, it's not there
>> yet, and even when it comes, will not have any hobbyist option available to
>> allow new users to get to know it. This needs to be fixed, either by
>> beating Oracle with a clue stick until something happens (unlikely) or by
>> updating the woefully outdated version of MariaDB currently available for
>> VMS.
>
> If you want adoption- and this will be unpopular here- skip RDB and port
> the current generation mainstream Oracle DB. There's already a huge
> installed base for that, having VMS under it could be a selling point.

I am not so sure.

Oracle is making a ton of money on Oracle DB. And relative few customers
are migrating away from Oracle DB.

But very few is migrating to or starting new from scratch on Oracle DB.
When they have the choice they go for more low cost options.

And migrating from Oracle DB on Linux to Oracle DB on VMS would probably
be a hard sell.

> Don't hold your breath waiting for that;

That would be unwise.

:-)

> My current preference would be:
> * VMS (implying modern networking, firewalling, logging to go with
> traditional clustering strengths)
> * WASD

If you need something VMSish then WASD would be a fine choice.

But if you need something industry standard then it is Apache or nginx.

> * OpenJDK.current-ish, Python, Ruby?

Hopefully VSI will get a OpenJDK 17 out soon as 8 is getting old (and I
do not see much point in 11 at this point in time).

Python exist.

Not sure about Ruby.

> * Postgres

Promised.

> Maybe the killer VMS platform is a rock-solid K8S / VM hosting platform.

Most want bare metal hypervisor for VM's in production.

And k8s very Linux centric. Microsoft support k8s for Windows. VSI could
support k8s for VMS. But Linux market share is sky high.

Arne

[Arne Vajhøj]

On 12/4/2021 5:46 PM, John Dallman wrote:
> In article <sogpfo$18sn$1...@gioia.aioe.org>,
> hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply))
> wrote:
>> At the application level, using standard SQL, does anyone even have to
>> know what database is running?
>
> Sadly, yes. SQL databases are not perfectly compatible:
>
> <https://en.wikipedia.org/wiki/SQL#Interoperability_and_standardization>

If applications are written with the intention to be database
independent and effort is put in to stick to standard SQL-92,
then migrating is not a problem.

But if no such requirements has been defined because "we will never
change database", then ...

> The big database product my employer produces has a non-trivial
> abstraction layer for SQL databases. It supports Oracle and MS SQL Server,
> and has supported DB/2 in the past. The abstraction layer needs regular
> maintenance to cope with changes in databases.

That sounds like a pretty crappy abstraction layer.

Common ORM frameworks like Hibernate, EclipseLink, OpenJPA, Entity
Framework, NHibernate etc. got support for almost everything
either from the framework itself or supplied by the database
vendor.

Arne

[William]

Some sort of free license so that open source developers can do OpenVMS ports and so that sites like GitLab can give projects an option of including OpenVMS in CI pipelines. Otherwise, why would someone want to start a project on an OS that few people know and that has an old and limited set of ported tools?

[Chris Townley]

On 05/12/2021 08:57, William wrote:
> Some sort of free license so that open source developers can do OpenVMS ports and so that sites like GitLab can give projects an option of including OpenVMS in CI pipelines. Otherwise, why would someone want to start a project on an OS that few people know and that has an old and limited set of ported tools?
>

Isn't the community/hobbyist license just that? Or the ISV scheme?

--
Chris

[John Dallman]

In article <61ac02f5$0$704$1472...@news.sunsite.dk>, ar...@vajhoej.dk

(Arne Vajhøj) wrote:

> That sounds like a pretty crappy abstraction layer.

When you need the best performance you can reasonably get, using
manufacturer extensions becomes necessary.


John

[Phillip Helbig (undress to reply)]

In article <soi3r4$8c2$1...@dont-email.me>, Chris Townley

<ne...@cct-net.co.uk> writes:

> On 05/12/2021 08:57, William wrote:
> > Some sort of free license so that open source developers can do
OpenVMS ports and so that sites like GitLab can give projects an option
of including OpenVMS in CI pipelines.

> Isn't the community/hobbyist license just that? Or the ISV scheme?

Hobbyist? No. ISV? Probably.

[Chris Townley]

From the VSI Website:

With the VSI Community Liсense Program, members of the community can
download OpenVMS for free for learning, open source development, and
exchanging knowledge of the operating system on Alpha and Integrity
systems. This program replaces the HP Hobbyist program.


--
Chris

[Phillip Helbig (undress to reply)]

In article <soibcl$1ee$1...@dont-email.me>, Chris Townley

<ne...@cct-net.co.uk> writes:

> On 05/12/2021 11:30, Phillip Helbig (undress to reply) wrote:
> > In article <soi3r4$8c2$1...@dont-email.me>, Chris Townley
> > <ne...@cct-net.co.uk> writes:
> >
> >> On 05/12/2021 08:57, William wrote:
> >>> Some sort of free license so that open source developers can do
> > OpenVMS ports and so that sites like GitLab can give projects an option
> > of including OpenVMS in CI pipelines.
> >
> >> Isn't the community/hobbyist license just that? Or the ISV scheme?
> >
> > Hobbyist? No. ISV? Probably.
> >
>
> From the VSI Website:
>

> With the VSI Community License Program, members of the community can

> download OpenVMS for free for learning, open source development, and
> exchanging knowledge of the operating system on Alpha and Integrity
> systems. This program replaces the HP Hobbyist program.

The wording has changed a bit. But it is still non-commercial.
Open-source can be non-commercial, but it can also be commercial.

[Arne Vajhøj]

Yes - some open source is commercial - and would not fall
under CL terms - but they would definitely fall under ISV
terms.

But the 99.9% of the open source for VMS that is non-commercial
will be fine with CL.

Arne

[Dave Froble]

Isn't there a difference between working on commercial software, such as porting
to VMS, and using commercial software?

[Phillip Helbig (undress to reply)]

In article <soin99$1ac$1...@dont-email.me>, Dave Froble

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

> >> With the VSI Community License Program, members of the community can
> >> download OpenVMS for free for learning, open source development, and
> >> exchanging knowledge of the operating system on Alpha and Integrity
> >> systems. This program replaces the HP Hobbyist program.
> >
> > The wording has changed a bit. But it is still non-commercial.
> > Open-source can be non-commercial, but it can also be commercial.
>
> Isn't there a difference between working on commercial software, such
> as porting to VMS, and using commercial software?

Using commercial software has nothing to do with VMS licenses.

AT BEST, the hobbyist license could be construed to be valid for working
on commercial software, but without receiving any sort of financial
reward, directly or indirectly, for doing so.

[Bill Gunshannon]

And yet, for at least the past 15 years, anytime I have brought
up a need for efficient programming I am told (by Professors
and practitioners of the trade) that it is irrelevant, "Just
throw more hardware at it!" Wouldn't the same apply to using
non-standard extensions to get "the best performance you can
reasonably get"? We're not running VAXen any more and we have
more than 12MB of memory.

bill

[John Dallman]

In article <j148c3...@mid.individual.net>, bill.gu...@gmail.com

(Bill Gunshannon) wrote:

> And yet, for at least the past 15 years, anytime I have brought
> up a need for efficient programming I am told (by Professors
> and practitioners of the trade) that it is irrelevant, "Just
> throw more hardware at it!" Wouldn't the same apply to using
> non-standard extensions to get "the best performance you can
> reasonably get"? We're not running VAXen any more and we have
> more than 12MB of memory.

When you're dealing with databases in the terrabyte range, efficiency
becomes important. Throwing more hardware at problems hits two
limitations: (a) customers' budgets, when you're an ISV, and (b) the lack
of available higher-powered hardware after a certain point.

Academics sometimes don't seem to understand the scale of the data sets
in commercial use. Were the "practitioners" people actually engaged in
delivering systems to customers, or were they consultants?

John

[Dave Froble]

On 12/5/2021 11:40 AM, Bill Gunshannon wrote:
> On 12/5/21 5:52 AM, John Dallman wrote:
>> In article <61ac02f5$0$704$1472...@news.sunsite.dk>, ar...@vajhoej.dk
>> (Arne Vajhøj) wrote:
>>
>>> That sounds like a pretty crappy abstraction layer.
>>
>> When you need the best performance you can reasonably get, using
>> manufacturer extensions becomes necessary.
>>
>>
>> John
>>
>
>
> And yet, for at least the past 15 years, anytime I have brought
> up a need for efficient programming I am told (by Professors

Those who can, do, those who can't, teach ...

> and practitioners of the trade)

Definitely not this one ...

> that it is irrelevant, "Just
> throw more hardware at it!" Wouldn't the same apply to using
> non-standard extensions to get "the best performance you can
> reasonably get"? We're not running VAXen any more and we have
> more than 12MB of memory.

That is an issue with multiple arguments, both for and con ...

[Phillip Helbig (undress to reply)]

In article <soivu2$202$1...@dont-email.me>, Dave Froble

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

> Those who can, do, those who can't, teach ...

Those who can't, teach gym.

[Dave Froble]

On 12/5/2021 10:55 AM, Phillip Helbig (undress to reply) wrote:
> In article <soin99$1ac$1...@dont-email.me>, Dave Froble
> <da...@tsoft-inc.com> writes:
>
>>>> With the VSI Community License Program, members of the community can
>>>> download OpenVMS for free for learning, open source development, and
>>>> exchanging knowledge of the operating system on Alpha and Integrity
>>>> systems. This program replaces the HP Hobbyist program.
>>>
>>> The wording has changed a bit. But it is still non-commercial.
>>> Open-source can be non-commercial, but it can also be commercial.
>>
>> Isn't there a difference between working on commercial software, such
>> as porting to VMS, and using commercial software?
>
> Using commercial software has nothing to do with VMS licenses.

Huh ?????????????????

> AT BEST, the hobbyist license could be construed to be valid for working
> on commercial software, but without receiving any sort of financial
> reward, directly or indirectly, for doing so.
>

[Phillip Helbig (undress to reply)]

In article <soj03a$202$2...@dont-email.me>, Dave Froble

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

> >>>> With the VSI Community License Program, members of the community can
> >>>> download OpenVMS for free for learning, open source development, and
> >>>> exchanging knowledge of the operating system on Alpha and Integrity
> >>>> systems. This program replaces the HP Hobbyist program.
> >>>
> >>> The wording has changed a bit. But it is still non-commercial.
> >>> Open-source can be non-commercial, but it can also be commercial.
> >>
> >> Isn't there a difference between working on commercial software, such
> >> as porting to VMS, and using commercial software?
> >
> > Using commercial software has nothing to do with VMS licenses.
>

> Huh ?????????????????

????????

Obviously VMS is commercial and its use involves VMS licenses. That is
not the topic here. The question is whether the hobbyist license would
cover open-source development on VMS for commercial software and, if so,
if that would be the case only if the developer received no
compensation.

[David Goodwin]

On Friday, December 3, 2021 at 5:44:29 PM UTC+13, Mike K. wrote:
> Honestly, I think that any new users will probably be coming out of the hobbyist/community license program. Basically people who try it out and use it for personal projects, and decide that they like it so much compared to Linux or Windows that when the time comes, they push to use VMS as a server platform for a work project. For this to happen, though, VMS needs to actually have all the pieces in place to serve a modern web app. This means not just a web server, but things like PHP, Python, node.js, etc. And a DBMS that folks can actually use. While Oracle may be planning to port Rdb to x86 eventually, it's not there yet, and even when it comes, will not have any hobbyist option available to allow new users to get to know it. This needs to be fixed, either by beating Oracle with a clue stick until something happens (unlikely) or by updating the woefully outdated version of MariaDB currently available for VMS.
>
> Documentation specifically aimed at new users who are primarily familiar with Linux of Window would be a help, as would a more reasonable set of default system parameter values that don't require an AUTOGEN cycle every time a significant software package is installed.

Even if OpenVMS had all the pieces in place and could run modern workloads, why choose OpenVMS over Linux? What can OpenVMS do that Linux simply cant? What can it do significantly better than Linux? What makes the license cost and yearly renewals worth it?

Right now Linux is the native platform for most major software development tools and a lot (most?) of the server software people care about. Here OpenVMS has some of the same challenges as Windows. Why pay extra to run Linux software on Windows when you could just do what everyone else is doing and run Linux software on Linux? Why pay extra for worse support and extra bugs because windows ports are often a lower priority?

Are there any proprietary operating systems that have actually managed to take market share from Linux besides MacOS on the desktop? All the proprietary unixes are basically dead (aside from Solaris which survives in open-source form as Illumos) and Windows Server is probably loosing market share outside of Azure. I doubt IBM is gaining a lot of new mainframe customers.

[Dave Froble]

Further down the list is "write for trade rags" ..

You know, the ones everyone listens to ..

[Arne Vajhøj]

On 12/5/2021 5:52 AM, John Dallman wrote:
> In article <61ac02f5$0$704$1472...@news.sunsite.dk>, ar...@vajhoej.dk

> (Arne Vajh�j) wrote:
>> That sounds like a pretty crappy abstraction layer.
>
> When you need the best performance you can reasonably get, using
> manufacturer extensions becomes necessary.

There are definitely such cases.

But there are way more cases where vendor specific extensions
are used without it being necessary.

Anyway the abstraction layer is obviously not improving
performance, so I assume that it is hiding the vendor
specific stuff.

But hiding vendor specific stuff and only support 2 vendors
is still not great.

Arne

[Bill Gunshannon]

What do you think? Some of them were here. :-)

bill

[Arne Vajhøj]

On 12/5/2021 11:40 AM, Bill Gunshannon wrote:

> On 12/5/21 5:52 AM, John Dallman wrote:
>> In article <61ac02f5$0$704$1472...@news.sunsite.dk>, ar...@vajhoej.dk
>> (Arne Vajhøj) wrote:
>>
>> When you need the best performance you can reasonably get, using
>> manufacturer extensions becomes necessary.
>

> And yet, for at least the past 15 years, anytime I have brought
> up a need for efficient programming I am told (by Professors
> and practitioners of the trade) that it is irrelevant, "Just
> throw more hardware at it!"  Wouldn't the same apply to using
> non-standard extensions to get "the best performance you can
> reasonably get"?  We're not running VAXen any more and we have
> more than 12MB of memory.

Performance is still very important today.

But what was relevant for good performance 30 years ago and
what is relevant for good performance today is very different.

Writing complex code to save a little CPU or memory is
typical a total waste of time today. Programming
languages and compiler settings are also often without
significance today. The change in HW prices has done that.

(typical is a normal business application - there are
exceptions in HPC simulations, cryptography brute force
attacks and other special cases)

But there are plenty of performance problems in the real
world. Data being stored the wrong way, data being accessed
the wrong way, too many network hops, something being interpreted
that should have been compiled, bad algorithms etc..

In general I will say that bad performance today is typical
due to error in architecture or high level design and rarely
due to error in coding.

Arne

[Arne Vajhøj]

On 12/5/2021 5:52 AM, John Dallman wrote:
> In article <61ac02f5$0$704$1472...@news.sunsite.dk>, ar...@vajhoej.dk

> (Arne Vajh�j) wrote:
>> That sounds like a pretty crappy abstraction layer.
>
> When you need the best performance you can reasonably get, using
> manufacturer extensions becomes necessary.

Can you reveal what the product is?

I was sort of thinking of LLBLGen, but it looks they support
a lot more databases.

Arne

[Arne Vajhøj]

On 12/5/2021 6:14 PM, David Goodwin wrote:
> Even if OpenVMS had all the pieces in place and could run modern
> workloads, why choose OpenVMS over Linux? What can OpenVMS do that
> Linux simply cant? What can it do significantly better than Linux?
> What makes the license cost and yearly renewals worth it?
>
> Right now Linux is the native platform for most major software
> development tools and a lot (most?) of the server software people
> care about. Here OpenVMS has some of the same challenges as Windows.
> Why pay extra to run Linux software on Windows when you could just do
> what everyone else is doing and run Linux software on Linux? Why pay
> extra for worse support and extra bugs because windows ports are
> often a lower priority?
>
> Are there any proprietary operating systems that have actually
> managed to take market share from Linux besides MacOS on the desktop?
> All the proprietary unixes are basically dead (aside from Solaris
> which survives in open-source form as Illumos) and Windows Server is
> probably loosing market share outside of Azure. I doubt IBM is
> gaining a lot of new mainframe customers.

Interesting question.

VSI is currently focusing a lot on existing VMS customers and
not so much on how to get new VMS customers.

But new VMS customers would of course be nice.

First I think we need to realize that VMS is unlikely to ever compete
with Linux in the commodity server market.

If you need to run thousands of servers with PHP or node.js, then
the OS price matters. VMS will be too expensive. Heck - a lot
of them consider RHEL too expensive and goes for RockyLinux or
Ubuntu.

VMS market will definitely be the slightly more expensive
server market where the OS cost is a very small part of
total cost and competitors are RHEL and Windows Server.

I don't think OS cost is a big issue in healthcare, finance,
defense etc..

Going after that market combined with a pricing that is competitive
with RHEL and Windows Server will handle the cost issue.

(HW cost of VMS boxes will be resolved by the x86-64 port)

And even though this market may only be like 10% of the server
market then the market is still more than big enough for VMS.

It is obviously a prerequisite for VMS to get new customers to
provide a decent set of tools and platform products - otherwise
VMS will be disqualified.

But that still leaves the question why would new customer choose
VMS over Linux.

Maybe VSI can try pushing "less is more" / "small is good". Both
Linux and Windows has become huge. To some extent they can be tailored
down (especially Linux), but still a simple OS may have a justification.

Arne

[John Dallman]

In article <61ad608b$0$706$1472...@news.sunsite.dk>, ar...@vajhoej.dk

(Arne Vajhøj) wrote:

> Can you reveal what the product is?
>
> I was sort of thinking of LLBLGen, but it looks they support
> a lot more databases.

No. I can speak more freely if I don't reveal who I work for. I can say
that that product does not run on VMS, and I don't think it ever has.

I work on a quite different product, which has run on VMS in the deep
past (VAX and Alpha, never Itanium). I started reading c.o.v because
there is a faint possibility of needing to revive VMS support for x86-64.



John

[John Dallman]

In article <61ad590c$0$700$1472...@news.sunsite.dk>, ar...@vajhoej.dk

(Arne Vajhøj) wrote:

> Anyway the abstraction layer is obviously not improving
> performance, so I assume that it is hiding the vendor
> specific stuff.
>
> But hiding vendor specific stuff and only support 2 vendors
> is still not great.

More vendors have been supported in the past; the current two are
apparently the only ones that many customers want to use, at present.

John

[Simon Clubley]

On 2021-12-03, Arne Vajhøj <ar...@vajhoej.dk> wrote:
> On 12/3/2021 1:55 PM, Simon Clubley wrote:
>> I think VSI have done some work with parameter defaults so at least
>> some of them will not be an issue on x86-64 VMS.
>>
>> Is there any option in VMS where, if a system/process goes past some
>> percentage (say 80% to 90%) of any system parameter designed to limit
>> use of a resource, VMS will issue an OPCOM warning about that system
>> parameter (and maybe log it elsewhere as well) ? If not, would this
>> be a good option to add to VMS ?
>
> I think VSI should get rid of >95% of SYSGEN and SYSUAF limits.
>
> They made sense with a 256 KB VAX but not so much on a 256 GB x86-64.
>

Agreed, but it would be nice if they would retain an overall ability
to stop a single runaway process from gobbling up all the resources
without at the same time having to micromanage resources as needed to
be done on the resource limited systems of old.

Simon.

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

[Arne Vajhøj]

On 12/6/2021 1:29 PM, Simon Clubley wrote:
> On 2021-12-03, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>> On 12/3/2021 1:55 PM, Simon Clubley wrote:
>>> I think VSI have done some work with parameter defaults so at least
>>> some of them will not be an issue on x86-64 VMS.
>>>
>>> Is there any option in VMS where, if a system/process goes past some
>>> percentage (say 80% to 90%) of any system parameter designed to limit
>>> use of a resource, VMS will issue an OPCOM warning about that system
>>> parameter (and maybe log it elsewhere as well) ? If not, would this
>>> be a good option to add to VMS ?
>>
>> I think VSI should get rid of >95% of SYSGEN and SYSUAF limits.
>>
>> They made sense with a 256 KB VAX but not so much on a 256 GB x86-64.
>
> Agreed, but it would be nice if they would retain an overall ability
> to stop a single runaway process from gobbling up all the resources
> without at the same time having to micromanage resources as needed to
> be done on the resource limited systems of old.

Sure. But if we have a MAXCPU, a MAXMEMORY and a MAXOPENFILES
then how much more do we need.

Arne

[Simon Clubley]

On 2021-12-05, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>
> But that still leaves the question why would new customer choose
> VMS over Linux.
>
> Maybe VSI can try pushing "less is more" / "small is good". Both
> Linux and Windows has become huge. To some extent they can be tailored
> down (especially Linux), but still a simple OS may have a justification.
>

The one thing you have not addressed in your reply are the time-limited
production licences.

Why would a manager new to the VMS world make a decision where you had
to pay for new licences for your newly acquired systems on a regular
basis or those new systems would suddenly stop working ?

It's one thing to stop getting support if a vendor goes bust.

It's a completely different thing to have your systems suddenly stop
working if a vendor previously unknown to them (VSI) suddenly goes bust.

Why would they ever agree to that ?

Even _if_ they could be convinced VSI would be taken over, how do you
address the question of a VSI collapse happening right when those new
licences were needed ?

How do you address the question that the customer has no control over the
price of the renewal licences and that a replacement company could charge
a stiff price increase because they have a captive market ?

Why would a manager new to the VMS world ever agree to that when they
have their job and their pension to think about and have several other
alternatives where this simply is not a concern to the manager ?

People here are looking at pushing VMS from a technical viewpoint.

I OTOH am looking at this from a human nature viewpoint of managers
in a client company who want to keep their jobs.

[Simon Clubley]

Not as many as we currently have. :-)

[Simon Clubley]

On 2021-12-05, Chris Townley <ne...@cct-net.co.uk> wrote:
> On 05/12/2021 11:30, Phillip Helbig (undress to reply) wrote:
>> In article <soi3r4$8c2$1...@dont-email.me>, Chris Townley
>> <ne...@cct-net.co.uk> writes:
>>
>>> On 05/12/2021 08:57, William wrote:
>>>> Some sort of free license so that open source developers can do
>> OpenVMS ports and so that sites like GitLab can give projects an option
>> of including OpenVMS in CI pipelines.
>>
>>> Isn't the community/hobbyist license just that? Or the ISV scheme?
>>
>> Hobbyist? No. ISV? Probably.
>>
>
> From the VSI Website:
>

> With the VSI Community Li?ense Program, members of the community can

> download OpenVMS for free for learning, open source development, and
> exchanging knowledge of the operating system on Alpha and Integrity
> systems. This program replaces the HP Hobbyist program.
>

I just had a look at the licence itself to see if it makes things
clearer and it actually confuses things a little bit more. :-)

|Non-commercial means not used for commercial advantage, direct monetary
|compensation, or indirect monetary compensation.

What is "indirect monetary compensation" ? Would this include getting
a pay rise or a new job as a result of learning a new skill using the
hobbyist licence ?

|You will not modify, reverse engineer, disassemble, decrypt, decompile, or
|make derivative works of the Software. Where You have other rights mandated
|under statute, You will provide VSI with reasonably detailed information
|regarding any intended modifications, reverse engineering, disassembly,
|decryption, or decompilation and the purposes therefore.

Does this prohibit some forms of security research ?

|In partial consideration of the license granted hereunder, you agree to
|provide reasonable participation in the online Software community forums,
|including without limitation such activities as answering questions and
|contributing articles and how-to videos to the OpenVMS online Software
|community at https://forum.vmssoftware.com.

Do VSI enforce this, or would taking part in comp.os.vms instead be enough ?

[Simon Clubley]

I think David might be asking if you can run commercial software on
a hobbyist system.

For example, if you could somehow get hold of a legal copy of Word Perfect
for Alpha, could you run it on a hobbyist system for personal use ?

If so, that's a good question, and I don't know the answer.

[Arne Vajhøj]

A license is an agreement between the software vendor and the software
user and can include limitations on software use.

VSI can and does limit use of VMS under CL. And given that an OS is
used for everything on a system then it practically limits the use
of the entire system.

So a CL licensed system cannot be used for anything commercial.

It does not make any sense for VSI to limit what SW can be installed
on the system and what licenses they come under. And I do not remember
there being anything like that in the CL license.

So I cannot see why you cannot install your paid WP software
on the system - just note that even though the WP license may
permit it then the VMS license still prohibits commercial
work including in WP as WP use the OS.

Arne

[Arne Vajhøj]

On 12/6/2021 2:07 PM, Simon Clubley wrote:
> On 2021-12-05, Chris Townley <ne...@cct-net.co.uk> wrote:
>> On 05/12/2021 11:30, Phillip Helbig (undress to reply) wrote:
>>> In article <soi3r4$8c2$1...@dont-email.me>, Chris Townley
>>> <ne...@cct-net.co.uk> writes:
>>>
>>>> On 05/12/2021 08:57, William wrote:
>>>>> Some sort of free license so that open source developers can do
>>> OpenVMS ports and so that sites like GitLab can give projects an option
>>> of including OpenVMS in CI pipelines.
>>>
>>>> Isn't the community/hobbyist license just that? Or the ISV scheme?
>>>
>>> Hobbyist? No. ISV? Probably.
>>
>> From the VSI Website:
>>
>> With the VSI Community Li?ense Program, members of the community can
>> download OpenVMS for free for learning, open source development, and
>> exchanging knowledge of the operating system on Alpha and Integrity
>> systems. This program replaces the HP Hobbyist program.
>>
>
> I just had a look at the licence itself to see if it makes things
> clearer and it actually confuses things a little bit more. :-)
>
> |Non-commercial means not used for commercial advantage, direct monetary
> |compensation, or indirect monetary compensation.
>
> What is "indirect monetary compensation" ? Would this include getting
> a pay rise or a new job as a result of learning a new skill using the
> hobbyist licence ?

Not likely. Impossible not to learn something that could potentially
be beneficial in some way.

Common sense says:

Commercial advantage => you cannot run your businesses accounting on the
system.

Direct monetary compensation => you cannot let your neighbor run his
business accounting on the system for 50 pound per month.

Indirect monetary compensation => you cannot let your neighbor run his
business accounting on the system and he clears snow from your driveway
in return.

> |You will not modify, reverse engineer, disassemble, decrypt, decompile, or
> |make derivative works of the Software. Where You have other rights mandated
> |under statute, You will provide VSI with reasonably detailed information
> |regarding any intended modifications, reverse engineering, disassembly,
> |decryption, or decompilation and the purposes therefore.
>
> Does this prohibit some forms of security research ?

Seems likely. The prohibition against disassemble and decompile could
be a problem.

> |In partial consideration of the license granted hereunder, you agree to
> |provide reasonable participation in the online Software community forums,
> |including without limitation such activities as answering questions and
> |contributing articles and how-to videos to the OpenVMS online Software
> |community at https://forum.vmssoftware.com.
>
> Do VSI enforce this,

I have never heard of it.

> or would taking part in comp.os.vms instead be enough ?

It is pretty broadly worded.

Arne

[Arne Vajhøj]

There is an outstanding issue about the expiring licenses.

But I think the OP was looking for something more general/fundamental -
this issue could be fixed in no time by VSI.

Arne

PS: And you know how I think VSI should fix this.

[Dave Froble]

On 12/6/2021 2:12 PM, Simon Clubley wrote:

No. David took one look at the statement:

"Using commercial software has nothing to do with VMS licenses."

Now, if Phillip meant to refer to hobbyist or CL or ISV licenses, he should have
written that. He didn't.

[Dave Froble]

On 12/6/2021 1:29 PM, Simon Clubley wrote:

> On 2021-12-03, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>> On 12/3/2021 1:55 PM, Simon Clubley wrote:
>>> I think VSI have done some work with parameter defaults so at least
>>> some of them will not be an issue on x86-64 VMS.
>>>
>>> Is there any option in VMS where, if a system/process goes past some
>>> percentage (say 80% to 90%) of any system parameter designed to limit
>>> use of a resource, VMS will issue an OPCOM warning about that system
>>> parameter (and maybe log it elsewhere as well) ? If not, would this
>>> be a good option to add to VMS ?
>>
>> I think VSI should get rid of >95% of SYSGEN and SYSUAF limits.
>>
>> They made sense with a 256 KB VAX but not so much on a 256 GB x86-64.
>>
>
> Agreed, but it would be nice if they would retain an overall ability
> to stop a single runaway process from gobbling up all the resources
> without at the same time having to micromanage resources as needed to
> be done on the resource limited systems of old.
>
> Simon.
>

I see no reason to remove system parameters. I think that many parameters could
be evaluated from the perspective of the anticipated systems being used, and
where there is no reason to not do so, set defaults to take advantage of those
anticipated systems. This should get rid of much of the need for AUTOGEN. This
should get rid of most of the need for any tuning.

However, getting rid of the parameters would then not allow specific tuning for
the perhaps handful of uses where such tuning might be necessary. I'm mainly
thinking of desired restrictions. Just because I cannot think of any such usage
doesn't mean it will never happen.

[abrsvc]

It would seem to me that the cost to update VMS to remove the parameters would be more than VSI wants to pay. Why remove them, make them arbitrarily high (or low) such that they are effectively "not there". Why make all kinds of source changes just to remove something that has little overhead? Do the values change over time, yes. It is difficult to make those changes, NO. Are changes made frequently, NO. Seems to me that this is one case where "if it aint broke..."

Haing the flexibility to alter how the system uses resources will always benefit someone somewhere. I would rather have control than not. But I have had the pleasure of using this system for 40 years and know it well enough to know how to use those controls effectively. Yes, in the hands of people that don't know better, bad things can happen, but the same can be said of other areas too.

[Dave Froble]

I'll always remember one event. A customer decided to adjust default working
set rather high, "for better performance". Unfortunately, when starting up the
process, the OS had to go and gather all that memory and assign it to the new
process. Not only was the login rather slow, but, the rest of the running
processes got real slow.

Me: Did you read the manual before playing with the system parameters?
Marie: No.
Me: Did you think the manual should be ignored?
Marie: No.

Now women are never wrong, right? I could see Marie was getting a bit upset
with my attitude. So I backed off, explained what she had done, and settled for
a promise (do women keep them?) that she would ask me before doing such things,
and saved my life.

:-)

[Phillip Helbig (undress to reply)]

In article <solnaj$uvn$5...@dont-email.me>, Simon Clubley

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

> >> > Using commercial software has nothing to do with VMS licenses.
> >>
> >> Huh ?????????????????
> >
> > ????????
> >
> > Obviously VMS is commercial and its use involves VMS licenses. That is
> > not the topic here. The question is whether the hobbyist license would
> > cover open-source development on VMS for commercial software and, if so,
> > if that would be the case only if the developer received no
> > compensation.
>

> I think David might be asking if you can run commercial software on
> a hobbyist system.
>
> For example, if you could somehow get hold of a legal copy of Word Perfect
> for Alpha, could you run it on a hobbyist system for personal use ?
>
> If so, that's a good question, and I don't know the answer.

VMS licenses are relevant for running VMS. It is clear that the
hobbyist license is not applicable if I use VMS, or anything running on
it, for commercial purposes, meaning that I profit from it. But if
someone write a program which does something which I, as a hobbyist,
want to do---say, just as an example, that it prints out prime
numbers---and charges money for it (thus it is commercial for that
programmer), what type of license VMS is running under is irrelevant.

I can barely conceive of the mindset which thinks that a commercial VMS
license would be needed to run a program which costs money.

[Phillip Helbig (undress to reply)]

In article <61ae67a5$0$693$1472...@news.sunsite.dk>,

=?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:

> >>>> Using commercial software has nothing to do with VMS licenses.
> >>>
> >>> Huh ?????????????????
> >>
> >> ????????
> >>
> >> Obviously VMS is commercial and its use involves VMS licenses. That is
> >> not the topic here. The question is whether the hobbyist license would
> >> cover open-source development on VMS for commercial software and, if so,
> >> if that would be the case only if the developer received no
> >> compensation.
> >
> > I think David might be asking if you can run commercial software on
> > a hobbyist system.
> >
> > For example, if you could somehow get hold of a legal copy of Word Perfect
> > for Alpha, could you run it on a hobbyist system for personal use ?
> >
> > If so, that's a good question, and I don't know the answer.
>

> A license is an agreement between the software vendor and the software
> user and can include limitations on software use.
>
> VSI can and does limit use of VMS under CL. And given that an OS is
> used for everything on a system then it practically limits the use
> of the entire system.
>
> So a CL licensed system cannot be used for anything commercial.
>
> It does not make any sense for VSI to limit what SW can be installed
> on the system and what licenses they come under. And I do not remember
> there being anything like that in the CL license.
>
> So I cannot see why you cannot install your paid WP software
> on the system - just note that even though the WP license may
> permit it then the VMS license still prohibits commercial
> work including in WP as WP use the OS.

Right. "Commercial" with regard to VMS licenses means whether the owner
of that license profits or not. What other software is running on the
system and what licenses it requires is completely irrelevant.

[Phillip Helbig (undress to reply)]

In article <som4dg$edu$1...@dont-email.me>, Dave Froble

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

> On 12/6/2021 2:12 PM, Simon Clubley wrote:

> No. David took one look at the statement:
>

> "Using commercial software has nothing to do with VMS licenses."
>

> Now, if Phillip meant to refer to hobbyist or CL or ISV licenses, he should have
> written that. He didn't.

The statement quoted immediately above was not written by me, merely
quoted by me.

[Phillip Helbig (undress to reply)]

In article <soms96$1nen$3...@gioia.aioe.org>,
hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply))
writes:

> In article <som4dg$edu$1...@dont-email.me>, Dave Froble

> <da...@tsoft-inc.com> writes:
>
> > On 12/6/2021 2:12 PM, Simon Clubley wrote:

> > No. David took one look at the statement:
> >

> > "Using commercial software has nothing to do with VMS licenses."
> >

> > Now, if Phillip meant to refer to hobbyist or CL or ISV licenses, he should have
> > written that. He didn't.
>
> The statement quoted immediately above was not written by me, merely
> quoted by me.

Sorry, it was. See below. But when I wrote it, the text I quoted makes
it clear that I was referring to the VSI community license.

----------8<--------------------------------------------------------------------

Path: aioe.org!1XSHQO4JptEK3D6Pc9T+OA.user.46.165.242.75.POSTED!not-for-mail
From: hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply))
Newsgroups: comp.os.vms
Subject: Re: What Will Drive More OpenVMS Adoption?
Date: Sun, 5 Dec 2021 15:55:48 -0000 (UTC)
Organization: Multivax C&R
Message-ID: <soine4$ovr$1...@gioia.aioe.org>
Xref: aioe.org comp.os.vms:132863

In article <soin99$1ac$1...@dont-email.me>, Dave Froble

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

> >> With the VSI Community License Program, members of the community can
> >> download OpenVMS for free for learning, open source development, and
> >> exchanging knowledge of the operating system on Alpha and Integrity
> >> systems. This program replaces the HP Hobbyist program.
> >
> > The wording has changed a bit. But it is still non-commercial.
> > Open-source can be non-commercial, but it can also be commercial.
>
> Isn't there a difference between working on commercial software, such
> as porting to VMS, and using commercial software?

Using commercial software has nothing to do with VMS licenses.

AT BEST, the hobbyist license could be construed to be valid for working
on commercial software, but without receiving any sort of financial
reward, directly or indirectly, for doing so.

[Phillip Helbig (undress to reply)]

> "Using commercial software has nothing to do with VMS licenses."

I guess that it can depend on what one means by "commercial software". I
took it to mean "software which has been purchased" and, indeed, as far
as I know, that has nothing to do with what type of VMS license one has.
Some might take it to mean "software with which money is earned"
(whether that software was purchased, obtained for free, or
self-written), in which case, of course, a commercial VMS license is
needed.

[Bill Gunshannon]

Cute. I may have told this one here before but just in case I haven't.
:-)

In another lifetime I worked with COBOL and Fortran on a Univac 1100
running Exec-8. Exec-8 has a control language like JCL. One of the
uses is to assign memory for processes before they are run. I got
assigned the project of finding out why a certain program would stop
running for an extended period of time and then just continue as if
nothing had happened. Turned out someone thought the COBOL SORT was
going to need a lot of memory so they added a parameter tot he ECL
stream to ask for all the system memory before running. So, every
time the process was run it would run fine until it hit that point.
Then it would sit there waiting for every other process to end grabbing
the memory each time until it had all the memory. Of course this also
stopped other jobs from coming out of the batch queue making it seem
like the machine was hung. Once it had all the memory the sort ran
in about a minute released the excess memory and the machine went
back to normal. Lesson learned: Parameters do matter. using a more
reasonable value in the memory allocation resulted in normal operations
and no more late night calls about a hung machine. :-)

bill

[Robert A. Brooks]

On 12/6/2021 11:40 PM, Dave Froble wrote:

> Now women are never wrong, right? I could see Marie was getting a
> bit upset with my attitude. So I backed off, explained what she had
> done, and settled for a promise (do women keep them?) that she would
> ask me before doing such things, and saved my life.

How about leaving the misogyny out of your narrative?

--

-- Rob

[Arne Vajhøj]

On 12/6/2021 10:20 PM, abrsvc wrote:

> It would seem to me that the cost to update VMS to remove the
> parameters would be more than VSI wants to pay. Why remove them,
> make them arbitrarily high (or low) such that they are effectively
> "not there". Why make all kinds of source changes just to remove
> something that has little overhead? Do the values change over time,
> yes. It is difficult to make those changes, NO. Are changes made
> frequently, NO. Seems to me that this is one case where "if it aint
> broke..."

It would be hard to justify removing them as a separate project. But if
some code needed a major overhaul for other reasons, then they could be
ripped out.

> Haing the flexibility to alter how the system uses resources will
> always benefit someone somewhere. I would rather have control than
> not. But I have had the pleasure of using this system for 40 years
> and know it well enough to know how to use those controls
> effectively. Yes, in the hands of people that don't know better, bad
> things can happen, but the same can be said of other areas too.

Old VMS'ers knows about this stuff.

But for any new VMS'ers it is just something that looks unnecessary
complicated.

Arne

[Dave Froble]

I didn't, and don't, and never will, consider it that word that isn't in my
vocabulary. Neither is "politically correct".

I'm a realist. Some things just are. Ignoring them doesn't change that.

[Phillip Helbig (undress to reply)]

In article <soo3ko$ulh$1...@dont-email.me>, Dave Froble

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

> Some things just are. Ignoring them doesn't change that.

Good self-description there. :-)

[Craig A. Berry]

On 12/7/21 10:54 AM, Dave Froble wrote:
> On 12/7/2021 10:40 AM, Robert A. Brooks wrote:
>> On 12/6/2021 11:40 PM, Dave Froble wrote:
>>
>>> Now women are never wrong, right?  I could see Marie was getting a
>>> bit upset with my attitude.  So I backed off, explained what she had
>>> done, and settled for a promise (do women keep them?) that she would
>>> ask me before doing such things, and saved my life.
>>
>> How about leaving the misogyny out of your narrative?
>>
>
> I didn't, and don't, and never will, consider it that word that isn't in
> my vocabulary.

Thanks for explaining that your misogyny is due to willful ignorance
rather than unconscious bias. It's still not welcome here.

[Simon Clubley]

On 2021-12-06, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>

> There is an outstanding issue about the expiring licenses.
>
> But I think the OP was looking for something more general/fundamental -
> this issue could be fixed in no time by VSI.
>

Yes, it could indeed be fixed very easily by VSI.

However, until they actually do fix it, this affects potential new
customers right here in 2021 and not just in some future time period.

Very few customers would agree to tie the future of their own companies
so closely to the future viability of VSI.

As things stand at the moment, those potential customers will be
severely damaged if VSI fails and their VMS systems suddenly stop
working one day when the licences expire.

Why would a manager in a customer company ever agree to such a thing
unless they were _absolutely_ forced to go to VSI ?

This policy by VSI could easily push more customers directly towards a
move to another OS, even if the cost of the port to the other OS
would otherwise have been greater than the customer was originally
willing to pay.

> Arne
>
> PS: And you know how I think VSI should fix this.
>

And that's the annoying thing. There are multiple ways to fix this,
but VSI are taking a short-term approach even with the long-term
effect it is going to have on potential customers thinking of going
to VSI. Just look at VSI's reaction to the French customers trying
to discuss this with VSI.

[Johnny Billquist]

Well, there is certainly a big difference between max physical memory
allowed, and max virtual memory.

And you probably want to have a reasonable working set at all times. And
if you go below some lower threshold, you probably don't want to
schedule the process, but instead wait until some more resources are
available. And meanwhile maybe swap out whatever is in there to free up
more resources for others.

And what does maxcpu mean? You're only allowed to use this much of a CPU
no matter how idle it is?

Johnny

[abrsvc]

On Wednesday, December 8, 2021 at 6:46:27 AM UTC-5, Johnny Billquist wrote:
.
>
> And what does maxcpu mean? You're only allowed to use this much of a CPU
> no matter how idle it is?
>
> Johnny

Yes. There is a CPU field within the authorize record that limits the CPU time available for a process or one session.
This goes back to the timesharing days when CPU time was charged and accounts were limited by time (both CPU and elapsed).
I haven't seen this used on over 35 years. All accounts these days have this set to 0 which is infinite time allowed.

Dan

[Henry Crun]

I have seen it used whan there is a danger of a batch job going into an unstoppable loop


--
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 12/8/2021 6:46 AM, Johnny Billquist wrote:
> On 2021-12-06 19:45, Arne Vajhøj wrote:
>> On 12/6/2021 1:29 PM, Simon Clubley wrote:
>>> On 2021-12-03, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>>>> On 12/3/2021 1:55 PM, Simon Clubley wrote:
>>>>> I think VSI have done some work with parameter defaults so at least
>>>>> some of them will not be an issue on x86-64 VMS.
>>>>>
>>>>> Is there any option in VMS where, if a system/process goes past some
>>>>> percentage (say 80% to 90%) of any system parameter designed to limit
>>>>> use of a resource, VMS will issue an OPCOM warning about that system
>>>>> parameter (and maybe log it elsewhere as well) ? If not, would this
>>>>> be a good option to add to VMS ?
>>>>
>>>> I think VSI should get rid of >95% of SYSGEN and SYSUAF limits.
>>>>
>>>> They made sense with a 256 KB VAX but not so much on a 256 GB x86-64.
>>>
>>> Agreed, but it would be nice if they would retain an overall ability
>>> to stop a single runaway process from gobbling up all the resources
>>> without at the same time having to micromanage resources as needed to
>>> be done on the resource limited systems of old.
>>
>> Sure. But if we have a MAXCPU, a MAXMEMORY and a MAXOPENFILES
>> then how much more do we need.
>
> Well, there is certainly a big difference between max physical memory
> allowed, and max virtual memory.

True. I was thinking physical.

But if necessary MAXPHYSMEM and MAXVIRTMEM.

The current number of WS parameters seems excessive.

> And you probably want to have a reasonable working set at all times. And
> if you go below some lower threshold, you probably don't want to
> schedule the process, but instead wait until some more resources are
> available. And meanwhile maybe swap out whatever is in there to free up
> more resources for others.

But does this need to be configurable and tunable or could we just leave
that to the OS?

I am advocating to leave it to the OS.

> And what does maxcpu mean? You're only allowed to use this much of a CPU
> no matter how idle it is?

That is usually a blocker of code going into infinite loop or
crazy stuff so yes.

Arne

[Johnny Billquist]

I had forgotten all about that. Thanks for reminding me. :-)

But that wouldn't be in system parameters, but accounting... And
potentially different for different users or processes. But I guess the
system parameter could give an upper limit, maybe... But some processes
obviously needs to be excluded from that limit.

Johnny

[Arne Vajhøj]

Very different.

But when I said that I think VSI could drop most of all those
limits I was thinking all limits (SYSGEN parameter, SYSUAF
limits etc.).

Arne