Discussion:
VMS still roaming in the wild? - Market Research Questions
(too old to reply)
Ron
2014-03-27 11:36:09 UTC
Permalink
Greetings all,

I am doing some market research for a customer. They market software and services related to VMS, OpenVMS and other legacy systems. We're trying fine-tune their messaging and market strategy by interviewing people who still work with these systems.

It has been tricky finding people to participate in this study, so I'm reaching out to a couple of online communities to see if I can recruit some interviewees.

If you work with/on VMS/OpenVMS in an enterprise environment or know of these systems in such an environment, and are willing to bounce some ideas around let me know and we can schedule some time to talk.

Sample questions:
- How are these systems holding up? Still highly available?
- Are they still running business-critical, mission-critical production systems?
- Have they been relegated to simply maintaining an archive of those systems?
- Are there challenges keeping the platforms running?
- Is there a movement afoot to get away from these systems?
- With all the market hype about the cloud and virtualization, how are these systems viewed by the staff? What about Management's perspective?

I promise that there is no hidden sales pitch in here. I won't even tell you who my customer is unless you ask.

Whaddya say? It would be great to speak by phone or Skype, but if you prefer, I would happily take responses on this forum.

Sincerely,
Ron
Jan-Erik Soderholm
2014-03-27 11:50:55 UTC
Permalink
Post by Ron
Greetings all,
I am doing some market research for a customer. They market software and services related to VMS, OpenVMS and other legacy systems. We're trying fine-tune their messaging and market strategy by interviewing people who still work with these systems.
It has been tricky finding people to participate in this study, so I'm reaching out to a couple of online communities to see if I can recruit some interviewees.
If you work with/on VMS/OpenVMS in an enterprise environment or know of these systems in such an environment, and are willing to bounce some ideas around let me know and we can schedule some time to talk.
- How are these systems holding up? Still highly available?
Yes.
Post by Ron
- Are they still running business-critical, mission-critical production systems?
Yes.
Post by Ron
- Have they been relegated to simply maintaining an archive of those systems?
No.
Post by Ron
- Are there challenges keeping the platforms running?
No particular.
Post by Ron
- Is there a movement afoot to get away from these systems?
Yes.
Post by Ron
- ... how are these systems viewed by the staff? What about Management's perspective?
Something the cat has draged in...
Gérard Calliet (pia-sofer)
2014-03-27 16:33:45 UTC
Permalink
Post by Ron
Greetings all,
I am doing some market research for a customer. They market software and services related to VMS, OpenVMS and other legacy systems. We're trying fine-tune their messaging and market strategy by interviewing people who still work with these systems.
It has been tricky finding people to participate in this study, so I'm reaching out to a couple of online communities to see if I can recruit some interviewees.
If you work with/on VMS/OpenVMS in an enterprise environment or know of these systems in such an environment, and are willing to bounce some ideas around let me know and we can schedule some time to talk.
- How are these systems holding up? Still highly available?
- Are they still running business-critical, mission-critical production systems?
- Have they been relegated to simply maintaining an archive of those systems?
- Are there challenges keeping the platforms running?
- Is there a movement afoot to get away from these systems?
- With all the market hype about the cloud and virtualization, how are these systems viewed by the staff? What about Management's perspective?
I promise that there is no hidden sales pitch in here. I won't even tell you who my customer is unless you ask.
Whaddya say? It would be great to speak by phone or Skype, but if you prefer, I would happily take responses on this forum.
Sincerely,
Ron
Have a look here : www.hp-interex.fr (french users club)
We launched a survey in q4 2013 about HP june 2013 letter about OpenVMS,
which seemed a little bit innapropriate.

I agree on all that have been said by Jan-Erik, except about "the
movement afoot to get away" :
1) this movement is constant for years, but slow, and for some
customers, it simply does'nt have any sense,
2) HP sees this movement in terms of wishfull thinking : they hope have
killed OpenVMS in a letter, and they are since in some troubles about
that, so it would be great for them if it were true,
3) it seems that alltogether for customers who want to go away it would
take big time, so there is around some work with OpenVMS for about 10 or
20 years, in a specialised market (old dependable systems).



---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Jan-Erik Soderholm
2014-03-27 16:50:15 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Post by Ron
Greetings all,
I am doing some market research for a customer. They market software and
services related to VMS, OpenVMS and other legacy systems. We're trying
fine-tune their messaging and market strategy by interviewing people who
still work with these systems.
It has been tricky finding people to participate in this study, so I'm
reaching out to a couple of online communities to see if I can recruit
some interviewees.
If you work with/on VMS/OpenVMS in an enterprise environment or know of
these systems in such an environment, and are willing to bounce some
ideas around let me know and we can schedule some time to talk.
- How are these systems holding up? Still highly available?
- Are they still running business-critical, mission-critical production systems?
- Have they been relegated to simply maintaining an archive of those systems?
- Are there challenges keeping the platforms running?
- Is there a movement afoot to get away from these systems?
- With all the market hype about the cloud and virtualization, how are
these systems viewed by the staff? What about Management's perspective?
I promise that there is no hidden sales pitch in here. I won't even tell
you who my customer is unless you ask.
Whaddya say? It would be great to speak by phone or Skype, but if you
prefer, I would happily take responses on this forum.
Sincerely,
Ron
Have a look here : www.hp-interex.fr (french users club)
We launched a survey in q4 2013 about HP june 2013 letter about OpenVMS,
which seemed a little bit innapropriate.
I agree on all that have been said by Jan-Erik,...
Maybe I was a bit unclear, those answers was based on my
current customer. It was not ment to be some general view
over all current OpenVMS users.

I'm sure that Gérard is just as right in his views. :-)

Regards,
Jan-Erik.
Post by Gérard Calliet (pia-sofer)
except about "the movement
1) this movement is constant for years, but slow, and for some customers,
it simply does'nt have any sense,
2) HP sees this movement in terms of wishfull thinking : they hope have
killed OpenVMS in a letter, and they are since in some troubles about
that, so it would be great for them if it were true,
3) it seems that alltogether for customers who want to go away it would
take big time, so there is around some work with OpenVMS for about 10 or 20
years, in a specialised market (old dependable systems).
---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant
parce que la protection avast! Antivirus est active.
http://www.avast.com
Ron
2014-03-27 18:02:05 UTC
Permalink
forgive me for any formatting issues... I'm kind of new to these Google Groups.

Follow-up questions & comments for Gerard and Jan-Erik (and anyone else who cares to weigh in:

1. I agree that HP is acting like they want VMS to go away. I also agree that it will be difficult for that to happen. My career started with Tandem and DEC, so I can completely relate that there are characteristics of these legacy platforms that are hard to replicate with "modern technology." (no question, here. just an observation.)

2. Question: What are the factors that drive companies to want to get off of these legacy platforms. Is it the cost of maintenance? The cost of hardware replacements? The business risk of an unsupported platform?

3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?

4. Does performance or availability become an issue with these platforms as they age?

5. If so, what is the source of the problem? (hardware failures, software glitches, incompatibility with new technology, etc.)

Thanks for continuing the conversation!

- Ron
David Froble
2014-03-27 18:48:13 UTC
Permalink
Post by Ron
forgive me for any formatting issues... I'm kind of new to these Google Groups.
Follow-up questions & comments for Gerard and Jan-Erik (and anyone
1. I agree that HP is acting like they want VMS to go away. I also
agree that it will be difficult for that to happen. My career
started with Tandem and DEC, so I can completely relate that there
are characteristics of these legacy platforms that are hard to
replicate with "modern technology." (no question, here. just an
observation.)
Some people really have no good options to port their solutions
elsewhere. They are looking at a total replacement. Some of these
solutions have specifics that are mandatory in their business, and
generic just won't work.

Most likely those still on VMS fit into the above catagory.

HP has basically a captive audience, and chooses to throw it away.
Post by Ron
2. Question: What are the factors that drive companies to want to
get off of these legacy platforms. Is it the cost of maintenance?
The cost of hardware replacements? The business risk of an
unsupported platform?
Probably many reasons. Some are products of the peter principal, and
make decisions they don't have the expertise to make, based upon what
they read on "trade journals" and such.

There is the difficulty in finding qualified people for some platforms.
Many VMS people have moved elsewhere, from the fear of not being able
to find jobs in VMS environments. A self fulfilling prophesy.
Post by Ron
3. Conversely: how does a company justify running their business on
a technology stack that is 15 years old, or perhaps even older?
That is such a poor question. Unix is older than VMS. It's not the age
of the environment, it's the capabilities. While VMS suffers from
neglect, it still has some things that other environments have never
achieved. That said, VMS could benefit from some attention and development.
Post by Ron
4. Does performance or availability become an issue with these platforms as they age?
Up until now performance and availability have not been major problems.
With IA-64 being discontinued, availability will soon be a major issue.

Have to say, a port of VMS to x86 would solve most of that issue.

A vendor who actually believes in their product would solve most issues.
Post by Ron
5. If so, what is the source of the problem? (hardware failures,
software glitches, incompatibility with new technology, etc.)
Thanks for continuing the conversation!
- Ron
Joukj
2014-03-28 08:16:27 UTC
Permalink
Post by David Froble
Have to say, a port of VMS to x86 would solve most of that issue.
Yes. Me too would like to see a version of OpenVMS for x86-64.
Bill Gunshannon
2014-03-28 12:28:48 UTC
Permalink
Post by Joukj
Post by David Froble
Have to say, a port of VMS to x86 would solve most of that issue.
Yes. Me too would like to see a version of OpenVMS for x86-64.
And here we go again!!!!

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
JF Mezei
2014-04-03 03:28:11 UTC
Permalink
Post by David Froble
HP has basically a captive audience, and chooses to throw it away.
HP is runned by beancounters who have to right the sinking ship. That
means streamlining product lines and getting rid of products with no future.

If VMS doesn't generate sufficient profits to sustain its engineering
costs, then should HP be expected to subsidize those captive customers ?

It did so for NSK, but that is partly because of how high profile they
are and the bad PR that would have been generated. But since VMS
customers have been quiet and disorganised since the demise of DECUS, HP
gets away with shutting down VMS with barely any media notice and no
customer backlash.

VMS customers are tired of the "<owner> is killing VMS", despite it
having been true all along. So when the actual announcement came, people
just stayed there and accepted it because we had been conditioned to
expect that announcement for a couple of decades.
JF Mezei
2014-03-27 19:12:47 UTC
Permalink
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.

It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.


If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
David Froble
2014-03-27 21:56:26 UTC
Permalink
Post by JF Mezei
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Post by JF Mezei
It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.
Well, if I was driving toward a bridge, and could see the bridge had
collapsed, I'll seriously consider turning the steering wheel.

It's not that VMS could not do the job, it's more that HP has abandoned
the market.
Post by JF Mezei
If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
If you're talking about serving pages for the point & click browsers,
it's true that VMS was not kept up to date with other environments. But
otherwise, for things such as computer to computer communications using
TCP/IP and sockets and such, well we're doing it all the time, and VMS
works quite well in the environment. (Yes, some more tender loving care
would be nice.) Much of our activity has changed from people on phones
manually entering customer orders on terminals to customer computers
placing orders via socket communications.
V***@SendSpamHere.ORG
2014-03-28 00:21:25 UTC
Permalink
Post by David Froble
Post by JF Mezei
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Post by JF Mezei
It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.
Well, if I was driving toward a bridge, and could see the bridge had
collapsed, I'll seriously consider turning the steering wheel.
It's not that VMS could not do the job, it's more that HP has abandoned
the market.
Post by JF Mezei
If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
If you're talking about serving pages for the point & click browsers,
it's true that VMS was not kept up to date with other environments. But
VMS hasn't? Serving web pages has nothing to do with the OS. It has all
to do with the web server. Sadly, the parochial, "all the world is unix"
mentality has created software that makes porting it to VMS hostile. The
C portability myth continues.
Post by David Froble
otherwise, for things such as computer to computer communications using
TCP/IP and sockets and such, well we're doing it all the time, and VMS
works quite well in the environment. (Yes, some more tender loving care
would be nice.) Much of our activity has changed from people on phones
manually entering customer orders on terminals to customer computers
placing orders via socket communications.
Well, I have clients who will very likely never move off of VMS. They've
got new Integrity boxes too but have code that has been built for and run-
ning on VMS for decades and they continure to develope it too! Hopelessly
Pathetic, despite their total disdain for VMS, will not stop its use.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.
Bill Gunshannon
2014-03-28 13:12:23 UTC
Permalink
Post by V***@SendSpamHere.ORG
Post by David Froble
Post by JF Mezei
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Post by JF Mezei
It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.
Well, if I was driving toward a bridge, and could see the bridge had
collapsed, I'll seriously consider turning the steering wheel.
It's not that VMS could not do the job, it's more that HP has abandoned
the market.
Post by JF Mezei
If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
If you're talking about serving pages for the point & click browsers,
it's true that VMS was not kept up to date with other environments. But
VMS hasn't?
Let's be serious here. VMS hasn't kept up with anything. It has seen
limited development beyond barely keeping it running and supporting
new but very limited hardware.
Post by V***@SendSpamHere.ORG
Serving web pages has nothing to do with the OS.
At least that part is true.
Post by V***@SendSpamHere.ORG
It has all
to do with the web server. Sadly, the parochial, "all the world is unix"
mentality has created software that makes porting it to VMS hostile.
Wait a minute. You are blaming Unix because no one writes VMS software?
Get serious. You don't have to port Unix software. VMS people are free
to write a true VMS webserver with all the capabilities of Apache. You
can write an Office Suite specifically for VMS. Windowing System, you
don't need to use X-11, write your own specifically targeted at VMS.
Etc. Etc. Etc.
There is probably a reason why that never got done. Economies of scale
comes immediately to mind.
Post by V***@SendSpamHere.ORG
The
C portability myth continues.
I have been using C for decades. I, for one, have never said it was any
more portable than any other language. If you write code in a portable
fashion it will be portable but that takes additional effort that most
programmers have no reason to take as they know their target when they
write their code. And, the portability issues for all the code you are
thinking of has nothing to do with C it has to do with API's. VMS has
always had two options. Support Unix-like (POSIX?) API's or write all
the software specifically using VMS API's. But is is not the fault of
C or Unix that non-VMS programmers don't write code that ports easily
to VMS.
Post by V***@SendSpamHere.ORG
Post by David Froble
otherwise, for things such as computer to computer communications using
TCP/IP and sockets and such, well we're doing it all the time, and VMS
works quite well in the environment. (Yes, some more tender loving care
would be nice.) Much of our activity has changed from people on phones
manually entering customer orders on terminals to customer computers
placing orders via socket communications.
Well, I have clients who will very likely never move off of VMS. They've
got new Integrity boxes too but have code that has been built for and run-
ning on VMS for decades and they continure to develope it too! Hopelessly
Pathetic, despite their total disdain for VMS, will not stop its use.
Wanna bet?

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
V***@SendSpamHere.ORG
2014-03-28 15:24:38 UTC
Permalink
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
Post by JF Mezei
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Post by JF Mezei
It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.
Well, if I was driving toward a bridge, and could see the bridge had
collapsed, I'll seriously consider turning the steering wheel.
It's not that VMS could not do the job, it's more that HP has abandoned
the market.
Post by JF Mezei
If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
If you're talking about serving pages for the point & click browsers,
it's true that VMS was not kept up to date with other environments. But
VMS hasn't?
Let's be serious here. VMS hasn't kept up with anything. It has seen
limited development beyond barely keeping it running and supporting
new but very limited hardware.
There had/have been a number of new OS features added in the past decade.
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Serving web pages has nothing to do with the OS.
At least that part is true.
Post by V***@SendSpamHere.ORG
It has all
to do with the web server. Sadly, the parochial, "all the world is unix"
mentality has created software that makes porting it to VMS hostile.
Wait a minute. You are blaming Unix because no one writes VMS software?
No, I'm blaming the third parties for having a parochial unix view.
Post by Bill Gunshannon
Get serious. You don't have to port Unix software. VMS people are free
to write a true VMS webserver with all the capabilities of Apache. You
There is OSU and WASD; however, the Apache crowd has made it the de jure
web server standard.
Post by Bill Gunshannon
can write an Office Suite specifically for VMS. Windowing System, you
There was ALL-IN-1, Wordperfect, DECwrite and others. The WEENDOZE crowd
opted for W(ie)RD. Gack. And it's nasty stuff if you've ever tasted it.
Post by Bill Gunshannon
don't need to use X-11, write your own specifically targeted at VMS.
Etc. Etc. Etc.
There is probably a reason why that never got done. Economies of scale
comes immediately to mind.
Post by V***@SendSpamHere.ORG
The
C portability myth continues.
I have been using C for decades. I, for one, have never said it was any
more portable than any other language. If you write code in a portable
fashion it will be portable but that takes additional effort that most
programmers have no reason to take as they know their target when they
write their code. And, the portability issues for all the code you are
thinking of has nothing to do with C it has to do with API's. VMS has
always had two options. Support Unix-like (POSIX?) API's or write all
the software specifically using VMS API's. But is is not the fault of
C or Unix that non-VMS programmers don't write code that ports easily
to VMS.
It's the fault of their parochial unix view and the "let's get it compiled
so we can ship it full of bugs" programmer.
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
otherwise, for things such as computer to computer communications using
TCP/IP and sockets and such, well we're doing it all the time, and VMS
works quite well in the environment. (Yes, some more tender loving care
would be nice.) Much of our activity has changed from people on phones
manually entering customer orders on terminals to customer computers
placing orders via socket communications.
Well, I have clients who will very likely never move off of VMS. They've
got new Integrity boxes too but have code that has been built for and run-
ning on VMS for decades and they continure to develope it too! Hopelessly
Pathetic, despite their total disdain for VMS, will not stop its use.
Wanna bet?
Wanna bet what?
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.
Bill Gunshannon
2014-03-28 15:08:47 UTC
Permalink
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
Post by JF Mezei
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Post by JF Mezei
It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.
Well, if I was driving toward a bridge, and could see the bridge had
collapsed, I'll seriously consider turning the steering wheel.
It's not that VMS could not do the job, it's more that HP has abandoned
the market.
Post by JF Mezei
If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
If you're talking about serving pages for the point & click browsers,
it's true that VMS was not kept up to date with other environments. But
VMS hasn't?
Let's be serious here. VMS hasn't kept up with anything. It has seen
limited development beyond barely keeping it running and supporting
new but very limited hardware.
There had/have been a number of new OS features added in the past decade.
And yet peopl ehere are continuously complaining about OS features that
are common in other systems and needed for modern computing that VMS
still doesn't have.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Serving web pages has nothing to do with the OS.
At least that part is true.
Post by V***@SendSpamHere.ORG
It has all
to do with the web server. Sadly, the parochial, "all the world is unix"
mentality has created software that makes porting it to VMS hostile.
Wait a minute. You are blaming Unix because no one writes VMS software?
No, I'm blaming the third parties for having a parochial unix view.
Why on earth would you expect anyone to care about making their software
work on an OS they have no knowledge of, and no interest in? They don't
have "a parochial unix view". They are Unix programmers. That's their
job or maybe just their interest. I certainly don't consider if something
I am writting will port to zOS when I am writting it. And I doubt anyone
writting VMS software goes out of their way to make it portable to Unix.
Them and their damn parochial VMS view.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Get serious. You don't have to port Unix software. VMS people are free
to write a true VMS webserver with all the capabilities of Apache. You
There is OSU and WASD; however, the Apache crowd has made it the de jure
web server standard.
The "Apache crowd" certainly has not. The IT industry has. The "Apache
crowd" has never come to my office with rubber hoses to make me use their
products. They egt used because at this point in time, they are the best
choice for the job. If VMS and WASD: or OSU were people would be installing
them.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
can write an Office Suite specifically for VMS. Windowing System, you
There was ALL-IN-1, Wordperfect, DECwrite and others.
And where are they today? Why? Two of them belonged to the owners of
VMS and even they didn't consider them worth the effort.
Post by V***@SendSpamHere.ORG
The WEENDOZE crowd
opted for W(ie)RD. Gack. And it's nasty stuff if you've ever tasted it.
Luckily, the rest of the IT world isn't as irrational as you. But even
putting that aside, there are numerous non-Windows Office Suites. For
most common OSes. Why did the VMS community never write one for their
favorite OS? Instead, it appears, they complain beecause the Unix
community didn't do it for them.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
don't need to use X-11, write your own specifically targeted at VMS.
Etc. Etc. Etc.
There is probably a reason why that never got done. Economies of scale
comes immediately to mind.
Post by V***@SendSpamHere.ORG
The
C portability myth continues.
I have been using C for decades. I, for one, have never said it was any
more portable than any other language. If you write code in a portable
fashion it will be portable but that takes additional effort that most
programmers have no reason to take as they know their target when they
write their code. And, the portability issues for all the code you are
thinking of has nothing to do with C it has to do with API's. VMS has
always had two options. Support Unix-like (POSIX?) API's or write all
the software specifically using VMS API's. But is is not the fault of
C or Unix that non-VMS programmers don't write code that ports easily
to VMS.
It's the fault of their parochial unix view and the "let's get it compiled
so we can ship it full of bugs" programmer.
Your pretty irrational on this issue as well, I see. Contrary to your
apparent belief everything not VMS is not buggy or not engineered. And
Unix people have the same educational backgrounds employment histories
as people who work on other systems, including VMS. The only loser in
this whole discussion is VMS. Maybe you should be looking at why it
failed if it was so good instead of trying to place the blame on other
systems.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
otherwise, for things such as computer to computer communications using
TCP/IP and sockets and such, well we're doing it all the time, and VMS
works quite well in the environment. (Yes, some more tender loving care
would be nice.) Much of our activity has changed from people on phones
manually entering customer orders on terminals to customer computers
placing orders via socket communications.
Well, I have clients who will very likely never move off of VMS. They've
got new Integrity boxes too but have code that has been built for and run-
ning on VMS for decades and they continure to develope it too! Hopelessly
Pathetic, despite their total disdain for VMS, will not stop its use.
Wanna bet?
Wanna bet what?
That VMS will continue to be used when HP winds it down. in a few years, I
expect, it will be the same as Primos. Yes, there actually are still people
using it and a company providing support. But no one in the IT world takes
it seriously and it will eventually just fade away.

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
David Froble
2014-03-28 16:59:25 UTC
Permalink
Now children, play nice ...

:-)
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
Post by JF Mezei
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Post by JF Mezei
It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.
Well, if I was driving toward a bridge, and could see the bridge had
collapsed, I'll seriously consider turning the steering wheel.
It's not that VMS could not do the job, it's more that HP has abandoned
the market.
Post by JF Mezei
If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
If you're talking about serving pages for the point & click browsers,
it's true that VMS was not kept up to date with other environments. But
VMS hasn't?
Let's be serious here. VMS hasn't kept up with anything. It has seen
limited development beyond barely keeping it running and supporting
new but very limited hardware.
There had/have been a number of new OS features added in the past decade.
I'm with Bill, and Steve, and others, on this point. Nor do I consider
modifications to the C-RTL for better Unix application portability as OS
features. Just the way I see things.
Post by Bill Gunshannon
And yet peopl ehere are continuously complaining about OS features that
are common in other systems and needed for modern computing that VMS
still doesn't have.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Serving web pages has nothing to do with the OS.
At least that part is true.
Post by V***@SendSpamHere.ORG
It has all
to do with the web server. Sadly, the parochial, "all the world is unix"
mentality has created software that makes porting it to VMS hostile.
Wait a minute. You are blaming Unix because no one writes VMS software?
No, I'm blaming the third parties for having a parochial unix view.
It's all a matter of scale, and to some extent, "monkey see, monkey do".
People ask their peers how they do things, get told to use Apache, and
then do so. Which causes Apache to get more development.

Now, if everyone who wants to run a web server on VMS would send the OSU
developer (damn senile old man cannot remember his name) and Mark a
twenty or a fifty, maybe they'd spend more time and add the features
that Apache gets, and then VMS users would have 2 fine options.

So, how about it? Can anyone spare a $20 or $50 ???

Not that running a pointy clicky web server is all there is to computing
these days.
Post by Bill Gunshannon
Why on earth would you expect anyone to care about making their software
work on an OS they have no knowledge of, and no interest in? They don't
have "a parochial unix view". They are Unix programmers. That's their
job or maybe just their interest. I certainly don't consider if something
I am writting will port to zOS when I am writting it. And I doubt anyone
writting VMS software goes out of their way to make it portable to Unix.
Them and their damn parochial VMS view.
Got to agree. I know my stuff won't run on Unix. Nor do I much care.
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Get serious. You don't have to port Unix software. VMS people are free
to write a true VMS webserver with all the capabilities of Apache. You
There is OSU and WASD; however, the Apache crowd has made it the de jure
web server standard.
The "Apache crowd" certainly has not. The IT industry has. The "Apache
crowd" has never come to my office with rubber hoses to make me use their
products. They egt used because at this point in time, they are the best
choice for the job. If VMS and WASD: or OSU were people would be installing
them.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
can write an Office Suite specifically for VMS. Windowing System, you
There was ALL-IN-1, Wordperfect, DECwrite and others.
Office stuff also is not the only reasons to use computers today.
However, it is the most visable stuff for all the office paper pusher
weenies.
Post by Bill Gunshannon
And where are they today? Why? Two of them belonged to the owners of
VMS and even they didn't consider them worth the effort.
Well, it's DEC's fault. They tried to maintain their little "thingy" in
the face of a broad move to more widespread and distributed computing,
and dare I say it, computers in the home, and the progression to tablets
and smart phones today.

When IBM wanted to introduce the PC, what if DEC had offered the VAX CPU
and VMS for the effort. Microsoft and Intel would have been the big
losers, and the world would have been the big winners.

"What, sell VMS for $98 a copy ??? Not on your life." Turns out it was
DEC's life that took the bullet.

I'm betting there was some Harvard MBAs counting their little beans who
were so much against things like the above happening.
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
The WEENDOZE crowd
opted for W(ie)RD. Gack. And it's nasty stuff if you've ever tasted it.
Luckily, the rest of the IT world isn't as irrational as you. But even
putting that aside, there are numerous non-Windows Office Suites. For
most common OSes. Why did the VMS community never write one for their
favorite OS? Instead, it appears, they complain beecause the Unix
community didn't do it for them.
Definition of a "leader". Someone who sees which direction the mob is
moving, runs in front and hollers "follow me".

Then there is the smug few who say "why should I follow (or lead) the
mob?" Simple answer. Seems once upon a time a bank robber was asked
"why do you rob banks", and the reply was "because that's where the
money is".
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
don't need to use X-11, write your own specifically targeted at VMS.
Etc. Etc. Etc.
There is probably a reason why that never got done. Economies of scale
comes immediately to mind.
Post by V***@SendSpamHere.ORG
The
C portability myth continues.
I have been using C for decades. I, for one, have never said it was any
more portable than any other language. If you write code in a portable
fashion it will be portable but that takes additional effort that most
programmers have no reason to take as they know their target when they
write their code. And, the portability issues for all the code you are
thinking of has nothing to do with C it has to do with API's. VMS has
always had two options. Support Unix-like (POSIX?) API's or write all
the software specifically using VMS API's. But is is not the fault of
C or Unix that non-VMS programmers don't write code that ports easily
to VMS.
It's the fault of their parochial unix view and the "let's get it compiled
so we can ship it full of bugs" programmer.
Your pretty irrational on this issue as well, I see. Contrary to your
apparent belief everything not VMS is not buggy or not engineered. And
Unix people have the same educational backgrounds employment histories
as people who work on other systems, including VMS. The only loser in
this whole discussion is VMS. Maybe you should be looking at why it
failed if it was so good instead of trying to place the blame on other
systems.
That's rather simple. Multiple moves by DEC, including having so little
faith in their own product that people in DEC were telling customers to
switch to Unix. I don't know if the English language has words to
describe just how stupid that was.
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
otherwise, for things such as computer to computer communications using
TCP/IP and sockets and such, well we're doing it all the time, and VMS
works quite well in the environment. (Yes, some more tender loving care
would be nice.) Much of our activity has changed from people on phones
manually entering customer orders on terminals to customer computers
placing orders via socket communications.
Well, I have clients who will very likely never move off of VMS. They've
got new Integrity boxes too but have code that has been built for and run-
ning on VMS for decades and they continure to develope it too! Hopelessly
Pathetic, despite their total disdain for VMS, will not stop its use.
Wanna bet?
Wanna bet what?
That VMS will continue to be used when HP winds it down. in a few years, I
expect, it will be the same as Primos. Yes, there actually are still people
using it and a company providing support. But no one in the IT world takes
it seriously and it will eventually just fade away.
Being useful to some people and being taken seriously by the mob are two
different things. The only question is, will those who find VMS useful
have any choices in the future?
Bill Gunshannon
2014-03-28 18:07:13 UTC
Permalink
Post by David Froble
Now children, play nice ...
:-)
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
Post by JF Mezei
Post by Ron
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Post by JF Mezei
It is a given that if the factory gets updated machinery (think paper
mill etc), it will get now control systems that won't be VMS.
Well, if I was driving toward a bridge, and could see the bridge had
collapsed, I'll seriously consider turning the steering wheel.
It's not that VMS could not do the job, it's more that HP has abandoned
the market.
Post by JF Mezei
If your applications are stable and don't require the latest and
greatest fads, then VMS can stay there. Just because it isn't up to date
doesn't mean it can't do the job you need it for. But if you need the
latest and greatest tools for web facing apps etc, then VMS has long ago
ceased to be a competitive offering.
If you're talking about serving pages for the point & click browsers,
it's true that VMS was not kept up to date with other environments. But
VMS hasn't?
Let's be serious here. VMS hasn't kept up with anything. It has seen
limited development beyond barely keeping it running and supporting
new but very limited hardware.
There had/have been a number of new OS features added in the past decade.
I'm with Bill, and Steve, and others, on this point. Nor do I consider
modifications to the C-RTL for better Unix application portability as OS
features. Just the way I see things.
Post by Bill Gunshannon
And yet peopl ehere are continuously complaining about OS features that
are common in other systems and needed for modern computing that VMS
still doesn't have.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Serving web pages has nothing to do with the OS.
At least that part is true.
Post by V***@SendSpamHere.ORG
It has all
to do with the web server. Sadly, the parochial, "all the world is unix"
mentality has created software that makes porting it to VMS hostile.
Wait a minute. You are blaming Unix because no one writes VMS software?
No, I'm blaming the third parties for having a parochial unix view.
It's all a matter of scale, and to some extent, "monkey see, monkey do".
Apache was not and is not the only option. There are a number of other
Web server programs for non-VMS machines. And there have been others
in the past. Apache is on top because it offers the most in needed
functionality, reliability, ease of administration, etc. If monkey see
server crash monkey not do no matter how many others are running it.
Post by David Froble
People ask their peers how they do things, get told to use Apache, and
then do so.
What would you expect them to do? Run the one no one else is using?
There's probably a reason for that!!
Post by David Froble
Which causes Apache to get more development.
Now, if everyone who wants to run a web server on VMS would send the OSU
developer (damn senile old man cannot remember his name) and Mark a
twenty or a fifty, maybe they'd spend more time and add the features
that Apache gets, and then VMS users would have 2 fine options.
Sorry, that ship has sailed. Anyone who would spend their time putting
in that kind of effort for an EOLed system is a fool.
Post by David Froble
So, how about it? Can anyone spare a $20 or $50 ???
Not that running a pointy clicky web server is all there is to computing
these days.
Pointy-clicky is what the masses want. In many cases it can handle the
majority of the needed configing. Of course, with Apache you can always
fall back to manually editing the config if that floats your boat.
Post by David Froble
Post by Bill Gunshannon
Why on earth would you expect anyone to care about making their software
work on an OS they have no knowledge of, and no interest in? They don't
have "a parochial unix view". They are Unix programmers. That's their
job or maybe just their interest. I certainly don't consider if something
I am writting will port to zOS when I am writting it. And I doubt anyone
writting VMS software goes out of their way to make it portable to Unix.
Them and their damn parochial VMS view.
Got to agree. I know my stuff won't run on Unix. Nor do I much care.
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Get serious. You don't have to port Unix software. VMS people are free
to write a true VMS webserver with all the capabilities of Apache. You
There is OSU and WASD; however, the Apache crowd has made it the de jure
web server standard.
The "Apache crowd" certainly has not. The IT industry has. The "Apache
crowd" has never come to my office with rubber hoses to make me use their
products. They egt used because at this point in time, they are the best
choice for the job. If VMS and WASD: or OSU were people would be installing
them.
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
can write an Office Suite specifically for VMS. Windowing System, you
There was ALL-IN-1, Wordperfect, DECwrite and others.
Office stuff also is not the only reasons to use computers today.
It is probably the what "computers" get used for the most. I expect the
ratio of servers to desktops is at least 1:10,000. Even when your talking
Windows servers. (As an aside, I have finally had a server go above 90% CPU.
The good news is I had anticipated replacing this one and the new boxes are
just about ready.)
Post by David Froble
However, it is the most visable stuff for all the office paper pusher
weenies.
See, there it is again. If you are going to spend your life insulting
potential users, why would you expect them to come to you for business?
Just because you think anything not done on VMS is done by "weenies" the
IT world doesn't agree and displaying an attitude like this can only hurt
your business. (Or your chances at getting a job in the IT world.)
Post by David Froble
Post by Bill Gunshannon
And where are they today? Why? Two of them belonged to the owners of
VMS and even they didn't consider them worth the effort.
Well, it's DEC's fault. They tried to maintain their little "thingy" in
the face of a broad move to more widespread and distributed computing,
and dare I say it, computers in the home, and the progression to tablets
and smart phones today.
When IBM wanted to introduce the PC, what if DEC had offered the VAX CPU
and VMS for the effort.
They were never even in consideration. And given the state of the CPU
at that time would have been a dismal failure.
Post by David Froble
Microsoft and Intel would have been the big
losers, and the world would have been the big winners.
Hardly. Intel was chosen because they had one foot in the grave. Contrary
to the recent publication by some computer museum the 8088 was not a popular
commodity chip at the time. it was a miserable hack which without IBM would
have never seen major production. Intel was not IBM's first choice, (neither
was VAX). And for an OS, MicroSoft was not IBM's first choice either.
Post by David Froble
"What, sell VMS for $98 a copy ??? Not on your life." Turns out it was
DEC's life that took the bullet.
I'm betting there was some Harvard MBAs counting their little beans who
were so much against things like the above happening.
I doubt it was anything that simple.
Post by David Froble
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
The WEENDOZE crowd
opted for W(ie)RD. Gack. And it's nasty stuff if you've ever tasted it.
Luckily, the rest of the IT world isn't as irrational as you. But even
putting that aside, there are numerous non-Windows Office Suites. For
most common OSes. Why did the VMS community never write one for their
favorite OS? Instead, it appears, they complain beecause the Unix
community didn't do it for them.
Definition of a "leader". Someone who sees which direction the mob is
moving, runs in front and hollers "follow me".
Your definition, I guess.
Post by David Froble
Then there is the smug few who say "why should I follow (or lead) the
mob?" Simple answer. Seems once upon a time a bank robber was asked
"why do you rob banks", and the reply was "because that's where the
money is".
That was Clyde Parker. And we all know where he ended up.
Post by David Froble
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
don't need to use X-11, write your own specifically targeted at VMS.
Etc. Etc. Etc.
There is probably a reason why that never got done. Economies of scale
comes immediately to mind.
Post by V***@SendSpamHere.ORG
The
C portability myth continues.
I have been using C for decades. I, for one, have never said it was any
more portable than any other language. If you write code in a portable
fashion it will be portable but that takes additional effort that most
programmers have no reason to take as they know their target when they
write their code. And, the portability issues for all the code you are
thinking of has nothing to do with C it has to do with API's. VMS has
always had two options. Support Unix-like (POSIX?) API's or write all
the software specifically using VMS API's. But is is not the fault of
C or Unix that non-VMS programmers don't write code that ports easily
to VMS.
It's the fault of their parochial unix view and the "let's get it compiled
so we can ship it full of bugs" programmer.
Your pretty irrational on this issue as well, I see. Contrary to your
apparent belief everything not VMS is not buggy or not engineered. And
Unix people have the same educational backgrounds employment histories
as people who work on other systems, including VMS. The only loser in
this whole discussion is VMS. Maybe you should be looking at why it
failed if it was so good instead of trying to place the blame on other
systems.
That's rather simple. Multiple moves by DEC, including having so little
faith in their own product that people in DEC were telling customers to
switch to Unix.
believe it or not, that was one of the things that contributed to the
demise of Primos as well. :-)
Post by David Froble
I don't know if the English language has words to
describe just how stupid that was.
Is it really being stupid to see which way the wind is blowing and to
decide not to try to buck it?
Post by David Froble
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
Post by David Froble
otherwise, for things such as computer to computer communications using
TCP/IP and sockets and such, well we're doing it all the time, and VMS
works quite well in the environment. (Yes, some more tender loving care
would be nice.) Much of our activity has changed from people on phones
manually entering customer orders on terminals to customer computers
placing orders via socket communications.
Well, I have clients who will very likely never move off of VMS. They've
got new Integrity boxes too but have code that has been built for and run-
ning on VMS for decades and they continure to develope it too! Hopelessly
Pathetic, despite their total disdain for VMS, will not stop its use.
Wanna bet?
Wanna bet what?
That VMS will continue to be used when HP winds it down. in a few years, I
expect, it will be the same as Primos. Yes, there actually are still people
using it and a company providing support. But no one in the IT world takes
it seriously and it will eventually just fade away.
Being useful to some people and being taken seriously by the mob are two
different things. The only question is, will those who find VMS useful
have any choices in the future?
I am still waiting for someone to send/post/point-me-to a real full VMS
license. Why? because I seriously expect there is a clause that let's
them withdraw permission to use the product any time they wish and I
doubt any legitimate business would continue to use it afterwards.
[I have seen such clauses in products that they wanted me to use here
and have recommended against them specifically because of those clauses.]

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
David Froble
2014-03-28 22:34:02 UTC
Permalink
Post by Bill Gunshannon
Post by David Froble
It's all a matter of scale, and to some extent, "monkey see, monkey do".
Apache was not and is not the only option. There are a number of other
Web server programs for non-VMS machines. And there have been others
in the past. Apache is on top because it offers the most in needed
functionality, reliability, ease of administration, etc. If monkey see
server crash monkey not do no matter how many others are running it.
If Apache is so good, then what's wrong with running it on VMS? It's
available there.
Post by Bill Gunshannon
Post by David Froble
People ask their peers how they do things, get told to use Apache, and
then do so.
What would you expect them to do? Run the one no one else is using?
There's probably a reason for that!!
This is where things get amusing. Here you accept the "follow the mob"
concept.
Post by Bill Gunshannon
Post by David Froble
Not that running a pointy clicky web server is all there is to computing
these days.
Pointy-clicky is what the masses want. In many cases it can handle the
majority of the needed configing. Of course, with Apache you can always
fall back to manually editing the config if that floats your boat.
I was referring to users running browsers ...
Post by Bill Gunshannon
It is probably the what "computers" get used for the most. I expect the
ratio of servers to desktops is at least 1:10,000. Even when your talking
Windows servers. (As an aside, I have finally had a server go above 90% CPU.
The good news is I had anticipated replacing this one and the new boxes are
just about ready.)
Post by David Froble
However, it is the most visable stuff for all the office paper pusher
weenies.
See, there it is again. If you are going to spend your life insulting
potential users, why would you expect them to come to you for business?
Just because you think anything not done on VMS is done by "weenies" the
IT world doesn't agree and displaying an attitude like this can only hurt
your business. (Or your chances at getting a job in the IT world.)
YOu miss the point that I think we all agree upon, as do you with the
ration mentioned above.
Post by Bill Gunshannon
Post by David Froble
When IBM wanted to introduce the PC, what if DEC had offered the VAX CPU
and VMS for the effort.
They were never even in consideration. And given the state of the CPU
at that time would have been a dismal failure.
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems you
used to play with.
Post by Bill Gunshannon
Post by David Froble
Microsoft and Intel would have been the big
losers, and the world would have been the big winners.
Hardly. Intel was chosen because they had one foot in the grave. Contrary
to the recent publication by some computer museum the 8088 was not a popular
commodity chip at the time. it was a miserable hack which without IBM would
have never seen major production. Intel was not IBM's first choice, (neither
was VAX). And for an OS, MicroSoft was not IBM's first choice either.
Well, there you go, and it was the applications that made the PC. If
the applications would have been written for the VAX ....
Post by Bill Gunshannon
Post by David Froble
"What, sell VMS for $98 a copy ??? Not on your life." Turns out it was
DEC's life that took the bullet.
I'm betting there was some Harvard MBAs counting their little beans who
were so much against things like the above happening.
I doubt it was anything that simple.
It's actually very simple. You either defend your market, or you lose
it. Lots of examples, free internet explorer, things like that. If you
keep competitors out of your market, and it's a decent market, then you
don't lose it and you prosper.
Post by Bill Gunshannon
Post by David Froble
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
The WEENDOZE crowd
opted for W(ie)RD. Gack. And it's nasty stuff if you've ever tasted it.
Luckily, the rest of the IT world isn't as irrational as you. But even
putting that aside, there are numerous non-Windows Office Suites. For
most common OSes. Why did the VMS community never write one for their
favorite OS? Instead, it appears, they complain beecause the Unix
community didn't do it for them.
Definition of a "leader". Someone who sees which direction the mob is
moving, runs in front and hollers "follow me".
Your definition, I guess.
But, but, above you advocated using the most popular web server. What's
the difference ?
Post by Bill Gunshannon
Post by David Froble
Then there is the smug few who say "why should I follow (or lead) the
mob?" Simple answer. Seems once upon a time a bank robber was asked
"why do you rob banks", and the reply was "because that's where the
money is".
That was Clyde Parker. And we all know where he ended up.
Doesn't mean anything to the concept. If you're in business, you
provide the services people want, if you want to continue in that business.
Post by Bill Gunshannon
Is it really being stupid to see which way the wind is blowing and to
decide not to try to buck it?
So, again you "follow the mob", huh?
Post by Bill Gunshannon
I am still waiting for someone to send/post/point-me-to a real full VMS
license. Why? because I seriously expect there is a clause that let's
them withdraw permission to use the product any time they wish and I
doubt any legitimate business would continue to use it afterwards.
[I have seen such clauses in products that they wanted me to use here
and have recommended against them specifically because of those clauses.]
bill
I do not think I have one easily available, but I'll look.

Can someone please send this guy a copy of a VMS license?

From memory, it is a license for perpetial use of the software on a
particular computer. Yeah, that is a problem when the computer no
longer works. But I've never seen anything that allows the owners of
VMS to say "stop using the software".
j***@gmail.com
2014-03-28 22:55:19 UTC
Permalink
Post by David Froble
If Apache is so good, then what's wrong with running it on VMS? It's
available there.
You'll consistently see better performance if you don't run Apache on VMS.

EJ
Jan-Erik Soderholm
2014-03-28 23:03:31 UTC
Permalink
Post by j***@gmail.com
Post by David Froble
If Apache is so good, then what's wrong with running it on VMS? It's
available there.
You'll consistently see better performance if you don't run Apache on VMS.
EJ
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy
does it at twice the performance.
j***@gmail.com
2014-03-28 23:13:16 UTC
Permalink
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy
does it at twice the performance.
I haven't compared the two, but based on comments, it sounds like the ngix http server is way faster.

http://nginx.org/en/

But I don't think that's available for VMS.

EJ
Jan-Erik Soderholm
2014-03-28 23:33:52 UTC
Permalink
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Post by j***@gmail.com
EJ
Simon Clubley
2014-03-28 23:43:17 UTC
Permalink
Post by Jan-Erik Soderholm
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
From the context above, Eric means Apache and WASD.
Post by Jan-Erik Soderholm
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Nginx is a web server.

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Jan-Erik Soderholm
2014-03-28 23:49:55 UTC
Permalink
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
From the context above, Eric means Apache and WASD.
Post by Jan-Erik Soderholm
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Nginx is a web server.
A web server on VMS?
Fine then. I read it otherwise.
Post by Simon Clubley
Simon.
Simon Clubley
2014-03-29 00:01:08 UTC
Permalink
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
From the context above, Eric means Apache and WASD.
Post by Jan-Erik Soderholm
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Nginx is a web server.
A web server on VMS?
Fine then. I read it otherwise.
Eric says Nginx is faster than WASD or Apache but that he thinks it's
_not_ available for VMS. That's the point he appears to be trying to make.

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Jan-Erik Soderholm
2014-03-29 00:40:45 UTC
Permalink
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
From the context above, Eric means Apache and WASD.
Post by Jan-Erik Soderholm
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Nginx is a web server.
A web server on VMS?
Fine then. I read it otherwise.
Eric says Nginx is faster than WASD or Apache but that he thinks it's
_not_ available for VMS. That's the point he appears to be trying to make.
Simon.
And my point was that, why then mention it at all !?
When comparing web servers for VMS, that seems pointless.
David Froble
2014-03-29 01:40:20 UTC
Permalink
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
From the context above, Eric means Apache and WASD.
Post by Jan-Erik Soderholm
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Nginx is a web server.
A web server on VMS?
Fine then. I read it otherwise.
Eric says Nginx is faster than WASD or Apache but that he thinks it's
_not_ available for VMS. That's the point he appears to be trying to make.
Simon.
And my point was that, why then mention it at all !?
When comparing web servers for VMS, that seems pointless.
Well, no.

Because the original discussion was about web servers in general, not
just VMS web servers.

Frankly, if a customer told me he had to implement a general purpose web
server, I'd have to suggest something other than VMS. Face it, VMS is
not the best at this task, and we always owe it to our customers to give
the best advice.

In your case, making available data from your VMS system, perhaps a VMS
based web server might be a valid solution.

But, really, just what is a web server, when you get down to the bottom
of it? It's a listener socket, with a bunch of capabilities that it can
use to service HTTP connection requests. (Actually, most likely two
listeners, one not SSL and one for SSL.)

So, if you REALLY need to serve connection requests for data from a VMS
system, you can kluge together something with a web server, or you can
implement application specific web services. That's what we do.
Nothing special about it, and it's not difficult. You got 10,000 people
who want to check on inventory availability? No problem. Just set up
an inventory inquiry web service. Hell of lot easier (for me at least)
and most likely much quicker than something that has to process HTTP
requests and build web pages for some browser. It's such services that
now process the vast majority of our customer inquiries and orders.

If anyone doesn't believe the above, just let me know, and I'll e-mail
you a copy of one of the early "proof of concept" programs that was
written to explore servicing a socket connection request.
Richard Maher
2014-03-29 03:34:06 UTC
Permalink
Post by David Froble
Frankly, if a customer told me he had to implement a general purpose web
server, I'd have to suggest something other than VMS. Face it, VMS is
not the best at this task, and we always owe it to our customers to give
the best advice.
Of course! But if your general purpose web/http server was a gateway to
your VMS-based application server then surely you have the best of all
worlds?
Post by David Froble
In your case, making available data from your VMS system, perhaps a VMS
based web server might be a valid solution.
SSE, WebSockets, Browser TCP Sockets (WebApps), WebRTC - all
functionality attempting to overcome the deficiencies, unreliability,
context and performance of HTTP.

And VMS was offered the solution in 1995 :-(
Post by David Froble
But, really, just what is a web server, when you get down to the bottom
of it? It's a listener socket, with a bunch of capabilities that it can
use to service HTTP connection requests. (Actually, most likely two
listeners, one not SSL and one for SSL.)
You know where the devil is?
Post by David Froble
If anyone doesn't believe the above, just let me know, and I'll e-mail
you a copy of one of the early "proof of concept" programs that was
written to explore servicing a socket connection request.
I see your deuces and raise you a production ready connection-oriented
context-rich solution that just requires the App Developer yo providde a
VMS DLL with 6 routines written in any 3GL.

But, if you're not in with the in-crowd at VMS or willing to bribe Fagan
then you have no chance.
David Froble
2014-03-29 13:42:31 UTC
Permalink
Post by Richard Maher
Post by David Froble
Frankly, if a customer told me he had to implement a general purpose web
server, I'd have to suggest something other than VMS. Face it, VMS is
not the best at this task, and we always owe it to our customers to give
the best advice.
Of course! But if your general purpose web/http server was a gateway to
your VMS-based application server then surely you have the best of all
worlds?
Post by David Froble
In your case, making available data from your VMS system, perhaps a VMS
based web server might be a valid solution.
SSE, WebSockets, Browser TCP Sockets (WebApps), WebRTC - all
functionality attempting to overcome the deficiencies, unreliability,
context and performance of HTTP.
And VMS was offered the solution in 1995 :-(
Post by David Froble
But, really, just what is a web server, when you get down to the bottom
of it? It's a listener socket, with a bunch of capabilities that it can
use to service HTTP connection requests. (Actually, most likely two
listeners, one not SSL and one for SSL.)
You know where the devil is?
Post by David Froble
If anyone doesn't believe the above, just let me know, and I'll e-mail
you a copy of one of the early "proof of concept" programs that was
written to explore servicing a socket connection request.
I see your deuces and raise you a production ready connection-oriented
context-rich solution that just requires the App Developer yo providde a
VMS DLL with 6 routines written in any 3GL.
But, if you're not in with the in-crowd at VMS or willing to bribe Fagan
then you have no chance.
I've seen you plugging your product for years, but I've never seen any
details (just marketing speak) of just what it does, or how to use it.
Are such details available?

What is a "VMS DLL" ????????????????????????????????
j***@gmail.com
2014-03-29 09:18:43 UTC
Permalink
Post by David Froble
So, if you REALLY need to serve connection requests for data from a VMS
system, you can kluge together something with a web server, or you can
implement application specific web services. That's what we do.
Nothing special about it, and it's not difficult. You got 10,000 people
who want to check on inventory availability? No problem. Just set up
an inventory inquiry web service. Hell of lot easier (for me at least)
and most likely much quicker [...]
Huge fan of this approach.

I think by segmenting it this way, you give yourself more modern choices for your
http(s) stack on the linux/windows side and you can also get better control over
the security/network relationship between the VMS backend and the web front end.

On some level, it stinks because poor VMS has been relegated to the backend and
leaves it further isolated from the modern world. But even with that, I think that's
a better way to go than fussing with WASD or Apache on VMS.

EJ
V***@SendSpamHere.ORG
2014-03-29 12:19:11 UTC
Permalink
Post by David Froble
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
From the context above, Eric means Apache and WASD.
Post by Jan-Erik Soderholm
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Nginx is a web server.
A web server on VMS?
Fine then. I read it otherwise.
Eric says Nginx is faster than WASD or Apache but that he thinks it's
_not_ available for VMS. That's the point he appears to be trying to make.
Simon.
And my point was that, why then mention it at all !?
When comparing web servers for VMS, that seems pointless.
Well, no.
Because the original discussion was about web servers in general, not
just VMS web servers.
Frankly, if a customer told me he had to implement a general purpose web
server, I'd have to suggest something other than VMS. Face it, VMS is
not the best at this task, and we always owe it to our customers to give
the best advice.
That's because VMS is NOT a web server; it's an operating system.
Post by David Froble
So, if you REALLY need to serve connection requests for data from a VMS
system, you can kluge together something with a web server, or you can
implement application specific web services. That's what we do.
Nothing special about it, and it's not difficult. You got 10,000 people
who want to check on inventory availability? No problem. Just set up
an inventory inquiry web service. Hell of lot easier (for me at least)
and most likely much quicker than something that has to process HTTP
requests and build web pages for some browser. It's such services that
now process the vast majority of our customer inquiries and orders.
If anyone doesn't believe the above, just let me know, and I'll e-mail
you a copy of one of the early "proof of concept" programs that was
written to explore servicing a socket connection request.
I wrote a small, specialized, web server for my UPShot APC UPS monitoring
utility. It was not difficult. It's also a single image and doesn't have
need for the umpteen gazillion heavyweigh detached processes which weight
down Apache's performance. So, servicing socket connections is nothing at
all too difficult. But then, you already knew that.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.
Jan-Erik Soderholm
2014-03-29 13:57:41 UTC
Permalink
Post by V***@SendSpamHere.ORG
Post by David Froble
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy does it
at twice the performance.
I haven't compared the two, but based on comments, it sounds like the
ngix http server is way faster.
Which "two"?
From the context above, Eric means Apache and WASD.
Post by Jan-Erik Soderholm
Post by j***@gmail.com
http://nginx.org/en/
But I don't think that's available for VMS.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Post by Jan-Erik Soderholm
Post by Simon Clubley
Post by Jan-Erik Soderholm
But then, I do not understand your post.
I need a web server to display data from my VMS server.
How does ngix help?
Nginx is a web server.
A web server on VMS?
Fine then. I read it otherwise.
Eric says Nginx is faster than WASD or Apache but that he thinks it's
_not_ available for VMS. That's the point he appears to be trying to make.
Simon.
And my point was that, why then mention it at all !?
When comparing web servers for VMS, that seems pointless.
Well, no.
Because the original discussion was about web servers in general, not
just VMS web servers.
Frankly, if a customer told me he had to implement a general purpose web
server, I'd have to suggest something other than VMS.
So would probably more or less everyone, incl me. If it doesn't
involve (or in other ways need) VMS, don't use VMS.

But if someone asked me to make some VMS data available from
a web-page, I'd certenly do it using WASD.
Post by V***@SendSpamHere.ORG
Post by David Froble
Face it, VMS is
not the best at this task,
What task? To serve/present VMS-data to "the web"?
Post by V***@SendSpamHere.ORG
Post by David Froble
and we always owe it to our customers to give
the best advice.
That's because VMS is NOT a web server; it's an operating system.
Post by David Froble
So, if you REALLY need to serve connection requests for data from a VMS
system, you can kluge together something with a web server,...
Or simply use standard featurs. No "kludges" there.
Post by V***@SendSpamHere.ORG
Post by David Froble
or you can
implement application specific web services. That's what we do.
So have I done. But there is no trouble with having WASD handling
the network part of it. Has worked "from boot to shutdown" for 6-7
years now with no sysman involment since setup. It just runs.
Post by V***@SendSpamHere.ORG
Post by David Froble
Nothing special about it, and it's not difficult. You got 10,000 people
who want to check on inventory availability? No problem. Just set up
an inventory inquiry web service. Hell of lot easier (for me at least)
and most likely much quicker than something that has to process HTTP
requests and build web pages for some browser.
There is no browser involved, of course. The SOAP request comes
from Excel using "Web Services Toolkit for Office" and the reply
is returned to the same Excel script.
Post by V***@SendSpamHere.ORG
Post by David Froble
It's such services that
now process the vast majority of our customer inquiries and orders.
If anyone doesn't believe the above, just let me know, and I'll e-mail
you a copy of one of the early "proof of concept" programs that was
written to explore servicing a socket connection request.
I know nothing about socket programming. Just built my VMS application
(Cobol business code and a C wrapper) using the supplied gSOAP build
exmaples and added a pass rule in the WASD config. That's it.

There is also a persistent gSOAP server for WASD for better
performance. See this page for a comparision with Apache:
http://wasd.vsm.com.au/SRC/GSOAP/READMORE.HTML

Yes, you can write the network part into every WS application,
but I do not realy see why when there are already specialized
applications (such as WASD) to handle all that.
Post by V***@SendSpamHere.ORG
I wrote a small, specialized, web server for my UPShot APC UPS monitoring
utility. It was not difficult. It's also a single image and doesn't have
need for the umpteen gazillion heavyweigh detached processes which weight
down Apache's performance.
We uses *2*, the main WASD server process and the process running the SOAP
application. Hardly "umpteen gazillion heavyweigh detached processes".

But then, that is why we do not run Apache.

Jan-Erik.
Paul Sture
2014-03-30 11:07:17 UTC
Permalink
Post by V***@SendSpamHere.ORG
Post by David Froble
So, if you REALLY need to serve connection requests for data from a VMS
system, you can kluge together something with a web server, or you can
implement application specific web services. That's what we do.
Nothing special about it, and it's not difficult. You got 10,000 people
who want to check on inventory availability? No problem. Just set up
an inventory inquiry web service. Hell of lot easier (for me at least)
and most likely much quicker than something that has to process HTTP
requests and build web pages for some browser. It's such services that
now process the vast majority of our customer inquiries and orders.
If anyone doesn't believe the above, just let me know, and I'll e-mail
you a copy of one of the early "proof of concept" programs that was
written to explore servicing a socket connection request.
I wrote a small, specialized, web server for my UPShot APC UPS monitoring
utility. It was not difficult. It's also a single image and doesn't have
need for the umpteen gazillion heavyweigh detached processes which weight
down Apache's performance. So, servicing socket connections is nothing at
all too difficult. But then, you already knew that.
That approach nicely removes the overhead of setting up and managing a
webserver as well. The KISS principle at work.

If you really want a webserver into this sort of stuff nginx and lighttpd
mentioned elsethread can act as front end gateways into it, and legacy
Apache based stuff as well.

Oops. I just mentioned legacy and Apache in the same sentence :-)

While we are on the subject Python has the ability to run your own webserver
so you don't even need stuff like MAMP[1] any more for web development.

Need to restart the server to effect config changes? CTRL-C, <UP> <CR>
does that, and there's stuff around which will monitor your config files
and do a server restart automatically for you.

[1] MAMP is an OS X application which you can run on demand to provide
a full LAMP stack. They have a Windows version on the way.

<http://www.mamp.info/en/>
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
Jan-Erik Soderholm
2014-03-30 12:01:43 UTC
Permalink
Post by Paul Sture
Post by V***@SendSpamHere.ORG
Post by David Froble
So, if you REALLY need to serve connection requests for data from a VMS
system, you can kluge together something with a web server, or you can
implement application specific web services. That's what we do.
Nothing special about it, and it's not difficult. You got 10,000 people
who want to check on inventory availability? No problem. Just set up
an inventory inquiry web service. Hell of lot easier (for me at least)
and most likely much quicker than something that has to process HTTP
requests and build web pages for some browser. It's such services that
now process the vast majority of our customer inquiries and orders.
If anyone doesn't believe the above, just let me know, and I'll e-mail
you a copy of one of the early "proof of concept" programs that was
written to explore servicing a socket connection request.
I wrote a small, specialized, web server for my UPShot APC UPS monitoring
utility. It was not difficult. It's also a single image and doesn't have
need for the umpteen gazillion heavyweigh detached processes which weight
down Apache's performance. So, servicing socket connections is nothing at
all too difficult. But then, you already knew that.
That approach nicely removes the overhead of setting up and managing a
webserver as well. The KISS principle at work.
What I find nice with a server such as WASD "in between" is that you
have one centralized place for your mapping and security setting.

How do you decide *who* can connect to the application builtin
web server?
Post by Paul Sture
If you really want a webserver into this sort of stuff nginx and lighttpd
mentioned elsethread can act as front end gateways into it, and legacy
Apache based stuff as well.
Oops. I just mentioned legacy and Apache in the same sentence :-)
While we are on the subject Python has the ability to run your own webserver
so you don't even need stuff like MAMP[1] any more for web development.
Need to restart the server to effect config changes? CTRL-C, <UP> <CR>
does that,...
But that is only if you run the "web-server" from a terminal session !?

WASD:

Reload the mapping files: $ httpd /do=map

No restart af the full server or detached CGI processes,
just an update of the mapping rules on-line visible for
the next request coming in.

Restart the full server: $ httpd /do=restart

That waits for active sessions to finish (while blocking
new sessions) and then does a clean restart without
breaking anything (apart from the block).

Restart the full server: $ httpd /do=restart=now

Restarts "now". Quits/breaks running scripts.

There is also: $ httpd /do=restart=quiet

That restarts the server when there have been a full second with
no incoming request at all. Even nicer. :-)

There are a lot of other "/do=" options to manage cache, SSL,
reopen of log files and so on.

This would be quite a job to implement in your own Pathon
based "web-server". But for single application specific
tasks or for simple proof-of-concepts, yes, maybe.


Jan-Erik.
Post by Paul Sture
and there's stuff around which will monitor your config files
and do a server restart automatically for you.
[1] MAMP is an OS X application which you can run on demand to provide
a full LAMP stack. They have a Windows version on the way.
<http://www.mamp.info/en/>
j***@gmail.com
2014-03-29 09:09:59 UTC
Permalink
Post by Simon Clubley
From the context above, Eric means Apache and WASD.
I was actually attempting to compare Apache and nginx. There was a point I was trying to convey in there, but it was rendered in a very confusing fashion. Sorry about that.

EJ
Paul Sture
2014-03-30 10:29:28 UTC
Permalink
Post by j***@gmail.com
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy
does it at twice the performance.
I haven't compared the two, but based on comments, it sounds like the ngix
http server is way faster.
http://nginx.org/en/
But I don't think that's available for VMS.
No, nginx (pronounced "eNGiNe-X") isn't available for VMS, but it does have
similarities with WASD in that it uses asynchronous processing rather than
the process based approach of Apache.

"Apache is like Microsoft Word, it has a million options but you only need
six. Nginx does those six things, and it does five of them 50 times faster
than Apache."

<https://www.wikivs.com/wiki/Apache_vs_nginx>

and the WASD technical details:

<http://wasd.vsm.com.au/wasd_root/src/httpd/readmore.txt>

P.S. another common high performance server available is lighttpd (pronounced
"lighty")
<https://www.wikivs.com/wiki/Apache_vs_Lighttpd>

A bit more reading: "6 Excellent Linux/Open Source Web Servers"
(except the last 2 are Windows or Linux under Wine ;-) )

<http://www.linuxplanet.com/linuxplanet/reviews/7239/1/>
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
Phillip Helbig---undress to reply
2014-03-30 11:41:16 UTC
Permalink
Post by Paul Sture
Post by Jan-Erik Soderholm
If you don't mind not "going with the mob", WASD is a better web
server on VMS. It can do most things Apache can and it usualy
does it at twice the performance.
Apache Server was originally "a patchy server". Really.
Post by Paul Sture
No, nginx (pronounced "eNGiNe-X") isn't available for VMS, but it does have
similarities with WASD in that it uses asynchronous processing rather than
the process based approach of Apache.
"Apache is like Microsoft Word, it has a million options but you only need
six. Nginx does those six things, and it does five of them 50 times faster
than Apache."
Many folks are happy with Dave Jones's "OSU" web server on VMS.
Simon Clubley
2014-03-28 23:39:38 UTC
Permalink
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
It's all a matter of scale, and to some extent, "monkey see, monkey do".
Apache was not and is not the only option. There are a number of other
Web server programs for non-VMS machines. And there have been others
in the past. Apache is on top because it offers the most in needed
functionality, reliability, ease of administration, etc. If monkey see
server crash monkey not do no matter how many others are running it.
If Apache is so good, then what's wrong with running it on VMS? It's
available there.
When security issues crop up in (for example) Apache, they get fixed
quickly in Linux.

You might wish to compare the time-to-fix for Internet facing software
running on Linux versus the time-to-fix for Internet facing software
such as Apache (and other software provided by HP) running on VMS.

I once spent the time doing that comparison when Kerry provoked me with
one of his out of touch comments. The results at the time were not
pretty for VMS.

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
David Froble
2014-03-29 01:41:52 UTC
Permalink
Post by Simon Clubley
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
It's all a matter of scale, and to some extent, "monkey see, monkey do".
Apache was not and is not the only option. There are a number of other
Web server programs for non-VMS machines. And there have been others
in the past. Apache is on top because it offers the most in needed
functionality, reliability, ease of administration, etc. If monkey see
server crash monkey not do no matter how many others are running it.
If Apache is so good, then what's wrong with running it on VMS? It's
available there.
When security issues crop up in (for example) Apache, they get fixed
quickly in Linux.
You might wish to compare the time-to-fix for Internet facing software
running on Linux versus the time-to-fix for Internet facing software
such as Apache (and other software provided by HP) running on VMS.
I once spent the time doing that comparison when Kerry provoked me with
one of his out of touch comments. The results at the time were not
pretty for VMS.
Simon.
I agree. We all know that VMS has been neglected.
Michael Kraemer
2014-03-29 01:05:42 UTC
Permalink
Post by David Froble
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems you
used to play with.
Would it have been possible to make a VS3100 at the price point
of a contemporary (ca 1987) PC?
Iirc the price ratio was an order of magnitude, if not more.
Post by David Froble
It's actually very simple. You either defend your market, or you lose
it. Lots of examples, free internet explorer, things like that. If you
keep competitors out of your market, and it's a decent market, then you
don't lose it and you prosper.
How successful is Blackberry these days in defending their market?
David Froble
2014-03-29 01:43:20 UTC
Permalink
Post by Michael Kraemer
Post by David Froble
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems
you used to play with.
Would it have been possible to make a VS3100 at the price point
of a contemporary (ca 1987) PC?
Iirc the price ratio was an order of magnitude, if not more.
Post by David Froble
It's actually very simple. You either defend your market, or you lose
it. Lots of examples, free internet explorer, things like that. If
you keep competitors out of your market, and it's a decent market,
then you don't lose it and you prosper.
How successful is Blackberry these days in defending their market?
You don't always win.

If you never try, then you surely will lose.
JF Mezei
2014-03-29 02:45:15 UTC
Permalink
Post by Michael Kraemer
Would it have been possible to make a VS3100 at the price point
of a contemporary (ca 1987) PC?
Iirc the price ratio was an order of magnitude, if not more.
Why not ? Where there is a will there is a way.

DEC was in a state of denial over the PC and still telling its employees
customers wanted to pay a premium for "DEC" quality. So designers
continued to build tanks without much emphasis on reducing the cost of
producing the units.

At that point, there weren't really good buses like PCI, so the
proprietary bus for the 3100 (although not really a bus) could be
excused, which also means proprietary SCSI interface and proprietary
ethernet interface.

The graphics card is the interesting problem. Develop your own, with
known interface and features, and spend considerable amout of time
reverse engineering el-cheapo PC cards who plug into an inferior bus
that the 3100 didn't have ?

However, DEC could have designed a single colour graphics card, instead
of the B/W and then the colour one. And if it had priced the system to
compete, the higher volumes would have spread the design cocts amongst
more units, reducing the cost per unit. But that takes management
guidance and vision, and at that point, DEC was aiming to price itself
like IBM instead of competing against Compaq.
Paul Sture
2014-03-30 11:11:48 UTC
Permalink
Post by Michael Kraemer
Post by David Froble
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems you
used to play with.
Would it have been possible to make a VS3100 at the price point
of a contemporary (ca 1987) PC?
Iirc the price ratio was an order of magnitude, if not more.
Dunno. In the mid-nineties I worked at a big IBM shop and I was very
surprised to find that the PS/2 systems they had had commanded a similar
price to the quotes I got for VS2000 systems in the late 80s.

We are talking IBM prices there of course.
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
Bill Gunshannon
2014-03-31 14:14:45 UTC
Permalink
Post by Paul Sture
Post by Michael Kraemer
Post by David Froble
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems you
used to play with.
Would it have been possible to make a VS3100 at the price point
of a contemporary (ca 1987) PC?
Iirc the price ratio was an order of magnitude, if not more.
Dunno. In the mid-nineties I worked at a big IBM shop and I was very
surprised to find that the PS/2 systems they had had commanded a similar
price to the quotes I got for VS2000 systems in the late 80s.
We are talking IBM prices there of course.
Products targeted at two different sectors. The PS/2 was a business
machine, the PC a home machine (which found its way into business
which probably contributed greatly to the demise of the PS/2.
Mandatory Anecdote: We used to do consulting for local businesses.
Blue Cross was an IBM Mainframe and, natuarally, PS/2 shop. With
research we determined that they need more memory for their PS/2's.
We recommended Kingston SIMMs. They balked and said the memory had
to be real IBM PS/2 memory. The cost was more than double. Opening
the boxes revealed Kingston SIMMs!)

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Bill Gunshannon
2014-03-31 14:07:17 UTC
Permalink
Post by Michael Kraemer
Post by David Froble
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems you
used to play with.
Would it have been possible to make a VS3100 at the price point
of a contemporary (ca 1987) PC?
Iirc the price ratio was an order of magnitude, if not more.
I think more research and expense went into deciding that price point
than into the actual engineering of the first PC. it held at that
point for a long time. No DEC product ever came close.
Post by Michael Kraemer
Post by David Froble
It's actually very simple. You either defend your market, or you lose
it. Lots of examples, free internet explorer, things like that. If you
keep competitors out of your market, and it's a decent market, then you
don't lose it and you prosper.
How successful is Blackberry these days in defending their market?
Some of us never really understood the success of Blackberry from the
beginning. I certainly don't understand the US Government (especially
DOD) reliance on it for so long. (Although corruption goes a long way
in explaining most of the crazy decisions made in that sector!!!)

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Bill Gunshannon
2014-03-29 20:39:23 UTC
Permalink
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
It's all a matter of scale, and to some extent, "monkey see, monkey do".
Apache was not and is not the only option. There are a number of other
Web server programs for non-VMS machines. And there have been others
in the past. Apache is on top because it offers the most in needed
functionality, reliability, ease of administration, etc. If monkey see
server crash monkey not do no matter how many others are running it.
If Apache is so good, then what's wrong with running it on VMS? It's
available there.
I'm not the one complaining about it. I see mo problem with running
Apache on VMS, even if it did originiate on Unix.
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
People ask their peers how they do things, get told to use Apache, and
then do so.
What would you expect them to do? Run the one no one else is using?
There's probably a reason for that!!
This is where things get amusing. Here you accept the "follow the mob"
concept.
Cars all have round wheels. Are yours square so you don't "follow the mob"?
Sometimes people "follow the mob" because the mob is right.
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
Not that running a pointy clicky web server is all there is to computing
these days.
Pointy-clicky is what the masses want. In many cases it can handle the
majority of the needed configing. Of course, with Apache you can always
fall back to manually editing the config if that floats your boat.
I was referring to users running browsers ...
What wold you have them run Lynx?
Post by David Froble
Post by Bill Gunshannon
It is probably the what "computers" get used for the most. I expect the
ratio of servers to desktops is at least 1:10,000. Even when your talking
Windows servers. (As an aside, I have finally had a server go above 90% CPU.
The good news is I had anticipated replacing this one and the new boxes are
just about ready.)
Post by David Froble
However, it is the most visable stuff for all the office paper pusher
weenies.
See, there it is again. If you are going to spend your life insulting
potential users, why would you expect them to come to you for business?
Just because you think anything not done on VMS is done by "weenies" the
IT world doesn't agree and displaying an attitude like this can only hurt
your business. (Or your chances at getting a job in the IT world.)
YOu miss the point that I think we all agree upon, as do you with the
ration mentioned above.
What point am I missing? Some people here have this wierd desire to
denigrate anything not VMS. Frequently to the extreme of openly insulting
anyone who uses anything else. Although some people may not have figured
it out yet, employers frequently check the INTERNET to learn about potential
employees. Just ask "Cisco Fatty".
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
When IBM wanted to introduce the PC, what if DEC had offered the VAX CPU
and VMS for the effort.
They were never even in consideration. And given the state of the CPU
at that time would have been a dismal failure.
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems you
used to play with.
And, it would very likely have been rejected for the same reason that
the M68K was. Hint: It had nothing to do with chip capabilities but
was based solely on the willingness of the chip vendor to screw its
other customers.
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
Microsoft and Intel would have been the big
losers, and the world would have been the big winners.
Hardly. Intel was chosen because they had one foot in the grave. Contrary
to the recent publication by some computer museum the 8088 was not a popular
commodity chip at the time. it was a miserable hack which without IBM would
have never seen major production. Intel was not IBM's first choice, (neither
was VAX). And for an OS, MicroSoft was not IBM's first choice either.
Well, there you go, and it was the applications that made the PC. If
the applications would have been written for the VAX ....
Who said it was any application. The original IBM-PC was selling like
hotcakes with no applications available at all. The first application
I ever saw was WordStar and being as that could already be used on CP/M
I hardly think it was a big sales pusher for the PC. There actually
would have been more application if they had gone with the M68K as IBM
already had a professional line using that CPU which included an OS and
applications.
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
"What, sell VMS for $98 a copy ??? Not on your life." Turns out it was
DEC's life that took the bullet.
I'm betting there was some Harvard MBAs counting their little beans who
were so much against things like the above happening.
I doubt it was anything that simple.
It's actually very simple. You either defend your market, or you lose
it. Lots of examples, free internet explorer, things like that. If you
keep competitors out of your market, and it's a decent market, then you
don't lose it and you prosper.
Post by Bill Gunshannon
Post by David Froble
Post by Bill Gunshannon
Post by V***@SendSpamHere.ORG
The WEENDOZE crowd
opted for W(ie)RD. Gack. And it's nasty stuff if you've ever tasted it.
Luckily, the rest of the IT world isn't as irrational as you. But even
putting that aside, there are numerous non-Windows Office Suites. For
most common OSes. Why did the VMS community never write one for their
favorite OS? Instead, it appears, they complain beecause the Unix
community didn't do it for them.
Definition of a "leader". Someone who sees which direction the mob is
moving, runs in front and hollers "follow me".
Your definition, I guess.
But, but, above you advocated using the most popular web server. What's
the difference ?
Apache didn't jump in front of any mob and holler "follow me". They
created (and continuously improved) a product that just happened to
out perform the competition. And at a better price. :-)
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
Then there is the smug few who say "why should I follow (or lead) the
mob?" Simple answer. Seems once upon a time a bank robber was asked
"why do you rob banks", and the reply was "because that's where the
money is".
That was Clyde Parker. And we all know where he ended up.
Doesn't mean anything to the concept. If you're in business, you
provide the services people want, if you want to continue in that business.
WHich I guess explains the demise of VMS. :-)
Post by David Froble
Post by Bill Gunshannon
Is it really being stupid to see which way the wind is blowing and to
decide not to try to buck it?
So, again you "follow the mob", huh?
You keep using "follow the mob" as a perjorative. You know, not everyone
who follows is a lemming and not everyone who leads does not deserve to be
followed. A wise man can tell the difference. Others go out of business.
Post by David Froble
Post by Bill Gunshannon
I am still waiting for someone to send/post/point-me-to a real full VMS
license. Why? because I seriously expect there is a clause that let's
them withdraw permission to use the product any time they wish and I
doubt any legitimate business would continue to use it afterwards.
[I have seen such clauses in products that they wanted me to use here
and have recommended against them specifically because of those clauses.]
bill
I do not think I have one easily available, but I'll look.
Can someone please send this guy a copy of a VMS license?
From memory, it is a license for perpetial use of the software on a
particular computer. Yeah, that is a problem when the computer no
longer works. But I've never seen anything that allows the owners of
VMS to say "stop using the software".
I will hold my judgement until I see one, but the one caveat you mentioned
is more than enough to predict the demise of VMS. When HP is no longer
willing to sell (or more likely interested in selling) transfers of VMS
licenses, what will customers do? And anyone who even considers waiting
for that day still running it is a fool.

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Paul Sture
2014-03-30 11:17:34 UTC
Permalink
Post by Bill Gunshannon
What point am I missing? Some people here have this wierd desire to
denigrate anything not VMS. Frequently to the extreme of openly insulting
anyone who uses anything else. Although some people may not have figured
it out yet, employers frequently check the INTERNET to learn about potential
employees. Just ask "Cisco Fatty".
Apparently a lot of the media reports on "Cisco Fatty" misconstrued what
really happened. No surprise there I suppose :-)

"Morgan Stanley Uses The "CiscoFatty" Story To Teach Its New Hires How NOT
To Use Twitter And Facebook

---quote ---
She told the website CP24, "The 'job' I 'lost' was merely a summer
internship I declined before I even tweeted about it," she said in an
email. "Any information you have pulled from other sources has largely
been made up by bloggers."
--- end quote ---

<http://www.businessinsider.com/morgan-stanley-uses-the-ciscofatty-story-to-teach-its-new-hires-how-to-use-twitter-and-facebook-2011-1>
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
Bill Gunshannon
2014-03-31 14:19:36 UTC
Permalink
Post by Paul Sture
Post by Bill Gunshannon
What point am I missing? Some people here have this wierd desire to
denigrate anything not VMS. Frequently to the extreme of openly insulting
anyone who uses anything else. Although some people may not have figured
it out yet, employers frequently check the INTERNET to learn about potential
employees. Just ask "Cisco Fatty".
Apparently a lot of the media reports on "Cisco Fatty" misconstrued what
really happened. No surprise there I suppose :-)
But that doesn't change the result.
Post by Paul Sture
"Morgan Stanley Uses The "CiscoFatty" Story To Teach Its New Hires How NOT
To Use Twitter And Facebook
---quote ---
She told the website CP24, "The 'job' I 'lost' was merely a summer
internship I declined before I even tweeted about it," she said in an
email. "Any information you have pulled from other sources has largely
been made up by bloggers."
--- end quote ---
<http://www.businessinsider.com/morgan-stanley-uses-the-ciscofatty-story-to-teach-its-new-hires-how-to-use-twitter-and-facebook-2011-1>
OK, so she "only" lost a summer internship. I don't know what it is like
on the left coast, but my students have to fight pretty hard for a decent
internship. Especially one that actually pays and isn't a "go get me a
cup of coffee" job. And do you not think this gaffe will have some affect
on other future possibilities and real jobs? Stupid is stupid at any level.
In this web ruled era one should be very careful about what they say in
public cause there are no takebacks or do overs. I am always left laughing
when I receive an email "recalling" a previous email. Ask Ollie North
how that works. :-)

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Johnny Billquist
2014-03-30 14:09:02 UTC
Permalink
Post by Bill Gunshannon
Post by David Froble
Post by Bill Gunshannon
Post by David Froble
When IBM wanted to introduce the PC, what if DEC had offered the VAX CPU
and VMS for the effort.
They were never even in consideration. And given the state of the CPU
at that time would have been a dismal failure.
I don't remember when the first PC was introduced, nor do I remember
when the C-VAX was introduced. However, the C-VAX would have been
perfect for the PC. It's what was in those VAXstation 3100 systems you
used to play with.
And, it would very likely have been rejected for the same reason that
the M68K was. Hint: It had nothing to do with chip capabilities but
was based solely on the willingness of the chip vendor to screw its
other customers.
This is a meaningless discussion.

The IBM PC was introduced in August 1981.

The 8088 was introduced in July 1979. It was the CPU used in the
original IBM PC.

The 68000 initial samples were available in February 1980, so it could
potentially have been selected, but IBM had no reason to go there. They
had used Intel stuff before, but not Motorola. IBM could have used its
own chips, like the 801, but decided not to, for various reasons.

The MicroVAX I was introduced in 1984, way after the IBM PC came out.
The C-VAX was yet another few years beyond that. Thus, the whole
argument about using a VAX for the PC is more than silly. It did not
exist back then. And it wasn't until the uVAX II that you actually had
the VAX (with limitations) on one chip. And that was in 1985.

The PDP-11 could have been a potential choice, but for that, DEC was not
a good partner (seem DEC was a bit protective about it, as well as
pricing it way high), and besides, it would have been a weird day when
IBM used an architecture from DEC.

Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: ***@softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Michael Kraemer
2014-03-31 01:42:09 UTC
Permalink
Post by Johnny Billquist
This is a meaningless discussion.
Not quite. Unlike the first VAXen,
the VS31xx mentioned by the OPs weren't too far
from contemporary PCs.
Except, of course, the price point,
and DECs lack of desire (or capability?)
to compete at the low end.
Post by Johnny Billquist
The 68000 initial samples were available in February 1980, so it could
potentially have been selected, but IBM had no reason to go there. They
had used Intel stuff before, but not Motorola. IBM could have used its
own chips, like the 801, but decided not to, for various reasons.
Iirc correctly, they used 68K in mainframe emulators,
and it would have been a better match if they'd used it in PCs
as well.
A descendant of the 801, ROMP, was used in mid-1980s RT PC,
but it wasn't too impressive compared to the AT.
Post by Johnny Billquist
The MicroVAX I was introduced in 1984, way after the IBM PC came out.
The C-VAX was yet another few years beyond that. Thus, the whole
argument about using a VAX for the PC is more than silly. It did not
exist back then. And it wasn't until the uVAX II that you actually had
the VAX (with limitations) on one chip. And that was in 1985.
Well, as long as people were happy to pay for expensive big VAXen,
there weren't too many reasons for DEC to go single chip.
Johnny Billquist
2014-03-31 11:52:16 UTC
Permalink
Post by Michael Kraemer
Post by Johnny Billquist
This is a meaningless discussion.
Not quite. Unlike the first VAXen,
the VS31xx mentioned by the OPs weren't too far
from contemporary PCs.
Except, of course, the price point,
and DECs lack of desire (or capability?)
to compete at the low end.
But contemporary PCs had already settled on the x86 for many years then.
Hardly an argument why they should/could have used VAX instead.
Post by Michael Kraemer
Post by Johnny Billquist
The 68000 initial samples were available in February 1980, so it could
potentially have been selected, but IBM had no reason to go there.
They had used Intel stuff before, but not Motorola. IBM could have
used its own chips, like the 801, but decided not to, for various
reasons.
Iirc correctly, they used 68K in mainframe emulators,
and it would have been a better match if they'd used it in PCs
as well.
Mainframe emulators? That was most likely after the PC had come out as
well, which then changes nothing of my point.
Post by Michael Kraemer
A descendant of the 801, ROMP, was used in mid-1980s RT PC,
but it wasn't too impressive compared to the AT.
IBM was too successful with the PC. As was Intel, it turned out. Nothing
have been able to replace it, no matter what technical merit there might
have been.
Post by Michael Kraemer
Post by Johnny Billquist
The MicroVAX I was introduced in 1984, way after the IBM PC came out.
The C-VAX was yet another few years beyond that. Thus, the whole
argument about using a VAX for the PC is more than silly. It did not
exist back then. And it wasn't until the uVAX II that you actually had
the VAX (with limitations) on one chip. And that was in 1985.
Well, as long as people were happy to pay for expensive big VAXen,
there weren't too many reasons for DEC to go single chip.
There were. DEC was working hard on getting to a single chip. But it
took them (and technology) that long before they managed with the VAX.

Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: ***@softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
glen herrmannsfeldt
2014-03-31 18:25:08 UTC
Permalink
Johnny Billquist <***@softjar.se> wrote:

(snip)
Post by Johnny Billquist
Post by Michael Kraemer
Iirc correctly, they used 68K in mainframe emulators,
and it would have been a better match if they'd used it in PCs
as well.
Mainframe emulators? That was most likely after the PC had come out as
well, which then changes nothing of my point.
IBM's emulation of S/370 goes back to the XT/370 and AT/370, card
sets that fit into slots in a PC/XT or PC/AT, respectively, and,
with appropriate software on both sides, emulate VM/CMS.

Most user mode S/370 instructions are executed by actual hardware,
but much supervisor mode is done by 8088 or 80286 instructions,
especially for I/O.

Usual VM/CMS runs VM on actual hardware, and CMS inside a virtual
machine, with CP (the part that actually makes virtual machines)
all written in S/370 code. VM/PC is partly in S/370 and partly in
8088 code, but runs ordinary CMS.

Later emulation of S/370, the P/370, runs only S/370 code for both
supervisor and user, but I/O devices are emulated in 80386 code,
running on OS/2. S/370 code will execute a channel program for disk
I/O with the SIO (Start I/O) instruction, the processing done by the
channel on actual S/370 hardware is done by the PS/2, when done
an I/O interrupt is passed to the S/370 system, and it continues
on just as it should.

I/O devices can be emulated, such as tapes using disk files,
or 9 track tapes can be emulated using SCSI tape drives.
Card readers are emulated using disk files as input, and line printers
as disk files for output.

-- glen
Mark Daniel
2014-03-28 23:47:43 UTC
Permalink
On 29/03/14 4:37 AM, Bill Gunshannon wrote:
8< snip 8<
Post by Bill Gunshannon
Post by David Froble
Now, if everyone who wants to run a web server on VMS would send the OSU
developer (damn senile old man cannot remember his name) and Mark a
twenty or a fifty, maybe they'd spend more time and add the features
that Apache gets, and then VMS users would have 2 fine options.
Sorry, that ship has sailed. Anyone who would spend their time putting
in that kind of effort for an EOLed system is a fool.
8< snip 8<

Anyone who would spend their time in a discussion of the (de)merits of
an EOLed system is a fool. (Oops.)
JF Mezei
2014-04-03 03:30:58 UTC
Permalink
Post by V***@SendSpamHere.ORG
Well, I have clients who will very likely never move off of VMS. They've
got new Integrity boxes too but have code that has been built for and run-
ning on VMS for decades and they continure to develope it too! Hopelessly
Pathetic, despite their total disdain for VMS, will not stop its use.
Performance-wise, how long can these customers remain stuck on Tukwila
era boxes ?

Long term, do they plan on going on x86 emulating itaniums to run VMS ?

or are you saying that they haven't had that discussion yet ?

When happens when they need a portion of IPv6 which doesn't work quite
well on VMS and there is no engineerinfg left to fix it ?
Simon Clubley
2014-03-28 18:16:13 UTC
Permalink
Post by David Froble
Post by JF Mezei
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Oh yes they do. :-)

What happens when a exploit is discovered and there's no-one around to
fix it ?

What happens when you have to replace existing hardware with something
different and there are no drivers available in your existing environment
for the new hardware ?

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
David Froble
2014-03-28 22:39:24 UTC
Permalink
Post by Simon Clubley
Post by David Froble
Post by JF Mezei
If you have a specialised factory that was built with machines running
VMS who control all the machinery, as long as the factory remains the
same, there is no real need to update the operating system. It runs, and
it runs well.
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Oh yes they do. :-)
Oh, no they don't. :-)
Post by Simon Clubley
What happens when a exploit is discovered and there's no-one around to
fix it ?
That in no way stops the software from doing what it has always done.
It just means that you might need to take action to stop such exploits.
Post by Simon Clubley
What happens when you have to replace existing hardware with something
different and there are no drivers available in your existing environment
for the new hardware ?
Now, there is a problem. Though, drivers have been written. How about
the LD driver, and the stuff to write CDs and DVDs?

A more difficult is hardware to interface with the rest of the computer.
An example might be S-ATA on an old VAX.

But, the software does not wear out, and it will continue to do what it
has always done. (If it can still find the hardware ..)
Simon Clubley
2014-03-28 23:31:54 UTC
Permalink
Post by David Froble
Post by Simon Clubley
Post by David Froble
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Oh yes they do. :-)
Oh, no they don't. :-)
Post by Simon Clubley
What happens when a exploit is discovered and there's no-one around to
fix it ?
That in no way stops the software from doing what it has always done.
It just means that you might need to take action to stop such exploits.
In essence it does if continuing to run your vulnerable software
leaves your system wide open. A example might help:

Imagine you are running a SMTP/FTP/HTTP/whatever server to receive
transactions or other data from customers or partners.

Now imagine someone finds a vulnerability within the SMTP/FTP/HTTP/whatever
implementation which is triggered by a protocol sequence containing
exploit triggering data.

How do you stop that exploit sequence without stopping all other legitimate
activity especially if there are multiple possible triggers (and which
could not be easily identified by a intelligent firewall) ?

The usual way to tell the vendor to fix the problem and issue a new
binary. What happens when the binary is frozen and you no longer have
a vendor to fix the new security issue ?
Post by David Froble
Post by Simon Clubley
What happens when you have to replace existing hardware with something
different and there are no drivers available in your existing environment
for the new hardware ?
Now, there is a problem. Though, drivers have been written. How about
the LD driver, and the stuff to write CDs and DVDs?
A more difficult is hardware to interface with the rest of the computer.
An example might be S-ATA on an old VAX.
With really old equipment, sometimes it can be as simple as putting in
a larger disk as the smaller ones are no longer available. I've gone
through multiple instances of PC BIOS lockups over the years due to disk
size limits being exceeded. (I have not seen this in recent equipment,
but don't forget some VAXes have had disk size limits as well.)

That's only one simple example; there are many ways you can be caught
out with trying to run newer hardware in a older system.

For example, what happens if you need to interface a new type of
printer for which drivers are not available for the software you want
to continue running ?

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
David Froble
2014-03-29 01:53:14 UTC
Permalink
Post by Simon Clubley
Post by David Froble
Post by Simon Clubley
Post by David Froble
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Oh yes they do. :-)
Oh, no they don't. :-)
Post by Simon Clubley
What happens when a exploit is discovered and there's no-one around to
fix it ?
That in no way stops the software from doing what it has always done.
It just means that you might need to take action to stop such exploits.
In essence it does if continuing to run your vulnerable software
Imagine you are running a SMTP/FTP/HTTP/whatever server to receive
transactions or other data from customers or partners.
Now imagine someone finds a vulnerability within the SMTP/FTP/HTTP/whatever
implementation which is triggered by a protocol sequence containing
exploit triggering data.
How do you stop that exploit sequence without stopping all other legitimate
activity especially if there are multiple possible triggers (and which
could not be easily identified by a intelligent firewall) ?
The usual way to tell the vendor to fix the problem and issue a new
binary. What happens when the binary is frozen and you no longer have
a vendor to fix the new security issue ?
1) Given specifics, I could probably find some type of fix for just
about anything you can think up.

but

2) what about a closed system, not networked outside the shop, such as
JF's paper mill, or someone else mentioned a steel mill in the past.

Not every job is based upon SMTP/FTP/HTTP and such. Also, Steve will
tell you that if you're depending on FTP, it's already hopelessly broke.

:-)
Post by Simon Clubley
Post by David Froble
Post by Simon Clubley
What happens when you have to replace existing hardware with something
different and there are no drivers available in your existing environment
for the new hardware ?
Now, there is a problem. Though, drivers have been written. How about
the LD driver, and the stuff to write CDs and DVDs?
A more difficult is hardware to interface with the rest of the computer.
An example might be S-ATA on an old VAX.
With really old equipment, sometimes it can be as simple as putting in
a larger disk as the smaller ones are no longer available. I've gone
through multiple instances of PC BIOS lockups over the years due to disk
size limits being exceeded. (I have not seen this in recent equipment,
but don't forget some VAXes have had disk size limits as well.)
That's only one simple example; there are many ways you can be caught
out with trying to run newer hardware in a older system.
For example, what happens if you need to interface a new type of
printer for which drivers are not available for the software you want
to continue running ?
I set up a cheap PC, with a service that accepts files to print, and I
use a client on the VMS side to send the stuff to print to the cheap PC.
Post by Simon Clubley
Simon.
Again, some thought and ingenuity can go a long way to solving many
problems.

But nothing you say here would cause software to just stop working as it
has in the past.
Simon Clubley
2014-03-29 09:46:50 UTC
Permalink
Post by David Froble
But nothing you say here would cause software to just stop working as it
has in the past.
Ok, what about when some internal time counter limit is reached ?

2038 is the most well known remaining one, but there are plenty of other
possibilities based on how the code may have been written.

For a past example, what about the VMS specific 10K days issue in 1997 ?

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
David Froble
2014-03-29 13:50:30 UTC
Permalink
Post by Simon Clubley
Post by David Froble
But nothing you say here would cause software to just stop working as it
has in the past.
Ok, what about when some internal time counter limit is reached ?
2038 is the most well known remaining one, but there are plenty of other
possibilities based on how the code may have been written.
For a past example, what about the VMS specific 10K days issue in 1997 ?
Simon.
You're really reaching now, thinking up possibilities, which may never
exist.

The VMS date format seems to be good for the forseeable future. If some
idiot used the "C" date instead, well, that was a mistake.

The way I see things, if software works, it will continue to do so. If
there is a problem, well, it's always been there, just undiscovered. I
don't think that "discovery" stops software from working. If anything
like that was true, then because it was "broken" from day one, it never
worked, and your thinking it was working until a problem was discovered
was just your imagination.

:-)
Paul Sture
2014-03-30 18:16:53 UTC
Permalink
Post by Simon Clubley
Post by David Froble
But nothing you say here would cause software to just stop working as it
has in the past.
Ok, what about when some internal time counter limit is reached ?
2038 is the most well known remaining one, but there are plenty of other
possibilities based on how the code may have been written.
For a past example, what about the VMS specific 10K days issue in 1997 ?
20-Jun-2007 marked another example of software "wearing out", though
this was to do with the OS version number rather than a date.

The date concerned was when OS X 10.4.10 was released[1], according to
<http://www.robservatory.com/?p=46>

As I recall, someone at Apple forgot to increase the size of the version
number string field in the relevant header files, so even recompiling an
app would end up with truncation.

Oops.

[1] The accompanying Apple article about that release is dated a little
later, 02-Jul-2007:

<http://support.apple.com/kb/DL237>
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
JF Mezei
2014-03-29 00:21:21 UTC
Permalink
Post by David Froble
Post by Simon Clubley
Post by David Froble
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Oh yes they do. :-)
Oh, no they don't. :-)
Oh, yes they do. :-)


Consider software that breaks when after running for a while, one of its
logfile hits the magical version number of 32,768 ? If it breaks after
running for a while, I would call that "wear out" :-)
V***@SendSpamHere.ORG
2014-03-29 12:10:42 UTC
Permalink
Post by JF Mezei
Post by David Froble
Post by Simon Clubley
Post by David Froble
YEs, such things as operating systems and software continue to do what
they were originally implemented to do. They don't wear out.
Oh yes they do. :-)
Oh, no they don't. :-)
Oh, yes they do. :-)
Consider software that breaks when after running for a while, one of its
logfile hits the magical version number of 32,768 ? If it breaks after
running for a while, I would call that "wear out" :-)
I'd call that a lack of system management/maintenance.

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

I speak to machines with the voice of humanity.
Michael Moroney
2014-03-29 13:21:00 UTC
Permalink
Post by V***@SendSpamHere.ORG
Post by JF Mezei
Consider software that breaks when after running for a while, one of its
logfile hits the magical version number of 32,768 ? If it breaks after
running for a while, I would call that "wear out" :-)
I'd call that a lack of system management/maintenance.
$ PURGE ;)
PURGE won't do it, it'll leave the remaining file(s) with a high version
number waiting for the magic ;32767 to roll around.

I once wrote a quickie utility to rename the remaining log files to ;1, ;2
etc. if they wanted a few versions around. Of course if you only needed
the latest a $ PURGE followed by a $ RENAME x.x ;1 is all you need.
Paul Sture
2014-03-29 20:31:05 UTC
Permalink
Post by Michael Moroney
Post by V***@SendSpamHere.ORG
Post by JF Mezei
Consider software that breaks when after running for a while, one of its
logfile hits the magical version number of 32,768 ? If it breaks after
running for a while, I would call that "wear out" :-)
I'd call that a lack of system management/maintenance.
$ PURGE ;)
PURGE won't do it, it'll leave the remaining file(s) with a high version
number waiting for the magic ;32767 to roll around.
I once wrote a quickie utility to rename the remaining log files to ;1, ;2
etc. if they wanted a few versions around. Of course if you only needed
the latest a $ PURGE followed by a $ RENAME x.x ;1 is all you need.
When you are creating tens of thousands of spool and temporary files a day
(with mostly different names, so 32767 takes a few months to come around)
something a little more robust is required.

But the one I wrote was more than a quickie but I suppose it didn't take
more than an hour or two to put together; obviously more than that for
testing before releasing into production.
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
V***@SendSpamHere.ORG
2014-03-29 22:26:21 UTC
Permalink
Post by Paul Sture
Post by Michael Moroney
Post by V***@SendSpamHere.ORG
Post by JF Mezei
Consider software that breaks when after running for a while, one of its
logfile hits the magical version number of 32,768 ? If it breaks after
running for a while, I would call that "wear out" :-)
I'd call that a lack of system management/maintenance.
$ PURGE ;)
PURGE won't do it, it'll leave the remaining file(s) with a high version
number waiting for the magic ;32767 to roll around.
I once wrote a quickie utility to rename the remaining log files to ;1, ;2
etc. if they wanted a few versions around. Of course if you only needed
the latest a $ PURGE followed by a $ RENAME x.x ;1 is all you need.
When you are creating tens of thousands of spool and temporary files a day
(with mostly different names, so 32767 takes a few months to come around)
something a little more robust is required.
But the one I wrote was more than a quickie but I suppose it didn't take
more than an hour or two to put together; obviously more than that for
testing before releasing into production.
I have mysql running on Linux. In /var/log/mysql are "binary" archives that
are generated daily of the form: mysql-bin.###### at several hundreds of MBs.
Not keeping after these is tantamount to not minding the version number of a
VMS log file. In time, these mysql-bin.###### consume an enormous amount of
storage space. A self-inflicted denial of service from running out of disk
space is just as much a software "wear out" as creating .LOG;32767.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.
JF Mezei
2014-03-29 22:18:03 UTC
Permalink
Post by V***@SendSpamHere.ORG
I have mysql running on Linux. In /var/log/mysql are "binary" archives that
are generated daily of the form: mysql-bin.###### at several hundreds of MBs.
mine are in /private/var/mysql

As a rule of thumb, which ones are safe to delete ? Anything older than
X, or are those incremental in nature so some contain stuff not
contained in more recent ones ?

(I have some large and some of only a few hundred bytes).
Paul Sture
2014-03-30 08:41:23 UTC
Permalink
Post by JF Mezei
Post by V***@SendSpamHere.ORG
I have mysql running on Linux. In /var/log/mysql are "binary" archives that
are generated daily of the form: mysql-bin.###### at several hundreds of MBs.
mine are in /private/var/mysql
Assuming you are using the default location for the MySQL config file, this
will show you where your data files are:

grep datadir /etc/my.cnf

and for the error log:

grep log-error /etc/my.cnf
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
V***@SendSpamHere.ORG
2014-03-29 16:19:29 UTC
Permalink
Post by Michael Moroney
Post by V***@SendSpamHere.ORG
Post by JF Mezei
Consider software that breaks when after running for a while, one of its
logfile hits the magical version number of 32,768 ? If it breaks after
running for a while, I would call that "wear out" :-)
I'd call that a lack of system management/maintenance.
$ PURGE ;)
PURGE won't do it, it'll leave the remaining file(s) with a high version
number waiting for the magic ;32767 to roll around.
I once wrote a quickie utility to rename the remaining log files to ;1, ;2
etc. if they wanted a few versions around. Of course if you only needed
the latest a $ PURGE followed by a $ RENAME x.x ;1 is all you need.
Sorry that my smiley didn't show up in your newsreader.

http://tmesis.org/downloads/REVERSION.COM
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.
Paul Sture
2014-03-29 20:51:04 UTC
Permalink
Post by V***@SendSpamHere.ORG
http://tmesis.org/downloads/REVERSION.COM
Thanks for posting that.

FWIW the one I did way back when had to work when production was in full
swing so had to cope with files being created while it was running.

BTW the reference number you use for each modification looks like
year.day_no.mod_no. Maybe the 2102 isn't the year, but something else :-)

$! Genesis: 2102.356.1 21-DEC-2012
$! History: 2102.356.2 21-DEC-2012
^^

Typos?
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
V***@SendSpamHere.ORG
2014-03-30 12:00:59 UTC
Permalink
Post by Paul Sture
Post by V***@SendSpamHere.ORG
http://tmesis.org/downloads/REVERSION.COM
Thanks for posting that.
FWIW the one I did way back when had to work when production was in full
swing so had to cope with files being created while it was running.
BTW the reference number you use for each modification looks like
year.day_no.mod_no. Maybe the 2102 isn't the year, but something else :-)
$! Genesis: 2102.356.1 21-DEC-2012
$! History: 2102.356.2 21-DEC-2012
^^
Typos?
No, I jumped inot my DeLorean, fired up the flux capacitor and sped off at
88 miles per hour to the year 2102. ;)

Yup, looks like dyslexic fingers to blame.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.
Jan-Erik Soderholm
2014-03-27 21:40:30 UTC
Permalink
Post by Ron
forgive me for any formatting issues... I'm kind of new to these Google Groups.
Follow-up questions & comments for Gerard and Jan-Erik (and anyone else
1. I agree that HP is acting like they want VMS to go away. I also
agree that it will be difficult for that to happen. My career started
with Tandem and DEC, so I can completely relate that there are
characteristics of these legacy platforms that are hard to replicate
with "modern technology." (no question, here. just an observation.)
2. Question: What are the factors that drive companies to want to get
off of these legacy platforms. Is it the cost of maintenance? The cost
of hardware replacements? The business risk of an unsupported
platform?
Again replying with background from my current costumer.

VMS is the "black sheep" that hardly noone in the IT organization
knows about, and less know what it serves in the factory. People
makes large eyes when told how depending they are on that system.
Activity on the assembly lines and the packing stations stops the
same second this system stops.
Post by Ron
3. Conversely: how does a company justify running their business on a
technology stack that is 15 years old, or perhaps even older?
Becuse it has run seccessfully for *30* years now. I regulary support
code that was originaly written 30 years ago. Same base Cobol code
and applications. Originaly written on a VAX 11/750.
Post by Ron
4. Does performance or availability become an issue with these platforms as they age?
Now, the code as such doesn't "age" just by itself...

And the hardware has changed manytimes and the code just
moved along. VAX 11/750, VAX6xxx, VAX8xxx, AlphaServer 800
and currently AlphaServer DS20e.
Post by Ron
5. If so, what is the source of the problem? (hardware failures,
software glitches, incompatibility with new technology, etc.)
Hardware support (we run Alpha) could become an issue.
But currently there are no particular "problems", it
runs just fine.

Regards,
Jan-Erik.
Post by Ron
Thanks for continuing the conversation!
- Ron
Gérard Calliet (pia-sofer)
2014-03-29 22:45:38 UTC
Permalink
Post by Ron
forgive me for any formatting issues... I'm kind of new to these Google Groups.
1. I agree that HP is acting like they want VMS to go away. I also agree that it will be difficult for that to happen. My career started with Tandem and DEC, so I can completely relate that there are characteristics of these legacy platforms that are hard to replicate with "modern technology." (no question, here. just an observation.)
2. Question: What are the factors that drive companies to want to get off of these legacy platforms. Is it the cost of maintenance? The cost of hardware replacements? The business risk of an unsupported platform?
3. Conversely: how does a company justify running their business on a technology stack that is 15 years old, or perhaps even older?
4. Does performance or availability become an issue with these platforms as they age?
5. If so, what is the source of the problem? (hardware failures, software glitches, incompatibility with new technology, etc.)
Thanks for continuing the conversation!
- Ron
I was not on the line, and I have to read all that have be written to
add some comments. I think you have got a lot of interesting things.

Before getting some rest, time and cleverness and have read all, I have
a very short answer for question 3 :
Ascending Compatibility : I think OpenVMS is one of few OSes with this
quality.

Some years ago, I have to give training for VMS 5.5-H2 on VAXes. I
haven't VAXes.

I prepared my training on a small rx1620 with OpenVMS 8.3-H1. I prepared
training, exercises. And then I got a VAX from my customer and tested my
training. I had about nothing to change.

Now, think about some big companies with certified software, very tights
links between software and function, ten or twenty years of development.

Could this customer survive with Windows x+1.y+2 every three years, or
new linux distributions every six months ?

It would be better for him to recompile from VAX to Alpha and from Alha
to Itanium (and from Itanium to x86 - yes, again -), with about the same
DCL around.


---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Simon Clubley
2014-03-30 01:47:34 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Could this customer survive with Windows x+1.y+2 every three years, or
new linux distributions every six months ?
There are no Linux distributions every six months in a business
environment. You purchase one of the long term support distributions
such as RHEL (with support timelines comparable to that of VMS) or
if your needs don't require paid support, then use one of the free
rebuilds of RHEL such as Scientific Linux (which is what I am
currently running on the desktop I am using to type this message).

I wish people would think twice before blindly making comments like
the above.

When you talk to people new to a environment, those people generally
evaluate what you say about things they don't know about by evaluating
what you say about things they _do_ know about. Therefore, when
people start saying the above about Linux in the business world (as well
as other types of comments), it reduces the value of any valid points
the person might be making.

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Gérard Calliet (pia-sofer)
2014-03-31 19:07:11 UTC
Permalink
Post by Simon Clubley
I wish people would think twice before blindly making comments like
the above.
You could be polite

---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Simon Clubley
2014-04-01 12:18:39 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
I wish people would think twice before blindly making comments like
the above.
You could be polite
I was not rude, but you obviously didn't like the message content.

That does not change the truth of what I was saying however.

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Gérard Calliet (pia-sofer)
2014-03-31 23:13:52 UTC
Permalink
Post by Simon Clubley
Post by Gérard Calliet (pia-sofer)
Could this customer survive with Windows x+1.y+2 every three years, or
new linux distributions every six months ?
There are no Linux distributions every six months in a business
environment. You purchase one of the long term support distributions
such as RHEL (with support timelines comparable to that of VMS) or
if your needs don't require paid support, then use one of the free
rebuilds of RHEL such as Scientific Linux (which is what I am
currently running on the desktop I am using to type this message).
I wish people would think twice before blindly making comments like
the above.
When you talk to people new to a environment, those people generally
evaluate what you say about things they don't know about by evaluating
what you say about things they _do_ know about. Therefore, when
people start saying the above about Linux in the business world (as well
as other types of comments), it reduces the value of any valid points
the person might be making.
Simon.
Sorry, I cannot see you on the map, can you help me ? :
http://upload.wikimedia.org/wikipedia/commons/1/1b/Linux_Distribution_Timeline.svg

Or perhaps you are there : http://en.wikipedia.org/wiki/Debian
("""Debian has made twelve major stable releases """" (from 1996))

"""support timelines comparable to that of VMS""" :
(2014 - 1977) / 8 = 4,63
(2014 - 1996) / 12 = 1,5

You are right : 1,5 and 4,6 are comparable : 4,6 is about three times
1,5, it is comparable. (It is just statistic, I know it, speaking in a
hurry, do take time to understand, however).

And for sure, R.H.E.L. can support with only 1 or 2 compilations an
application written with debian kernel version 1, on some spoutnic
distribution.

Perhaps you are just a CEO, and R.H.E.L. company sold you their product
with "ascendant compatibility". Just hire a lawyer : they were laughing
at you :=(

Or perhaps you were aware once upon a time about the words "ascendant
compatibility" would mean, and I am sorry to say you have got a memory
desease.

And, yes, I have some knoweledge of some industrial distribution of
Linux. Is it the point ?

As a little bit forgetful CEO you can go on writing here with or without
memory, with or without competence, I do love fairytales.

Gérard Calliet

Gérard Calliet






---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Simon Clubley
2014-04-01 12:39:24 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Post by Gérard Calliet (pia-sofer)
Could this customer survive with Windows x+1.y+2 every three years, or
new linux distributions every six months ?
There are no Linux distributions every six months in a business
environment. You purchase one of the long term support distributions
such as RHEL (with support timelines comparable to that of VMS) or
if your needs don't require paid support, then use one of the free
rebuilds of RHEL such as Scientific Linux (which is what I am
currently running on the desktop I am using to type this message).
I wish people would think twice before blindly making comments like
the above.
When you talk to people new to a environment, those people generally
evaluate what you say about things they don't know about by evaluating
what you say about things they _do_ know about. Therefore, when
people start saying the above about Linux in the business world (as well
as other types of comments), it reduces the value of any valid points
the person might be making.
Simon.
http://upload.wikimedia.org/wikipedia/commons/1/1b/Linux_Distribution_Timeline.svg
Or perhaps you are there : http://en.wikipedia.org/wiki/Debian
("""Debian has made twelve major stable releases """" (from 1996))
I can't work out if you are trolling or if you simply don't understand
the Linux ecosystem.

What on earth makes you think Debian has anything to do with Red Hat ?

As I said above, in business there are no six month Linux distributions.
You use one of the long term support distributions designed for use in
such a environment, with the leader being RHEL (and it's free rebuilds
such as CentOS and Scientific Linux).
Post by Gérard Calliet (pia-sofer)
(2014 - 1977) / 8 = 4,63
(2014 - 1996) / 12 = 1,5
Red Hat still actively support RHEL 5 which came out in the 2006/2007
timeframe. The only _active_ support for VMS is for 8.3/8.4 (yes, there
is prior version support for some older specific releases, but it's not
the same thing).
Post by Gérard Calliet (pia-sofer)
You are right : 1,5 and 4,6 are comparable : 4,6 is about three times
1,5, it is comparable. (It is just statistic, I know it, speaking in a
hurry, do take time to understand, however).
And for sure, R.H.E.L. can support with only 1 or 2 compilations an
application written with debian kernel version 1, on some spoutnic
distribution.
What the hell does that even mean ?
Post by Gérard Calliet (pia-sofer)
Perhaps you are just a CEO, and R.H.E.L. company sold you their product
with "ascendant compatibility". Just hire a lawyer : they were laughing
at you :=(
On this Scientific Linux 5.x desktop (based around RHEL 5.x) I am using,
the latest extended support release of Firefox, compiled for SL 5, works
just fine. I'd say userland backwards compatibility is pretty good.

And going the other way, in the past, I've run binaries from old RH9
systems unchanged on RHEL 5.x.
Post by Gérard Calliet (pia-sofer)
Or perhaps you were aware once upon a time about the words "ascendant
compatibility" would mean, and I am sorry to say you have got a memory
desease.
And, yes, I have some knoweledge of some industrial distribution of
Linux. Is it the point ?
As a little bit forgetful CEO you can go on writing here with or without
memory, with or without competence, I do love fairytales.
I think your desire to see VMS succeed has blinded you to the capabilities
of other operating systems.

Like I said before (but a little more directly now), if you talk nonsense
about things people know about (such as Linux), they are going to assume,
rightly or wrongly, you are talking nonsense about things they don't know
about.

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Gérard Calliet (pia-sofer)
2014-04-01 16:56:58 UTC
Permalink
Post by Simon Clubley
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Post by Gérard Calliet (pia-sofer)
Could this customer survive with Windows x+1.y+2 every three years, or
new linux distributions every six months ?
There are no Linux distributions every six months in a business
environment. You purchase one of the long term support distributions
such as RHEL (with support timelines comparable to that of VMS) or
if your needs don't require paid support, then use one of the free
rebuilds of RHEL such as Scientific Linux (which is what I am
currently running on the desktop I am using to type this message).
I wish people would think twice before blindly making comments like
the above.
When you talk to people new to a environment, those people generally
evaluate what you say about things they don't know about by evaluating
what you say about things they _do_ know about. Therefore, when
people start saying the above about Linux in the business world (as well
as other types of comments), it reduces the value of any valid points
the person might be making.
Simon.
http://upload.wikimedia.org/wikipedia/commons/1/1b/Linux_Distribution_Timeline.svg
Or perhaps you are there : http://en.wikipedia.org/wiki/Debian
("""Debian has made twelve major stable releases """" (from 1996))
I can't work out if you are trolling or if you simply don't understand
the Linux ecosystem.
I am just a little bit ironic, and I just point out the huge differences
of philosophy between Linux ecosystem and VMS ecosystem.
Post by Simon Clubley
What on earth makes you think Debian has anything to do with Red Hat ?
As I said above, in business there are no six month Linux distributions.
You use one of the long term support distributions designed for use in
such a environment, with the leader being RHEL (and it's free rebuilds
such as CentOS and Scientific Linux).
Post by Gérard Calliet (pia-sofer)
(2014 - 1977) / 8 = 4,63
(2014 - 1996) / 12 = 1,5
Red Hat still actively support RHEL 5 which came out in the 2006/2007
timeframe. The only _active_ support for VMS is for 8.3/8.4 (yes, there
is prior version support for some older specific releases, but it's not
the same thing).
Post by Gérard Calliet (pia-sofer)
You are right : 1,5 and 4,6 are comparable : 4,6 is about three times
1,5, it is comparable. (It is just statistic, I know it, speaking in a
hurry, do take time to understand, however).
And for sure, R.H.E.L. can support with only 1 or 2 compilations an
application written with debian kernel version 1, on some spoutnic
distribution.
What the hell does that even mean ?
It means comparing styles of life cycle in Linux and VMS.
You compare things that are from very distant worlds, and you are doing
a mixup.
Post by Simon Clubley
Post by Gérard Calliet (pia-sofer)
Perhaps you are just a CEO, and R.H.E.L. company sold you their product
with "ascendant compatibility". Just hire a lawyer : they were laughing
at you :=(
On this Scientific Linux 5.x desktop (based around RHEL 5.x) I am using,
the latest extended support release of Firefox, compiled for SL 5, works
just fine. I'd say userland backwards compatibility is pretty good.
And going the other way, in the past, I've run binaries from old RH9
systems unchanged on RHEL 5.x.
Post by Gérard Calliet (pia-sofer)
Or perhaps you were aware once upon a time about the words "ascendant
compatibility" would mean, and I am sorry to say you have got a memory
desease.
And, yes, I have some knoweledge of some industrial distribution of
Linux. Is it the point ?
As a little bit forgetful CEO you can go on writing here with or without
memory, with or without competence, I do love fairytales.
I think your desire to see VMS succeed has blinded you to the capabilities
of other operating systems.
Like I said before (but a little more directly now), if you talk nonsense
about things people know about (such as Linux), they are going to assume,
rightly or wrongly, you are talking nonsense about things they don't know
about.
You suppose I don't know anything about Linux. You don't ever want to
think about what I mean.

I do know there are some sectors in Linux realm which can offer what you
describe. I was just pointing the general way of development in this
ecosystem.

I have not figures, but what you talk about is a minority in Linux
world, and a very little set of researched skills in this community.

I am absolutely not blind about what other ecosystems can do, and that a
lot of things can be done in other ecosystems that VMS cannot do.

Like a majority of good guys from Berkeley, I do think about ecology.
And what was given from the "unix world", the "open world" is very
important, and I don't think it could had be produced by "big heads"
from MIT. (There are cultural differences, even in a federal state :=) ).

And, from this peculiar ideas, I do think it is very important to have
DIVERSITY in the IT world.

Why ? Because I don't believe all the issues are the same. What can be
addressed by VMS is not the same that that can be addressed by Linux or
Windows (or is not ALL the same).

So, I agree some things can be translated from VMS to others OS. But
some other things will encounter very bad troubles in translation.

One of the ideas I never agreed about Unixes was their pretention about
universality (it is in their name). I have always thought it is a false
good idea.

What I pointed here, and you didn't say anything about it is : what
about "ascendant compatibility", which was very important for many in
the 80s, and which is very rarely of interest nowadays. VMS, for this
issue has continued, and it is very valuable for peculiar sectors.

And perhaps we don't agree on that, but I think this continued value in
VMS is one thing which justifies going on with VMS for a class of users,
and saying "we can do this with x (Unix, LInux, ???)" is not realistic
in a simple way.

Did you replace "VMS is the only OS" by "Linux is the only OS" ? The
bug, here, is the world ONLY.

I don't know if VMS would "succeed". I do think our IT world needs for
diversity.

Gérard Calliet
Post by Simon Clubley
Simon.
---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Paul Sture
2014-04-01 19:54:49 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Post by Gérard Calliet (pia-sofer)
Could this customer survive with Windows x+1.y+2 every three years, or
new linux distributions every six months ?
There are no Linux distributions every six months in a business
environment. You purchase one of the long term support distributions
such as RHEL (with support timelines comparable to that of VMS) or
if your needs don't require paid support, then use one of the free
rebuilds of RHEL such as Scientific Linux (which is what I am
currently running on the desktop I am using to type this message).
http://upload.wikimedia.org/wikipedia/commons/1/1b/Linux_Distribution_Timeline.svg
Scroll to the bottom of the page then back up a little and you will see
Red Hat (not very clearly). If you follow the red line to the right you
will see "Red Hat Enterprise" (aka RHEL) appearing in 2002. Look further
along the red line and you will see CentOS in late 2003 and "Scientific" in
2004.

Please have a look at the "Red Hat Enterprise Linux Product Life Cycle":

<https://en.wikipedia.org/wiki/Red_Hat_Enterprise_Linux#Product_life-cycle>

For what it's worth I also use Scientific Linux and appreciate it for its
stability when compared with other distributions of Linux.

salutations...
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
Simon Clubley
2014-04-02 12:10:34 UTC
Permalink
Post by Paul Sture
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
There are no Linux distributions every six months in a business
environment. You purchase one of the long term support distributions
such as RHEL (with support timelines comparable to that of VMS) or
if your needs don't require paid support, then use one of the free
rebuilds of RHEL such as Scientific Linux (which is what I am
currently running on the desktop I am using to type this message).
http://upload.wikimedia.org/wikipedia/commons/1/1b/Linux_Distribution_Timeline.svg
Scroll to the bottom of the page then back up a little and you will see
Red Hat (not very clearly). If you follow the red line to the right you
will see "Red Hat Enterprise" (aka RHEL) appearing in 2002. Look further
along the red line and you will see CentOS in late 2003 and "Scientific" in
2004.
Actually, that's a very good point.

Because Gerard's been comparing Linux to VMS, I just assumed he knew,
at least in general terms, about the requirements for Linux in business
and reacted accordingly.

Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.

You absolutely cannot compare Linux to VMS based only on knowledge of
how Linux is used at a low end consumer desktop/tablet level.

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Gérard Calliet (pia-sofer)
2014-04-02 12:36:09 UTC
Permalink
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.

And now, have you some little time to answer about what I say ?

What about "ascendant compatibility", now (I speak for tens of years) ?

And I apologize, as such an ignorant I am, but I do insist : it is a
matter of diversity.

I know some lands or langages have God on their side, but if I remember,
the Babel project has been refused by a very high authority,in a very
old time.

It would be very interesting if I could hear a conversation between you
and some big company representatives about the (real) VMS problem now.
It could be funny if you said them : "it is not a so big issue, we have
RHEL".

What I try to say is : there is a problem, and the ideological answers
are of interest, but we are not wainting for such answers.

Gérard Calliet

---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Paul Sture
2014-04-02 16:42:03 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.
And now, have you some little time to answer about what I say ?
What about "ascendant compatibility", now (I speak for tens of years) ?
Sorry Gérard but I am not sure I understand what you mean by "ascendant
compatibility". I *think* that's the same as "Upwards compatibility"
but my French (et aussi mon Franglais!) is rusty and I am not sure.
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
Jan-Erik Soderholm
2014-04-02 17:16:58 UTC
Permalink
Post by Paul Sture
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.
And now, have you some little time to answer about what I say ?
What about "ascendant compatibility", now (I speak for tens of years) ?
Sorry Gérard but I am not sure I understand what you mean by "ascendant
compatibility". I *think* that's the same as "Upwards compatibility"
but my French (et aussi mon Franglais!) is rusty and I am not sure.
Probably "backwards compatibility", which might be just as
important (or the same), depending on who you ask.
Bill Gunshannon
2014-04-02 17:39:26 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.
And now, have you some little time to answer about what I say ?
What about "ascendant compatibility", now (I speak for tens of years) ?
Could you post the french for what you are calling "ascendant compatibility"?
The only thing I find for it here refers to astrology.
Post by Gérard Calliet (pia-sofer)
And I apologize, as such an ignorant I am, but I do insist : it is a
matter of diversity.
I know some lands or langages have God on their side, but if I remember,
the Babel project has been refused by a very high authority,in a very
old time.
It would be very interesting if I could hear a conversation between you
and some big company representatives about the (real) VMS problem now.
It could be funny if you said them : "it is not a so big issue, we have
RHEL".
Where do you propose to find a "big company representative" who even knows
what VMS was.
Post by Gérard Calliet (pia-sofer)
What I try to say is : there is a problem, and the ideological answers
are of interest, but we are not wainting for such answers.
Gérard Calliet
---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
I love people who's messages contain this crap no matter what language it
is in. What possible reason would I have to believe them?


bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Paul Sture
2014-04-02 18:21:06 UTC
Permalink
Post by Bill Gunshannon
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.
And now, have you some little time to answer about what I say ?
What about "ascendant compatibility", now (I speak for tens of years) ?
Could you post the french for what you are calling "ascendant compatibility"?
The only thing I find for it here refers to astrology.
I've got the Dictionnaire de l'Academie Française here, which I suppose is
somewhat authoritative :-)

* Climbing, ascending movement or force. Med: Ascending aorta.
* The astrology usage
* Jurisprudence and geneology: indicates the persons you are descended from
* Ascendancy as in occupation of a position of dominant power or influence:
the ascendancy of good over evil
Post by Bill Gunshannon
Post by Gérard Calliet (pia-sofer)
Ce courrier électronique ne contient aucun virus ou logiciel malveillant
parce que la protection avast! Antivirus est active.
http://www.avast.com
I love people who's messages contain this crap no matter what language it
is in. What possible reason would I have to believe them?
Well if we are going to be Properly Pedantic, it's Probably false
advertising :-) Its claim is that because Avast is active, the post
contains no viruses or malware-. "No viruses or malware detected" is
more accurate. :-)
--
Paul Sture

"In that case," said Napoleon, "let us wait twenty minutes; when the enemy
is making a false movement we must take good care not to interrupt him."
Gérard Calliet (pia-sofer)
2014-04-02 22:22:49 UTC
Permalink
Post by Paul Sture
Post by Bill Gunshannon
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.
And now, have you some little time to answer about what I say ?
What about "ascendant compatibility", now (I speak for tens of years) ?
Could you post the french for what you are calling "ascendant compatibility"?
The only thing I find for it here refers to astrology.
I've got the Dictionnaire de l'Academie Française here, which I suppose is
somewhat authoritative :-)
* Climbing, ascending movement or force. Med: Ascending aorta.
* The astrology usage
* Jurisprudence and geneology: indicates the persons you are descended from
the ascendancy of good over evil
Post by Bill Gunshannon
Post by Gérard Calliet (pia-sofer)
Ce courrier électronique ne contient aucun virus ou logiciel malveillant
parce que la protection avast! Antivirus est active.
http://www.avast.com
I love people who's messages contain this crap no matter what language it
is in. What possible reason would I have to believe them?
Well if we are going to be Properly Pedantic, it's Probably false
advertising :-) Its claim is that because Avast is active, the post
contains no viruses or malware-. "No viruses or malware detected" is
more accurate. :-)
« « L'Ane vint à son tour et dit : J'ai souvenance
Qu'en un pré de Moines passant,
La faim, l'occasion, l'herbe tendre, et je pense
Quelque diable aussi me poussant,
Je tondis de ce pré la largeur de ma langue.
Je n'en avais nul droit, puisqu'il faut parler net.
A ces mots on cria haro sur le baudet. » »

Jean de Lafontaine, les animaux maladies de la peste.

I don’t know if everyone can understand here.

For the ignorants :

“”It came the Ass's turn. "I recollect," he said,
"That once in spring I crossed a field
Of grass so sweet and tender I commenced to yield
To devilish desires that popped into my head
And took a bite broad as my tongue of that good hay.
I had no right. My conscience warned me to say nay!"
At that, the assembly shouted, "Shame upon the Ass!" » »

Jean de Lafontaine, The animal stricken with the plague.

“A bite broad of my tongue”. I don’t know if there is the same idiomatic
relation in English between “tongue" and language. The sort of fury
around my bad English in last mails is in a sense very funny.

Because all our issue here is about language, translation, idiomatics,
cultural misunderstanding.

The Ass did conduct the discussion into its meadow : yes there are
cultural misunderstanding, yes it is all about necessity of diversity in
IT science.

I don’t remember where I got it in the mails, but “pedantic” is the
right world. Shall I be as pedantic as some bunch of oldest VMS “has
been” ? I have to say IT science was born neighbor to the efforts of
refunding of mathematics, and as they wanted to have an unique basis,
they discover all the results about non proof of consistency, paradoxes
and so on. The sense : foundation of mathematics encounters paradoxes,
linguistics discover central function of idiomatics, and for sure the
technology which is sort a result of all these researches has to
understand Unicity of language is not only (market) utopia, it is wrong
and false.

I do understand, as the animal of the fable, that a younger French VMS
activist, is a sort of offense against such of noble assembly of old VMS
captains, who, at some good time, abandon the ship before it sinks, and
it is a sort of honor for me to be anathemised about “my tongue”.

CQFD (Ce Qu’il Fallait Démontrer)

Gérard Calliet



---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Bill Gunshannon
2014-04-02 22:34:54 UTC
Permalink
Post by Bill Gunshannon
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.
And now, have you some little time to answer about what I say ?
What about "ascendant compatibility", now (I speak for tens of years) ?
Could you post the french for what you are calling "ascendant compatibility"?
The only thing I find for it here refers to astrology.
I've got the Dictionnaire de l'Academie Française here, which I suppose is
somewhat authoritative :-)
* Climbing, ascending movement or force. Med: Ascending aorta.
* The astrology usage
* Jurisprudence and geneology: indicates the persons you are descended from
the ascendancy of good over evil
Post by Bill Gunshannon
Post by Gérard Calliet (pia-sofer)
Ce courrier électronique ne contient aucun virus ou logiciel malveillant
parce que la protection avast! Antivirus est active.
http://www.avast.com
I love people who's messages contain this crap no matter what language it
is in. What possible reason would I have to believe them?
Well if we are going to be Properly Pedantic, it's Probably false
advertising :-) Its claim is that because Avast is active, the post
contains no viruses or malware-. "No viruses or malware detected" is
more accurate. :-)
« « L'Ane vint à son tour et dit : J'ai souvenance
Qu'en un pré de Moines passant,
La faim, l'occasion, l'herbe tendre, et je pense
Quelque diable aussi me poussant,
Je tondis de ce pré la largeur de ma langue.
Je n'en avais nul droit, puisqu'il faut parler net.
A ces mots on cria haro sur le baudet. » »
Jean de Lafontaine, les animaux maladies de la peste.
I don’t know if everyone can understand here.
“”It came the Ass's turn. "I recollect," he said,
"That once in spring I crossed a field
Of grass so sweet and tender I commenced to yield
To devilish desires that popped into my head
And took a bite broad as my tongue of that good hay.
I had no right. My conscience warned me to say nay!"
At that, the assembly shouted, "Shame upon the Ass!" » »
Jean de Lafontaine, The animal stricken with the plague.
“A bite broad of my tongue”. I don’t know if there is the same idiomatic
relation in English between “tongue" and language. The sort of fury
around my bad English in last mails is in a sense very funny.
Because all our issue here is about language, translation, idiomatics,
cultural misunderstanding.
The Ass did conduct the discussion into its meadow : yes there are
cultural misunderstanding, yes it is all about necessity of diversity in
IT science.
I don’t remember where I got it in the mails, but “pedantic” is the
right world. Shall I be as pedantic as some bunch of oldest VMS “has
been” ? I have to say IT science was born neighbor to the efforts of
refunding of mathematics, and as they wanted to have an unique basis,
they discover all the results about non proof of consistency, paradoxes
and so on. The sense : foundation of mathematics encounters paradoxes,
linguistics discover central function of idiomatics, and for sure the
technology which is sort a result of all these researches has to
understand Unicity of language is not only (market) utopia, it is wrong
and false.
I do understand, as the animal of the fable, that a younger French VMS
activist, is a sort of offense against such of noble assembly of old VMS
captains, who, at some good time, abandon the ship before it sinks, and
it is a sort of honor for me to be anathemised about “my tongue”.
CQFD (Ce Qu’il Fallait Démontrer)
Gérard Calliet
I think we have, yet another, lost in translation. I saw no "fury" over
your use of english. In my case I merely wanted to know what French
phrase you were translating as "ascendant compatibility" as I am sure
that is a mistaken translation. I often return to the original language
when trying to understand things I read that are translations as they
are frequently mis-translated or biased by some mis-understanding of
the translator.

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Simon Clubley
2014-04-02 22:35:21 UTC
Permalink
On 2014-04-02, Gérard Calliet (pia-sofer) <***@pia-sofer.fr> wrote:

[big, big, snip]
Post by Gérard Calliet (pia-sofer)
I do understand, as the animal of the fable, that a younger French VMS
activist, is a sort of offense against such of noble assembly of old VMS
captains, who, at some good time, abandon the ship before it sinks, and
it is a sort of honor for me to be anathemised about “my tongue”.
$ set response/mode=good_natured

Gerard,

Have you ever answered a simple question with a simple answer ? :-)

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Gérard Calliet (pia-sofer)
2014-04-02 23:07:38 UTC
Permalink
Post by Simon Clubley
[big, big, snip]
Post by Gérard Calliet (pia-sofer)
I do understand, as the animal of the fable, that a younger French VMS
activist, is a sort of offense against such of noble assembly of old VMS
captains, who, at some good time, abandon the ship before it sinks, and
it is a sort of honor for me to be anathemised about “my tongue”.
$ set response/mode=good_natured
Gerard,
Have you ever answered a simple question with a simple answer ? :-)
(Is it a question or an answer ?)

My answer (1) : yes it was about "upward compatibility", I think so.
My question (2) : do you really think this issue, upward compatibility,
is addressed at the same level now in other OS that it was and it is
with VMS ?

My answer (2) : I think reading the context of my mails, and not
hurrying into polemics, you could have "translate it"
My question (2) : Don't you think general difficulties in translation
witnesses plurality of languages is the real image of complexicity of
reality ?
Post by Simon Clubley
Simon.
---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Gérard Calliet (pia-sofer)
2014-04-02 22:45:31 UTC
Permalink
Post by Bill Gunshannon
Where do you propose to find a "big company representative" who even knows
what VMS was.
Severstal North American
Manor (emea)
Renault
Peugeot
Volvo
Ikea
France Télécom
(I'm just a minor consultant in France, you know, and I don't know
anything I cannot get by my very little strengths, and there are some
customers I know who don't want to be cited ; it is just a minor example ;
I know that for all these customers june HP bug about i4 is just huge
problems, lot loss of money and lot of risks.
Do phone them about RHEL).

Gérard Calliet

---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
Simon Clubley
2014-04-02 20:13:40 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world and are much closer to
the VMS style support requirements.
Ok, I will have a look.
And now, have you some little time to answer about what I say ?
What about "ascendant compatibility", now (I speak for tens of years) ?
I don't understand what you mean by that phrase. I'm going to assume you
mean can binaries from older versions run on newer versions ?

One of the goals in the Linux kernel itself is to maintain upwards
compatibility at userland level so at that level the answer is yes.
However, Linux has a vast ecosystem of libraries which do not have
any equivalent in the core VMS distribution.

If the application was built against these libraries in shared mode,
instead as a static binary, then versions of these libraries also need
to be present on the later system. The exact same thing would be true
on VMS if it had the same range of library functionality as Linux does.

At work, for support reasons, I just use the versions of packages shipped
with the distribution so I don't have any experience with really old
binaries there.

At home, most of the things which are not distribution supplied are
built from source. However, I do have experience with a couple of
third party binary packages:

1) Although Firefox is a part of Linux distributions, for various reasons
I have sometimes used the Mozilla supplied Firefox ESR build instead.
Up until a few months ago (until Mozilla rebased the library versions
used to build the ESR releases from ESR 24.x) the ESR releases ran just
fine on Scientific Linux 5.x. Those same binaries ran just fine on
different and later Linux distributions as well.

I have no reason to believe those earlier ESR releases would not run for
the forseeable future on later Linux versions, however you could not do
that for security reasons.

2) There was a commercial version of Tuxracer released about a decade
ago, and I have the free to download demo version. Although for some
reason, it won't perform a new install on SL 5.x due to what looks like
some scripting compatibility issue, the binary from my existing installed
version works just fine under SL 5.x (I just tried it :-))
Post by Gérard Calliet (pia-sofer)
And I apologize, as such an ignorant I am, but I do insist : it is a
matter of diversity.
Sometimes the correct tool for the job is VMS, other times it's Linux,
Windows, a RTOS or another OS, or even bare metal. You just to know
what is the correct tool for the job.
Post by Gérard Calliet (pia-sofer)
I know some lands or langages have God on their side, but if I remember,
the Babel project has been refused by a very high authority,in a very
old time.
Huh ?
Post by Gérard Calliet (pia-sofer)
It would be very interesting if I could hear a conversation between you
and some big company representatives about the (real) VMS problem now.
It could be funny if you said them : "it is not a so big issue, we have
RHEL".
RHEL is the solution to some problems. It is not the solution to every
problem - and neither is VMS.
Post by Gérard Calliet (pia-sofer)
What I try to say is : there is a problem, and the ideological answers
are of interest, but we are not wainting for such answers.
My answers are practical and realistic; I think it's you who has the
ideological approach. :-)

Simon.
--
Simon Clubley, ***@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world
Bill Gunshannon
2014-04-02 16:37:52 UTC
Permalink
Post by Simon Clubley
Post by Paul Sture
Post by Gérard Calliet (pia-sofer)
Post by Simon Clubley
There are no Linux distributions every six months in a business
environment. You purchase one of the long term support distributions
such as RHEL (with support timelines comparable to that of VMS) or
if your needs don't require paid support, then use one of the free
rebuilds of RHEL such as Scientific Linux (which is what I am
currently running on the desktop I am using to type this message).
http://upload.wikimedia.org/wikipedia/commons/1/1b/Linux_Distribution_Timeline.svg
Scroll to the bottom of the page then back up a little and you will see
Red Hat (not very clearly). If you follow the red line to the right you
will see "Red Hat Enterprise" (aka RHEL) appearing in 2002. Look further
along the red line and you will see CentOS in late 2003 and "Scientific" in
2004.
Actually, that's a very good point.
Because Gerard's been comparing Linux to VMS, I just assumed he knew,
at least in general terms, about the requirements for Linux in business
and reacted accordingly.
Gerard, if you don't know that RHEL is, then you really need to get up to
speed on Linux in the business world because the standards are very
different from Linux in the consumer world
As is also true in the Windows world wether people want to believe it or
not.
Post by Simon Clubley
and are much closer to
the VMS style support requirements.
And VMS is the same. Requirements for a commercial VMS installation
are considerably different from a Hobbyist system.
Post by Simon Clubley
You absolutely cannot compare Linux to VMS based only on knowledge of
how Linux is used at a low end consumer desktop/tablet level.
You can't compare serious business systems to home systems no matter
what the OS is.

bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
***@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Gérard Calliet (pia-sofer)
2014-03-31 21:39:56 UTC
Permalink
Le 27/03/2014 19:02, Ron a écrit :> forgive me for any formatting
issues... I'm kind of new to these Google Groups.
Post by Ron
Follow-up questions & comments for Gerard and Jan-Erik (and anyone
1. I agree that HP is acting like they want VMS to go away. I also
agree that it will be difficult for that to happen. My career started
with Tandem and DEC, so I can completely relate that there are
characteristics of these legacy platforms that are hard to replicate
with "modern technology." (no question, here. just an observation.)
Post by Ron
2. Question: What are the factors that drive companies to want to
get off of these legacy platforms. Is it the cost of maintenance? The
cost of hardware replacements? The business risk of an unsupported
platform?
You had already a lot of answers.

As said Jan-Eric VMS is often a sort of "black sheep" : not so fashioned
technology, written in old days, very valuable and so very difficult to
move it. The CEO knows all about excel, big data, cloud, is invited in
forum by big sellers, ans is a little bit ashamed having to say "all my
central descriptions for my factory are on a vax with a 9G hard drive".
" - Put it on the cloud !! Give your robots HD video !! Be Agile !!!"

I translate : often VMS applications are the ones which remain on VMS
because there are valuable, but this type of value is not so tractable
as over less important new data like video marketing, for example.

First idea : VMS IS the problem because it became impossible to say VMS
IS the value.

Second idea : VMS was written by people who are now retiring, there is
not or very few young people knowing it ; it is a big risk factor.
Post by Ron
3. Conversely: how does a company justify running their business on
a technology stack that is 15 years old, or perhaps even older?

Take this : """VMS IS the problem because it became impossible to say
VMS IS the value""", and you have got the answer : even if it became
impossible to say its value, VMS is there BECAUSE of its value.
Post by Ron
4. Does performance or availability become an issue with these platforms as they age?
For a very few customers performance could be an issue : they were
waiting for i4 for this reason, and they are in trouble. (Don't worry
about them : HP would give them special conditions).
Post by Ron
5. If so, what is the source of the problem? (hardware failures,
software glitches, incompatibility with new technology, etc.)

By decrescent order of importance :
1) end of sale of hardware (2015 is tomorow),
2) end of support for hardware too soon, problem getting parts,...
3) end of legal support for software (I underline "legal" : everybody
knows 2 or 3 India engineers are not "support", but companies need to
officially say "we have support").

All other issues you noted are irrelevant (I think so) : you got here a
lot of answers about these issues, some opposed thinking, but anything
"wrong" in software can be addressed, and there is around A LOT of
solutions to interface VMS with new technologies (gSOAP, one example
among hundreds).
Post by Ron
Thanks for continuing the conversation!
Thank you giving us opportunity. And go on yourselves !
Post by Ron
- Ron
And now, some "philosophie".

David Frobe :
"""
Being useful to some people and being taken seriously by the mob are two
different things. The only question is, will those who find VMS useful
have any choices in the future? """

As said some American poet : "the time there are a changing".

There were times where growing IT was "hand by hand" useful IT, big
profits. Innovation was generally innovation for new uses of IT.

Now in the huge world of IT, some things are more useful than others,
some things generate more profit than others, and the "hand by hand" is
closed.

It is a very normal cycle in industry : large and very useful
innovations at the beginnings, and, after that, search for huge
benefits, differences between main stream and specialized markets,
etc... (Think about cars or planes).

In the 70s or 80s critical systems were the top of the mountain, the
glory for IBM, DEC, HP... For now, critical systems are a specialized
market, and not a so big market. I talk about dependable applications
for energy, industry, health, transport. Not so big data, software very
difficult to change, problems of certification...
Critical systems is no more the top of the mountain, the glory for HP or
IBM, etc... Now the top is huge data centers for huge clouds, HPC to do
big data mining, very large networks, new format of smartphone every
year, and so on.

Here is some big trend. I don't say it is not important. I say with
critical systems, we are no more on the mainstream. We can deny it, cry
for it, insult people who cry for it, deny their situation, predict the
boat is sinking, being the first or the last to save her soul, scuttle
the boat to save it, I don't know, everything is possible. But the fact
is : system critical ITs are no more on the main stream.

So the only question is : what can be done, what have to be done for
customers who NEED critical systems ?

My answer : we have to re-invent the gearshift (is it the word in
english ?) :
We have to think about different levels of speeds, different levels for
life cycle, and so on.
Try to start a car without different levels of speed, you cannot start.

VMS is no more an universal OS, and also times of wars between VMS and
Unix and ... to be THE universal software are gone.

We need OS diversity, with VMS, Unix, Linux,... because the IT world is
no more a new continent. We live in Massachusetts AND in West Coast, we
don't need same things in New-York and in Las Vegas.

Gérard Calliet

---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
JF Mezei
2014-04-03 03:22:45 UTC
Permalink
Post by Gérard Calliet (pia-sofer)
1) this movement is constant for years, but slow, and for some
customers, it simply does'nt have any sense,
Those who could easily migrate from VMS did so ages ago.

Then there were those who were forced to migrate from VMS. SWIFT
terminals is one example whenere drop of VMS support forced them to
migrate to new platforms at a specific time frame. Totally new apps were
provided along with the transition to a new version of the SWIFT
network. ST400 was not ported.


Another example is the CERNER software. Once Cerner stopped providing
upgrades to its VMS software, and VMS system wouldn't get upgrades
required whenever law/reporting requirements changed.


What is left are those still with VMS because they like it and just
refuse to accept its demise, and those who are still on VMS because they
really do need the unique features provided by VMS and moving to another
platform requires a major rething/re-engineering of apps.


I'd be willing to bet a chocolate bar that most of the remaining VMS
installed base has a lot of homebrewed code.
Gérard Calliet (pia-sofer)
2014-04-03 05:21:24 UTC
Permalink
Post by JF Mezei
Post by Gérard Calliet (pia-sofer)
1) this movement is constant for years, but slow, and for some
customers, it simply does'nt have any sense,
Those who could easily migrate from VMS did so ages ago.
Then there were those who were forced to migrate from VMS. SWIFT
terminals is one example whenere drop of VMS support forced them to
migrate to new platforms at a specific time frame. Totally new apps were
provided along with the transition to a new version of the SWIFT
network. ST400 was not ported.
Another example is the CERNER software. Once Cerner stopped providing
upgrades to its VMS software, and VMS system wouldn't get upgrades
required whenever law/reporting requirements changed.
What is left are those still with VMS because they like it and just
refuse to accept its demise, and those who are still on VMS because they
really do need the unique features provided by VMS and moving to another
platform requires a major rething/re-engineering of apps.
I'd be willing to bet a chocolate bar that most of the remaining VMS
installed base has a lot of homebrewed code.
Perhaps one more reason for companies or departments which did not go
out : productivity.

Companies using VMS experimented a very low need of human maintenance
for the OS and simply use it for their home software. And for the
central subsystems which need more productivity and stability than
innovation it has been a reason to 1) not go out 2) freeze.

So, with the perspective of big money to translate, they have also
intuition about where they needed 10 VMS engineers, they will have to
pay 20 Linux or AIX ingineers plus 30 Windows ingineers : 2 bad news.
(not to have polemic : the numbers and names I use are there as "way of
talking" to exprim a difference in productivity).

---
Ce courrier électronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com

JohnF
2014-03-28 13:28:09 UTC
Permalink
Post by Ron
Greetings all,
I am doing some market research for a customer.
They market software and services related to VMS, OpenVMS
and other legacy systems. [...]
It has been tricky finding people to participate in this study, [...]
I promise that there is no hidden sales pitch in here.
I won't even tell you who my customer is unless you ask. [...]
Sincerely, Ron
Just curious. Consider yourself asked: who's your customer?
--
John Forkosh ( mailto: ***@f.com where j=john and f=forkosh )
Doc Trins O'Grace
2014-03-28 18:11:32 UTC
Permalink
Post by Ron
Greetings all,
My responses: Yes. Yes. No. No. Yes. Antiquated.
Stanley F. Quayle
2014-03-30 04:32:46 UTC
Permalink
I had sent the following directly to Ron. But it might be of interest to the group at large.
Post by Ron
If you work with/on VMS/OpenVMS in an enterprise environment or know
of these systems in such an environment, and are willing to bounce
some ideas around let me know and we can schedule some time to talk.
My clients are all running VMS, from 1-person companies to >1000
employees.
Post by Ron
- How are these systems holding up? Still highly available?
The only problem is getting parts and hardware support. That's why I
sell the CHARON line of emulators. Most have switched (I have several
that only use me for VMS software support). See
http://www.stanq.com/wf1.html for an example. They not only saved lots
of money, but got improved performance, while running the same
applications without change.
Post by Ron
- Are they still running business-critical, mission-critical
production systems?
- Have they been relegated to simply maintaining an archive of those systems?
About 25% of the emulators I sell are for archival systems. The rest
are for mission-critical applications. These include government
entities such as the US Army, Air Force, Navy, and local/county
agencies. Other companies include nuclear power plants, pharmaceutical
manufacturers, paper manufacturers, hardware manufacturers. While I
can't disclose their names, I have several customers that are Fortune 50
companies.
Post by Ron
- Are there challenges keeping the platforms running?
Development of VMS (which is still in active development, 36 years and
counting) was moved to India a few years ago. These guys are sharp,
many of which already have extensive VMS experience -- DEC had a
development center there for many years already. The challenge in the
USA that I see is HB-1 visa'd people are willing to work for $25/hour.
That has caused many of my colleagues to move to some other platform --
Linux is a popular choice.
Post by Ron
- Is there a movement afoot to get away from these systems?
If I had a nickel for every customer that says they're going to move,
but don't, I would be a millionaire. Most companies are thinking of
moving, but it's very expensive, generally a complete re-write, for
mission-critical applications. Most companies have a move way down the
priority list, more like a wish than a plan.
Post by Ron
- With all the market hype about the cloud and virtualization, how are
these systems viewed by the staff?
Since the emulators run on Windows (or Linux), they fit in most data
centers. The biggest problem is that Windows guys reboot immediately
when any problem arises. That's not a good idea when running VMS under
emulation.

The current crop of emulators can run on VMware. VMware is the way that
many big customers partition a big machine -- and that fits into the
data center plan. Also, emulators can use SAN storage, while most
physical VMS systems cannot.
Post by Ron
What about Management's perspective?
Management is generally clueless about VMS. They understand (or think
they do) Windows. I've made several emulator sales because the VMS guys
can say that they went to Windows -- not mentioning that the application
is still running on VMS.

Management is generally happy about emulation because it saves them big
money in hardware support. Many of the Windows servers running
emulators even come with a 3-year hardware warranty.

I had one big customer move to emulation because he didn't want the
hardware to fail -- a career-ending move for him. If it died, hundreds
of people would be idled.
Post by Ron
I won't even tell you who my customer is unless you ask.
Please do.
JF Mezei
2014-03-31 19:41:13 UTC
Permalink
Post by Stanley F. Quayle
Post by Ron
- Are there challenges keeping the platforms running?
Development of VMS (which is still in active development, 36 years and
counting) was moved to India a few years ago. These guys are sharp,
many of which already have extensive VMS experience
This borders on the propaganda level. VMS is no longer actively
developped, no new versions. It is in maintenance mode with a few
patches scheduled to be released to support new disk hardware, not new
computer systems.

There are a few in India who have been there for a long time, but most
would be newbies.
Continue reading on narkive:
Loading...