|
|
|
|
|
|
| Author |
Message |
Nessie
*nix forums beginner
Joined: 07 Feb 2005
Posts: 3
|
 Posted: Wed Feb 09, 2005 5:50 am Post subject:
Re: Why aren't more universities doing this?
|
|
|
Sorry to butt-in, but I spent 8 years programming mostly cobol on a VMS
system in the late 80's/early 90's, and another 9 years programming
cobol on IBM 3600's, DOS and MVS.
They may be slow, but boy are they development friendly
environments...give me a VAX to work with any day above an IBM...!!!
You can understand what it wants, instead of having to learn letter
commands like unix / AIX use.
Enough ranting, you get the point ! |
|
| Back to top |
|
 |
Tom Linden
*nix forums Guru
Joined: 06 May 2002
Posts: 841
|
| Posted: Wed Feb 09, 2005 12:13 am Post subject:
Re: Why aren't more universities doing this?
|
|
|
On 8 Feb 2005 20:35:39 GMT, Bill Gunshannon <bill@cs.uofs.edu> wrote:
| Quote: | In article <YsmdnTP2Faolj5TfRVn-qQ@igs.net>,
"John Smith" <a@nonymous.com> writes:
Bill Gunshannon wrote:
In article <cuaqk6$qqm$1@naig.caltech.edu>,
David Mathog <mathog@caltech.edu> writes:
bob@instantwhip.com wrote:
http://www.shannonknowshpc.com/stories.php?story=05/02/07/1720316
That's an easy one. Because porting open source software
to a moribund OS is not a particularly worthy goal and nobody
is going to give you very much credit for having done it.
Not exactly true. It would have to be examined like any other
project, but I was successful in getting 4 porting projects
approved by our faculty committee a few years ago. I asked one
of the professors who has visited a number of other schools as
a member of an ABET Acreditation Team and he assures me that
about 50% of colleges with CS programs (a higher % for acredited
programs) have similar senior projects courses. The biggest
thing holding up projects like these is the lack of VMS systems
in academia. (As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.) I look forward to trying to get some VMS porting
projects in the works again. The students are always looking
for something more meaningful than the typical doctor's office
Excel/Access billing system.
I used to do rather a lot of porting to OpenVMS and believe me,
my time is much better spent doing other things and just
running the code as is on Linux. When I work on code now it's
to make it faster or better, not just to get it to run in
the first place on a platform that nobody in academia
cares about or uses anymore.
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department. It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!! :-)
Push HP to give you some 'current' technology, like an Alpha or one of
those
prototype IA64 boxes (they'll be making those until the pre-production
line
closes when Intel EOL's the cpu).
Actually, Compaq did give me an Alpha. A small one, but adequate.
That was in the days when they actually seemed interested. But
interest seemed to dry up about the time of the merger. I have
had limited contact since then, but it seems I am mostly back on
my own again.
Father: "So what did you do in CS class today, son?
Son: "I got to work some goofy fossilized computer Professor Bill still
has
hanging around the lab."
Father: "Really? Which one?"
Son: "Something called WAX or VAX, or something like that."
Father: "I used to program on those too."
Son: "Like I said, some fossil. Slow too."
So tell me, how would the average user writing program in a
high-level language (think under-grads here) even tell what
the architecture was? Why would he care? And I hardly expect
that the 4 processor machines I have are going to be slow
supporting only a small handful of students. At it's peak
it will have maybe 15 simultaneous users runnging the COBOL
Compiler. I think it will offer acceptable performance.
And anyway, they are hardly fossilized computers when compared to
the PDP-11/44's standing next to them. 
bill
Bill |
I would be pleased to provide you with a PL/I license so your students can
see
how a real sophisticated programming language works and behaves. Write me
offline
if you want a pak.
Tom
--
Using Opera's revolutionary e-mail client: http://www.opera.com/m2/ |
|
| Back to top |
|
|
Tom Linden
*nix forums Guru
Joined: 06 May 2002
Posts: 841
|
| Posted: Wed Feb 09, 2005 12:09 am Post subject:
Re: Why aren't more universities doing this?
|
|
|
On 8 Feb 2005 18:41:01 GMT, Bill Gunshannon <bill@cs.uofs.edu> wrote:
Bill,
I would be pleased to donate a PL/I license so your students can
see how a real programming language works and behaves.
Tom
| Quote: | In article <cuaqk6$qqm$1@naig.caltech.edu>,
David Mathog <mathog@caltech.edu> writes:
bob@instantwhip.com wrote:
http://www.shannonknowshpc.com/stories.php?story=05/02/07/1720316
That's an easy one. Because porting open source software
to a moribund OS is not a particularly worthy goal and nobody
is going to give you very much credit for having done it.
Not exactly true. It would have to be examined like any other
project, but I was successful in getting 4 porting projects
approved by our faculty committee a few years ago. I asked one
of the professors who has visited a number of other schools as
a member of an ABET Acreditation Team and he assures me that
about 50% of colleges with CS programs (a higher % for acredited
programs) have similar senior projects courses. The biggest
thing holding up projects like these is the lack of VMS systems
in academia. (As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.) I look forward to trying to get some VMS porting
projects in the works again. The students are always looking
for something more meaningful than the typical doctor's office
Excel/Access billing system.
I used to do rather a lot of porting to OpenVMS and believe me,
my time is much better spent doing other things and just
running the code as is on Linux. When I work on code now it's
to make it faster or better, not just to get it to run in
the first place on a platform that nobody in academia
cares about or uses anymore.
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department. It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!! :-)
bill
|
--
Using Opera's revolutionary e-mail client: http://www.opera.com/m2/ |
|
| Back to top |
|
|
JF Mezei
*nix forums Guru
Joined: 21 Jul 2005
Posts: 2556
|
| Posted: Tue Feb 08, 2005 9:49 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
Bill Gunshannon wrote:
| Quote: | Not exactly true. It would have to be examined like any other
project, but I was successful in getting 4 porting projects
approved by our faculty committee a few years ago.
|
At a bank, I was, for long time, prevented from buying a $6000 C
compiler. Reason being that since VMS was not strategic, no development
should be done to it, especially since the bank was spending $300,000 on
a STUDY to replace the VMS systems. When that study failed, the bank was
forced to instiutute a disaster recoivery plan for this mission critical
system (SWIFT funds transfers). For a while, the purse was easy to dip into.
However, the day SWIFT announced it was widthdrawing support for VMS,
the bank again froze all development on that system since it would have
to be replaced as soon as possible. (that ended up taking years before
the unix software had the same functionality as the old ST400).
During those years, Digital/Compaq still saw maintenance revenus, so
they may have considered that customer safe even though the decisions
had been taken to dump the VMS systems. And yes, Sun got the business.
Even though its disaster recovery scheme was nowhere as good as what I
had setup, it was good enough and besides the bank had no choice since
VMS was no longer a solution for SWIFT.
When a company decides that a system is to be disposed of eventually,
don't expect it to be easy to obtain anything to improve that system
during that period. The grunts at the low level do see these decisions
being taken, but the folks at HP/compaq/Digital don't, until it is too late.
So when someone reports here that they are not allowed to buy an extra
disk drive for a VMS system, HP employees should see this as an early
warning of an impending loss of customer and intervene to try to save
the account before the replacement system is implemented.
You can't salvage an account the day the customer doesn't renew
maintenance contract since that happens when the cutsomer already has
the replacement system in production.
The story of VMS is very similar to that of IBM. In his book "Who says
elephants can't dance", Gerstner outlines how he had to drive employees
not to accept loss of an account and fight hard to keep accounts and win
new sales, instead of just pushing papers to process orders that come
from loyal customers whom IBM never feared losing before. |
|
| Back to top |
|
|
John Smith
*nix forums Guru
Joined: 15 Oct 2003
Posts: 561
|
| Posted: Tue Feb 08, 2005 9:47 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
Bill Gunshannon wrote:
| Quote: | In article <cub66o$1i4$1@naig.caltech.edu>,
David Mathog <mathog@caltech.edu> writes:
Bill Gunshannon wrote:
(As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.)
Killer VAXen??? Your cell phone probably has a faster processor.
"Killer" when compared to the VAXen we have had in the department
in the past. :-)
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department.
But why does Cobol imply VMS? There are Cobol
compilers for other platforms, including
ones the students already own (Windows, Linux).
Windows compilers are expensive and frequently lack particular
features needed for the course. I am working very hard to keep
the Linux garbage out of my computer room.
|
I can see it now....Bill as Monty Python's 'Black Knight' standing at the
computer room door saying "No Linux shall pass". :-)
Or perhaps in 9999 A.D. when somebody is worried about the Y10K crisis
they'll break into the ruin of the CS building just to find Bill's skeleton
chained to the door of the server room and the VAXcluster still working away
inside the room. |
|
| Back to top |
|
|
Bill Gunshannon
*nix forums Guru
Joined: 21 Jul 2005
Posts: 1019
|
| Posted: Tue Feb 08, 2005 7:50 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
In article <cub66o$1i4$1@naig.caltech.edu>,
David Mathog <mathog@caltech.edu> writes:
| Quote: | Bill Gunshannon wrote:
(As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.)
Killer VAXen??? Your cell phone probably has a faster processor.
|
"Killer" when compared to the VAXen we have had in the department
in the past. :-)
| Quote: |
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department.
But why does Cobol imply VMS? There are Cobol
compilers for other platforms, including
ones the students already own (Windows, Linux).
|
Windows compilers are expensive and frequently lack particular
features needed for the course. I am working very hard to keep
the Linux garbage out of my computer room.
| Quote: | Some of
these compilers are free. For instance, one can obtain
a free Fujitsu compiler that runs on Windows, Solaris,
and HPUX.
|
I looked at it quite some time ago. It wasn't of much real use.
| Quote: | There are also commercial ones,
like Acucobol. I have no idea how good any of these are, although
I'd expect that at least the Fujitsu offering is an industry grade
compiler.
|
The commercial ones are too expensive and the "Free" Fujitsu one
(unless it has changed) was a student edition and seriously crippled.
Interesting that someone here is busy trying to convince me not to
use VMS and suggesting Windows and linux. The use of COBOL is just
a "foot int he door" thing. As long as the VMS version is the best
solution for the task I get to keep VMS in the department (as long
as it has a zero cost).
| Quote: |
It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!! :-)
No, the part that sounds funny is you pushing VMS to run Cobol
|
As I said above, it is the use of COBOL that lets me keep VMS around.
We have looked at some of the "free" COBOL for Unix systems and they
have all proven to be too buggy for any real use. VMS COBOL is a "no
additional cost" solution. At least at the moment. That could change.
| Quote: | instead of one of the IBM platforms. Cobol is available for VMS
but all the Cobol programmers I know (including my wife) worked
on IBMs.
|
Yeah, I have done COBOL on IBM's, Univacs, Primes, PDP's, VAXen as well
as numerous Microcomputers. Actually, it is probably the one language
that can truly claim portability between a large number of different
systems. I always enjoyed working with it and hope they continue to
use it here so the students get at least a little exposure to it. It
is not going away any time soon.
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h> |
|
| Back to top |
|
|
John Smith
*nix forums Guru
Joined: 15 Oct 2003
Posts: 561
|
| Posted: Tue Feb 08, 2005 7:45 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
Keith Cayemberg wrote:
| Quote: | David Mathog wrote:
Bill Gunshannon wrote:
(As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.)
Killer VAXen??? Your cell phone probably has a faster processor.
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a course that
still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department.
But why does Cobol imply VMS? There are Cobol
compilers for other platforms, including
ones the students already own (Windows, Linux). Some of
these compilers are free. For instance, one can obtain
a free Fujitsu compiler that runs on Windows, Solaris,
and HPUX. There are also commercial ones,
like Acucobol. I have no idea how good any of these are, although
I'd expect that at least the Fujitsu offering is an industry grade
compiler.
It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!! :-)
No, the part that sounds funny is you pushing VMS to run Cobol
instead of one of the IBM platforms. Cobol is available for VMS
but all the Cobol programmers I know (including my wife) worked
on IBMs.
Regards,
David Mathog
mathog@caltech.edu
Well, I work for IBM, however, I'm responsible for several CIM
applications mostly programmed by us in COBOL (mixed with up to 3
other languages in the same image) running on over one hundred
OpenVMS Systems in factories all over (new and old) Europe. This is
only one of several customers I'm aware of using COBOL on OpenVMS. So
I suppose it all depends on what part of the world you exist in, or
of which you have experienced.
|
Question: Who chose to use VMS in those apps? IBM or the end customer? |
|
| Back to top |
|
|
Keith Cayemberg
*nix forums addict
Joined: 26 May 2005
Posts: 80
|
| Posted: Tue Feb 08, 2005 7:42 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
David Mathog wrote:
| Quote: | Bill Gunshannon wrote:
(As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.)
Killer VAXen??? Your cell phone probably has a faster processor.
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a course that
still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department.
But why does Cobol imply VMS? There are Cobol
compilers for other platforms, including
ones the students already own (Windows, Linux). Some of
these compilers are free. For instance, one can obtain
a free Fujitsu compiler that runs on Windows, Solaris,
and HPUX. There are also commercial ones,
like Acucobol. I have no idea how good any of these are, although
I'd expect that at least the Fujitsu offering is an industry grade
compiler.
It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!! :-)
No, the part that sounds funny is you pushing VMS to run Cobol
instead of one of the IBM platforms. Cobol is available for VMS
but all the Cobol programmers I know (including my wife) worked
on IBMs.
Regards,
David Mathog
mathog@caltech.edu
|
Well, I work for IBM, however, I'm responsible for several CIM
applications mostly programmed by us in COBOL (mixed with up to 3 other
languages in the same image) running on over one hundred OpenVMS Systems
in factories all over (new and old) Europe. This is only one of several
customers I'm aware of using COBOL on OpenVMS. So I suppose it all
depends on what part of the world you exist in, or of which you have
experienced.
Cheers!
Keith Cayemberg |
|
| Back to top |
|
|
Bill Gunshannon
*nix forums Guru
Joined: 21 Jul 2005
Posts: 1019
|
| Posted: Tue Feb 08, 2005 7:35 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
In article <YsmdnTP2Faolj5TfRVn-qQ@igs.net>,
"John Smith" <a@nonymous.com> writes:
| Quote: | Bill Gunshannon wrote:
In article <cuaqk6$qqm$1@naig.caltech.edu>,
David Mathog <mathog@caltech.edu> writes:
bob@instantwhip.com wrote:
http://www.shannonknowshpc.com/stories.php?story=05/02/07/1720316
That's an easy one. Because porting open source software
to a moribund OS is not a particularly worthy goal and nobody
is going to give you very much credit for having done it.
Not exactly true. It would have to be examined like any other
project, but I was successful in getting 4 porting projects
approved by our faculty committee a few years ago. I asked one
of the professors who has visited a number of other schools as
a member of an ABET Acreditation Team and he assures me that
about 50% of colleges with CS programs (a higher % for acredited
programs) have similar senior projects courses. The biggest
thing holding up projects like these is the lack of VMS systems
in academia. (As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.) I look forward to trying to get some VMS porting
projects in the works again. The students are always looking
for something more meaningful than the typical doctor's office
Excel/Access billing system.
I used to do rather a lot of porting to OpenVMS and believe me,
my time is much better spent doing other things and just
running the code as is on Linux. When I work on code now it's
to make it faster or better, not just to get it to run in
the first place on a platform that nobody in academia
cares about or uses anymore.
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department. It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!! :-)
Push HP to give you some 'current' technology, like an Alpha or one of those
prototype IA64 boxes (they'll be making those until the pre-production line
closes when Intel EOL's the cpu).
|
Actually, Compaq did give me an Alpha. A small one, but adequate.
That was in the days when they actually seemed interested. But
interest seemed to dry up about the time of the merger. I have
had limited contact since then, but it seems I am mostly back on
my own again.
| Quote: |
Father: "So what did you do in CS class today, son?
Son: "I got to work some goofy fossilized computer Professor Bill still has
hanging around the lab."
Father: "Really? Which one?"
Son: "Something called WAX or VAX, or something like that."
Father: "I used to program on those too."
Son: "Like I said, some fossil. Slow too."
|
So tell me, how would the average user writing program in a
high-level language (think under-grads here) even tell what
the architecture was? Why would he care? And I hardly expect
that the 4 processor machines I have are going to be slow
supporting only a small handful of students. At it's peak
it will have maybe 15 simultaneous users runnging the COBOL
Compiler. I think it will offer acceptable performance.
And anyway, they are hardly fossilized computers when compared to
the PDP-11/44's standing next to them.
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h> |
|
| Back to top |
|
|
David Mathog
*nix forums Guru Wannabe
Joined: 23 Feb 2005
Posts: 145
|
| Posted: Tue Feb 08, 2005 7:07 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
Bill Gunshannon wrote:
| Quote: | (As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.)
|
Killer VAXen??? Your cell phone probably has a faster processor.
| Quote: |
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department.
|
But why does Cobol imply VMS? There are Cobol
compilers for other platforms, including
ones the students already own (Windows, Linux). Some of
these compilers are free. For instance, one can obtain
a free Fujitsu compiler that runs on Windows, Solaris,
and HPUX. There are also commercial ones,
like Acucobol. I have no idea how good any of these are, although
I'd expect that at least the Fujitsu offering is an industry grade
compiler.
| Quote: | It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!!
|
No, the part that sounds funny is you pushing VMS to run Cobol
instead of one of the IBM platforms. Cobol is available for VMS
but all the Cobol programmers I know (including my wife) worked
on IBMs.
Regards,
David Mathog
mathog@caltech.edu |
|
| Back to top |
|
|
John Smith
*nix forums Guru
Joined: 15 Oct 2003
Posts: 561
|
| Posted: Tue Feb 08, 2005 6:27 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
Bill Gunshannon wrote:
| Quote: | In article <cuaqk6$qqm$1@naig.caltech.edu>,
David Mathog <mathog@caltech.edu> writes:
bob@instantwhip.com wrote:
http://www.shannonknowshpc.com/stories.php?story=05/02/07/1720316
That's an easy one. Because porting open source software
to a moribund OS is not a particularly worthy goal and nobody
is going to give you very much credit for having done it.
Not exactly true. It would have to be examined like any other
project, but I was successful in getting 4 porting projects
approved by our faculty committee a few years ago. I asked one
of the professors who has visited a number of other schools as
a member of an ABET Acreditation Team and he assures me that
about 50% of colleges with CS programs (a higher % for acredited
programs) have similar senior projects courses. The biggest
thing holding up projects like these is the lack of VMS systems
in academia. (As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.) I look forward to trying to get some VMS porting
projects in the works again. The students are always looking
for something more meaningful than the typical doctor's office
Excel/Access billing system.
I used to do rather a lot of porting to OpenVMS and believe me,
my time is much better spent doing other things and just
running the code as is on Linux. When I work on code now it's
to make it faster or better, not just to get it to run in
the first place on a platform that nobody in academia
cares about or uses anymore.
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department. It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!!
|
Push HP to give you some 'current' technology, like an Alpha or one of those
prototype IA64 boxes (they'll be making those until the pre-production line
closes when Intel EOL's the cpu).
Father: "So what did you do in CS class today, son?
Son: "I got to work some goofy fossilized computer Professor Bill still has
hanging around the lab."
Father: "Really? Which one?"
Son: "Something called WAX or VAX, or something like that."
Father: "I used to program on those too."
Son: "Like I said, some fossil. Slow too." |
|
| Back to top |
|
|
Bill Gunshannon
*nix forums Guru
Joined: 21 Jul 2005
Posts: 1019
|
| Posted: Tue Feb 08, 2005 5:41 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
In article <cuaqk6$qqm$1@naig.caltech.edu>,
David Mathog <mathog@caltech.edu> writes:
Not exactly true. It would have to be examined like any other
project, but I was successful in getting 4 porting projects
approved by our faculty committee a few years ago. I asked one
of the professors who has visited a number of other schools as
a member of an ABET Acreditation Team and he assures me that
about 50% of colleges with CS programs (a higher % for acredited
programs) have similar senior projects courses. The biggest
thing holding up projects like these is the lack of VMS systems
in academia. (As soon as the electricians get my new power in I
will have a couple of killer VAXen to make available to the
students again.) I look forward to trying to get some VMS porting
projects in the works again. The students are always looking
for something more meaningful than the typical doctor's office
Excel/Access billing system.
| Quote: |
I used to do rather a lot of porting to OpenVMS and believe me,
my time is much better spent doing other things and just
running the code as is on Linux. When I work on code now it's
to make it faster or better, not just to get it to run in
the first place on a platform that nobody in academia
cares about or uses anymore.
|
It's usually a fight, but I still try to keep VMS at least
on the horizon here. I have one professor who teaches a
course that still includes some COBOL programming and as
long as that persists I will, at least, have enough support
to keep VMS running in the department. It is up to me to try
and expand that usage and once I have a couple machines with
real performance I will get back to work on that. (Must sound
funny having the groups biggest Unix defender wrtitting about
pushing VMS in academia!! :-)
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h> |
|
| Back to top |
|
|
David Mathog
*nix forums Guru Wannabe
Joined: 23 Feb 2005
Posts: 145
|
| Posted: Tue Feb 08, 2005 3:50 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
bob@instantwhip.com wrote:
| Quote: | http://www.shannonknowshpc.com/stories.php?story=05/02/07/1720316
|
That's an easy one. Because porting open source software
to a moribund OS is not a particularly worthy goal and nobody
is going to give you very much credit for having done it.
I used to do rather a lot of porting to OpenVMS and believe me,
my time is much better spent doing other things and just
running the code as is on Linux. When I work on code now it's
to make it faster or better, not just to get it to run in
the first place on a platform that nobody in academia
cares about or uses anymore.
And if you weren't paying attention and reading my posts
all of those years back, the two main reasons nobody cares
about OpenVMS is that it's:
1. expensive
2. slow
The open source code in particular tends to run like a
dog on OpenVMS. Factors of 20x slower on VMS versus
linux (both on DS10) were common. CPU bound stuff ran
at the same speed, but a lot of this code is I/O bound
and there's a world of difference between forcing the data
all the way to disk vs. just to disk cache. Plus even
with ramdisks RMS was socking you for a factor of 2 or 3
vs. linux.
Regards,
David Mathog
mathog@caltech.edu |
|
| Back to top |
|
|
Bill Gunshannon
*nix forums Guru
Joined: 21 Jul 2005
Posts: 1019
|
| Posted: Tue Feb 08, 2005 1:10 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
In article <M5adnT9Wpb9ivJXfRVn-gQ@igs.net>,
"John Smith" <a@nonymous.com> writes:
| Quote: | bob@instantwhip.com wrote:
http://www.shannonknowshpc.com/stories.php?story=05/02/07/1720316
The entire program was flawed from the start - launched nearly a year ago,
the University of Turn is one of perhaps 2-3 educational institutions that
decided to participate - a far cry from the 20 or so that HP wanted to have.
|
Something about a lamp under a bushelbasket comes to mind.
| Quote: |
Think about it....how many universities are going to give a course credit to
a student for porting open source to VMS?
|
Pretty much any one that has a CS program as they likely all include
a required "senior projects" course.
| Quote: |
I have repeatedly asked HP to *fund* a dedicated team of under-employed
experienced VMS developers to port the top 5 open source apps in each
category (however that's defined is up for discussion) which currently don't
have VMS versions. How much could that cost? ...a few million dollars.....
chicken feed in the whole scheme of things...maybe 200 apps in 1 year in
addition to those from commercial ISV's. A dedicated open-source porting
team is far better than relying on students in academia who have Okotberfest
and girlfriends to attend to when not in class.
|
Don't know much about colleges today, do you? We had a single
student develop an entire Secure Wireless system with accounts
and authentication tied to our user database. The system was
based on recent papers on the subject and we are hoping to go
production with it this next summer when time allows revamping
our whole wireless system. The students flock to "real" projects
over the various contrived applications that make up most of the
offerings.
Many major current systems started out as "Student Projects".
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h> |
|
| Back to top |
|
|
Bill Gunshannon
*nix forums Guru
Joined: 21 Jul 2005
Posts: 1019
|
| Posted: Tue Feb 08, 2005 1:01 pm Post subject:
Re: Why aren't more universities doing this?
|
|
|
In article <1107821396.025828.137400@z14g2000cwz.googlegroups.com>,
bob@instantwhip.com writes:
| Quote: | http://www.shannonknowshpc.com/stories.php?story=05/02/07/1720316
|
Some of us talked about this long before Italy came into the
picture. The reception from Compaq and later HP was somewhat
less than lukewarm. It's hard enough selling an idea to the
administration of a University. It's pretty much impossible
without even token support from the vendor.
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h> |
|
| Back to top |
|
|
Google
|
|
| Back to top |
|
|
|
|
The time now is Thu Jan 08, 2009 5:02 am | All times are GMT
|
|
Free Advertising | Xbox Mod Chips | Western Union Locations | Debt Consolidation | Remortgages
|
|
Copyright © 2004-2005 DeniX Solutions SRL
|
|
|
|
Other DeniX Solutions sites:
Unix/Linux blog |
electronics forum |
medicine forum |
science forum |
|
|
Privacy Policy
|
Powered by phpBB © 2001, 2005 phpBB Group
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Forum index
»
Not Unix
»
VMS
