[usl-v-bsdi-ucb/include/usenet-header.html]Xref: gmd.de comp.org.decus:2401 comp.os.vms:41053 comp.sys.dec:13632
comp.unix.advocacy:676
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Path: gmd.de!newsserver.jvnc.net!darwin.sura.net!howland.reston.ans.net!
news.moneng.mei.com!uwm.edu!mixcom.com!Craig.Dedo
From: Craig Dedo <Craig...@mixcom.mixcom.com>
Subject: Is VMS Dying?
Message-ID: <1994Feb26.204059.4068@mixcom.mixcom.com>
Sender: eec...@mixcom.mixcom.com (Craig Dedo)
Organization: Milwaukee Internet Xchange BBS, Milwaukee, WI (414) 241-5469
Date: Sat, 26 Feb 1994 20:40:59 GMT
Lines: 52
February 26, 1994
This week a software engineer told me that VMS is a dying operating
system. I would like to know what the readers of this forum think. I have
some thoughts on this issue, but first I would like to see what other
people think. Some of the following questions involve business rather than
technical issues. Some people may feel that discussing business issues is
inappropriate in a technical forum. However, the future of a technical
product like VMS depends heavily on BOTH.
Please consider the following questions, as well as any related matters.
1. Does VMS have much of a future?
2. Is VMS losing out to Unix? If so, why? If not, what evidence is
there that this is not the case?
3. Does VMS provide enough bang for the buck? How does it compare to
the competition in this area?
4. What do you think are the strengths of VMS that would make it an
attractive alternative to someone who is used to Unix? To MS-DOS?
5. If the big objection to VMS is that it is proprietary, why don't we
hear such complaints about MS-DOS and Windows?
6. Should DEC consider franchising VMS to independent software
developers (ISVs) who could add their own enhancements, the way Microsoft
has done with MS-DOS or McDonald's has done with their restaurants? Would
this answer the objection that VMS is proprietary while Unix is not?
7. How can DEC make VMS attractive to the nation's computer science
departments?
8. How can DEC make VMS attractive to software developers?
9. Does VMS have features that are not found or are not nearly as well
developed in competing operating systems? If so, what are they?
10. Are these features (which are better developed in VMS) very much
appreciated by the people who actually make the operating system selection
decisions? If not, why not?
11. What market should VMS serve?
12. Should DEC develop a standard GUI interface to VMS, such as
Microsoft Windows or X-Windows?
Sincerely,
--
Craig T. Dedo Craig...@mixcom.com
17130 W. Burleigh Place (414) 783-5869
Brookfield, WI 53005
Xref: gmd.de comp.org.decus:2503 comp.os.vms:41241 comp.sys.dec:13709
comp.unix.advocacy:725
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Path: gmd.de!xlink.net!fauern!news.th-darmstadt.de!zib-berlin.de!netmbx.de!
Germany.EU.net!EU.net!ieunet!curia!csvax1.ucc.ie!CHRIS
From: ch...@csvax1.ucc.ie (Chris Higgins - System Administrator)
Subject: Re: Is VMS Dying?
Message-ID: <CLvuJ0.9Cn@curia.ucc.ie>
Sender: use...@curia.ucc.ie
Reply-To: ch...@csvax1.ucc.ie
Organization: Computer Science Dept. University College, Cork, Ireland.
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
Date: Sun, 27 Feb 1994 12:25:47 GMT
Lines: 108
In article <1994Feb26....@mixcom.mixcom.com>,
Craig Dedo <Craig...@mixcom.mixcom.com> writes:
> February 26, 1994
>
> This week a software engineer told me that VMS is a dying operating
>system. I would like to know what the readers of this forum think. I have
>some thoughts on this issue, but first I would like to see what other
>people think. Some of the following questions involve business rather than
>technical issues. Some people may feel that discussing business issues is
>inappropriate in a technical forum. However, the future of a technical
>product like VMS depends heavily on BOTH.
>
> Please consider the following questions, as well as any related matters.
> 1. Does VMS have much of a future?
Yes. Lots.
>
> 2. Is VMS losing out to Unix? If so, why? If not, what evidence is
>there that this is not the case?
Unix seems to be the choice of universities, so graduates don't know "VMS".
On the other hand, most of the businesses that I know that want a reliable
system go VMS, and VMS only.
>
> 3. Does VMS provide enough bang for the buck? How does it compare to
>the competition in this area?
As an Operating system, VMS drives Unix into the ground. I run a VMS system
(and several Unix system so I'm not biased :-). Over the last couple of months
I've witnessed numerous occasions where other local admins (Unix machines) have
proclaimed that feature X should be part of Unix, and why don't all Unix
machines come fitted with feature Y, and I've taken great pleasure in pointing
out that those features are part of VMS, and have been for a long time.
>
> 4. What do you think are the strengths of VMS that would make it an
>attractive alternative to someone who is used to Unix? To MS-DOS?
People on Unix who want to get more out of the machine than just compliation,
people who want security and control over their systems should realise that VMS
has all the controls necessary.
>
> 5. If the big objection to VMS is that it is proprietary, why don't we
>hear such complaints about MS-DOS and Windows?
To get a VMS machine you need to admin it.. so you need to know computers...
so you'll know if something is wrong... whereas with Windoze you can give it to
any business man, throw some jargon about "standards", "user-interface",
"progressive-technology", etc etc etc
You should talk to people 'in the know' about Windoze... the amount of hardware
that you need to throw at windoze to get reasonable response time is criminal.
If you threw equivalent hardware at any other system then you'd need Cray level
power in an average Alpha.
>
> 6. Should DEC consider franchising VMS to independent software
>developers (ISVs) who could add their own enhancements, the way Microsoft
>has done with MS-DOS or McDonald's has done with their restaurants? Would
>this answer the objection that VMS is proprietary while Unix is not?
Unix is *** NON *** STANDARD ***. How many versions of Unix exist, how big is
the source code to various packages because they have to support hundreds of
variations of Unix.
>
> 7. How can DEC make VMS attractive to the nation's computer science
>departments?
Provide shells other than DCL, so that people can have the flexibility to do as
they wish.... (I love DCL, but TCSH straight on the OS (no posix) would be
nice.
>
> 8. How can DEC make VMS attractive to software developers?
Shells.
>
> 9. Does VMS have features that are not found or are not nearly as well
>developed in competing operating systems? If so, what are they?
File ACLs, Accounting, Resource control......
>
> 10. Are these features (which are better developed in VMS) very much
>appreciated by the people who actually make the operating system selection
>decisions? If not, why not?
Only if the people making the decision 'know' what they are looking for, and
where they may expand.
>
> 11. What market should VMS serve?
All of them, why not :-)
>
> 12. Should DEC develop a standard GUI interface to VMS, such as
>Microsoft Windows or X-Windows?
Haven't you heard of DECwindows ???
>
>Sincerely,
>--
>Craig T. Dedo Craig...@mixcom.com
>17130 W. Burleigh Place (414) 783-5869
>Brookfield, WI 53005
Disclaimer:
1/ All comments are my own.
2/ None of my comments may be printed/ published without my explicit agreement.
3/ I run VMS machines, Linux machines, OSF/AXP machines, SUN-OS machines,
Ultrix machines, and avoid DOS/Windows machines :-)
4/ I lecture Unix courses, but still think that VMS is a 'better' operating
system, esp from a sys admin point of view.
Chris.
+ J.C. Higgins, + Ch...@csvax1.ucc.ie + If you love something, set it +
+ VMS Sys. Admin, + Ch...@cs.ucc.ie + free. If it doesn't come back +
+ Comp.Sc.Dept. + Ch...@odyssey.ucc.ie + to you, hunt it down and +
+ UCC, Ireland + C.Hi...@bureau.ucc.ie + KILL it. +
Xref: gmd.de comp.org.decus:2417 comp.os.vms:41083 comp.sys.dec:13649
comp.unix.advocacy:689
Path: gmd.de!xlink.net!howland.reston.ans.net!cs.utexas.edu!convex!
news.utdallas.edu!feenix.metronet.com!tachyon!montagar!davidc
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Subject: Re: Is VMS Dying?
Message-ID: <1994Feb27.120007.18724@avatar.montagar.com>
From: dav...@avatar.montagar.com (David L. Cathey)
Date: 27 Feb 94 12:00:07 CDT
References: <1994Feb26.204059.4068@mixcom.mixcom.com> <CLvuJ0.9Cn@curia.ucc.ie>
Organization: Montagar Software Concepts, Plano TX
Lines: 151
In article <CLvuJ...@curia.ucc.ie>,
ch...@csvax1.ucc.ie (Chris Higgins - System Administrator) writes:
> In article <1994Feb26....@mixcom.mixcom.com>,
> Craig Dedo <Craig...@mixcom.mixcom.com> writes:
>> February 26, 1994
>>
>> This week a software engineer told me that VMS is a dying operating
>>system. I would like to know what the readers of this forum think. I have
>>some thoughts on this issue, but first I would like to see what other
>>people think. Some of the following questions involve business rather than
>>technical issues. Some people may feel that discussing business issues is
>>inappropriate in a technical forum. However, the future of a technical
>>product like VMS depends heavily on BOTH.
>>
>> Please consider the following questions, as well as any related matters.
>> 1. Does VMS have much of a future?
> Yes. Lots.
Yes it does. I feel that the AXP has put the price/performance
back into VMS.
>> 2. Is VMS losing out to Unix? If so, why? If not, what evidence is
>>there that this is not the case?
> Unix seems to be the choice of universities, so graduates don't know "VMS".
> On the other hand, most of the businesses that I know that want a reliable
> system go VMS, and VMS only.
The UNIX people would have you to believe that, and in some cases
that may be true. However, I saw an article in a mag (not one of the normal
"DEC-centric" ones) that VMS is still a growing part of Digitals revenues.
It seems that they are seen as the choice for down-sized mainframes.
>> 3. Does VMS provide enough bang for the buck? How does it compare to
>>the competition in this area?
> As an Operating system, VMS drives Unix into the ground. I run a VMS system
> (and several Unix system so I'm not biased :-).
Despite the previous response, your "bang for the buck" will vary.
Not all systems are suited for all purposes. However, given the flexibility
of VMS, it provide excellent price/performance. And for some performance,
it AXP or a Cray. :-)
>> 4. What do you think are the strengths of VMS that would make it an
>>attractive alternative to someone who is used to Unix? To MS-DOS?
Security. Internetworking (TCP/IP, OSI, LanMan, Novell, XNS, X.25
and all CONCURRENTLY!). Robust O/S (doesn't "panic" often (UNIX) or have odd
memory lockups (MS-DOS); have known systems to run over a year without
rebooting!).
>> 5. If the big objection to VMS is that it is proprietary, why don't we
>>hear such complaints about MS-DOS and Windows?
Here's a hint: Except for a very few (like Linux), all O/S's are
propriatary. Why else are they trademarked??? AIX, HP-UX, Solaris, Sun-OS,
Ultrix, MVS, OpenVMS, MS-DOS, MS-Windows, WNT, ... all are proprietary.
You cannot buy vendor A's hardware and run vendor B's O/S on it. Oh, but
the UNIX crowd says that the your more portable. Really? Did you know that
the only operating system to meet the toughest UNIX standard (FIPS 151-2)
is OpenVMS??? I've ported many applications from the net to VMS (including
network and X-window stuff) with little to trouble. And the code is usually
sprinkled with #ifdef's for various UNIX O/S's. If UNIX is so standard, then
why do you need "#ifdef BSD", ... ???
> To get a VMS machine you need to admin it.. so you need to know computers...
Not so fast! You need to "admin" any system.
> If you threw equivalent hardware at any other system then you'd need Cray level
> power in an average Alpha.
True. I can run a VMS system with Motif with 16Mb and have better
performance than a 16Mb PC with WNT... :-) Though I'd still want more memory..
>> 6. Should DEC consider franchising VMS to independent software
>>developers (ISVs) who could add their own enhancements, the way Microsoft
>>has done with MS-DOS or McDonald's has done with their restaurants? Would
>>this answer the objection that VMS is proprietary while Unix is not?
>
> Unix is *** NON *** STANDARD ***. How many versions of Unix exist, how big is
> the source code to various packages because they have to support hundreds of
> variations of Unix.
True. This would create more confusion than VMS -> OpenVMS did.
>> 7. How can DEC make VMS attractive to the nation's computer science
>>departments?
>
> Provide shells other than DCL, so that people can have the flexibility to do as
> they wish.... (I love DCL, but TCSH straight on the OS (no posix) would be
> nice.
(So compile tcsh on POSIX!) DEC needs to do what AT&T did with UNIX:
(Listen DEC, this is important) GIVE UNIVERSITIES LICENSES AND HARDWARE.
DEC has TONS of hardware that they throw away in Arizona because it is still
on the books. Give this stuff and Campus-Wide-Software-Licenses to Colleges,
High Schools, Universities, ...
Here's why: UNIX is big because AT&T and Berkley got students used
to using it 10 years ago. Guess what - Those same people are NOW making
purchasing decisions about what they know: UNIX. Kids come out of college
knowing only what: UNIX. C. C++. Why? Because that's *all* they learned to
use, and it because the "standard". Put VMS boxes in there with lots of fun
software, and when they graduate, they will want to put VMS in there.
>> 8. How can DEC make VMS attractive to software developers?
> Shells.
Actually, DEC does do a good job on this, but it might be one of their
better kept secrets. DEC's Independent Software Vendor program has been
invaluable to me.
>> 9. Does VMS have features that are not found or are not nearly as well
>>developed in competing operating systems? If so, what are they?
> File ACLs, Accounting, Resource control......
ACL's on everything, Process Class scheduling (V6.0), file structures,
I/O through the command language, VMSclusters, locking ($ENQ), logicals
names, ...
>> 10. Are these features (which are better developed in VMS) very much
>>appreciated by the people who actually make the operating system selection
>>decisions? If not, why not?
> Only if the people making the decision 'know' what they are looking for, and
> where they may expand.
In short: NO. Typically, the people making the purchase don't know
squat about the things they are buying. Managment hears "Open Systems", and
somewhere they read that means "UNIX". VMS provides just as open of an
environment as any UNIX, and provides a lot more for those that want it.
Me, yes I appreciate all the features.
>> 11. What market should VMS serve?
> All of them, why not :-)
I've seen VMS systems control shop floors, point-of-sale for
Blockbuster Video, number crunching/simulations. DEC's AXP systems are
even used to produce "Beavis and Butthead"!
>> 12. Should DEC develop a standard GUI interface to VMS, such as
>>Microsoft Windows or X-Windows?
>
> Haven't you heard of DECwindows ???
Actually, it's "Motif" - which is already on VMS. Motif is a toolkit
that sits on top of X-windows.
>>Craig T. Dedo Craig...@mixcom.com
> + J.C. Higgins, + Ch...@csvax1.ucc.ie + If you love something, set it +
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
David L. Cathey |INET: dav...@montagar.com
Montagar Software Concepts |UUCP: ...!montagar!davidc
P. O. Box 260772, Plano TX 75026-0772 |Fone: (214)-618-2117
Xref: gmd.de comp.org.decus:2409 comp.os.vms:41068 comp.sys.dec:13640
comp.unix.advocacy:683
Path: gmd.de!newsserver.jvnc.net!howland.reston.ans.net!europa.eng.gtefsd.com!
news.umbc.edu!haven.umd.edu!nova.umd.edu!nova!rockwell
From: rock...@nova.umd.edu (Raul Deluth Miller)
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Subject: Re: Is VMS Dying?
Date: 27 Feb 1994 20:08:53 -0500
Organization: University of Maryland University College
Lines: 78
Distribution: world
Message-ID: <ROCKWELL.94Feb27200852@nova.umd.edu>
References: <1994Feb26.204059.4068@mixcom.mixcom.com> <CLvuJ0.9Cn@curia.ucc.ie>
<1994Feb27.120007.18724@avatar.montagar.com>
NNTP-Posting-Host: nova.umd.edu
In-reply-to: davidc@avatar.montagar.com's message of 27 Feb 94 12:00:07 CDT
>>Craig T. Dedo
>J.C. Higgins:
David L. Cathey:
Security. Internetworking (TCP/IP, OSI, LanMan, Novell,
XNS, X.25 and all CONCURRENTLY!). Robust O/S (doesn't "panic"
often (UNIX) or have odd memory lockups (MS-DOS); have known
systems to run over a year without rebooting!).
Interesting. UMUC moved me off of the VMS platform I was on, and onto
Unix, because of problems I was having with the TCP/IP implementation.
[Of course, these problems have probably been fixed, since then. I've
not found any compelling reason to go back.]
>> 5. If the big objection to VMS is that it is proprietary, why
>>don't we hear such complaints about MS-DOS and Windows?
Here's a hint: Except for a very few (like Linux), all
O/S's are propriatary. Why else are they trademarked??? AIX,
HP-UX, Solaris, Sun-OS, Ultrix, MVS, OpenVMS, MS-DOS, MS-Windows,
WNT, ... all are proprietary. You cannot buy vendor A's hardware
and run vendor B's O/S on it. Oh, but the UNIX crowd says that the
your more portable. Really?
Yeah, runs on more machines.
(So compile tcsh on POSIX!) DEC needs to do what AT&T did
with UNIX: (Listen DEC, this is important) GIVE UNIVERSITIES
LICENSES AND HARDWARE. DEC has TONS of hardware that they throw
away in Arizona because it is still on the books. Give this stuff
and Campus-Wide-Software-Licenses to Colleges, High Schools,
Universities, ...
I don't know what this would solve. Every college I've been to has
had both vms and unix [and some ibm os, such as cms] available for
student use [and dos and macintoshes and nexts]. For some reason, the
unix systems have been more pleasant to use [you can have a larger
storage quota, you can run more processes, etc...].
Which may be why I've never invested the time to master dcl.
>> 8. How can DEC make VMS attractive to software developers?
> Shells.
Actually, DEC does do a good job on this, but it might be
one of their better kept secrets. DEC's Independent Software
Vendor program has been invaluable to me.
Bleah. The vms systems I've been on have had a variety of packages
that LOOK neat -- but when I try to use them, I find only superficial
documentation on little things like "why is this system fundamentally
different other things like it, on other machines?"
Without widely available information/examples, it's slow going on any
machine. "Better kept secrets" aren't going to make things easier for
the people who have to use the machine.
[That, and all to often I run into parts of the help system that are
unable to send information a screen at a time. Which can be
inconvenient, or crippling, depending on my situation at the time.]
>> 9. Does VMS have features that are not found or are not nearly
>>as well developed in competing operating systems? If so, what
>>are they?
> File ACLs, Accounting, Resource control......
ACL's on everything, Process Class scheduling (V6.0), file
structures, I/O through the command language, VMSclusters,
locking ($ENQ), logicals names, ...
Which might be why VMS systems are so hard for a person to use --
there's all this administrative bullshit you have to go through to get
permission to use the system. Better to use a system that relies on
social mechanisms than one which relies on red tape.
Micro-management is not necessarily better management. And armor
plating is not always the best approach to security.
Raul D. Miller
<rock...@nova.umd.edu>
Xref: gmd.de comp.org.decus:2406 comp.os.vms:41064 comp.sys.dec:13637
comp.unix.advocacy:680
Path: gmd.de!newsserver.jvnc.net!howland.reston.ans.net!gatech!swrinde!
elroy.jpl.nasa.gov!nntp-server.caltech.edu!SOL1.GPS.CALTECH.EDU!CARL
From: ca...@SOL1.GPS.CALTECH.EDU (Carl J Lydick)
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Subject: Re: Is VMS Dying?
Date: 28 Feb 1994 08:33:50 GMT
Organization: HST Wide Field/Planetary Camera
Lines: 57
Distribution: world
Message-ID: <2ksa9e$run@gap.cco.caltech.edu>
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
<CLvuJ0.9Cn@curia.ucc.ie> <1994Feb27.120007.18724@avatar.montagar.com>,
<ROCKWELL.94Feb27200852@nova.umd.edu>
Reply-To: ca...@SOL1.GPS.CALTECH.EDU
NNTP-Posting-Host: sol1.gps.caltech.edu
In article <ROCKWELL.94...@nova.umd.edu>,
rock...@nova.umd.edu (Raul Deluth Miller) writes:
=>>Craig T. Dedo
=>J.C. Higgins:
=David L. Cathey:
= Security. Internetworking (TCP/IP, OSI, LanMan, Novell,
= XNS, X.25 and all CONCURRENTLY!). Robust O/S (doesn't "panic"
= often (UNIX) or have odd memory lockups (MS-DOS); have known
= systems to run over a year without rebooting!).
=
=Interesting. UMUC moved me off of the VMS platform I was on, and onto
=Unix, because of problems I was having with the TCP/IP implementation.
Which implentation was that? UCX, at a guess? Or equally bad, WIN? If it was
either of those, you could've saved yourself some money and switched to CMUIP,
which is free, and better than either of the above.
=
=I don't know what this would solve. Every college I've been to has
=had both vms and unix [and some ibm os, such as cms] available for
=student use [and dos and macintoshes and nexts]. For some reason, the
=unix systems have been more pleasant to use [you can have a larger
=storage quota, you can run more processes, etc...].
Have you ever encountered the term "circular argument"? I didn't think so;
otherwise you'd've realized that that what you made above.
= >> 9. Does VMS have features that are not found or are not nearly
= >>as well developed in competing operating systems? If so, what
= >>are they?
= > File ACLs, Accounting, Resource control......
= ACL's on everything, Process Class scheduling (V6.0), file
= structures, I/O through the command language, VMSclusters,
= locking ($ENQ), logicals names, ...
=
=Which might be why VMS systems are so hard for a person to use --
=there's all this administrative bullshit you have to go through to get
=permission to use the system. Better to use a system that relies on
=social mechanisms than one which relies on red tape.
VMS does *NOT* rely on red tape. If the folks who own the system want to do it
that way, they can; if they don't, they don't have to. You're blaming the O/S
for local political problems. Of course, the other idiocies you've committed
make it unsurprising that you'd fuck up on this, too.
=Micro-management is not necessarily better management.
No, it's not. And VMS doesn't require it. However, NO management is not
necessarily good management. And many Unix implementations *DO* require that.
Your claim amounts to: Flexibility is evil. Of course, I don't expect you to
understand that.
--------------------------------------------------------------------------------
Carl J Lydick | INTERnet: CA...@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL
Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My
understanding of astronomy is purely at the amateur level (or below). So
unless what I'm saying is directly related to VAX/VMS, don't hold me or my
organization responsible for it. If it IS related to VAX/VMS, you can try to
hold me responsible for it, but my organization had nothing to do with it.
Xref: gmd.de comp.org.decus:2522 comp.os.vms:41278 comp.sys.dec:13733
comp.unix.advocacy:740
Path: gmd.de!xlink.net!rz.uni-karlsruhe.de!news.uni-stuttgart.de!
news.belwue.de!news.dfn.de!scsing.switch.ch!swidir.switch.ch!univ-lyon1.fr!
jussieu.fr!univ-lille1.fr!zaphod.crihan.fr!warwick!pipex!howland.reston.ans.net!
europa.eng.gtefsd.com!news.umbc.edu!haven.umd.edu!nova.umd.edu!nova!rockwell
From: rock...@nova.umd.edu (Raul Deluth Miller)
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Subject: Re: Is VMS Dying?
Date: 28 Feb 1994 07:07:04 -0500
Organization: University of Maryland University College
Lines: 59
Distribution: world
Message-ID: <ROCKWELL.94Feb28070703@nova.umd.edu>
References: <1994Feb26.204059.4068@mixcom.mixcom.com> <CLvuJ0.9Cn@curia.ucc.ie>
<1994Feb27.120007.18724@avatar.montagar.com>,
<ROCKWELL.94Feb27200852@nova.umd.edu>
<2ksa9e$run@gap.cco.caltech.edu>
NNTP-Posting-Host: nova.umd.edu
In-reply-to: carl@SOL1.GPS.CALTECH.EDU's message of 28 Feb 1994 08:33:50 GMT
Carl J Lydick:
. Which implentation was that? UCX, at a guess? Or equally bad,
. WIN? If it was either of those, you could've saved yourself some
. money and switched to CMUIP, which is free, and better than either
. of the above.
I haven't a clue what those stand for. Nor could I have saved myself
any money -- I was a *user*, not an *administrator*.
I think one implementation was from Wollogong [sp?]. [There was an
upgrade before I moved to another system.]
. =I don't know what this would solve. Every college I've been to
. =has had both vms and unix [and some ibm os, such as cms] available
. =for student use [and dos and macintoshes and nexts]. For some
. =reason, the unix systems have been more pleasant to use [you can
. =have a larger storage quota, you can run more processes, etc...].
. Have you ever encountered the term "circular argument"? I didn't
. think so; otherwise you'd've realized that that what you made
. above.
Really? At one school there were a lot of vms resources lying around
unused [several idle machines, even].
. = >> 9. Does VMS have features that are not found or are not
. = >>nearly as well developed in competing operating systems? If
. = >>so, what are they?
. = > File ACLs, Accounting, Resource control......
. = ACL's on everything, Process Class scheduling (V6.0), file
. = structures, I/O through the command language, VMSclusters,
. = locking ($ENQ), logicals names, ...
. =
. =Which might be why VMS systems are so hard for a person to use --
. =there's all this administrative bullshit you have to go through to
. =get permission to use the system. Better to use a system that
. =relies on social mechanisms than one which relies on red tape.
. VMS does *NOT* rely on red tape. If the folks who own the system
. want to do it that way, they can; if they don't, they don't have
. to. You're blaming the O/S for local political problems. Of
. course, the other idiocies you've committed make it unsurprising
. that you'd fuck up on this, too.
I've run across hostile system administrators on both vms and unix
systems. They seem to have a more pervasive effect on vms systems.
. =Micro-management is not necessarily better management.
. No, it's not. And VMS doesn't require it. However, NO management
. is not necessarily good management. And many Unix implementations
. *DO* require that. Your claim amounts to: Flexibility is evil. Of
. course, I don't expect you to understand that.
You mean you're not trying to communicate with me, you're just trying
to make me look bad?
Raul D. Miller
<rock...@nova.umd.edu>
Xref: gmd.de comp.org.decus:2511 comp.os.vms:41256 comp.sys.dec:13719
comp.unix.advocacy:732
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Path: gmd.de!newsserver.jvnc.net!netnews.upenn.edu!msuinfo!agate!
howland.reston.ans.net!cs.utexas.edu!swrinde!ihnp4.ucsd.edu!library.ucla.edu!
csulb.edu!csus.edu!netcom.com!laurel
From: lau...@netcom.com (Laurel Edgecomb)
Subject: Re: Is VMS Dying?
Message-ID: <laurelCM0J1s.L59@netcom.com>
Followup-To: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Organization: Northgate Equestrian Center, Walnut Creek, CA
X-Newsreader: TIN [version 1.2 PL1]
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
Date: Wed, 2 Mar 1994 01:05:52 GMT
Lines: 168
> 1. Does VMS have much of a future?
Yes, but its end is now in sight. Not final extinction but the
RSXification/RSTSification of VMS where only sites that need not
or can not change will still have VMS.
> 2. Is VMS losing out to Unix? If so, why? If not, what evidence is
> there that this is not the case?
Yes. Everything seems to be pointing towards U*X like OS's or to
a new OS. There are more U*X sites, applications, and
programmers. The net and US schools are mostly a U*X world. Rumors
from from DEC sales indicate that they are being instructed to
market OSF/1 to new sites...VMS is just not competitive.
Developer after developer is dumping VMS support and many new
applications (even from DEC) are not being developed for VMS.
For example FUSION is DEC's software development system for U*X.
It is one generation better than DECSET (at least). The DEC
developers would love to move it to VMS but there is no market.
VMS sites doing software development already have DECSET or
homebrew and don't need to change, and there are not enough new
sites doing VMS software development to be worth porting FUSION.
There is on U*X.
It is yet to be seen if Windows/NT will be a major player, but
unless they screw up badly it looks very good for NT. Windows/NT
promises (but has not yet delivered by a very long shot) all the
advantages that VMS delivers and more. A huge _potential_
application and programmer base, an even cleaner programmer
interface to the OS, security, reliability, support, and it is
truly multi-platform (2 is not multi).
For VMS (no matter how good it is) to remain viable must continue
to have a critical mass of sites running VMS and constant flow of
new sites. I believe that VMS has already crossed over the border
to below the critical mass for viability, and will, inevitably,
follow that long, slow, spiral to OS oblivion.
> 3. Does VMS provide enough bang for the buck? How does it compare to
> the competition in this area?
Yes if you need what VMS gives. VMS gives Security, Reliability,
Support, and A Clean (but a bit slow) Calling Standard. If you
don't need these features (and many don't or don't want to pay
for them) VMS is not the best bang for the buck.
> 4. What do you think are the strengths of VMS that would make it an
> attractive alternative to someone who is used to Unix? To MS-DOS?
It depends on the application. The application requirements drive the
hardware and OS selection.
> 5. If the big objection to VMS is that it is proprietary, why don't we
> hear such complaints about MS-DOS and Windows?
That is _not_ the big objection to VMS. The objection to VMS is
cost and "openness". By "openness" is meant the ability to run
applications written for U*X. The problem for VMS is there is
more U*X out there than VMS, more sites, more seats, more
developers, more applications (remember BETA vs VHS...some people
still have Beta).
There are plenty of complaints about MS-DOS and Windows. The
difference is only $ and the sophistication of the users. DOS
and Windows users are mostly application end users. They use a
package from some vendor. If the package does not work they tell
the vendor and the vendor fixes it. The software developers for
DOS and windows know that the OS is a _very_ simple OS. They
don't expect much and it did not cost big bucks. So they can't
yell very loud. When there is a new version they are happy.
VMS users expect, and pay for, a robust and sophisticated OS and
ask a lot. Businesses depend on the OS working as expected. If
it doesn't, they want a workaround or patch _now_. Under U*X the
source is often available and a systems programmer can make the
changes. There is often little support from the OS vender. Under
VMS the sources are available only for a price and most sites
don't have source. Most sites running VMS need the security of
VMS or its reliability and also have a DEC service contract.
Many modifications programmers would like to make to VMS seem to
cause problems later with getting good service from DEC (they
want to see the problem happen with your changes out). Of course
if you have VMS sources and don't care about DEC service, then
nothing stops one from modifying VMS.
> 6. Should DEC consider franchising VMS to independent software
> developers (ISVs) who could add their own enhancements, the way Microsoft
> has done with MS-DOS or McDonald's has done with their restaurants? Would
> this answer the objection that VMS is proprietary while Unix is not?
They should consider it, even try it. They have nothing to lose.
But that won't help. It is too late.
> 7. How can DEC make VMS attractive to the nation's computer science
> departments?
Give it away free along with all layered products, support it,
and sell hardware at a good discount.
It would have been a very good investment -- too late now though.
> 8. How can DEC make VMS attractive to software developers?
Sell VMS to more sites. ya, right.
> 9. Does VMS have features that are not found or are not nearly as well
> developed in competing operating systems? If so, what are they?
Yes.
RMS
Robust Security Features
Good Multi-Language Calling Standards
Relatively consistent API to system (compared to U*X & DOS)
Clusters
Relatively bug free and moderately powerful tools
(DOS tools are not powerful but relatively bug free
U*X tools are powerful but buggy and inconsistent)
> 10. Are these features (which are better developed in VMS) very much
> appreciated by the people who actually make the operating system selection
> decisions? If not, why not?
Yes.
> 11. What market should VMS serve?
VMS seems to make sense for applications that require large
amounts of custom programming in a secure, reliable environment
which is evolving, and thus requiring constant software updates.
> 12. Should DEC develop a standard GUI interface to VMS, such as
> Microsoft Windows or X-Windows?
It is being developed, but far from complete. (DECWindows/Motif
itself is not an O/S interface). It will be complete when a
system manager/programmer never needs to click in a decterm.
Should it be developed? I don't think so. VMS is not likely to
expand greatly into new sites, and existing sites seem happy with
the ASCII based tools. Nobody uses the GUI interfaces for most of
the tools that are in place now.
REGARDING COMMENTS ABOUT CARL LYDICK:
I have seen a few comments in this thread about people adding
Lydick to their kill file. Just so you know, that would be a
mistake (unless you just can't take being called, or seeing others
called, shit-heads). Except about religious or preference issues
Lydick is usually right. This is not because he is smarter or
knows more about VMS than anyone else on the net, but that he
knows enough to try a solution before spouting it to the net.
He, quite justifiably, acts pissed when someone spends the time to
ask a question but does not take the time to include what they
are trying to do, what they really did, and exactly what the
problem really is (including error messages, examples, and/or
logs). Given that those who fail to do this really are not
shit-heads, they are inconsiderate to those who are spending time
reading these posts to help them. I don't know why Lydick
chooses to use colorful, aggressive, personal attacks. I don't
really care why. I am more amused at the people that reply by,
first, indicating how childish Lydick is for doing such things,
then throw in a childish jab of their own. Lydick is, at least,
consistent (consistently useful and consistently acerbic). So if
he flames you - ignore the shit - get the message - or go crawl
under a rock somewhere and add him to you kill file.
Xref: gmd.de comp.org.decus:2531 comp.os.vms:41297 comp.sys.dec:13745
comp.unix.advocacy:747
Path: gmd.de!xlink.net!howland.reston.ans.net!cs.utexas.edu!convex!
news.utdallas.edu!feenix.metronet.com!tachyon!montagar!davidc
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Subject: Re: Is VMS Dying?
Message-ID: <1994Mar2.145103.18732@avatar.montagar.com>
From: dav...@avatar.montagar.com (David L. Cathey)
Date: 2 Mar 94 14:51:03 CDT
Followup-To: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
References: <1994Feb26.204059.4068@mixcom.mixcom.com> <laurelCM0J1s.L59@netcom.com>
Organization: Montagar Software Concepts, Plano TX
Lines: 44
In article <laurelCM...@netcom.com>, lau...@netcom.com (Laurel Edgecomb) writes:
>> 2. Is VMS losing out to Unix? If so, why? If not, what evidence is
>> there that this is not the case?
>
> Yes. Everything seems to be pointing towards U*X like OS's or to
> a new OS. There are more U*X sites, applications, and
> programmers. The net and US schools are mostly a U*X world. Rumors
> from from DEC sales indicate that they are being instructed to
> market OSF/1 to new sites...VMS is just not competitive.
> Developer after developer is dumping VMS support and many new
> applications (even from DEC) are not being developed for VMS.
Not always. There are some shops that switched to Unix are now
wanting to go back to VMS - Unix just didn't give them what they needed.
> For VMS (no matter how good it is) to remain viable must continue
> to have a critical mass of sites running VMS and constant flow of
> new sites. I believe that VMS has already crossed over the border
> to below the critical mass for viability, and will, inevitably,
> follow that long, slow, spiral to OS oblivion.
I forget the magazine, but VMS is still a growing revenue for
DEC. People are still buying more of it, and there is still new business.
>> 7. How can DEC make VMS attractive to the nation's computer science
>> departments?
>
> Give it away free along with all layered products, support it,
> and sell hardware at a good discount.
I have also posted something along those lines. DEC, are your
listening?
>> 8. How can DEC make VMS attractive to software developers?
>
> Sell VMS to more sites. ya, right.
Some of the are. However, the message from DEC-salespeople is that
VMS is dying - not something that VMS Engineering has been told.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
David L. Cathey |INET: dav...@montagar.com
Montagar Software Concepts |UUCP: ...!montagar!davidc
P. O. Box 260772, Plano TX 75026-0772 |Fone: (214)-618-2117
Xref: gmd.de comp.org.decus:2884 comp.os.vms:42107 comp.sys.dec:14171
Path: gmd.de!Germany.EU.net!netmbx.de!zrz.TU-Berlin.DE!cs.tu-berlin.de!
math.fu-berlin.de!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!
cs.utexas.edu!uunet!meteor.syscon.hii.com!nebula!itsmrwk
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec
Subject: Re: Is VMS Dying?
Message-ID: <1994Mar4.010658.1@nebula>
From: its...@nebula.syscon.hii.com
Date: 4 Mar 94 01:06:58 -0700
Followup-To: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
<laurelCM0J1s.L59@netcom.com> <1994Mar2.145103.18732@avatar.montagar.com>
Organization: SYSCON Corporation
Nntp-Posting-Host: nebula.syscon.hii.com
Lines: 30
In article <1994Mar2.1...@avatar.montagar.com>,
dav...@avatar.montagar.com (David L. Cathey) writes:
> In article <laurelCM...@netcom.com>, lau...@netcom.com (Laurel Edgecomb) writes:
>>
>>> 7. How can DEC make VMS attractive to the nation's computer science
>>> departments?
>>
>> Give it away free along with all layered products, support it,
>> and sell hardware at a good discount.
>
> I have also posted something along those lines. DEC, are your
> listening?
>
I couldn't agree more. I learned VMS in college and actually managed a
VMS system for over a year. Due to this I had several system management
jobs to choose from. Today's students need to gain experience in as many
aspects of computers as possible to be more marketable. However, I am
confident that anyone who has the chance to learn both Unix and VMS in
college will realize how much better VMS is.
Unfortunately, in todays computer market technical capability seems to be
less important than marketing and market share. For example, when Unix
vendors started touting clustering it was suddenly all the rage. VMS has had
significantly greater cluster abilities for many years.
--
+----------------------------------------------------------------------------+
|Ronald W. Kaltenbaugh SYSCON Corporation|
|RKALTE...@SYSCON.HII.COM 1000 Thomas Jefferson St., NW|
+(202) 342-4393 Washington, DC 20007+
Xref: gmd.de comp.org.decus:2577 comp.os.vms:41443 comp.sys.dec:13825
comp.unix.advocacy:789
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Path: gmd.de!urmel.informatik.rwth-aachen.de!tornado.oche.de!rnihd.rni.sub.org!
subnet.sub.net!xlink.net!howland.reston.ans.net!pipex!uknet!EU.net!CERN.ch!
dxcern!lishka
From: lis...@dxcern.cern.ch (Chris Lishka)
Subject: Re: Is VMS Dying?
Message-ID: <1994Mar4.110926.16860@dxcern.cern.ch>
Organization: CERN European Laboratory for Particle Physics
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
<laurelCM0J1s.L59@netcom.com> <1994Mar2.145103.18732@avatar.montagar.com>
<1994Mar4.010658.1@nebula>
Date: Fri, 4 Mar 1994 11:09:26 GMT
Lines: 28
Ronald W. Kaltenbaugh writes:
> However, I am confident that anyone who has the chance to
> learn both Unix and VMS in college will realize how much
> better VMS is.
I've managed a medium-sized VMS cluster for three years now. I have
also managed Unix systems, Macintoshes, and PCs. If I had to choose
between VMS and Unix, I would choose Unix simply because I like the
design philosophy and the way one interacts with it more than I like
VMS' strengths. This is only my opinion, but it does contradict what
you proposed above.
I propose that anyone who has the chance to learn both Unix and VMS will
choose whichever O/S meets their needs and style. Both have strengths
and weaknesses. VMS is not inherently "better" than Unix, nor is Unix
inherently "better" than VMS. "Better" is a personal preference, not
an incontestable fact.
.oO Chris Oo.
--
I hear the endless murmur, Christopher Lishka
Every blade of grass that shimmers in the breeze PPE Division, CERN
And the sound that comes to carry me lis...@dxcern.cern.ch
Across the land and over the sea. -- Crowded House vxaluw::lishka
Xref: gmd.de comp.org.decus:2570 comp.os.vms:41424 comp.sys.dec:13814
comp.unix.advocacy:783
Path: gmd.de!xlink.net!howland.reston.ans.net!math.ohio-state.edu!usc!
elroy.jpl.nasa.gov!nntp-server.caltech.edu!SOL1.GPS.CALTECH.EDU!CARL
From: ca...@SOL1.GPS.CALTECH.EDU (Carl J Lydick)
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Subject: Re: Is VMS Dying?
Date: 5 Mar 1994 17:29:27 GMT
Organization: HST Wide Field/Planetary Camera
Lines: 19
Distribution: world
Message-ID: <2lafhn$46u@gap.cco.caltech.edu>
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
<laurelCM0J1s.L59@netcom.com> <1994Mar2.145103.18732@avatar.montagar.com>
<1994Mar4.010658.1@nebula>,<1994Mar4.110926.16860@dxcern.cern.ch>
Reply-To: ca...@SOL1.GPS.CALTECH.EDU
NNTP-Posting-Host: sol1.gps.caltech.edu
In article <1994Mar4.1...@dxcern.cern.ch>,
lis...@dxcern.cern.ch (Chris Lishka) writes:
=I've managed a medium-sized VMS cluster for three years now. I have
=also managed Unix systems, Macintoshes, and PCs. If I had to choose
=between VMS and Unix, I would choose Unix simply because I like the
=design philosophy and the way one interacts with it more than I like
=VMS' strengths. This is only my opinion, but it does contradict what
=you proposed above.
A singularly meaningless paragraph: "I like unix because I like unix."
Perhaps you'd be so kind as to say something with some semantic content?
Instead of uselessly vague generalities, how about a few specifics?
--------------------------------------------------------------------------------
Carl J Lydick | INTERnet: CA...@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL
Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My
understanding of astronomy is purely at the amateur level (or below). So
unless what I'm saying is directly related to VAX/VMS, don't hold me or my
organization responsible for it. If it IS related to VAX/VMS, you can try to
hold me responsible for it, but my organization had nothing to do with it.
Xref: gmd.de comp.org.decus:2609 comp.os.vms:41564 comp.sys.dec:13862
comp.unix.advocacy:816
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Path: gmd.de!newsserver.jvnc.net!netnews.upenn.edu!msuinfo!agate!
howland.reston.ans.net!pipex!sunic!EU.net!CERN.ch!dxcern!lishka
From: lis...@dxcern.cern.ch (Chris Lishka)
Subject: Re: Is VMS Dying?
Message-ID: <1994Mar6.154156.17891@dxcern.cern.ch>
Organization: CERN European Laboratory for Particle Physics
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
<1994Mar4.010658.1@nebula> <1994Mar4.110926.16860@dxcern.cern.ch>
<2lafhn$46u@gap.cco.caltech.edu>
Date: Sun, 6 Mar 1994 15:41:56 GMT
Lines: 107
I wrote:
> =I've managed a medium-sized VMS cluster for three years
> =now. I have also managed Unix systems, Macintoshes, and
> =PCs. If I had to choose between VMS and Unix, I would
> =choose Unix simply because I like the design philosophy and
> =the way one interacts with it more than I like VMS'
> =strengths. This is only my opinion, but it does contradict
> =what you proposed above.
Carl J Lydick replied:
> A singularly meaningless paragraph: "I like unix because I
> like unix." Perhaps you'd be so kind as to say something
> with some semantic content? Instead of uselessly vague
> generalities, how about a few specifics?
Actually, "I like unix because I like unix" is exactly what I meant to
imply: that one's preferred O/S is a matter of taste. This was in
contradiction to the original author's posting stating (paraphrasing
here): if a college student had learned both Unix and VMS, then they
would choose VMS as a superior O/S. If you had included the second
paragraph in my article, this would have been clear (I assumed you did
>read< the second paragraph...).
However, if you want reasons why I prefer Unix, I'll certainly provide them:
* Better interactive shells. Give me ksh or tcsh over DCL any
day. Give me command-line editting that works properly
(i.e. can backspace beyond the wrap-around boundary without
hassles). I like how ksh or tcsh work. After three years
using DCL, I still don't like it.
* Better shell programming. Give me modern day control
constructs, as I prefer them over if-then-goto. Give me
short commands strung together with pipes. Give me Unix-style
quoting mechanisms. Give me Unix-style environment variables.
I like these features over what DCL has to offer.
* A different filesystem view philosophy. Unix takes the view
that there is one filesystem which starts at root. I like
this. DCL uses a view that has seperate disks, with
directory trees starting on each disk. I don't like this.
* Program-centerred command syntax. VMS uses a mechanism to
hide the actual programs being run from the user. For
example, the "SET" and "ANALYZE" commands actually run
several different executables depending on what you choose
(i.e. "ANALYZE DISK" vs. "ANALYZE CRASH"). I prefer the
Unix system: when you run a command named "foo", you know
you will be in the environment the foo's programmer
designed. I also know that the first word in a command line
is the name of the program, and is not a "verb" which may
run different programs depending on other parameters.
Furthermore, I like the Unix concept of a command path to
search over the DCLTABLES method -- the Unix method is more
apparent and easier to search, and I don't need to resort to
a public-domain program like VERB to find out what
executables are being run when I type a command.
* Tapes. Oddly enough, although VMS has more powerful tape
handling that is better integrated into the filesystem view,
I prefer how Unix does it. With Unix I know that a tape is
treated like a tape. I know that a tape has one file
followed by another. I don't have to figure out what
Files-11 differences a tape has (no directories, the ability
to store the same file with the same version number multiple
times, etc.). In Unix a tape is a tape, it can only be
accessed from a tape device (/dev/rmt*), and there is no
integration into the filesystem. Oddly, I prefer this.
* Many other little things. Symbolic links. The very tight
integration of C with Unix which makes it a great C
programming environment. stdin, stdout, and stderr
which work more consistently than VMS' SYS$INPUT,
SYS$OUTPUT, SYS$COMMAND, and SYS$ERROR. The relative
simplicity (even though there are many arcane things) of
Unix. etc.
I have found that through the years I have come to like the
concepts and mechanisms that Unix offers. Before a flame-war starts
about my choices above, let me say that these are >MY PERSONAL
OPINIONS<, which Carl asked me to elaborate on. Yes, there are ways
to do many of the things listed above in VMS (e.g. retrofitting VMS to
include support for pipes, using a different shell or POSIX, etc.).
However, I have tried some of the alternatives and find that >I
PREFER< Unix (at least BSD Unix with System V extensions, ala Ultrix).
Don't get me wrong: VMS has many benefits over Unix, but >IN MY
OPINION< Unix has more of what I am looking for in a programming and
general environment. If you think VMS is better, that is your choice.
I like Unix better.
To tie this back to the original message which prompted me to reply:
some people who have learned both VMS and Unix may actually prefer
Unix more. Others will prefer VMS more. I see this as a matter of
taste, and refuse to believe that VMS is "better" than Unix overall
(or Unix is "better" than VMS overall).
.oO Chris Oo.
--
I hear the endless murmur, Christopher Lishka
Every blade of grass that shimmers in the breeze PPE Division, CERN
And the sound that comes to carry me lis...@dxcern.cern.ch
Across the land and over the sea. -- Crowded House vxaluw::lishka
Xref: gmd.de comp.org.decus:2655 comp.os.vms:41669 comp.sys.dec:13929
comp.unix.advocacy:866
Path: gmd.de!newsserver.jvnc.net!yale.edu!yale!gumby!wupost!cs.utexas.edu!
convex!seas.smu.edu!mic!tachyon!ondec!montagar!davidc
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Subject: Re: Is VMS Dying?
Message-ID: <1994Mar7.082703.18762@avatar.montagar.com>
From: dav...@avatar.montagar.com (David L. Cathey)
Date: 7 Mar 94 08:27:02 CDT
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
<1994Mar4.010658.1@nebula> <1994Mar6.154156.17891@dxcern.cern.ch>
Organization: Montagar Software Concepts, Plano TX
Lines: 82
In article <1994Mar6.1...@dxcern.cern.ch>,
lis...@dxcern.cern.ch (Chris Lishka) writes:
> * Program-centerred command syntax. VMS uses a mechanism to
> hide the actual programs being run from the user. ...
> Furthermore, I like the Unix concept of a command path to
> search over the DCLTABLES method -- the Unix method is more
> apparent and easier to search, and I don't need to resort to
> a public-domain program like VERB to find out what
> executables are being run when I type a command.
The disadvantage is that if you don't know the path, you can't
execute the command, either. Also, for your last point, a "CTRL-T" will
include the program name in the status line (along with additional
information). I can't tell you how many times I miss CTRL-T on an Unix system
I've been using lately...
> * Tapes. Oddly enough, although VMS has more powerful tape
> handling that is better integrated into the filesystem view,
> I prefer how Unix does it. With Unix I know that a tape is
> treated like a tape.
On VMS, a tape is treated as a tape. Tapes contains files as per the
ANSI standard. Does your Unix support ANSI tapes via all file-oriented
utilities? I like that I can use the same commands for tapes and/or disks
and not be concerned over details like how to do it. If you don't want to
use ANSI tapes, you are free to use tar or any other utility of your choice
to perform tape processing. Even the shell (SET MAGTAPE, READ, WRITE, ...).
> In Unix a tape is a tape, it can only be
> accessed from a tape device (/dev/rmt*),
Same for VMS...
> and there is no
> integration into the filesystem. Oddly, I prefer this.
>
> * Many other little things. Symbolic links.
I prefer logicals. The user can create them to point them at anything
at anytime.
> The very tight
> integration of C with Unix which makes it a great C
> programming environment.
Of course, this goes without saying...
> stdin, stdout, and stderr
> which work more consistently than VMS' SYS$INPUT,
> SYS$OUTPUT, SYS$COMMAND, and SYS$ERROR. The relative
> simplicity (even though there are many arcane things) of
> Unix. etc.
>
> I have found that through the years I have come to like the
> concepts and mechanisms that Unix offers. Before a flame-war starts
> about my choices above, let me say that these are >MY PERSONAL
> OPINIONS<, which Carl asked me to elaborate on. ...
>
> Don't get me wrong: VMS has many benefits over Unix, but >IN MY
> OPINION< Unix has more of what I am looking for in a programming and
> general environment. If you think VMS is better, that is your choice.
> I like Unix better.
And in this you are correct. As I stated before, there is no system
that is the end-all-be-all of O/S, certainly not Unix. Even DEC never pushed
VMS into every environment: ELN for real-time (XLN for AXP), now OSF/1 and WNT.
But you are also a programmer. And I've noticed that the biggest
proponents of Unix are programmers. Most users I've seen want DOS/Windows. :-(
For me, it's easier to switch between a PC and VMS and back, which if more
common for me than Unix.
> I see this as a matter of
> taste, and refuse to believe that VMS is "better" than Unix overall
> (or Unix is "better" than VMS overall).
This I definitely agree with...
> .oO Chris Oo.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
David L. Cathey |INET: dav...@montagar.com
Montagar Software Concepts |UUCP: ...!montagar!davidc
P. O. Box 260772, Plano TX 75026-0772 |Fone: (214)-618-2117
Xref: gmd.de comp.org.decus:2684 comp.os.vms:41729 comp.sys.dec:13967
comp.unix.advocacy:901
Newsgroups: comp.org.decus,comp.os.vms,comp.sys.dec,comp.unix.advocacy
Path: gmd.de!newsserver.jvnc.net!yale.edu!nigel.msen.com!caen!
saimiri.primate.wisc.edu!ames!elroy.jpl.nasa.gov!swrinde!cs.utexas.edu!
math.ohio-state.edu!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!CERN.ch!
dxcern!lishka
From: lis...@dxcern.cern.ch (Chris Lishka)
Subject: Re: Is VMS Dying?
Message-ID: <1994Mar8.130206.888@dxcern.cern.ch>
Organization: CERN European Laboratory for Particle Physics
References: <1994Feb26.204059.4068@mixcom.mixcom.com>
<1994Mar4.010658.1@nebula> <1994Mar6.154156.17891@dxcern.cern.ch>
<1994Mar7.082703.18762@avatar.montagar.com>
Date: Tue, 8 Mar 1994 13:02:06 GMT
Lines: 106
I wrote (lines beginning with >)
David L. Cathey replied (lines beginning with #)
> * Program-centerred command syntax. VMS uses a
> mechanism to hide the actual programs being run from
> the user. ... Furthermore, I like the Unix concept
> of a command path to search over the DCLTABLES
> method -- the Unix method is more apparent and
> easier to search, and I don't need to resort to a
> public-domain program like VERB to find out what
> executables are being run when I type a command.
# The disadvantage is that if you don't know the path,
# you can't execute the command, either.
Unix and VMS should come configured with a reasonable path or command
table. In both Unix and VMS, the system manager is usually
responsible for setting a decent path or adding site-specific commands
to the command table. In Unix, if you don't know the path to a
command, you cannot execute it. Similarily, in VMS if you don't know
the location of the .CLD file, you can't SET COMMAND to it (and simply
typing RUN FOO.EXE won't work if you need / parameters). I see little
difference.
# Also, for your last point, a "CTRL-T" will include the
# program name in the status line (along with additional
# information). I can't tell you how many times I miss CTRL-T
# on an Unix system I've been using lately...
Yes, ^T does include the program/image name.
Agreed: Unix does really need a ^T equivalent. I remember back in
college [insert violin music] using Unix on a Gould machine, in which
somebody had hacked ^T to run uptime.
# On VMS, a tape is treated as a tape. Tapes contains
# files as per the ANSI standard. Does your Unix support ANSI
# tapes via all file-oriented utilities?
Nope. I don't see a need to, though. A tape is a tape, and is used
differently from a disk.
# I like that I can use the same commands for tapes and/or
# disks and not be concerned over details like how to do it.
I can see where having the same commands for disks and tapes would be
preferable for some (many? most?) folks. It is a good point, but I
think both views are valid. My personal preference is to have disks
and tapes decoupled.
I have had to explain the difference in FILES-11 disks and FILES-11
tapes to users enough times now that I personally feel it is best to
treat the two as completely seperate concepts. I feel that tapes are
different enough beasts from disks that having seperate command spaces
for each is a good thing. Hence, I like Unix's simple handling of
tapes better.
# If you don't want to use ANSI tapes, you are free to use tar
# or any other utility of your choice to perform tape
# processing. Even the shell (SET MAGTAPE, READ, WRITE, ...).
Agreed.
> In Unix a tape is a tape, it can only
> be accessed from a tape device (/dev/rmt*),
# Same for VMS...
OK, I goofed on the semantics here. I meant to imply that Unix tries
to "seemlessly" integrate disks (and their partitions) into the
overall filesystem view (it doesn't always succeed, but it is good
enough for me). On the other hand, Unix presents tape drives as a
device "file", and doesn't bother trying to extend the filesystem view
to a tape. As far as Unix is concerned, there is >no< filesystem on a
tape. One has to use tape-oriented commands to determine the
structure on a tape, whereas the Unix filesystem is infused in the OS
(and in the Unix "philosophy").
# But you are also a programmer. And I've noticed that
# the biggest proponents of Unix are programmers.
Agreed. Unix presents a very nice C programming environment, with
great shell-level programming in addition.
I am a programmer, but I am also a regular user and a system manager.
My home machines are an Apple PowerBook 165c and a Commodore Amiga --
i.e. as a normal home-user I prefer the Mac or AmigaOS. For C
programming Unix wins hands-down (in my humble opinion), although
Think C on the Mac is a nice small environment. For system
management, I lean towards Unix, although VMS is very functional as
well.
# Most users I've seen want DOS/Windows. :-( For me, it's
# easier to switch between a PC and VMS and back, which if
# more common for me than Unix.
To each their own! I can see where the PC/VMS relationship is strong.
.oO Chris Oo.
--
I hear the endless murmur, Christopher Lishka
Every blade of grass that shimmers in the breeze PPE Division, CERN
And the sound that comes to carry me lis...@dxcern.cern.ch
Across the land and over the sea. -- Crowded House vxaluw::lishka
[usl-v-bsdi-ucb/include/usenet-footer.html]