Home > Archive > Cobol > August 2005 > Will Linux on the mainframe kill COBOL?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Will Linux on the mainframe kill COBOL?
|
|
| Frank Swarbrick 2005-08-11, 4:59 pm |
| The following question was posed on the search390 website
(http://search390.techtarget.com/ate...5,sid10_cid9568
05_tax293308,00.html?track=NL-16&ad=524762).
------------
Q: I would also like to know what is the future of COBOL in the Linux
environment?
This question posed on 25 July 2005
A: Non-existant, other than as a porting tool from other environments. If
you're doing a crash conversion from another system (eg z/OS), you may use
one of the Linux COBOL compilers to get things running ASAP, but it is
almost unheard-of to do additional development on COBOL applications on
Linux. C is still the preferred language (although with decent
Java-to-native code compilers coming in newer versions of GCC (gcj), Java is
seeing some additional use), and any new work is still more likely to be C
than any other language.
-------------
I just wrote the following in response:
-------------
Recently a question was posed to you "I would also like to know what is the
future of COBOL in the Linux environment?"
This question posed on 25 July 2005
You reply
"Non-existant, other than as a porting tool from other environments. If
you're doing a crash conversion from another system (eg z/OS), you may use
one of the Linux COBOL compilers to get things running ASAP, but it is
almost unheard-of to do additional development on COBOL applications on
Linux. C is still the preferred language (although with decent
Java-to-native code compilers coming in newer versions of GCC (gcj), Java is
seeing some additional use), and any new work is still more likely to be C
than any other language."
While I would agree that Linux development *currently* is mostly done in C
and Java, I don't see any reason why this *needs* to be the case.
MicroFocus offers it's Server Express COBOL application environment on many
flavors of UNIX/Linux, including Linux for zSeries. See
http://supportline.microfocus.com/p...levels/unix.asp
I personally do not yet have any experience using this product, but I intend
to look at it when time allows.
Assuming the MicroFocus product (or one by another vendor) functions as
advertised it seems to me that there is no good reason not to both port
existing COBOL business software to Linux nor is there reason not to write
*new* business software for Linux using COBOL. If more COBOL software is
written for Linux perhaps COBOL can become " " again (or for the first
time!).
--------------
Comments? Am I believe naive to believe that COBOL can be a viable
programming language for Linux applications?
Also, last w I posed a question the IBM COBOL guy (Tom Ross) if they were
planning on releasing a z/Linux COBOL compiler. No response as of yet.
Frank
---
Frank Swarbrick
Senior Developer/Analyst - Mainframe Applications
FirstBank Data Corporation - Lakewood, CO USA
| |
| William M. Klein 2005-08-11, 4:59 pm |
| FYI,
IBM has recently ACCEPTED a SHARE requirement to port a COBOL compiler to
Linux. They have also marked as "RECOGNIZED" the SHARE requirement to provide a
"full" LE environment under Linux.
What this means (or doesn't mean) for "general" use of COBOL under Linux
(especially in an IBM environment) is anyone's guess.
NOTE WELL:
Even an "ACCEPT" requirement does NOT guarantee that such a product will ever
be available - much less when. However, it is certainly MUCH more "positive"
than a REJECT <G>.
--
Bill Klein
wmklein <at> ix.netcom.com
"Frank Swarbrick" <Frank.Swarbrick@efirstbank.com> wrote in message
news:3m1hfsF14tiihU1@individual.net...
> The following question was posed on the search390 website
> (http://search390.techtarget.com/ate...5,sid10_cid9568
> 05_tax293308,00.html?track=NL-16&ad=524762).
>
> ------------
> Q: I would also like to know what is the future of COBOL in the Linux
> environment?
> This question posed on 25 July 2005
>
>
> A: Non-existant, other than as a porting tool from other environments. If
> you're doing a crash conversion from another system (eg z/OS), you may use
> one of the Linux COBOL compilers to get things running ASAP, but it is
> almost unheard-of to do additional development on COBOL applications on
> Linux. C is still the preferred language (although with decent
> Java-to-native code compilers coming in newer versions of GCC (gcj), Java is
> seeing some additional use), and any new work is still more likely to be C
> than any other language.
> -------------
>
> I just wrote the following in response:
> -------------
> Recently a question was posed to you "I would also like to know what is the
> future of COBOL in the Linux environment?"
>
> This question posed on 25 July 2005
>
> You reply
> "Non-existant, other than as a porting tool from other environments. If
> you're doing a crash conversion from another system (eg z/OS), you may use
> one of the Linux COBOL compilers to get things running ASAP, but it is
> almost unheard-of to do additional development on COBOL applications on
> Linux. C is still the preferred language (although with decent
> Java-to-native code compilers coming in newer versions of GCC (gcj), Java is
> seeing some additional use), and any new work is still more likely to be C
> than any other language."
>
> While I would agree that Linux development *currently* is mostly done in C
> and Java, I don't see any reason why this *needs* to be the case.
>
> MicroFocus offers it's Server Express COBOL application environment on many
> flavors of UNIX/Linux, including Linux for zSeries. See
> http://supportline.microfocus.com/p...levels/unix.asp
>
> I personally do not yet have any experience using this product, but I intend
> to look at it when time allows.
>
> Assuming the MicroFocus product (or one by another vendor) functions as
> advertised it seems to me that there is no good reason not to both port
> existing COBOL business software to Linux nor is there reason not to write
> *new* business software for Linux using COBOL. If more COBOL software is
> written for Linux perhaps COBOL can become "" again (or for the first
> time!).
> --------------
>
> Comments? Am I believe naive to believe that COBOL can be a viable
> programming language for Linux applications?
>
> Also, last w I posed a question the IBM COBOL guy (Tom Ross) if they were
> planning on releasing a z/Linux COBOL compiler. No response as of yet.
>
> Frank
>
>
> ---
> Frank Swarbrick
> Senior Developer/Analyst - Mainframe Applications
> FirstBank Data Corporation - Lakewood, CO USA
| |
| Peter Lacey 2005-08-11, 4:59 pm |
| Frank Swarbrick wrote:
>
> The following question was posed on the search390 website
> (http://search390.techtarget.com/ate...5,sid10_cid9568
> 05_tax293308,00.html?track=NL-16&ad=524762).
>
> ------------
> Q: I would also like to know what is the future of COBOL in the Linux
> environment?
> This question posed on 25 July 2005
>
>
The question is incorrectly posed. The operating system has nothing
whatever to do with the suitability or otherwise of language A or B.
The Linux "environment" is no different so far as a language is
concerned from any other - UNIX, Windows, mainframe O/S, or what have
you. I defy the original poster to point out an operating system which
positively makes it impossible for a given language to be ported to it.
If what s/he meant was - will Cobol be used in Linux - the answer is
yes. If it has been ported from some other system and works well -
what's the need to rewrite it in some other language? And if a
compiler's available for the Linux system and works well - it can be
used.
I won't argue there are many programmers who would turn up their noses
at the thought of using COBOL - who prefer C or Java. That has nothing
to do with the language, just the individual preferences.
PL
| |
| Richard 2005-08-11, 4:59 pm |
| > The Linux "environment" is no different so far as a language is
> concerned from any other - UNIX, Windows, mainframe O/S,
There is the very real issue that with Linux C, C++, Java, Perl,
Python, PHP, and several others are already installed and available
ready to use for free with extensive libraries while Cobol has to be
purchased and installed, and usually also requires run-time licencing.
AFAIK on mainframe O/S (eg MVS) each language has to be licenced and
paid for, often at tens of thousands of dollars per year. This means
that sites will tend to use as few languages as possible and Cobol may
be used for everything.
On Linux with many languages available, the best language (of those
available) for a particular job can be used without cost penalty -
except Cobol because that is the one that costs and isn't already
there.
| |
| Frank Swarbrick 2005-08-11, 4:59 pm |
| Encouraging!
Frank
FYI,
IBM has recently ACCEPTED a SHARE requirement to port a COBOL compiler to
Linux. They have also marked as "RECOGNIZED" the SHARE requirement to
provide a
"full" LE environment under Linux.
What this means (or doesn't mean) for "general" use of COBOL under Linux
(especially in an IBM environment) is anyone's guess.
NOTE WELL:
Even an "ACCEPT" requirement does NOT guarantee that such a product will
ever
be available - much less when. However, it is certainly MUCH more
"positive"
than a REJECT <G>.
--
Bill Klein
wmklein <at> ix.netcom.com
"Frank Swarbrick" <Frank.Swarbrick@efirstbank.com> wrote in message
news:3m1hfsF14tiihU1@individual.net...[color=darkred]
> The following question was posed on the search390 website
>
(http://search390.techtarget.com/ate...5,sid10_cid9568
> 05_tax293308,00.html?track=NL-16&ad=524762).
>
> ------------
> Q: I would also like to know what is the future of COBOL in the Linux
> environment?
> This question posed on 25 July 2005
>
>
> A: Non-existant, other than as a porting tool from other environments. If
> you're doing a crash conversion from another system (eg z/OS), you may
use
> one of the Linux COBOL compilers to get things running ASAP, but it is
> almost unheard-of to do additional development on COBOL applications on
> Linux. C is still the preferred language (although with decent
> Java-to-native code compilers coming in newer versions of GCC (gcj), Java
is
> seeing some additional use), and any new work is still more likely to be
C
> than any other language.
> -------------
>
> I just wrote the following in response:
> -------------
> Recently a question was posed to you "I would also like to know what is
the
> future of COBOL in the Linux environment?"
>
> This question posed on 25 July 2005
>
> You reply
> "Non-existant, other than as a porting tool from other environments. If
> you're doing a crash conversion from another system (eg z/OS), you may
use
> one of the Linux COBOL compilers to get things running ASAP, but it is
> almost unheard-of to do additional development on COBOL applications on
> Linux. C is still the preferred language (although with decent
> Java-to-native code compilers coming in newer versions of GCC (gcj), Java
is
> seeing some additional use), and any new work is still more likely to be
C
> than any other language."
>
> While I would agree that Linux development *currently* is mostly done in
C
> and Java, I don't see any reason why this *needs* to be the case.
>
> MicroFocus offers it's Server Express COBOL application environment on
many
> flavors of UNIX/Linux, including Linux for zSeries. See
> http://supportline.microfocus.com/p...levels/unix.asp
>
> I personally do not yet have any experience using this product, but I
intend
> to look at it when time allows.
>
> Assuming the MicroFocus product (or one by another vendor) functions as
> advertised it seems to me that there is no good reason not to both port
> existing COBOL business software to Linux nor is there reason not to
write
> *new* business software for Linux using COBOL. If more COBOL software is
> written for Linux perhaps COBOL can become "" again (or for the first
> time!).
> --------------
>
> Comments? Am I believe naive to believe that COBOL can be a viable
> programming language for Linux applications?
>
> Also, last w I posed a question the IBM COBOL guy (Tom Ross) if they
were
> planning on releasing a z/Linux COBOL compiler. No response as of yet.
>
> Frank
>
>
> ---
> Frank Swarbrick
> Senior Developer/Analyst - Mainframe Applications
> FirstBank Data Corporation - Lakewood, CO USA
| |
| Frank Swarbrick 2005-08-11, 4:59 pm |
| >
>
>Frank Swarbrick wrote:
(http://search390.techtarget.com/ate...5,sid10_cid9568
[color=darkred]
>
>
>The question is incorrectly posed. The operating system has nothing
>whatever to do with the suitability or otherwise of language A or B.
>The Linux "environment" is no different so far as a language is
>concerned from any other - UNIX, Windows, mainframe O/S, or what have
>you. I defy the original poster to point out an operating system which
>positively makes it impossible for a given language to be ported to it.
>
>If what s/he meant was - will Cobol be used in Linux - the answer is
>yes. If it has been ported from some other system and works well -
>what's the need to rewrite it in some other language? And if a
>compiler's available for the Linux system and works well - it can be
>used.
>
>I won't argue there are many programmers who would turn up their noses
>at the thought of using COBOL - who prefer C or Java. That has nothing
>to do with the language, just the individual preferences.
Totally agree. But even if the question had been phrased differently I'd
probably have taken issue with the answer. And I suspect you would as
well.
As for C and Java, Richard had a good point that there are good free C and
Java compilers for Linux, but I haven't heard that any of the free COBOL
compilers are worth much. Haven't been able to try one myself, yet, so I
could be wrong.
Frank
---
Frank Swarbrick
Senior Developer/Analyst - Mainframe Applications
FirstBank Data Corporation - Lakewood, CO USA
| |
| Peter Lacey 2005-08-11, 9:59 pm |
| Richard wrote:
>
>
> There is the very real issue that with Linux C, C++, Java, Perl,
> Python, PHP, and several others are already installed and available
> ready to use for free with extensive libraries while Cobol has to be
> purchased and installed, and usually also requires run-time licencing.
>
The question as I understood it was what effect Linux would have - not
the licensing costs. I wouldn't want to say that there isn't a free
Cobol product available.
PL
| |
| Richard 2005-08-12, 3:59 am |
| > The question as I understood it was what effect Linux would have - not
> the licensing costs.
And licensing costs may cause part of the effect. On mainframe z/OS
you write in whatever is installed and paid for: JCL and Cobol. There
may be C, or possibly Java, or maybe not. If they aren't there it may
be because no one has paid for them yet.
It would be the other way around on z/Linux. C/C++, Java and several
others will be there and paid for. If no one buys Cobol it won't be on
the list of languages to use.
> I wouldn't want to say that there isn't a free Cobol product available.
There are tinyCobol and OpenCobol which are free but somewhat limited.
Kobol is also cheap enough. However these are on Intel and the
question was z/Linux which is quite a different thing. It may be ported
to z/Linux, it may already work.
While these are Cobol it is unlikely they would work to create a CICS
service and may be too different from mainframe Cobols to take on batch
streams without a lot of work. In fact last time I looked there was no
usable file sharing with record locking which means they are pretty
much learning tools rather than production.
I would have thought that the whole point of putting in z/Linux was to
run services that connect to the databases. Application service, web
services, content management, ... This mean JBoss or Zope or Websphere.
| |
|
|
COBOL will be used on Linux for _new_ work under any of the following
conditions (there may be more and they are my opinion to which I am sure
people can object):
1) An existing COBOL app running under Linux exists that needs enhancements.
Somewhat likely but small scope.
2) When COBOL is the cheapest option. Not likely, it's now a marginalized
skill.
3) Requirements _fit_ a procedural structure and the people in the
development team are COBOL bigots. Not likely.
4) Real bona fide Legacy Transformation. This one I find _highly_
probably - a lot of people are looking to modernize, cut costs, transform
their business...and this is a legitimate option.
There is, in my opinion, a large market for LT. I see IBM seriously working
on the COBOL environment solely because of the market for bullet 4 based on
the mantra Evolution not Revolution.
A comment on the post you answered:
I don't see any reason why _new_ development would be done in C anymore in a
"corporate" environment _either_.
I still don't doubt a new OS coming down the pike anyway. The IT shift is
more than just software, it's a hardware shift. I'm not sure at the moment
who's the slave to the master in the software/hardware arena.
With regard to the "freeness" of compilers. I'm sure that a deal can be had
if you wanted to buy a large server :-)
The issue with the question is that a Linux environment is vastly scalable,
and different from one location to the next, from one distribution to the
next. Will 13 years old code in COBOL for fun on linux? I doubt it. Will
professionals code in COBOL for money on Linux, I'm sure that they will.
With regard to the 13 year old, I think it's a shame that they won't. I
think COBOL can teach you things that other languages can't. COBOL
encourages you not to be sloppy. It encourages you to think things through
linearly (which is not always bad!) and more than that gives you a grounding
in another way of thinking - more so, I believe than any other commonly used
language.
JCE
"Frank Swarbrick" <Frank.Swarbrick@efirstbank.com> wrote in message
news:3m1hfsF14tiihU1@individual.net...
> The following question was posed on the search390 website
> (http://search390.techtarget.com/ate...5,sid10_cid9568
> 05_tax293308,00.html?track=NL-16&ad=524762).
>
> ------------
> Q: I would also like to know what is the future of COBOL in the Linux
> environment?
> This question posed on 25 July 2005
>
>
> A: Non-existant, other than as a porting tool from other environments. If
> you're doing a crash conversion from another system (eg z/OS), you may use
> one of the Linux COBOL compilers to get things running ASAP, but it is
> almost unheard-of to do additional development on COBOL applications on
> Linux. C is still the preferred language (although with decent
> Java-to-native code compilers coming in newer versions of GCC (gcj), Java
> is
> seeing some additional use), and any new work is still more likely to be C
> than any other language.
> -------------
>
> I just wrote the following in response:
> -------------
> Recently a question was posed to you "I would also like to know what is
> the
> future of COBOL in the Linux environment?"
>
> This question posed on 25 July 2005
>
> You reply
> "Non-existant, other than as a porting tool from other environments. If
> you're doing a crash conversion from another system (eg z/OS), you may use
> one of the Linux COBOL compilers to get things running ASAP, but it is
> almost unheard-of to do additional development on COBOL applications on
> Linux. C is still the preferred language (although with decent
> Java-to-native code compilers coming in newer versions of GCC (gcj), Java
> is
> seeing some additional use), and any new work is still more likely to be C
> than any other language."
>
> While I would agree that Linux development *currently* is mostly done in C
> and Java, I don't see any reason why this *needs* to be the case.
>
> MicroFocus offers it's Server Express COBOL application environment on
> many
> flavors of UNIX/Linux, including Linux for zSeries. See
> http://supportline.microfocus.com/p...levels/unix.asp
>
> I personally do not yet have any experience using this product, but I
> intend
> to look at it when time allows.
>
> Assuming the MicroFocus product (or one by another vendor) functions as
> advertised it seems to me that there is no good reason not to both port
> existing COBOL business software to Linux nor is there reason not to write
> *new* business software for Linux using COBOL. If more COBOL software is
> written for Linux perhaps COBOL can become "" again (or for the first
> time!).
> --------------
>
> Comments? Am I believe naive to believe that COBOL can be a viable
> programming language for Linux applications?
>
> Also, last w I posed a question the IBM COBOL guy (Tom Ross) if they
> were
> planning on releasing a z/Linux COBOL compiler. No response as of yet.
>
> Frank
>
>
> ---
> Frank Swarbrick
> Senior Developer/Analyst - Mainframe Applications
> FirstBank Data Corporation - Lakewood, CO USA
| |
| Richard 2005-08-12, 3:59 am |
| > Will 13 years old code in COBOL for fun on linux? I doubt it.
I have code running on Linux some of which started 20 years or more
ago. Small parts are left of what started under CP/M and DRX 8 bit and
has moved via multiuser-DOS (non-MS) and Unix to Linux.
This, of course, is nothing like mainframe code being X/Open
interactive.
| |
|
| I never could understand that licensing thing with IBM. Especially with the
availability of free, open source programs. But companies are paying to
license it anyway, there must be a reason why.
> AFAIK on mainframe O/S (eg MVS) each language has to be licenced and
> paid for, often at tens of thousands of dollars per year. This means
> that sites will tend to use as few languages as possible and Cobol may
> be used for everything.
| |
| Richard 2005-08-12, 4:59 pm |
| > I never could understand that licensing thing with IBM.
> Especially with the availability of free, open source programs.
Please provide evidence of this purported 'availability of free' in
respect of a language system that would be suitable to run, say, a CICS
service on a mainframe. And also list those available in each decade
from the 1950s.
| |
|
| Peter Lacey wrote:
> Richard wrote:
>
>
> The question as I understood it was what effect Linux would have - not
> the licensing costs. I wouldn't want to say that there isn't a free
> Cobol product available.
Are you aware of one that's worth two shakes?
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
~ / \ / ~ Live from Montgomery, AL! ~
~ / \/ o ~ ~
~ / /\ - | ~ daniel@thebelowdomain ~
~ _____ / \ | ~ http://www.djs-consulting.com ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ GEEKCODE 3.12 GCS/IT d s-:+ a C++ L++ E--- W++ N++ o? K- w$ ~
~ !O M-- V PS+ PE++ Y? !PGP t+ 5? X+ R* tv b+ DI++ D+ G- e ~
~ h---- r+++ z++++ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
| |
| HeyBub 2005-08-12, 9:59 pm |
| Matt wrote:
> I never could understand that licensing thing with IBM. Especially
> with the availability of free, open source programs. But companies
> are paying to license it anyway, there must be a reason why.
>
Nobody ever got fired for using IBM, no matter the cost.
First, because the cost of the software or hardware is trivial compared to
the result. Whether a hammer costs $1.98 or $20.00 is not even noise in
considering the price of a $200,000 house.
I remember a leasing agent telling me the Blue Bell Creamery in Palestine,
Texas used IBM disk drives on their mainframe despite having half the speed
and five times the cost of plug-to-plug replacements.
Why?
Because there was an IBM repair technician in their town. Oh he spent most
of his time fixing typewriters and wouldn't know a bit from a bug, but he
had a toolkit with "IBM" printed on it, a list of telephone numbers in his
pocket, spoke IBM-ese fluently, and, most importantly, would be at their
site within 30 minutes of a call.
The nearest repairman for Telex (or other brands of disk drives) was in
Houston or Dallas, each 200 miles away.
Point is, cost is basically irrelevent.
| |
| HeyBub 2005-08-12, 9:59 pm |
| Frank Swarbrick wrote:
> The following question was posed on the search390 website
> (http://search390.techtarget.com/ate...5,sid10_cid9568
> 05_tax293308,00.html?track=NL-16&ad=524762).
>
> ------------
> Q: I would also like to know what is the future of COBOL in the Linux
> environment?
> This question posed on 25 July 2005
>
>
> A: Non-existant, other than as a porting tool from other
> environments. If you're doing a crash conversion from another system
> (eg z/OS), you may use one of the Linux COBOL compilers to get things
> running ASAP, but it is almost unheard-of to do additional
> development on COBOL applications on Linux. C is still the preferred
> language (although with decent Java-to-native code compilers coming
> in newer versions of GCC (gcj), Java is seeing some additional use),
> and any new work is still more likely to be C than any other language.
> -------------
Giggle.
| |
|
| "HeyBub" <heybubNOSPAM@gmail.com> wrote in message
news:11fqf0qjnrrf855@news.supernews.com...
> Matt wrote:
>
> Nobody ever got fired for using IBM, no matter the cost.
>
> First, because the cost of the software or hardware is trivial compared to
> the result. Whether a hammer costs $1.98 or $20.00 is not even noise in
> considering the price of a $200,000 house.
And yet, people are paid thousands of dollars to come in to ask questions
like "exactly _why_ does you organization have two $400,000 dollar licences
that you don't use?"
Software price is _not_ trivial - check out how much it would cost to
licence a full CRM package, a messaging platform, a dbms (a proper one for
real work), SAP. One should examine the costs involved in implementing some
solutions. The comparison is not a $20.00 hammer on a $200,000 house. The
comparison is sometimes whether you want carpet or marble in your house.
Some software charges $20,000 per CPU....they cound a dual core chip as
TWO...that't $40,000.
Any _serious_ app, with _serious_ support is _not_ cheap. Even Visual
Studio costs a few thousand for an enterprise copy and that _doesn't_
include SQL Server (that I am aware of).......
JCE
| |
| Peter Lacey 2005-08-12, 9:59 pm |
| Matt wrote:[color=darkred]
>
> I never could understand that licensing thing with IBM. Especially with the
> availability of free, open source programs. But companies are paying to
> license it anyway, there must be a reason why.
>
I wonder who is actualy paying for all this "free" software? Linus
Torwalds is employed by some company. Most of the people who work on
open source software must be making money somehow, somewhere. I can't
see how a company that distributes software for free is going to stay in
business unless it's making its bread somewhere else.
TANSTAAFL, you know.
PL
Iconoclast
| |
| Peter Lacey 2005-08-12, 9:59 pm |
| LX-i wrote:
>
> Peter Lacey wrote:
>
> Are you aware of one that's worth two shakes?
>
Are you aware that you're changing the subject?
PL
| |
|
| And answering a question with a question is what tactic, precisely?
JCE
"Peter Lacey" <lacey@mts.net> wrote in message
news:42FD4880.4DEDE31B@mts.net...
> LX-i wrote:
>
>
> Are you aware that you're changing the subject?
>
> PL
| |
| Peter Lacey 2005-08-12, 9:59 pm |
| jce wrote:
>
> And answering a question with a question is what tactic, precisely?
>
> JCE
Realized that just after I'd sent it. My answer should have been: I've
never had cause to look into it so I have no answer. Then "changing the
subject".
I've had a good deal of this, mind you, where I've posed a question and
people have refused to answer BY changing the subject.
Apologies but I'll stay with my point. He is changing the subject.
PL[color=darkred]
>
> "Peter Lacey" <lacey@mts.net> wrote in message
> news:42FD4880.4DEDE31B@mts.net...
| |
| Richard 2005-08-12, 9:59 pm |
| > I can't see how a company that distributes software for free is going to stay in
> business unless it's making its bread somewhere else.
Some are subsidised by profitable businesses, manly because they want
particular features incorporated. Some do it on support and by selling
books on the subject. Some also sell commercial versions of the free
software.
| |
|
| Peter Lacey wrote:
> LX-i wrote:
>
>
> Are you aware that you're changing the subject?
You said that you wouldn't want to say that there isn't a free Cobol
product available. If you wouldn't want to say that, then you're either
one who denies reality (which I doubt), or you have something in mind
that keeps you from agreeing with that statement. I'm asking, because
I'd really like to know if there *is* one, based in Linux, that's
production quality *and* free.
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
~ / \ / ~ Live from Montgomery, AL! ~
~ / \/ o ~ ~
~ / /\ - | ~ daniel@thebelowdomain ~
~ _____ / \ | ~ http://www.djs-consulting.com ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ GEEKCODE 3.12 GCS/IT d s-:+ a C++ L++ E--- W++ N++ o? K- w$ ~
~ !O M-- V PS+ PE++ Y? !PGP t+ 5? X+ R* tv b+ DI++ D+ G- e ~
~ h---- r+++ z++++ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
| |
|
| jce wrote:
> And answering a question with a question is what tactic, precisely?
I'm illustrating an alternative to the auto-responder, in an attempt to
show that rational discourse *can* continue in spite of it. Depending
on his response, we'll see how that experiment pans out. ;)
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
~ / \ / ~ Live from Montgomery, AL! ~
~ / \/ o ~ ~
~ / /\ - | ~ daniel@thebelowdomain ~
~ _____ / \ | ~ http://www.djs-consulting.com ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ GEEKCODE 3.12 GCS/IT d s-:+ a C++ L++ E--- W++ N++ o? K- w$ ~
~ !O M-- V PS+ PE++ Y? !PGP t+ 5? X+ R* tv b+ DI++ D+ G- e ~
~ h---- r+++ z++++ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
| |
|
| I'm thinking small scale. As if one could burn a copy of OS/390 and then
look for a key number online!
"Richard" <riplin@Azonic.co.nz> wrote in message
news:1123877333.198140.135900@g47g2000cwa.googlegroups.com...
>
> Please provide evidence of this purported 'availability of free' in
> respect of a language system that would be suitable to run, say, a CICS
> service on a mainframe. And also list those available in each decade
> from the 1950s.
>
| |
| Peter Lacey 2005-08-13, 3:59 am |
| LX-i wrote:
>
> Peter Lacey wrote:
>
> You said that you wouldn't want to say that there isn't a free Cobol
> product available. If you wouldn't want to say that, then you're either
> one who denies reality (which I doubt), or you have something in mind
> that keeps you from agreeing with that statement. I'm asking, because
> I'd really like to know if there *is* one, based in Linux, that's
> production quality *and* free.
>
Look up at the paragraph that starts "Python, PHP,...": it is stated
therein that Cobol has to be purchased and installed. That is an
unconditional statement. I replied "I wouldn't want to say ..." because
I don't think the statement is true. As I stated in another post, I
shouldn't have just replied to your question with another question. My
answer should have been "I don't know because I've never looked into
it. Are you aware that you're changing the subject?". Because the
original question, AS I UNDERSTOOD IT, was what effect the Linux
environment would have on the use of Cobol under Linux. The cost of a
compiler and its quality have nothing to do with the environment it's
running under. My answer concerned what I thought to be the essential
meaning of the question: I still think so. And I still think that
bringing in the cost and quality is changing the subject.
Still, there have been so many occasions when I've posted a question to
this group and had people replying by changing the subject that I get
tetchy at times. I'm not going to apologize further for that and I
wouldn't want to say that I'll never do it again.
PL
| |
| Joe Zitzelberger 2005-08-13, 3:59 am |
| In article <11fqf0qjnrrf855@news.supernews.com>,
"HeyBub" <heybubNOSPAM@gmail.com> wrote:
> Matt wrote:
>
> Nobody ever got fired for using IBM, no matter the cost.
Except those 300,000 poor fools that bet their farms on Micro$oft not
punking IBM in the NT/OS2 mess...
| |
|
| Peter Lacey wrote:
> LX-i wrote:
>
>
> Look up at the paragraph that starts "Python, PHP,...": it is stated
> therein that Cobol has to be purchased and installed. That is an
> unconditional statement. I replied "I wouldn't want to say ..." because
> I don't think the statement is true.
Okay - but there's no specific compiler you have in mind that causes you
to say that? (I promise I'm not trying to pick a debate or anything - I
really enjoy Linux, and if there is a no-cost COBOL that runs using the
gcc runtime (or whatever), that would be sweet.)
> As I stated in another post, I
> shouldn't have just replied to your question with another question.
If you think I'm concerned about that, you've got the wrong guy. :)
> And I still think that
> bringing in the cost and quality is changing the subject.
It's a logical extension of it. Most mainframe environments of which
I'm aware include a COBOL runtime environment (usually the same runtime
environment as C and ForTran - Bill's recent post regarding LE on IBM
seems to indicate this, and the Unisys IX (2200) series has the
"Universal Runtime System" (URTS)). It may not be the exact same thing,
but it's included.
> Still, there have been so many occasions when I've posted a question to
> this group and had people replying by changing the subject that I get
> tetchy at times. I'm not going to apologize further for that and I
> wouldn't want to say that I'll never do it again.
No offense taken (initially, and afterward). Thanks for the explanation
on why you thought I was changing the subject.
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
~ / \ / ~ Live from Montgomery, AL! ~
~ / \/ o ~ ~
~ / /\ - | ~ daniel@thebelowdomain ~
~ _____ / \ | ~ http://www.djs-consulting.com ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ GEEKCODE 3.12 GCS/IT d s-:+ a C++ L++ E--- W++ N++ o? K- w$ ~
~ !O M-- V PS+ PE++ Y? !PGP t+ 5? X+ R* tv b+ DI++ D+ G- e ~
~ h---- r+++ z++++ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
| |
| docdwarf@panix.com 2005-08-13, 8:59 am |
| In article <ad807$42fd653d$45491c57$26571@KNOLOGY.NET>,
LX-i <lxi0007@netscape.net> wrote:
>jce wrote:
>
>I'm illustrating an alternative to the auto-responder, in an attempt to
>show that rational discourse *can* continue in spite of it.
What '*can* continue' (emphasis original) might be a variety of things...
programs - programs that do what the users want and continue providing the
desired results in a Production environment - can be written with a
mixture of sections, paragraphs, PERFORMs, PERFORM THRUs, periods/full
stops, GO TOs and scope-delimiters; I try to do my best to make sure that
the code I write does not continue such conditions.
There! I've brought it back to COBOL... my first impulse was to be
Gravely Historical and say 'What '*can* continue' (emphasis original)
might be a variety of things... in Egypt there was a thriving, innovative
society that *did* continue for a good many years, in spite of an
hereditary monarchy which indulged in a bit of incest and chattel slavery.
I try to do my best to make sure those conditions are not continued.
DD
| |
| HeyBub 2005-08-13, 4:59 pm |
| jce wrote:
> "HeyBub" <heybubNOSPAM@gmail.com> wrote in message
> news:11fqf0qjnrrf855@news.supernews.com...
>
> And yet, people are paid thousands of dollars to come in to ask
> questions like "exactly _why_ does you organization have two $400,000
> dollar licences that you don't use?"
In some organizations, that's like asking "Why do you have two pencils in
your desk?"
>
> Software price is _not_ trivial - check out how much it would cost to
> licence a full CRM package, a messaging platform, a dbms (a proper
> one for real work), SAP. One should examine the costs involved in
> implementing some solutions. The comparison is not a $20.00 hammer
> on a $200,000 house. The comparison is sometimes whether you want
> carpet or marble in your house.
Agreed. Sometimes it's a choice between a dirt floor and marble. Inasmuch as
both "work" (i.e., you can walk on it), the choice is based on something
other than cost.
The cost of a full-bore CRM *IS* insignificant, trivial, and irrelevant,
even if, in absolute terms it runs into millions, compared to the result it
achieves. If an inventory system costs $100 million per year to run, that's
a pittance compared to $200 billion in sales for Wal-Mart.
>
> Some software charges $20,000 per CPU....they cound a dual core chip
> as TWO...that't $40,000.
And two firetrucks cost twice as much as one but put the fire out twice as
fast. Your point being? Nobody compelled Tijuana to buy Ward-LaFrance
firetrucks - the city could have organized a bucket-brigade.
>
> Any _serious_ app, with _serious_ support is _not_ cheap. Even Visual
> Studio costs a few thousand for an enterprise copy and that _doesn't_
> include SQL Server (that I am aware of).......
Nope. Is cheap. I know one organization that licenses VS, SQL Server, and
the rest. When they sell ONE copy of their software - and they sell over 100
per year - they recoup their entire license costs and then some. The cost of
their development tools represent less than one percent of their revenue.
One of their programmers actually lobbied for them switch to the knock-off
of a 40-year-old operating system originally designed by a money-losing
division of the local telephone company. They fired his heretical ass and,
at last report, he was desiging plug-in graphic flames for otherwise
ignorable web sites.
As for support, I agree that in absolute terms support may not be cheap. But
consider the alternative: If I have an airline reservation system built
entirely with, say, IBM products, and it develops serious flaws, IBM could
put 100 hardware and software technicians in my facility in 24 hours. This
repair project may cost me $2 million but compared to the lost revenue of
$50 million per day, IBM's ultimate bill garners about as much attention as
a two-martini lunch. (Well, maybe a LITTLE more attention.)
On the other hand, if one is a dilettante programmer, paying for VS out of
one's own pocket and with no hope of selling the result, $10,000 *IS* a big
deal.
Around my company, we have no hesitation buying $20.00 hammers.
| |
|
| "HeyBub" <heybubNOSPAM@gmail.com> wrote in message
news:11fsb06p4qq9d42@news.supernews.com...
> jce wrote:
>
> In some organizations, that's like asking "Why do you have two pencils in
> your desk?"
If you have $400,000 dollar pencils, yes.
>
> Agreed. Sometimes it's a choice between a dirt floor and marble. Inasmuch
> as both "work" (i.e., you can walk on it), the choice is based on
> something other than cost.
I agree that the choice is often based on something other than cost. Is it
scalable, how many people do I need to maintain to support this...can a
trained monkey do it? However, to suggest that companies don't look at
open source solutions is naive at best. The rationale may be that there is
more open knowledge base (and hence more skilled workers = cheaper labour)
but the fact is that some companies don't want to pay for anything. Most IT
centers within corporations are COST centers. Todays profits are more
cutting costs than increasing revenue. IT centers that are the corporation
are REVENUE centers (in an ideal world) and the rules change significantly.
A lot of Linux adoption is outside of the "regular report chain" on machines
that have become lost (not allowed anymore in the US in public companies, of
course). Web Logic, WebSphere, and the like are taking great hits to the
likes of JBoss. To say price isn't a factor in these examples is just nuts.
Perhaps the better analogy is that the Building company doesn't care about
$20.00 hammers....but the hired help *do* care about their $20.00 hammers.
I once had a workman call me up to see if I still had the card board paint
brush cover at my house so he could use it to protect his $20 brush....
> The cost of a full-bore CRM *IS* insignificant, trivial, and irrelevant,
> even if, in absolute terms it runs into millions, compared to the result
> it achieves. If an inventory system costs $100 million per year to run,
> that's a pittance compared to $200 billion in sales for Wal-Mart.
Tell that to the man who is in the position of analyzing the costs of the
cost center. I don't think Wal-Mart gives him a blank cheque. I understand
cost is relative....but cost is cost and not one company (even Wal-Mart)
likes cost.
> And two firetrucks cost twice as much as one but put the fire out twice as
> fast. Your point being? Nobody compelled Tijuana to buy Ward-LaFrance
> firetrucks - the city could have organized a bucket-brigade.
Point is that a dual core chip is _not_ twice as fast as two yet the
software is now twice as expensive. The fact that this occurs and the fact
that this is not _cheap_ means that someone somewhere is worried about the
cost of it. I don't equate this cost to a hammer.
>
> Nope. Is cheap. I know one organization that licenses VS, SQL Server, and
> the rest. When they sell ONE copy of their software - and they sell over
> 100 per year - they recoup their entire license costs and then some. The
> cost of their development tools represent less than one percent of their
> revenue.
Yes, and one full time employee represents less that one percent of just
about most companies revenues - yet they still make layoffs. Why? Because
it cuts not _insignificant_ costs.
> One of their programmers actually lobbied for them switch to the knock-off
> of a 40-year-old operating system originally designed by a money-losing
> division of the local telephone company. They fired his heretical ass and,
> at last report, he was desiging plug-in graphic flames for otherwise
> ignorable web sites.
This would almost be grounds for unfair dismissal. I say almost because in
the US, I think that's as close as you can get.
> As for support, I agree that in absolute terms support may not be cheap.
> But consider the alternative: If I have an airline reservation system
> built entirely with, say, IBM products, and it develops serious flaws, IBM
> could put 100 hardware and software technicians in my facility in 24
> hours. This repair project may cost me $2 million but compared to the lost
> revenue of $50 million per day, IBM's ultimate bill garners about as much
> attention as a two-martini lunch. (Well, maybe a LITTLE more attention.)
Knowing customers, including some IBM customers, I know for a *fact* that
this isn't true.
> On the other hand, if one is a dilettante programmer, paying for VS out of
> one's own pocket and with no hope of selling the result, $10,000 *IS* a
> big deal.
I think that this is more the market we were discussing here. I'd love to
develop for z/OS for a living...unfortunately I cannot afford the
development environment costs (though there are partner programs that
release laptop versions). I know I cannot really justify spending the money
on a Microsoft subscription....so I do without.
> Around my company, we have no hesitation buying $20.00 hammers.
I wish I worked for your company :-)
I don't disagree with you completely. I think you make many valid and
salient arguments, but I don't think it's the complete picture...and not all
of us have Microsoft stock ;)
JCE
| |
| HeyBub 2005-08-14, 3:59 am |
| jce wrote:
>
> Yes, and one full time employee represents less that one percent of
> just about most companies revenues - yet they still make layoffs. Why?
> Because it cuts not _insignificant_ costs.
Trust me on this. No Fortune 500 company lays off one employee as a
cost-cutting measure.
>
[color=darkred]
> This would almost be grounds for unfair dismissal. I say almost
> because in the US, I think that's as close as you can get.
He who has the gold makes the rules (except in California which is more like
Amsterdam writ large).
Years ago Peter Lawford and his wife tried to buy an apartment in New York.
The residence committee overseeing sales refused his offer on the basis that
"He is an actor and his wife is a Democrat."
I've been fired for offering to punch out my boss. I've also fired workers
for having the "wrong" political outlook, personal grooming not up to my
arbitrary standards, complaining about my smoking, and having a teen-ager
that got arrested just about once a w. I've refused to hire people who
were under indictment for burglary, had bones in their nose or some other
obvious mutilations, had monikers like "Latashatoyo" or females with
hyphenated last-names.
Unfair? Probably. Business ain't bean-bag.
Speaking of "unfair," did you know in the U.S. you get one free "grope?"
Yes. A claim for sexual harrassment has to be based on repetitive words or
actions AND the offending activity must be complained of. Give them hooters
a squeeze and see what happens. Worst case, you have to fire her for chronic
mopery.
There are underground personnel manuals detailing this and hundreds of other
neat little things one can do.
>
[color=darkred]
> Knowing customers, including some IBM customers, I know for a *fact*
> that this isn't true.
Having done the equivalent on a smaller scale, I know for a fact that it IS
true. I had a 2938 Array Processor punctured by a air freight forklift. This
was a new, rare device and IBM flew $600,000 worth of spare parts, two
hardware engineers, and one of the original designers into our shop to fix
the sucker. On another occasion, IBM bought a first-class airplane ticket
for a power supply to get it from Philadelphia to Houston in the most
expeditious fashion. And this was on Christmas Eve.
>
> I think that this is more the market we were discussing here. I'd
> love to develop for z/OS for a living...unfortunately I cannot afford
> the development environment costs (though there are partner programs
> that release laptop versions). I know I cannot really justify
> spending the money on a Microsoft subscription....so I do without.
Sell blood. Wash cars. Make the sacrifice. What's art without suffering.
>
[color=darkred]
> I wish I worked for your company :-)
You'd probably mention Linux without spitting and I'd fire you within a
w.
>
> I don't disagree with you completely. I think you make many valid and
> salient arguments, but I don't think it's the complete picture...and
> not all of us have Microsoft stock ;)
Ah, another difference. You can actually BUY Micros~1 stock.
| |
|
| "HeyBub" <heybubNOSPAM@gmail.com> wrote in message
news:11ftird5ierio12@news.supernews.com...
> jce wrote:
> Trust me on this. No Fortune 500 company lays off one employee as a
> cost-cutting measure.
Working for one, I know they do. They just do it in lots of places.
> He who has the gold makes the rules (except in California which is more
> like Amsterdam writ large).
> Years ago Peter Lawford and his wife tried to buy an apartment in New
> York. The residence committee overseeing sales refused his offer on the
> basis that "He is an actor and his wife is a Democrat."
> I've been fired for offering to punch out my boss. I've also fired workers
> for having the "wrong" political outlook, personal grooming not up to my
> arbitrary standards, complaining about my smoking, and having a teen-ager
> that got arrested just about once a w. I've refused to hire people who
> were under indictment for burglary, had bones in their nose or some other
> obvious mutilations, had monikers like "Latashatoyo" or females with
> hyphenated last-names.
And I'm sure that if I looked there are people who complained that they once
had to fire an XXXXXXX, and many who had worked for an XXXXXXX too.
> Unfair? Probably. Business ain't bean-bag.
> Speaking of "unfair," did you know in the U.S. you get one free "grope?"
> Yes. A claim for sexual harrassment has to be based on repetitive words or
> actions AND the offending activity must be complained of. Give them
> hooters a squeeze and see what happens. Worst case, you have to fire her
> for chronic mopery.
> There are underground personnel manuals detailing this and hundreds of
> other neat little things one can do.
I did a google for "how to be an XXXXXXX" I came up with too many links.
>
>
> Having done the equivalent on a smaller scale, I know for a fact that it
> IS true. I had a 2938 Array Processor punctured by a air freight forklift.
> This was a new, rare device and IBM flew $600,000 worth of spare parts,
> two hardware engineers, and one of the original designers into our shop to
> fix the sucker. On another occasion, IBM bought a first-class airplane
> ticket for a power supply to get it from Philadelphia to Houston in the
> most expeditious fashion. And this was on Christmas Eve.
I didn't question IBM doing it. I questioned those paying the bills. I'm
sure your companies invoice cleared no questions asked the next month.
You'd be the exception I think.
>
> Sell blood. Wash cars. Make the sacrifice. What's art without suffering.
>
I take it all back....
[color=darkred]
> You'd probably mention Linux without spitting and I'd fire you within a
> w.
>
>
> Ah, another difference. You can actually BUY Micros~1 stock.
Biggest difference is that in real life I'm actually a nice guy.
JCE
| |
| Howard Brazee 2005-08-15, 8:59 am |
|
On 13-Aug-2005, "HeyBub" <heybubNOSPAM@gmail.com> wrote:
> Speaking of "unfair," did you know in the U.S. you get one free "grope?"
> Yes. A claim for sexual harrassment has to be based on repetitive words or
> actions AND the offending activity must be complained of. Give them hooters
> a squeeze and see what happens. Worst case, you have to fire her for chronic
> mopery.
Look at the news and see what celebrities get in trouble for. Sports
celebrities can do whatever they want and get chances to redeem themselves.
But if they *say* the wrong thing, they're toast.
I'm not sure that business is that much different.
| |
| HeyBub 2005-08-15, 4:59 pm |
| Howard Brazee wrote:
> On 13-Aug-2005, "HeyBub" <heybubNOSPAM@gmail.com> wrote:
>
>
> Look at the news and see what celebrities get in trouble for. Sports
> celebrities can do whatever they want and get chances to redeem
> themselves. But if they *say* the wrong thing, they're toast.
>
> I'm not sure that business is that much different.
You mean, like: "I know I wasn't supposed to squeeze the Charmin's, but I
couldn't help it. They just kept saying 'Squeeze me! Squeeze me!' "
|
|
|
|
|