Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

It's Not About the Language

9 views
Skip to first unread message

Frank Swarbrick

unread,
Feb 1, 2012, 6:03:21 PM2/1/12
to

Pete Dashwood

unread,
Feb 1, 2012, 10:31:41 PM2/1/12
to
Frank Swarbrick wrote:
> http://www.gonzobanker.com/2012/01/cobol-not-about-language/

I agree with his conclusion: It IS about Functionality.

After that we diverge to opposite poles. I posted a couple of responses but
it is a moderated group so they may censor anything which doesn't toe the
party line... :-)

Pete

--
"I used to write COBOL...now I can do anything."


iNFO_rene

unread,
Feb 2, 2012, 1:45:26 AM2/2/12
to
It is a very nice article.

Arnold Trembley

unread,
Feb 2, 2012, 3:42:41 AM2/2/12
to
On 2/1/2012 5:03 PM, Frank Swarbrick wrote:
> http://www.gonzobanker.com/2012/01/cobol-not-about-language/

My favorite quote from the reader comments:

"If truly 80ish% of the business world runs on COBOL, where are all the
programmers? I’ve never personally met one. How is all this code
maintained?"

A few years ago someone asked me what I did for a living. When I told
him I was a COBOL programmer, he was surprised. He thought everybody
had stopped using COBOL around 1990.

--
http://www.arnoldtrembley.com/

Alistair Maclean

unread,
Feb 2, 2012, 1:42:19 PM2/2/12
to
hat did he think the millennium fiasco was about?

Robert Doerfler

unread,
Feb 3, 2012, 3:51:03 AM2/3/12
to
I just meet one person around the age of 65 who was a cobol/pl1 programmer
and he told me to go the cobol way of programming as cobol developers are
always needed. But in the past years i've just seen just some very rare job
offers for cobol programmers around in my area. So where are all the
cobol programmers? Are they hidden in some basement room? :o)

Pete Dashwood

unread,
Feb 3, 2012, 5:56:30 AM2/3/12
to
It's an interesting question.

In its heyday there were COBOL programmers everywhere. Most of your
colleagues were COBOL guys, you went to User Groups and met other COBOL guys
from other sites, and it was quite usual to go for drinks with guys who
lived COBOL. There were Bulletin Boards dedicated to COBOL back in the days
of 2400 bps modems (before the Internet, but after PCs became available). By
the end of the 80s though the signs were starting to show. There were fewer
jobs being advertised (although nothing to get anxious about) and people
were talking about new fangled stuff like Objects and Classes that didn't
make much sense to COBOL guys. Languages like Java were dismissed as rubbish
because they weren't self-documenting and English-like... Throughout the 90s
COBOL continued to lose ground receiving some CPR for the Y2K problem, but
declining again after that.

So, what DID happen to all the COBOL guys?

Hard to say. I think some of them saw the writing on the wall and moved to
other things. Quite a few were "kicked upstairs" and became Team Leaders,
Business Analysts, and line managers, and some simply decided that they were
born to program and so they expanded their skill sets and avoided
redundancy.

As was noted in some of the comments on the OP's link, there are very few
COBOL programmers around who are under 30. (There was ONE guy who responded
who was 27 but he only knows COBOL, so his experience is limited.)

The ones who frequent this forum are largely the die-hards who are moving to
retirement (or have already) and they just love to think the Glory Days are
still with us. I see no harm in that and participate in it myself
occasionally. After 40 odd years I have as much "Glory" as anybody else :-)

There was a time here when we dealt with problems in COBOL and discussed
them, supporting Newbies and encouraging them to do their own homework...
:-) That hasn't happened for a very long time and the kind of questions we
get now are much more specialised and usually involve other things besides
COBOL. The decline in questions (and Newbies) is just one indication of the
decline in COBOL. If the language declines, then the number of people using
it will decline and that may account for the fact that COBOL guys are thin
on the ground.

The determined promotion of COBOL vendors helped to prolong the life of it
but even they are starting to realise it isn't going anywhere. IBM are not
investing large sums in COBOL to provide modern facilities for it, Micro
Focus have nailed their colours to the mast of Visual COBOL, which, although
it is an excellent product, I believe is doomed in the long run (say 3 or 4
years), and Fujitsu have already divested themselves of their COBOL product,
including their excellent NetCOBOL for .NET, which is doomed to share the
fate of MF's Visual COBOL.

(It really is sad to see two such excellent products that simply missed the
boat. As people using these products become more familiar and comfortable
with Objects and Layers, and start to realise that many of the precepts that
apply traditionally to COBOL are not relevant to Object COBOL or .Net or
Mono; when the penny drops that OBJECT code (encapsulated functionality) is
infinitely more important than SOURCE code, then they will start to wonder
why they are writing and maintaining thousands of lines of COBOL when they
could be using other languages that don't require that, and are provided
free. Pretty soon they move away from COBOL. (Whether you agree with this
reasoning or not, the fact is that both Micro Focus and IBM are looking to
diversify to other products than COBOL and Fujitsu has already left the
field.)

It is the stupid and misleading propaganda spread about by vested interest
companies like Gartner that "80% of the world's computing runs on COBOL"
that then causes people to ask where the programmers are. It's a very fair
question.

Another major factor that comes into this is that today's programmers are
generally better trained than we were and they have a grounding in Computer
Science (Principles and Practice of Programming, across paradigms, which we
never really had handed to us) These guys are capable of picking up COBOL
legacy and dealing with it until it can be replaced. They don't consider
themselves "COBOL Programmers" but they do consider themselves "computer
programmers" who are perfectly capable of dealing with most things. Their
Toolboxes are much fuller than ours were because the world they live in is
no place for "one trick ponies".

(I am currently working with one such person (there were two, but one of
them "completed" and got what she came for) who is doing Computer Science at
one of the leading NZ universities, and not being formally taught COBOL as
part of it. They both wanted to get some COBOL knowledge purely on their own
behalf, to improve their skill if they should be asked to deal with Legacy,
and one of them actually plans to make a living in the same niche market
where PRIMA is working: Moving legacy COBOL on and salvaging what it makes
sense to salvage.)

There are some time warp installations (particularly in the USA,
apparently...) where they still maintain fossilized COBOL core systems and
use the Waterfall to develop, just as they did in the last century. And
there are still some "one trick pony" COBOL codgers who ply their trade just
as they have always done. There's nothing wrong with that, but it is not a
place for young people who are trying to make a career. And it DOESN'T mean
that COBOL is alive and well and will be around for the next 50 years...

There are few people today who can claim to be a "COBOL programmer" and
fewer yet who would want to. (Putting COBOL experience on a CV these days
can actually lower your chances.) But for those of us, the increasngly
diminishing band who were there when commerce started using computers on a
grand scale, who learned the stuff by the seat of our pants and were driven
by our own passion for programming and curiosity, and then saw what we
learned become largely obsolete as the pace of technology increased
dramatically, we have nothing to be ashamed of.

I'm proud to be a "COBOL programmer" even though I have acquired many other
skills. I'll always have an affinity for COBOL; the simple beauty of it
reminds me of a simpler time when computing wasn't "Awesome", it was just
bread and butter...

Now, the notebook on my lap has many times the power of the machines I wrote
(non-Object) COBOL on. I string together components like Lego blocks and
build structures of grace and beauty with user interfaces that could only be
dreamed of in COBOL. I have learned new programming tricks, facilities and
techniques (functional Lambdas, LINQ, Reflection, Delegation, Regular
Expressions, to name just some...) that were never a part of my COBOL
world. My productivity has soared; my maintenance time has dived and I am a
happy developer... but I still remember when we did it line by line. :-)

"For the sin that ye do
by two and by two
ye must pay for
one by one."

(Kipling, "Tomlinson".)

Pete.

Bill Gunshannon

unread,
Feb 3, 2012, 8:03:58 AM2/3/12
to
In article <slrnjin7v...@seeteufel.bloerp.de>,
While the lack of IBM/CICS experience has kept me from getting back into
COBOL fulltime (that is where most of the work seems to be, and I can
easily find dozens if not hundreds of jobs) I am happy to say that as
of the end of this month I will be goign back to at least partial COBOL
programming for a living. The job required COBOL on IBM, but not on
a mainframe. I will be doing COBOL and AIX. A language and an environment
that meets my skillset perfectly.

If you really want a COBOL job, they are out there. I suspect the biggest
problem finding the jobs is that the places hiring know the market for
programmers is thin and tend to limit their searches to places where they
actually expect to find candidates (that means no Monster, Dice or local
newspapers!) or the are doing things like General Dynamics and setting
up internships in order to create their own.

Blame academia!! Even after repeatedly showing that the demand was there
and providing pointers to people who might be glad to have formal intern-
ship agreements (these can be used as a large selling point for an academic
program) I was unable to get people I was associated with to even consider
the possibility. Trying to drive the bus rather than meeting the needs
of the industry is going to come back to bite academia. Signs of this
are already showing up in the press.

bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

SkippyPB

unread,
Feb 3, 2012, 10:06:09 AM2/3/12
to
On 3 Feb 2012 13:03:58 GMT, bill...@cs.uofs.edu (Bill Gunshannon)
wrote:
My employer has nearly 30,000 employees. At least 25% of them are
COBOL programmers. Most of us have been with the company (in one of
its many forms) for 25 years or more. I suspect that at any company
that has a large programming staff the same statistics would hold.

What that means is that as us baby boomers start to retire, there will
be more openings in COBOL. My employer, and I suspect others as well,
are starting to panic. They look at how many COBOL programmers they
are going to lose over the next 5-10 years, look at what they have as
backfill and realize they are going to be woefully short. Worse, the
pool of COBOL programmers coming out of colleges is very small as
well. It is almost the 70's all over again where demand outstrips
supply.

My employer, and others I suspect, have started to recruit college
grads from other disciplines and train them in COBOL programming and
then partner them up with some of us old timers so they can learn the
language and the applications. It is a re-startup of a similar
program we had 20 years ago.

In short, the jobs in COBOL are out there and more will be coming. You
may have to relocate to land one, but the need is still there and will
be for some time.

Regards,
--

////
(o o)
-oOO--(_)--OOo-

"If the shoe fits, get another one just like it."
--Unknown
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Remove nospam to email me.

Steve

Howard Brazee

unread,
Feb 3, 2012, 12:03:46 PM2/3/12
to
On 3 Feb 2012 08:51:03 GMT, Robert Doerfler <ro...@zlug.org> wrote:

>I just meet one person around the age of 65 who was a cobol/pl1 programmer
>and he told me to go the cobol way of programming as cobol developers are
>always needed. But in the past years i've just seen just some very rare job
>offers for cobol programmers around in my area. So where are all the
>cobol programmers? Are they hidden in some basement room? :o)

My workplace told us to retire.

--
"In no part of the constitution is more wisdom to be found,
than in the clause which confides the question of war or peace
to the legislature, and not to the executive department."

- James Madison

Pete Dashwood

unread,
Feb 3, 2012, 6:34:56 PM2/3/12
to
Bill Gunshannon wrote:
> In article <slrnjin7v...@seeteufel.bloerp.de>,
> Robert Doerfler <ro...@zlug.org> writes:
>> On 2012-02-02, Arnold Trembley <arnold....@att.net> wrote:
>>> On 2/1/2012 5:03 PM, Frank Swarbrick wrote:
>>>> http://www.gonzobanker.com/2012/01/cobol-not-about-language/
>>>
>>> My favorite quote from the reader comments:
>>>
>>> "If truly 80ish% of the business world runs on COBOL, where are all
>>> the programmers? Iā?Tve never personally met one. How is all this
I remember having exactly the same discussions with representatives of
"Acadaemia" back in the 1980s. We knew then that the people being produced
from Universities and Technical Institutes were not much use in the real
world and had to be trained into the world of Commerce.

Fortunately, most of them were bright young people who were capabale of
learning quickly, but industry generally felt it was being let down by
Acadaemia.

Nowadays, people who arrive with a Computer Science degree are generally
much better equipped that the people who arrived then and just because they
don't know COBOL does not mean that they are of no use.

I have enjoyed working with the "new breed" of IT graduate. They are more
open minded, less evangelical, and much more capable than generally used to
be the case.

I believe what is being churned out from Acadaemia now, is a far better and
more useful product, than what was being churned out through the 60s, 70s,
and 80s.

Pete Dashwood

unread,
Feb 3, 2012, 6:44:06 PM2/3/12
to
SkippyPB wrote:
> On 3 Feb 2012 13:03:58 GMT, bill...@cs.uofs.edu (Bill Gunshannon)
> wrote:
>
>> In article <slrnjin7v...@seeteufel.bloerp.de>,
>> Robert Doerfler <ro...@zlug.org> writes:
>>> On 2012-02-02, Arnold Trembley <arnold....@att.net> wrote:
>>>> On 2/1/2012 5:03 PM, Frank Swarbrick wrote:
>>>>> http://www.gonzobanker.com/2012/01/cobol-not-about-language/
>>>>
>>>> My favorite quote from the reader comments:
>>>>
>>>> "If truly 80ish% of the business world runs on COBOL, where are
>>>> all the programmers? Iā?Tve never personally met one. How is all
So, the company has over 7000 people invested in COBOL and NOW they start to
worry?

They have two options:

1. Get more people. (Whatever it takes; training if necessary...)
2. Reduce dependency on COBOL.

They are going for option 1, but it would be foolish in the extreme not to
also use option 2. (Watch what happens as new managers come in...)

In fact, just focussing on EITHER of these options is not a good strategy,
for ANY COBOL dependent company.

Eventually, option 2 supersedes option 1, but I agree that it won't happen
this week... :-)

Lüko Willms

unread,
Feb 3, 2012, 6:48:02 PM2/3/12
to
Am 03.02.2012 09:51, schrieb Robert Doerfler:
> So where are all the
> cobol programmers? Are they hidden in some basement room?

they are on retirement or early retirement.

Sure, because there are still quite some legacy applications built in
COBOL out there, and the breed of COBOL programmers is dying out, there
might be a chance for old COBOL hacks like me to make some fast bucks.
I'll attend a Micro Focus roadshow here in Frankfurt on February 12;
maybe I can learn something, make some contacts, maybe even meet
somebody from the participants in this forum?

But nobody should think of building a life-long career on COBOL and
nothing but COBOL.


Cheers,
L.W.

Howard Brazee

unread,
Feb 3, 2012, 8:41:58 PM2/3/12
to
On Sat, 04 Feb 2012 00:48:02 +0100, Lüko Willms
<lueko....@domain.invalid> wrote:

> they are on retirement or early retirement.
>
> Sure, because there are still quite some legacy applications built in
>COBOL out there, and the breed of COBOL programmers is dying out, there
>might be a chance for old COBOL hacks like me to make some fast bucks.
>I'll attend a Micro Focus roadshow here in Frankfurt on February 12;
>maybe I can learn something, make some contacts, maybe even meet
>somebody from the participants in this forum?
>
> But nobody should think of building a life-long career on COBOL and
>nothing but COBOL.

As one of those retired CoBOL programmers - we were predominately
CoBOL, but never "nothing but CoBOL".

iNFO_rene

unread,
Feb 3, 2012, 10:14:30 PM2/3/12
to
>   But nobody should think of building a life-long career on COBOL and
> nothing but COBOL.
>
> As one of those retired CoBOL programmers - we were predominately
> CoBOL, but never "nothing but CoBOL".
>

Very well mentioned. It's all mixed-up PLs right now.

Experienced Cobol programmers are up to their late 40s, and are
thinking of "early" retirement. Nothing fancy here, but most of these
guys are rich (working with Cobol) and can afford to retire (in the
Philippines). Unexpectedly, to my surprise... some companies are
training their "new" programmers Cobol! Blame academia?!? Maybe, they
opted to teach students opensource (free) coding because it is costly
teaching them Cobol (compilers fee is soaring).

Cobol functionality is a fact. Thats why we are Cobol programmers
ourselves... but most of us unfortunately hate Cobol (I knew Pete is
one) because of the compiler cost. For me, mix the Cobol up with
opensource technologies and you will get more functionalities. Cost
effectiveness of PLs (especially Cobol) does not depend on the IDEs
itself but on business operations upfront.

SkippyPB

unread,
Feb 4, 2012, 11:13:20 AM2/4/12
to
It really isn't like they JUST started to panic. The senior
management, like that of many other software companies, bought into
the soup du jour language myth and thought COBOL would just fade away.
Instead, they are finding that our clients, while supplementing their
"big iron" applications with server based technologies, are not
willing to give up the superior file handling and speed that is
critical in their businesses that mainframe gives them. In addition,
the regulatory changes that come down the pike every month or so are
far easier to implement in our COBOL code than in the JAVA or whatever
else we happen to be running on the server side of the house. Hence,
we continue to develop and maintain legacy and new COBOL code. We do
offer server based systems and support them as well but the number of
clients is small and so is the revenue. I'm not sure what language
they use.

Option 2 is no option at all in our world and won't be a very long

Pete Dashwood

unread,
Feb 4, 2012, 5:29:55 PM2/4/12
to
Perception.

> Instead, they are finding that our clients, while supplementing their
> "big iron" applications with server based technologies, are not
> willing to give up the superior file handling and speed that is
> critical in their businesses that mainframe gives them.

Perception. While you may believe it, there is no basis in reality for
thinking that mainframes are faster or more reliable than server networks.
(They were for a while, but technology evolves.) The Internet evolved from a
network that was invented to provide reliability in the event of nuclear
attack... But, at the end of the day it comes down to perception. If you and
your clients believe it, then it is true for you.

> In addition,
> the regulatory changes that come down the pike every month or so are
> far easier to implement in our COBOL code than in the JAVA or whatever
> else we happen to be running on the server side of the house.

Perception. Ask the Java guys if it is easier in COBOL... :-)


> Hence,
> we continue to develop and maintain legacy and new COBOL code. We do
> offer server based systems and support them as well but the number of
> clients is small and so is the revenue. I'm not sure what language
> they use.
>
> Option 2 is no option at all in our world and won't be a very long
> time.

We'll see... :-)

(Let's dicuss this again in 5 years...)

My point here, Steve, is that the premises for the discussion are based on
perception, and not on independent fact. It doesn't matter, most people buy
stuff based on their perception of it and there is a huge industry based in
Madison Avenue which is designed to change our perception of products and
services, in an effort to make us buy them.

At the moment in your company you have managers who have been there a long
time, you have a solid core of COBOL people (whose perception is often
limited to "the COBOL way" (and while that may be fine in some cases it is
not always the BEST way...))

But,as I'm sure you are aware, the world and companies do not stand still.
New people come in and their experience is different. They perceive things
differently and are not necessarily satisfied with the status quo. They ask
questions, they know about other ways to do things, and they eventually
become managers. The same thing happens at the Clients. Change is actually
inevitable UNLESS you can show independently and unemotionally, that the
current service is the best available for the price. You use COBOL because
it works for you and provides the services you sell, at a good price, not
because you have 7000 COBOL programmers. :-) It may well be, that in your
company, COBOL is a good model. Perhaps the underlying services don't change
a lot or amendments are trivial in the scheme of things. Everything ticks
over nicely just as it has for many years and there is a core of code that
the business is based on and which nobody really wants to meddle with. There
is therefore no push to change things.. If that is the case then it may
indeed be some time before change is implemented.

But, eventually, it will be.

Lüko Willms

unread,
Feb 5, 2012, 6:27:43 AM2/5/12
to
Am 04.02.2012 23:29, schrieb Pete Dashwood:
> The Internet evolved from a
> network that was invented to provide reliability in the event of nuclear
> attack...

Myth.


Cheers,
L.W.

SkippyPB

unread,
Feb 5, 2012, 11:58:48 AM2/5/12
to
On Sun, 5 Feb 2012 11:29:55 +1300, "Pete Dashwood"
Oh there are plenty of facts to support the mainframe still be way
ahead of a server network in terms of speed, file handling and I'll
even throw in security. When was the last time a mainframe was hacked
and had customer/client data stolen?

And show me a server network that can support a full mortgage loan
apllication and accept input 24/7 from around the world and process
that in 3 seconds or less. And that's just a couple of examples. It
is far more than perception..it is reality.


>> In addition,
>> the regulatory changes that come down the pike every month or so are
>> far easier to implement in our COBOL code than in the JAVA or whatever
>> else we happen to be running on the server side of the house.
>
>Perception. Ask the Java guys if it is easier in COBOL... :-)
>

Only because the JAVA guys aren't bright enough to code in COBOL. :)
It's like asking a native French speaker to write a novel in English
and then again in French. Which is he going to be more comfortable
doing and finding easier?

>
>> Hence,
>> we continue to develop and maintain legacy and new COBOL code. We do
>> offer server based systems and support them as well but the number of
>> clients is small and so is the revenue. I'm not sure what language
>> they use.
>>
>> Option 2 is no option at all in our world and won't be a very long
>> time.
>
>We'll see... :-)
>
>(Let's dicuss this again in 5 years...)
>

Crap I hope I'm retired and lving out my days on a beach in Koh Samui
in 5 years instead of worrying if COBOL is still around or not!


>My point here, Steve, is that the premises for the discussion are based on
>perception, and not on independent fact. It doesn't matter, most people buy
>stuff based on their perception of it and there is a huge industry based in
>Madison Avenue which is designed to change our perception of products and
>services, in an effort to make us buy them.
>

You might be right if we were talking about after shave. But I'm
talking about finance and banking and other large business
applications and while server technology was being pushed by your
Madison Avenue types as being a better, cheaper way to do big
business, it never lived up to the hype. The nice for server networks
in the environment I work in is as a front end system. It does a
great job with GUI and paints pretty screens and makes it all user
friendly but the real work is done behind the scenes on the mainframe.


>At the moment in your company you have managers who have been there a long
>time, you have a solid core of COBOL people (whose perception is often
>limited to "the COBOL way" (and while that may be fine in some cases it is
>not always the BEST way...))
>

The average age of our Senior managers is around 45. They are for the
most part younger than me. They were pushing the server network
answer but soon learned it had too many limitations. We do have
complete applications, like credit risk, some card (credit, ATM etc)
management and a few other things that are well suited and run just
fine on the server networks. We even have banking applications
written to run on the small Big Iron boxes and are run by community
bank type industries.

>But,as I'm sure you are aware, the world and companies do not stand still.
>New people come in and their experience is different. They perceive things
>differently and are not necessarily satisfied with the status quo. They ask
>questions, they know about other ways to do things, and they eventually
>become managers. The same thing happens at the Clients. Change is actually
>inevitable UNLESS you can show independently and unemotionally, that the
>current service is the best available for the price. You use COBOL because
>it works for you and provides the services you sell, at a good price, not
>because you have 7000 COBOL programmers. :-) It may well be, that in your
>company, COBOL is a good model. Perhaps the underlying services don't change
>a lot or amendments are trivial in the scheme of things. Everything ticks
>over nicely just as it has for many years and there is a core of code that
>the business is based on and which nobody really wants to meddle with. There
>is therefore no push to change things.. If that is the case then it may
>indeed be some time before change is implemented.
>
>But, eventually, it will be.
>

Perhaps that will come true but long after we have gone on to Cyber
Heaven.


>Pete.

SkippyPB

unread,
Feb 5, 2012, 12:08:03 PM2/5/12
to
That's true. It came from ARPA Network which went online in 1969 and
connected 4 Universities. Designed for research, education, and
government organizations, it provides a communications network linking
the country in the event that a military attack destroys conventional
communications systems.

It was financed by the US Department of Defense. Email didn't happen
for another 3 years. TCP/IP was designed in 1973 but didn't become
reality until 1983. This included FTP. The word "Internet" didn't
come into use until 1982 and in 1984 DNS is established. The first
provider of dial-up Internet services debut in 1989 (world.std.com).

Things have escalated and grown exponentially since then. Personally,
I got on the Internet for the first time from Thailand in 1993 or 4.

Lüko Willms

unread,
Feb 5, 2012, 2:52:53 PM2/5/12
to
Am 05.02.2012 18:08, schrieb SkippyPB:
>>>>> The Internet evolved from a
>>>>> network that was invented to provide reliability in the event of nuclear
>>>>> attack...

>> Myth

> That's true.

Ah you agree with me?

> It came from ARPA Network which went online in 1969 and
> connected 4 Universities.

More and more universities and other research institutions

> Designed for research, education, and
> government organizations, it provides a communications network linking
> the country in the event that a military attack destroys conventional
> communications systems.

No, this was the only real communication system which went beyond
point-to-point modem connections.

Truth is that the ARPANET was built for military research, i.e. for
research to enable the USA to militarily attack any country in the world.

The network ran on a variety of the protocol which is internationally
known as X.25. It was certainly not safe to withstand a nuclear attack
on the switching nodes.

When local area networks came up in the early or mid 1970ies, posing
the problem for interconnecting those LANs, the solution of using
connection-less network service (CLNS) overall prevailed over the idea
to put connection-oriented network service (CONS) like X.25 over the
LANs. From this arose IP and TCP, and with it the Internet (with a
capital I).

The myth that this had to do with trying to withstand a nuclear attack
arose several decades later. It would be interesting how this myth came
up, who first put this gossip into the air.


Cheers,
L.W.

Pete Dashwood

unread,
Feb 6, 2012, 6:52:46 PM2/6/12
to
I worked with DOD in NZ in the 1970s. We had instant computer-based
communiication with Washington and Canberra via a military net which may
have had ARPANET nodes. The important thing was that the networking changed
the way in which people perceived computers; they became devices for
connecting people rather than just very powerful abacuses.

Although details of how it worked were not freely available, I was told that
in the event of the network going down (euphemism for nuclear attack)the
system could re-route through other nodes and was therefore considered to be
a reliable command and control system.

It may be myth to you, but I was that soldier... :-)

Pete Dashwood

unread,
Feb 6, 2012, 7:01:02 PM2/6/12
to
SkippyPB wrote:
> On Sun, 05 Feb 2012 12:27:43 +0100, Lüko Willms
> <lueko....@domain.invalid> wrote:
>
>> Am 04.02.2012 23:29, schrieb Pete Dashwood:
>>> The Internet evolved from a
>>> network that was invented to provide reliability in the event of
>>> nuclear attack...
>>
>> Myth.
>>
>>
>> Cheers,
>> L.W.
>
> That's true. It came from ARPA Network which went online in 1969 and
> connected 4 Universities. Designed for research, education, and
> government organizations, it provides a communications network linking
> the country in the event that a military attack destroys conventional
> communications systems.
>
> It was financed by the US Department of Defense. Email didn't happen
> for another 3 years. TCP/IP was designed in 1973 but didn't become
> reality until 1983. This included FTP. The word "Internet" didn't
> come into use until 1982 and in 1984 DNS is established. The first
> provider of dial-up Internet services debut in 1989 (world.std.com).
>
> Things have escalated and grown exponentially since then. Personally,
> I got on the Internet for the first time from Thailand in 1993 or 4.
>
> Regards,

I don't remember the year, but I was in Europe when the first public ISP
(Compuserve) offered their service. At the time I joined we were all
terribly impressed because there were around 16 million people using the
Internet... :-). I actually was a volunteer Compuserve support guy for a
while and still have the treasured T-Shirt. I wrote my first personal web
site using Notepad, after teaching myself HTML in a rainy afternoon, and it
was hosted on Compuserve until the very end. Sadly, they were absorbed a
few years ago (I think it was AOL who took them over, not sure...) and I,
for one, was sorry to see their demise.

(I still have the website complete with pictures, animations, music, and
what today would be called a Blog. One of these days, as a retirement
project, I'll republish it as part of a more modern web presence.)

Pete Dashwood

unread,
Feb 6, 2012, 7:03:41 PM2/6/12
to
Probably sthe kind of person who would believe that the USA were building a
Network so they could attack the world... :-)

Pete Dashwood

unread,
Feb 6, 2012, 9:34:00 PM2/6/12
to
Your post prompted me to take a break and run some searches. I found a lot
of very interesting stuff on both sides. Some of my own perception was
modified as a result. :-)

Here is one of the most coherent posts I found:
<quoting from an internet forum>

The problem is that the term 'mainframe' does not actually describe either a
computer or an operating system.

The IBM 9370 running AIX/370 that sat under a desk at one of my previous
jobs was a (baby) 'mainframe'. The 3090s running VM/CMS and RETAIN (an OS in
itself) that I used when in IBM were 'mainframes'. The Amdahl 5890E running
UTS and AT&T RDS UNIX was a 'mainframe'. The Honeywell 6180 running MULTICS
was a 'mainframes'. LEO was a 'mainframe'. The IBM 370/168 running MTS I
used at University was a 'mainframe'. The ICL 1904 and 2904 running George
that many Universities had were 'mainframes'. The DEC Systems 10 and 20
running TOPS were 'mainframes'. I could dig around and find a lot more
'mainframe' systems.

Now. Were none of these hacked? I can tell you for a fact that I hacked an
Amdahl running R&D UNIX as part of my job more than once, and I must admit
to breaking into accounts on MTS on the 370/168 while at University to get
more computing budget to play the original Adventure (come on, it was 30
years ago. There must be a statute of limitations on this, surely!).

This article probably means an IBM mainframe running z/OS or its ancestors,
probably using RACF. Even this, I'm sure, can not claim to never have been
hacked! I have just found this
http://www.os390-mvs.freesurf.fr/tenflaws.htm, in which item 9 clearly
states that the author gained key 0 protection from a non supervisor account
on MVS. Sounds like hacking to me.

I will freely admit that current mainframes running z/OS are incredibility
secure, but I ask again. Where is the references that state a mainframe has
never been hacked!

Posted Wednesday 22nd June 2011 11:59 GMT

Peter Gathercole.
</quoting from an internet forum>

Sound common sense I think, and a timely reminder that there is no reason to
be smug about mainframe security.

Now back to our own discussion...

> Oh there are plenty of facts to support the mainframe still be way
> ahead of a server network in terms of speed,

Given the emergence of HPC clusters into commerce it is a bit pointless to
talk abut high performance on either side. Lets agree that modern computers
are "fast"...

Here are some figures. Notice that IBM don't rate z90 as a high performance
machine. Given that some of these systems are now getting into 5 teraflops
country (and above) , it is really a bit academic. You can believe whatever
you like but the gap between mainframe and non-mainframe server performance
has been closing for some time now. Eventually it crosses a boundary (1
teraflops?) where it is no longer relevant.

http://icl.cs.utk.edu/hpcc/hpcc_results.cgi

I like the networked approach because it allows as many CPUs as you find
useful to be applied to a single problem. Parallel computing is the "next
big thing" for programming and facilities to support it are already being
incorporated into modern languages. A mainframe has a finite number of
cores; a network has as many as you want/need.

>file handling

Agreed that moving data around quickly and reliably is of high priority for
any computer based solution but again, both mainframes and network servers
solve this problem very well.

An enterprise Z10 system using its InfiniBand facility gets an impressive
6GB/s of bandwith for file transfer. (IBM figures). An equivalent Intel Xeon
cluster gets 20 GB/s ... It really doesn't matter and it is only one
indication because there are many other things that affect data throughput
in reality. BOTH solutions have adequate data transfer facilities, just as
BOTH systems have adequate processing facilities. (The z10 has a maximum 64
cores; the Intel architecture allows any number... see comment above
regarding my preference for networks.).

The ONLY thing I'm taking issue with is your statement that " there are
plenty of facts to support the mainframe still be way ahead..." . It isn't.
Not in ANY of the things you mentioned. (And I note ruefully that you did
not provide any of these "facts"...)

As I noted originally, your belief stems from historical perception (I
shared it for a while until I actually was exposed to some modern networking
technology...) and is not true in today's world. The performance lines of
these systems have been converging for over a decade. They are now so close
as to be blurred. Excluding Super Computers, I believe that any performance
figures you can show for a given mainframe system can be matched or exceeded
by an equivalent modern server network. What's the point?

What IS true is that the PRICE of mainframes has had to fall in order to
compete with the alternative solutions. Otherwise we'd see mainframe systems
getting creamed instead of enjoying a resurgence.

> and I'll
> even throw in security.

You need to get out more, Steve... :-)

Take a look at this: http://www.sdsusa.com/netqdocs/SNA.Security.090721.pdf

While it is being published by people with a vested security interest, there
is no denying the need for their products...

> When was the last time a mainframe was hacked
> and had customer/client data stolen?

In large Banks and Insurance companies such events are not publicised for
obvious reasons. However, they DO happen and, occasionally, it gets leaked.

I know of two instances from my own personal experience working inside some
major household name Banks but I obviously am not at liberty to discuss them
and both of them occurred more than 15 years ago so are probably not
relevant for this discussion.

Banks are losing money all the time to infiltration and hacking. Try a
search on "bank loses data"... Apparently 2008 was a pretty bad year in
general. Not all of these losses are down to mainframe failure but it is
fair to say that some of them are using manframes as their major data
respoitories.

The most recent PUBLICLY DOCUMENTED Event (involving IBM hardware) I could
find is:

http://spectrum.ieee.org/riskfactor/computing/it/south-korean-banks-weeklong-system-failure-affecting-30-million-an-inside-job



>
> And show me a server network that can support a full mortgage loan
> apllication and accept input 24/7 from around the world and process
> that in 3 seconds or less. And that's just a couple of examples. It
> is far more than perception..it is reality.

I agree it is your reality... until such time as you think about it and
change your perception. (Or not... :-))

>
>
>>> In addition,
>>> the regulatory changes that come down the pike every month or so are
>>> far easier to implement in our COBOL code than in the JAVA or
>>> whatever else we happen to be running on the server side of the
>>> house.
>>
>> Perception. Ask the Java guys if it is easier in COBOL... :-)
>>
>
> Only because the JAVA guys aren't bright enough to code in COBOL. :)
> It's like asking a native French speaker to write a novel in English
> and then again in French. Which is he going to be more comfortable
> doing and finding easier?
>
>>
>>> Hence,
>>> we continue to develop and maintain legacy and new COBOL code. We
>>> do offer server based systems and support them as well but the
>>> number of clients is small and so is the revenue. I'm not sure
>>> what language they use.
>>>
>>> Option 2 is no option at all in our world and won't be a very long
>>> time.
>>
>> We'll see... :-)
>>
>> (Let's dicuss this again in 5 years...)
>>
>
> Crap I hope I'm retired and lving out my days on a beach in Koh Samui
> in 5 years instead of worrying if COBOL is still around or not!
>

"Amen!" to that!

Hope you make it.
I don't mind where I go when I die; I have friends in both places... :-)

Pete Dashwood

unread,
Feb 7, 2012, 4:18:37 AM2/7/12
to
iNFO_rene wrote:
>> But nobody should think of building a life-long career on COBOL and
>> nothing but COBOL.
>>
>> As one of those retired CoBOL programmers - we were predominately
>> CoBOL, but never "nothing but CoBOL".
>>
>
> Very well mentioned. It's all mixed-up PLs right now.
>
> Experienced Cobol programmers are up to their late 40s, and are
> thinking of "early" retirement. Nothing fancy here, but most of these
> guys are rich (working with Cobol) and can afford to retire (in the
> Philippines). Unexpectedly, to my surprise... some companies are
> training their "new" programmers Cobol! Blame academia?!? Maybe, they
> opted to teach students opensource (free) coding because it is costly
> teaching them Cobol (compilers fee is soaring).
>
> Cobol functionality is a fact. Thats why we are Cobol programmers
> ourselves... but most of us unfortunately hate Cobol (I knew Pete is
> one) because of the compiler cost.

I do NOT "hate" COBOL. If you actually read what I post you'd know that.

I don't even object to it being expensive; I understand the reasons why it
is.

BUT, if somebody offers me all the facilities I can get with COBOL (and
more...) and lets me have it for FREE, it is not a hard choice to make.

All it costs me is the time and effort to learn a new language. For the
Microsoft platform there are abundant opportunities to get whatever training
I need (mostly for free...) and support, help, and sample code are instantly
available 24/7 at no cost other than an Internet connection.(I never thought
they'd keep these offers open indefinitely and when I grabbed at the chance
I really thought it would not be there forever. To their credit, they STILL
offer free introductory videos on VS and the Net Languages, and people are
still taking them up.) The only people still using COBOL are corporates; it
is way too expensive for SMEs or independent contractors, when there are
other options that are better and free.

I'm not stupid; I can learn new stuff. (In fact, I actually enjoy expanding
my education).

(I think that is true for most computer programmers...at least it is if they
have passion for what they do.)

I spent years of my life learning technologies that were obselete, or
useless to me within five years., so learning new stuff is not a problem for
me. I sweated blood learning 1400 Autocode, Machine language for NCR 500,
NEAT3, IBM Assembler (BAL) and ICL PLAN. I wrote RPG on AS 400 (for a few
weeks :-)), learned System 2000, IMS DB/DC, VanDL, ROSCOE, REXX, and managed
to become proficient in all of them even though I might as well not have
bothered. I learned most of the major dialects of COBOL although I'm still
using some of that so I guess it can't be discounted. Why would I baulk at
learning C# or Java? The time I invested in doing this has repaid itself in
actual money many times over.

And COBOL does not have "functionality". You can build functionality with
it, but you can with many things.

"COBOL functionality is a fact". Yes, it is in many corporations and there
is much COBOL code that provides essential functionality to the business.
However, there are now other options to provide that functionality (for many
years there weren't; COBOL was the only game in town) and there are even
options to lever that Legacy code onto modern platforms at a price that is
way less than rewriting it.

> For me, mix the Cobol up with
> opensource technologies and you will get more functionalities.

But you don't. I would move to Open COBOL in a heartbeat for clients who are
migrating but it doesn't support OOP, so I can't.

I want Legacy wrapped as objects and built into layers that can be used
with anything else that is currently being developed, irrespective of what
languages the new development is being done in. I can't get it by "mixing
COBOL with Open Source". And who do I go to when the Open Source stuff
doesn't work? (Already been bitten on that one...PRIMA is now standardised
on SQL Server instead of PostgreSQL or MySQL, both of which were tried and
found wanting. SQL Server is free just like the others mentioned, but at
least there is proper support for it. And the ODBC drivers do actually
work...)

I don't have an emotional aversion to using Microsoft software and it serves
me very well.


>Cost
> effectiveness of PLs (especially Cobol) does not depend on the IDEs
> itself but on business operations upfront.

Certainly the business opertions being computerised will have some effect on
the cost effectiveness of the languages being used to implement that
computerization. But, all application things being equal, the cost
effectiveness of a language can be measured in what it costs you to buy,
support, maintain and use.

(If you implemented the same business functionality in COBOL, VB.NET, Java,
and C#, then monitored the costs of those systems for the next 5 years to
get a total cost of ownership, the results will not even be in the same
ballpark (and you would reasonably expect they might be; it is all just
generating computer code...) COBOL does not do well in this kind of cost
comparison. Even although it is theoretically designed to be easy to
maintain.)

HansJ

unread,
Feb 7, 2012, 6:46:18 AM2/7/12
to
Lueko,

would you mind to let us know how many people did attend for the COBOL
part? Just curious.

You might want to check for the date, Sunday Feb. 12th 2012 might not
be a good idea for Micro Focus to adress their audience.

Regards Hans

Alistair Maclean

unread,
Feb 7, 2012, 1:05:31 PM2/7/12
to
Good God!!! I can't believe that you believe that Mayan pseudo-
religious clap-trap. ;-)

docd...@panix.com

unread,
Feb 7, 2012, 1:13:31 PM2/7/12
to
In article <8ed9c093-df9f-496a...@s7g2000vby.googlegroups.com>,
Alistair Maclean <alistair.j...@gmail.com> wrote:
>On Feb 7, 11:46?am, HansJ <hans.i...@googlemail.com> wrote:

[snip]

>> You might want to check for the date, Sunday Feb. 12th 2012 might not
>> be a good idea for Micro Focus to adress their audience.
>>
>
>Good God!!! I can't believe that you believe that Mayan pseudo-
>religious clap-trap. ;-)

Hmmmmmm... *someone* has shown how they are going to spend Eternity. Let
those who dust off their feathered head-dresses and await Quetzalcoatl's
return show appropriate wariness!

(decades on back I was fascinated by the various native cultures of
Central and South America)

DD

SkippyPB

unread,
Feb 7, 2012, 1:31:01 PM2/7/12
to
I was on Compuserve in the US in the mid-80's when they were not much
more than a huge BBS. They had a Comodore area and that's where I
spent my time.

At some point they stopped supporting or posting new Comodore stuff so
the Comodore people took they toys and stated a Comodor only BBS
called Q-Link. It had graphics, sound and just about everything else
PCs didn't have or cost a ton to have.

Q-Link evolved into AOL.

Regards,
--

////
(o o)
-oOO--(_)--OOo-

"If you saw a heat wave, would you wave back?"
--Steven Wright

Pete Dashwood

unread,
Feb 7, 2012, 5:23:00 PM2/7/12
to
It's an interesting date because it is almost palindromic (at least for the
cultures who use DDMMYYYY). February 12, 2021 will be fully palindromic, but
of course we won't last that long because the World will have been destroyed
by then. :-)

I read a very interesting book on famous prophecies where they explained
that it is just one of the great cycles of the Mayan calendar that ends on
this date. The Mayans never claimed the world would end then; they just
start another cycle that goes on for around 64000 years.

If you step back a bit and think, it is pretty obvious that the Universe is
completely unaware of the way we structure our calendars and planetary
motion just goes on the way it always has, in accordance with the Laws of
Physics, until some natural event causes it to change or stop.

It is only the arrogance of Man that lets us think our superstitions would
have any effect on it.

Pete.
.--

Howard Brazee

unread,
Feb 7, 2012, 6:32:29 PM2/7/12
to
On Tue, 7 Feb 2012 03:46:18 -0800 (PST), HansJ
<hans...@googlemail.com> wrote:

>You might want to check for the date, Sunday Feb. 12th 2012 might not
>be a good idea for Micro Focus to adress their audience.

So there will be a bunch of bad jokes. I was at a local science
fiction night last year at a different end-of-the-world. It was a
night of bad joke references to The End. Same thing will happen on
December 21 (Mayan Calendar Roll Over Day).

docd...@panix.com

unread,
Feb 7, 2012, 9:31:39 PM2/7/12
to
In article <9pdmi7...@mid.individual.net>,
Pete Dashwood <dash...@removethis.enternet.co.nz> wrote:

[snip]

>If you step back a bit and think, it is pretty obvious that the Universe is
>completely unaware of the way we structure our calendars and planetary
>motion just goes on the way it always has, in accordance with the Laws of
>Physics, until some natural event causes it to change or stop.

If you step back even farther the prospect of a human being or two
attempting to ascribe states such as 'awareness' to the Universe might be
a bit of a curiousity. I barely know what *I* am aware of, let alone
anyone else.

>
>It is only the arrogance of Man that lets us think our superstitions would
>have any effect on it.

It might also be a similar arrogance which might cause one to conclude
what would *not* have an effect, as well... both sides of the balanced
equation, as it were.

DD

Robert Doerfler

unread,
Feb 8, 2012, 2:57:26 AM2/8/12
to
It is 14th Feb. :

http://eventonweb.org/mf_evolution_2012/

Too bad that Frankfurt is a bit too far away from me. Sounds interesting.

Robert Doerfler

unread,
Feb 8, 2012, 3:15:25 AM2/8/12
to
Btw... do you know Q-Link Reloaded? http://orrtech.us/qlink/index.html ;)

Lüko Willms

unread,
Feb 8, 2012, 4:37:06 AM2/8/12
to
Am 07.02.2012 01:03, schrieb Pete Dashwood:
> Probably sthe kind of person who would believe that the USA were building a
> Network so they could attack the world...

The idea of cyberwar arose very few years ago.


Cheers,
L.W.

Lüko Willms

unread,
Feb 8, 2012, 4:52:00 AM2/8/12
to
Am 07.02.2012 00:52, schrieb Pete Dashwood:

> have had ARPANET nodes.

> Although details of how it worked were not freely available,

The details of the network switches of the ARPANET were and are
freely available. After Honeywell 516 and 316, they were BBN C-30 and
C-300 (BBN = Bolt, Beranek and Newman;
> see <http://en.wikipedia.org/wiki/BBN_Technologies>
)
On the protocol see
> <http://en.wikipedia.org/wiki/BBN_Report_1822>

> I was told that
> in the event of the network going down (euphemism for nuclear attack)the
> system could re-route through other nodes and was therefore considered to be
> a reliable command and control system.

Sure, that is the nature of a network. Dynamic routing. Error tolerant.

The problems requiring rerouting originate from the most various
resources, one of the most drastic an earthquake severing a data cable.
A nuclear attack would in all probability so devastating on a wide
scale, that no rerouting would be possible.

Also, ARPANET was established for military _research,_ in the
academia, not for actual warfare. A nuclear war would have made any
research obsolete by eradicating mankind as a whole.


Cheers,
L.W.



Lüko Willms

unread,
Feb 8, 2012, 5:02:36 AM2/8/12
to
Am 04.02.2012 02:41, schrieb Howard Brazee:
>> But nobody should think of building a life-long career on COBOL and
>> >nothing but COBOL.
> As one of those retired CoBOL programmers - we were predominately
> CoBOL, but never "nothing but CoBOL".

I just looked up the printed "Draft Proposed Revised X3.23 American
National Standard Programming Language COBOL" as the most standardized
document on COBOL which is at this moment on my desk; this official
COBOL document always spells COBOL in capital letters.


Cheers,
L.W.

Lüko Willms

unread,
Feb 8, 2012, 5:05:17 AM2/8/12
to
Am 07.02.2012 12:46, schrieb HansJ:
>> I'll attend a Micro Focus roadshow here in Frankfurt on February 12;
>> > maybe I can learn something, make some contacts, maybe even meet
>> > somebody from the participants in this forum?

> would you mind to let us know how many people did attend for the COBOL
> part? Just curious.
>
> You might want to check for the date, Sunday Feb. 12th 2012 might not
> be a good idea for Micro Focus to adress their audience.

Sure, the event is on Tuesday, February 14.

Sorry for the confusion.


Cheers,
L.W.



Pete Dashwood

unread,
Feb 8, 2012, 6:16:15 AM2/8/12
to
I just received today a note saying that the monthly Dot NET user group
meeting, which WAS scheduled for February 14th has had to be moved to the
15th... The group protested the date.

Computer geeks in this part of the world DO have wives and girlfriends...

Only a very brave (or foolish...) man would tell his partner he would be
unavailable on Valentine's Day, due to a computer event being higher
priority...

Pete.

SkippyPB

unread,
Feb 8, 2012, 12:47:39 PM2/8/12
to
No I didn't. Thanks for the link. If I ever get my C-64 set back up
and running, I'll definitely be logging into that site.

docd...@panix.com

unread,
Feb 8, 2012, 5:36:36 PM2/8/12
to
In article <v7d5j79cd4tl6kesk...@4ax.com>,
SkippyPB <swie...@Nospam.neo.rr.com> wrote:
>On 8 Feb 2012 08:15:25 GMT, Robert Doerfler <ro...@zlug.org> wrote:

[snip]

>>Btw... do you know Q-Link Reloaded? http://orrtech.us/qlink/index.html ;)
>
>No I didn't. Thanks for the link. If I ever get my C-64 set back up
>and running, I'll definitely be logging into that site.

Hmmmmmm... might be my own sense of humor but 'If I ever get my C-64 set
back up and running' strikes me as a more polite equivalent of 'After I
finish that snowball-fight in the Ninth Circle'.

DD

HansJ

unread,
Feb 9, 2012, 6:58:24 AM2/9/12
to
On 8 Feb., 08:57, Robert Doerfler <r...@zlug.org> wrote:
Robert,

I did check on their web site and did read the agenda, though not
living too far away from Frankfurt/Main, it was not attracting me too
much, but that might be a personal point of view.

Regards Hans

HansJ

unread,
Feb 9, 2012, 7:01:10 AM2/9/12
to
On 8 Feb., 12:16, "Pete Dashwood" <dashw...@removethis.enternet.co.nz>
wrote:
... ohh Valentines day, did not even think about that.

Robert Doerfler

unread,
Feb 9, 2012, 7:06:26 AM2/9/12
to
I havent heard much about it in general so thats why it is interesting ;)
I wouldnt use VisualCOBOL anyway, sticking to OpenCOBOL ;)

SkippyPB

unread,
Feb 9, 2012, 11:43:16 AM2/9/12
to
Given my current free time status and a lack of sutiable area to
perform such a task, your timeframe is accurate. :)

Pete Dashwood

unread,
Feb 9, 2012, 5:51:10 PM2/9/12
to
That's a bit like saying: "I wouldn't eat apples because I'm sticking to
pears..."

They are two different products.

Until such time as Open COBOL implements OO support (if it ever does) it
cannot be a viable commercial product for anything but batch processing and
possibly some primitive CGI web pages.

Visual COBOL is a CIL generating compiler that produces code designed to run
as managed CLR code in the .Net environment . While I personally have some
reservations about people doing this with procedural COBOL, at least you CAN
build .Net Classes with it. You can't with Open COBOL.

(That doesn't mean that Open COBOL is worthless; it is an excellent attempt
to provide a COBOL compiler without the fierce costs normally associated
with such a product and reports from people using it say it is robust and
reliable. I would love to be able ot use it, but it doesn't meet our current
needs.)

I did consider attending the Frankfurt event because PRIMA is currently
talking to people in Germany about the Migration Toolset and I've just
finished reading all our documentation translated into German. :-) I had
thought a trip to Germany might be useful ( I lived there for some years and
really love the country) but decided against it at this stage for the
following reasons:

1. It is bloody cold in Europe right now and I am writing this in 26 degrees
with glorious sunshine blazing into my lounge. :-) (Summer in Tauranga...)

2. Flights to Europe from here are very tedious even with the few days
stopover I usually make in California. (I'm getting old; there was a time
when I used to jump at it... :-))

3. Neither the Company nor myself can really afford such a trip right now.
If we close a deal in Germany it might be different.

4. I don't think we would have a use for Visual COBOL and I probably
wouldn't learn much by attending. But it would be fun and it is possible we
might have some future clients who are using it. I'd enjoy talking to people
who have been using it for a while.

On balance (at least for now) I'm staying home... :-)

docd...@panix.com

unread,
Feb 9, 2012, 8:35:05 PM2/9/12
to
In article <nqt7j7pf78djgmv9c...@4ax.com>,
SkippyPB <swie...@Nospam.neo.rr.com> wrote:
>On Wed, 8 Feb 2012 22:36:36 +0000 (UTC), docd...@panix.com () wrote:
>
>>In article <v7d5j79cd4tl6kesk...@4ax.com>,
>>SkippyPB <swie...@Nospam.neo.rr.com> wrote:

[snip]

>>>No I didn't. Thanks for the link. If I ever get my C-64 set back up
>>>and running, I'll definitely be logging into that site.
>>
>>Hmmmmmm... might be my own sense of humor but 'If I ever get my C-64 set
>>back up and running' strikes me as a more polite equivalent of 'After I
>>finish that snowball-fight in the Ninth Circle'.
>
>Given my current free time status and a lack of sutiable area to
>perform such a task, your timeframe is accurate. :)

Wow... I got one right? Quick, someone call the folks at MovieTone and
have them send a newsreel crew out!

DD

Alistair Maclean

unread,
Feb 12, 2012, 6:55:34 AM2/12/12
to
On Feb 9, 10:51 pm, "Pete Dashwood"
<dashw...@removethis.enternet.co.nz> wrote:
> Robert Doerfler wrote:
>
> > I havent heard much about it in general so thats why it is
> > interesting ;) I wouldnt use VisualCOBOL anyway, sticking to
> > OpenCOBOL ;)
>
> That's a bit like saying: "I wouldn't eat apples because I'm sticking to
> pears..."
>
> They are two different products.
>

I think the important bits were that Robert already uses OpenCobol
(sticking to...) and possibly that there may be something about the
development environment. Why should he tie himself in to the .NET
environment, however superior that may be to any other environment
(I'm not saying it is superior so please folks no spam or environment
wars)?


> 1. It is bloody cold in Europe right now and I am writing this in 26 degrees
> with glorious sunshine blazing into my lounge. :-) (Summer in Tauranga...)
>

But the Europeans have lobsters. How is the lobster drought in NZ?

Pete Dashwood

unread,
Feb 12, 2012, 7:45:57 AM2/12/12
to
Alistair Maclean wrote:
> On Feb 9, 10:51 pm, "Pete Dashwood"
> <dashw...@removethis.enternet.co.nz> wrote:
>> Robert Doerfler wrote:
>>
>>> I havent heard much about it in general so thats why it is
>>> interesting ;) I wouldnt use VisualCOBOL anyway, sticking to
>>> OpenCOBOL ;)
>>
>> That's a bit like saying: "I wouldn't eat apples because I'm
>> sticking to pears..."
>>
>> They are two different products.
>>
>
> I think the important bits were that Robert already uses OpenCobol
> (sticking to...) and possibly that there may be something about the
> development environment. Why should he tie himself in to the .NET
> environment, however superior that may be to any other environment
> (I'm not saying it is superior so please folks no spam or environment
> wars)?

It's not about the fact that Visual COBOL addresses the .Net environment
(although it does) it is about the fact that Visual COBOL is Object COBOL
and Open COBOL is not. Hence apples and pears... two different products,
similar in some regards, but not the same.

Writing Object COBOL doesn't tie you in to ANY environment OTHER THAN
environments that support Objects. (That would be most of the Networks
available in the world, along with Windows and Unix based OSes)

The .Net (and Mono) Frameworks both provide a level playing field, in that
many different languages can be used and their classes and objects are all
treated equally. There are no "second class citizens" in .Net - if it
implements .Net interfaces you can write it in Bendix assembler and it can
still call classes written in COBOL, VB.Net, C# and so on.

I may have mentioned before that the SOURCE is not important; it is the
functionality that matters. You can build .Net applications in kiddie
script, VB and even Javascript. In fact, Javascript is shaping to be an
important player for the future because it is quickly learned, simple to
use, and millions of Web Developers already know it...

Personally, I have no axe to grind about whatever Robert (or anybody
else...) uses. But let's be clear that just because Visual COBOL has "COBOL"
in the name does not mean it is the same as Open COBOL, any more than
Javascript is the same as Java. The paradigms supported define and determine
the usefulness of the products.

If you wouldn't try something because you think it is the same as what you
already have, that's your prerogative. But if you won't eat apples because
you like pears, that is less than logical.
>
>
>> 1. It is bloody cold in Europe right now and I am writing this in 26
>> degrees with glorious sunshine blazing into my lounge. :-) (Summer
>> in Tauranga...)
>>
>
> But the Europeans have lobsters. How is the lobster drought in NZ?

I can get lobsters any time I'm prepared (and have the wherewithal) to pay
$100 a kilo for them. Alternatively, I can get a friend with a boat to pass
me one or two at "mates rates" (usually around $15 a kilo), assuming they
have some spare. (there are limits on the number you can take, but it is
quite generous)

With a bouyant export market to Asia the commercial lobster fishing industry
is doing very nicely. The demand is being met and the resource is being
managed but as an old-timer I notice there are not as many around as in my
youth, and the prices are out of reach of normal people.

Fortunately, I consider them a treat and don't expect to eat them often. (I
didn't eat them any more often when I was in Europe than I do here.)

My local supermarket was offering Canadian lobsters (small) at 2 for $30 or
$20 each.

They sold out pretty quickly. :-)

jo...@wexfordpress.com

unread,
Feb 15, 2012, 4:54:44 PM2/15/12
to
On Feb 5, 6:27 am, Lüko Willms <lueko.wil...@domain.invalid> wrote:
> Am 04.02.2012 23:29, schrieb Pete Dashwood:
>
> >   The Internet evolved from a
> > network that was invented to provide reliability in the event of nuclear
> > attack...
>
>    Myth.
>
> Cheers,
> L.W.

No, fact. It was called ARPANET and it used brief packets to transmit
files. If one path was interrupted it
would switch to another path. The software at the other end would
reassemble the packets in correct order and request
any missing packets. It involved Defense Advanced Research Projects
Agency (DARPA).

jo...@wexfordpress.com

unread,
Feb 15, 2012, 4:48:37 PM2/15/12
to
On Feb 3, 6:34 pm, "Pete Dashwood"
<dashw...@removethis.enternet.co.nz> wrote:
> Bill Gunshannon wrote:
> > In article <slrnjin7vn.a9t.r...@seeteufel.bloerp.de>,
> > Robert Doerfler <r...@zlug.org> writes:
> >> On 2012-02-02, Arnold Trembley <arnold.tremb...@att.net> wrote:
> >>> On 2/1/2012 5:03 PM, Frank Swarbrick wrote:
> >>>>http://www.gonzobanker.com/2012/01/cobol-not-about-language/
>
> >>> My favorite quote from the reader comments:
>
> >>> "If truly 80ish% of the business world runs on COBOL, where are all
> >>> the programmers? Iâ?Tve never personally met one. How is all this
> >>> code maintained?"
>
> >>> A few years ago someone asked me what I did for a living.  When I
> >>> told him I was a COBOL programmer, he was surprised.  He thought
> >>> everybody had stopped using COBOL around 1990.
>
> >> I just meet one person around the age of 65 who was a cobol/pl1
> >> programmer and he told me to go the cobol way of programming as
> >> cobol developers are always needed. But in the past years i've just
> >> seen just some very rare job offers for cobol programmers around in
> >> my area. So where are all the
> >> cobol programmers? Are they hidden in some basement room? :o)
>
> > While the lack of IBM/CICS experience has kept me from getting back
> > into COBOL fulltime (that is where most of the work seems to be, and
> > I can easily find dozens if not hundreds of jobs) I am happy to say
> > that as
> > of the end of this month I will be goign back to at least partial
> > COBOL programming for a living.  The job required COBOL on IBM, but
> > not on
> > a mainframe.  I will be doing COBOL and AIX.  A language and an
> > environment that meets my skillset perfectly.
>
> > If you really want a COBOL job, they are out there.  I suspect the
> > biggest problem finding the jobs is that the places hiring know the
> > market for programmers is thin and tend to limit their searches to
> > places where they actually expect to find candidates (that means no
> > Monster, Dice or local newspapers!) or the are doing things like
> > General Dynamics and setting
> > up internships in order to create their own.
>
> > Blame academia!!  Even after repeatedly showing that the demand was
> > there and providing pointers to people who might be glad to have
> > formal intern- ship agreements (these can be used as a large selling
> > point for an academic program) I was unable to get people I was
> > associated with to even consider the possibility.  Trying to drive
> > the bus rather than meeting the needs
> > of the industry is going to come back to bite academia.  Signs of this
> > are already showing up in the press.
>
> > bill
>
> I remember having exactly the same discussions with representatives of
> "Acadaemia" back in the 1980s. We knew then that the people being produced
> from Universities and Technical Institutes were not much use in the real
> world and had to be trained into the world of Commerce.
>
> Fortunately, most of them were bright young people who were capabale of
> learning quickly, but industry generally felt it was being let down by
> Acadaemia.
>
> Nowadays, people who arrive with a Computer Science degree are generally
> much better equipped that the people who arrived then and just because they
> don't know COBOL does not mean that they are of no use.
>
> I have enjoyed working with the "new breed" of IT graduate. They are more
> open minded, less evangelical, and much more capable than generally used to
> be the case.
>
> I believe what is being churned out from Acadaemia now, is a far better and
> more useful product, than what was being churned out through the 60s, 70s,
> and 80s.
>
> Pete.
> --
> "I used to write COBOL...now I can do anything."

Hopefully they never took a college course in COBOL. In my day the
profs who taught COBOL hated the
language and communicated that hatred to their students.

John Culleton

Pete Dashwood

unread,
Feb 15, 2012, 9:35:51 PM2/15/12
to
jo...@wexfordpress.com wrote:
> On Feb 3, 6:34 pm, "Pete Dashwood"
> <dashw...@removethis.enternet.co.nz> wrote:
>> Bill Gunshannon wrote:
>>> In article <slrnjin7vn.a9t.r...@seeteufel.bloerp.de>,
>>> Robert Doerfler <r...@zlug.org> writes:
>>>> On 2012-02-02, Arnold Trembley <arnold.tremb...@att.net> wrote:
>>>>> On 2/1/2012 5:03 PM, Frank Swarbrick wrote:
>>>>>> http://www.gonzobanker.com/2012/01/cobol-not-about-language/
>>
>>>>> My favorite quote from the reader comments:
>>
>>>>> "If truly 80ish% of the business world runs on COBOL, where are
>>>>> all the programmers? Iā?Tve never personally met one. How is all
That's sad.

Like there isn't enough prejudice and hatred in the world...:-)

Maybe its's a good thing they stopped teaching it.

Robert Doerfler

unread,
Feb 16, 2012, 10:35:15 AM2/16/12
to
On 2012-02-16, Pete Dashwood <dash...@removethis.enternet.co.nz> wrote:
> jo...@wexfordpress.com wrote:
>> Hopefully they never took a college course in COBOL. In my day the
>> profs who taught COBOL hated the
>> language and communicated that hatred to their students.
>
> That's sad.
>
> Like there isn't enough prejudice and hatred in the world...:-)
>
> Maybe its's a good thing they stopped teaching it.

I had a whole semestre of cobol in 2009 and my prof was completly
different. He was hating everything else than cobol.

Bill Gunshannon

unread,
Feb 16, 2012, 10:47:01 AM2/16/12
to
In article <slrnjjq8h...@seeteufel.bloerp.de>,
COBOL classes were taught by at least 5 of the professors who are still
at the University. I can't say any of them hated or even disliked COBOL.
But pretty much all of them bought into the notion that COBOL was gone.
One used to talk about a local shop that was "converting all of their
COBOL into Java". Everytime I mentioned this to ione of the managers
of the COBOL side of the house they rolled on the floor laughing.

And, here I am leavnng the University to, you guessed it, go back to
doing COBOL. And not on an IBM Mainframe, either. :-)

bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

Robert Doerfler

unread,
Feb 16, 2012, 11:08:38 AM2/16/12
to
On 2012-02-16, Bill Gunshannon <bill...@cs.uofs.edu> wrote:
> In article <slrnjjq8h...@seeteufel.bloerp.de>,
> Robert Doerfler <ro...@zlug.org> writes:
>> On 2012-02-16, Pete Dashwood <dash...@removethis.enternet.co.nz> wrote:
>>> jo...@wexfordpress.com wrote:
>>>> Hopefully they never took a college course in COBOL. In my day the
>>>> profs who taught COBOL hated the
>>>> language and communicated that hatred to their students.
>>>
>>> That's sad.
>>>
>>> Like there isn't enough prejudice and hatred in the world...:-)
>>>
>>> Maybe its's a good thing they stopped teaching it.
>>
>> I had a whole semestre of cobol in 2009 and my prof was completly
>> different. He was hating everything else than cobol.
>
> COBOL classes were taught by at least 5 of the professors who are still
> at the University. I can't say any of them hated or even disliked COBOL.
> But pretty much all of them bought into the notion that COBOL was gone.
> One used to talk about a local shop that was "converting all of their
> COBOL into Java". Everytime I mentioned this to ione of the managers
> of the COBOL side of the house they rolled on the floor laughing.
>
> And, here I am leavnng the University to, you guessed it, go back to
> doing COBOL. And not on an IBM Mainframe, either. :-)

He even didnt told us that companies are converting their cobol stuff
to java. He repeated - mantra-like - that 80% of the worlds written
source lines are made of cobol. Anyway i liked the practical part which
had much to do my AcuCOBOL and Databases.

HansJ

unread,
Feb 16, 2012, 11:24:17 AM2/16/12
to
Lüko, now 2 days after the MF event, would you mind sharing your
experience?

Lüko Willms

unread,
Feb 16, 2012, 12:18:47 PM2/16/12
to
Am 15.02.2012 22:54, schrieb jo...@wexfordpress.com:
>>> The Internet evolved from a
>>> > > network that was invented to provide reliability in the event of nuclear
>>> > > attack...

> No, fact. It was called ARPANET and it used brief packets to transmit
> files.

That's what one calls packet switching. It used a variant of what
later became known as X.25 (this being the ID of a "recommendation" of
the CCITT, later ITU).

> If one path was interrupted it would switch to another path.

Though only netween the switching nodes; the connection of a client
host to the network was more or less fixed.

> The software at the other end would reassemble the packets in correct
order

Again this might apply to the switching nodes, of which the protocols
between them would be invisible to the client hosts.

ARPANET used what we call CONS, i.e. Connection Oriented Networking
Service, instead of CLNS or Connection-Less Networking SErvice; the IP
of TCP/IP is a CLNS.

> and request any missing packets.
> It involved Defense Advanced Research Projects Agency (DARPA).

Sure, which says that it was and is controlled by the US-american
military, but not that they designed it to withstand a nuclear attack.
That myth is quite idiotic, if I dare speak frankly.

Mind you, after beginning my careed in computing as a COBOL
programmer, I worked in the networking support department of a mainframe
manufacturer.


Cheers,
L.W.


Lüko Willms

unread,
Feb 16, 2012, 12:35:11 PM2/16/12
to
Am 16.02.2012 17:24, schrieb HansJ:

>>>> I'll attend a Micro Focus roadshow here in Frankfurt on
>> > Tuesday, February 14.

> Lüko, now 2 days after the MF event, would you mind sharing your
> experience?

I liked it. It was time well spent (besides the good food at
beginning and end). The venue was changed from the "Squaire" )as they
call the big office and hotel building on top of the long distance
railway station at the Frankfurt airport) to the old Steigenberger Hotel
which is in a short distance from the airport; this had to be done,
Microfocus told me, because there too many people registered for the
event. A total of somewhat more than 200 people attended (besides the MF
personnel).

The technical presentations were held in three tracks:
- Mainframe solutions
- COBOL solutions
- Borland testing solutions.

This sequence is also the ranking in terms of attendance. About 150
had registered for Mainframe solutions (which centered around Mainframe
Express), 90 for COBOL, i.e. the workstation COBOL compiler and
development environment, and the remainder for the testing solutions.
These are registrations, but some did not show up.

Borland was acquired by Micro Focus, as previously Acucobol and RM
Cobol. Also a part of Microware. Borland testing solutions are also
being integrated into MF's mainframe business; they announced a new
product in that area.

The programme can be found on the website for the event
> <http://eventonweb.org/mf_evolution_2012>

The presentations are also online for download (for registered
users...).

At the end was again a common session with a presentation by Werner
Tiki Küstenmacher, a caricaturist and author, especially known for his
"Simplify your life" (available also in an engish translation).

One of his more important suggestions: "Don't start your day's work
by checking your email, but start with something productive, and look at
the email only after having accomplished something". And switch off the
automatic check for new email!



Cheers,
L.W.


Lüko Willms

unread,
Feb 16, 2012, 12:41:32 PM2/16/12
to
Am 16.02.2012 18:35, schrieb Lüko Willms:
> The technical presentations were held in three tracks:
> - Mainframe solutions

that is the track which I attended.


Cheers,
L.W.

Pete Dashwood

unread,
Feb 16, 2012, 9:52:46 PM2/16/12
to
I am always in favour of people speaking frankly, and that is one of the
reasons I see this (unmoderated) forum as being priceless in a world
increasingly manipulated by vested interests.

However, the idea of a redundant network for a command and control system,
which would be resistant to nuclear attck, is not as "idiotic" as the idea
that:

"Truth is that the ARPANET was built for military research, i.e. for
research to enable the USA to militarily attack any country in the world."

Outside of the rigid disciplines of Logic, it would seem that the word
"Truth" has many different meanings to different people... :-)

Lüko Willms

unread,
Feb 17, 2012, 8:04:08 PM2/17/12
to
Am 17.02.2012 03:52, schrieb Pete Dashwood:
> However, the idea of a redundant network for a command and control system,
> which would be resistant to nuclear attck, is not as "idiotic" as the idea
> that:

No, but singling out the utter destruction of everything as the reason
is really an idiotic invention.

The main task to be solved by ARPANET was to link the various
institutions doing research for the US military together. That's all.
But that is not a nuclear attack.


Cheerio,
L.W.

Pete Dashwood

unread,
Feb 17, 2012, 9:24:08 PM2/17/12
to
It seems that DOD personnel at the time did not share your opinion. I know
from first hand what THEIR perception was, as I was working in that area in
the early '70s. To be fair, I seldom heard "ARPANET" referred to by name,
and I understand there were other networks involved as well, which were
secret and "military only" at the time. People generally spoke about the
"connections" or the "network" and it was understood it had been designed to
re-route in the event of nodes being destroyed by nuclear attack.

To be honest, Lueko, I don't really care :-) It was a long time ago and,
speaking as one of a generation of people who grew up in the cold war with
the shadow of nuclear annihilation hanging over us, I'm just glad that we
moved away from that particular abyss, and the much-maligned nuclear weapons
which were designed to be a deterrent, actually achieved that.

Perhaps the cycle will repeat all over again and perhaps someone in Iran or
North Korea or somewhere else is currently working on a redundant,
nuclear-war resistant, command and control network, in preparation for the
Glorious Day when they can bring down the civilized world... I don't know;
but I hope it doesn't happen in the little that remains of my lifetime.
0 new messages