Home > Archive > Cobol > June 2007 > cobol on pc
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]
|
|
| wltiii 2007-06-06, 3:55 am |
| I have googled, read COBUG, and tried other resources, but cannot
determine which of the compilers/IDE's best support the latest COBOL
standard. I am especially interested in those that best support the OO
features. I have little to no need to write code that will run on the
mainframe (at least, at this time). Any pointers or suggestions would
be greatly appreciated!
TIA!
bill
| |
|
|
| Alistair 2007-06-07, 7:55 am |
| On 6 Jun, 04:52, wltiii <goo...@changent.com> wrote:
> I have googled, read COBUG, and tried other resources, but cannot
> determine which of the compilers/IDE's best support the latest COBOL
> standard. I am especially interested in those that best support the OO
> features. I have little to no need to write code that will run on the
> mainframe (at least, at this time). Any pointers or suggestions would
> be greatly appreciated!
>
> TIA!
>
> bill
Perhaps you should do some gap analysis to determine which compiler
most closely meets with the standard defined. I think that no compiler
will meet with 100% of the standards but I am probably wrong on that
one.
You could try Fujitsu, Microfocus or Cobolscript.
| |
| James J. Gavan 2007-06-11, 6:55 pm |
| Alistair wrote:
> On 6 Jun, 04:52, wltiii <goo...@changent.com> wrote:
>
>
>
> Perhaps you should do some gap analysis to determine which compiler
> most closely meets with the standard defined. I think that no compiler
> will meet with 100% of the standards but I am probably wrong on that
> one.
You are in fact now 100% right, (even if you were pissed on Saturday
night :-) ). How could any COBOL compiler be compatible when only *just*
- J4 have determined that 'modules' included in COBOL 2002 are now
OPTIONAL.
>
> You could try Fujitsu, Microfocus or Cobolscript.
>
>
| |
| William M. Klein 2007-06-11, 6:55 pm |
| Come now, Jimmy,
You are in error when you state
"How could any COBOL compiler be compatible when only *just* J4 have determined
that 'modules' included in COBOL 2002 are now > OPTIONAL"
J4 made one vote at one meeting on what they would RECOMMEND to WG4 - if the
question comes up on what (if anything) should be optional.
--
Bill Klein
wmklein <at> ix.netcom.com
"James J. Gavan" <jgavandeletethis@shaw.ca> wrote in message
news:okibi.14025$NV3.8474@pd7urf2no...[color=darkred]
> Alistair wrote:
>
> You are in fact now 100% right, (even if you were pissed on Saturday night
> :-) ). How could any COBOL compiler be compatible when only *just* - J4 have
> determined that 'modules' included in COBOL 2002 are now OPTIONAL.
| |
| James J. Gavan 2007-06-12, 9:55 pm |
| William M. Klein wrote:
> Come now, Jimmy,
>
> You are in error when you state
>
> "How could any COBOL compiler be compatible when only *just* J4 have determined
> that 'modules' included in COBOL 2002 are now > OPTIONAL"
>
> J4 made one vote at one meeting on what they would RECOMMEND to WG4 - if the
> question comes up on what (if anything) should be optional.
>
And your latest thread on 'More on COBOL Standards' ??? Let's be
realistic, the penny has dropped and reluctantly they realize COBOL is a
sturdy horse which is no longer a contender for the Kentucky/Derby or
otherwise.
Seems to me what's left on J4, i.e. Hitachi (indirectly), IBM - I'm
losing track of who now doesn't show. Assuming Micro Focus was still in
the game, that makes three (which includes the taken over AcuCorp). And
for all the effort others might put in it's down to the compiler vendors
what truly happens.
Suppose WG4 conclude, 'We reject J4's suggestions of
curtailing/optionalizing COBOL'. It doesn't matter a cuss what the
formalized procedures are, if those three above retort, 'Get knotted. We
ain't spending the money on unnecessary enhancesmets, even if the
Mongolian Republic is pleading for them via WG4', game's over.
Quite naturally Bill, you are looking at the options with your
background steeped in the formalized procedures. This Joe is just trying
to visualize the real world.
Jimmy
| |
| William M. Klein 2007-06-12, 9:55 pm |
| If you look at implementations and not the "theoretical", then I think it is
fair (or at least "medium fair") to say that the game has been over AT LEAST
since the '02 Standard was approved 5 years ago. I have (recently) heard a
RUMOR that Hitachi has implemented much (still not all) of the '02 Standard, but
there has been little or no interest from most vendors in implementing what was
"new" in the '02 Standard. Even when it comes to "OO" - most of the vendors
that had/have ANY OO support have used some pre-2000 version (or in IBM's case
part of a version) and have shown little interest in the Standard.
As far as "optional" goes, I think the COBOL user-base and the implementors have
treated the entire '02 Standard as "optional" (and close to irrelevant).
I, personally, will be VERY interested in what comes out of the WG4 meeting in
June and the J4 meeting in August. I don't personally see how (even through
rose-colored glasses) they can claim that the Standard is pertinent or being
enhanced, but I have been surprised before by what they can and do say -
regardless of the facts that are facing them.
--
Bill Klein
wmklein <at> ix.netcom.com
"James J. Gavan" <jgavandeletethis@shaw.ca> wrote in message
news:avnbi.16026$1i1.1606@pd7urf3no...
> William M. Klein wrote:
> And your latest thread on 'More on COBOL Standards' ??? Let's be realistic,
> the penny has dropped and reluctantly they realize COBOL is a sturdy horse
> which is no longer a contender for the Kentucky/Derby or otherwise.
>
> Seems to me what's left on J4, i.e. Hitachi (indirectly), IBM - I'm losing
> track of who now doesn't show. Assuming Micro Focus was still in the game,
> that makes three (which includes the taken over AcuCorp). And for all the
> effort others might put in it's down to the compiler vendors what truly
> happens.
>
> Suppose WG4 conclude, 'We reject J4's suggestions of curtailing/optionalizing
> COBOL'. It doesn't matter a cuss what the formalized procedures are, if those
> three above retort, 'Get knotted. We ain't spending the money on unnecessary
> enhancesmets, even if the Mongolian Republic is pleading for them via WG4',
> game's over.
>
> Quite naturally Bill, you are looking at the options with your background
> steeped in the formalized procedures. This Joe is just trying to visualize the
> real world.
>
>
> Jimmy
| |
| Pete Dashwood 2007-06-12, 9:55 pm |
|
"William M. Klein" <wmklein@nospam.netcom.com> wrote in message
news:iNnbi.97951$dg3.47335@fe10.news.easynews.com...
> If you look at implementations and not the "theoretical", then I think it
> is fair (or at least "medium fair") to say that the game has been over AT
> LEAST since the '02 Standard was approved 5 years ago.
I agree. The war was lost as far as Standards were concerned when there was
nothing by 1995.
(To take 10 years for the next release of a standard, plus all the bickering
and inefficiency that surrounded it, meant serious doubts were raised as to
the the role of ANSI. That the process still dragged on another 7 years
after that, simply meant the process lost all credibilility.) The people who
WERE still interested in a viable standards process simply had the
enthusiasm knocked out of them. I don't blame the vendors who walked. (In
fact, I believe it was a sensible businsess decision and exactly what I wuld
have done myself if I had that responsibility, and it had become apparent it
was going nowhere and no amount of effort was going to fix it...)
The only reason there is any kind of reasonable COBOL available today is
because the major vendors have a committment to their respective customer
bases, and there is no clear exit strategy for them anyway (apart from
simply ceasing to trade, which is not an option for most companies...) I see
their dilemma as follows:
1. IBM have a major legacy support committment that must not be eroded.
(Nevertheless, it IS being eroded and IBM will be looking to steer it
towards SOA and slowly move the COBOL base to Java.)
ASSESSMENT: IBM will stop supporting COBOL altogether by 2015. (It will
still be available on a "use at your own risk" basis.) Don't expect major
enhancements, implementation of new standards, or rewrites from them in the
interim. The move towards SOA and Object technology will gather momentum.
They will implement bridges to DotNET, and SOA and Web Services will be a
major part of their long term strategy. It will be effective and we can
expect to see them around for the foreseeable future.
2. MicroFocus saw an opportunity to offer mainframe development off the
mainframe, and this has been good business for them. Unfortunately, they
have been beset by internal problems and change of ownership of the company
and these upheavals have meant that their core business has become tenuous.
Added to this is a decline in mainframe COBOL development and general
dissatisfaction with the viability of COBOL into the future (not helped by
the Standards debacle). They are moving now to get COBOLWeb Enabled and
persuade people this is good. (It is...SOA is an excellent approach and Web
Services are a major part of it. Unfortunately, COBOL is not the best tool
for this job and even with the new lease of life that people migrating COBOL
to the Web brings, it is still a limited future for the Language., and, by
association, the MF implementation of it.
ASSESSMENT: MicroFocus will pick up most of the current COBOL base that is
running on Client/Server. They will then be taken over by (or sell out
to...) IBM (or another major player that has eyes on their customer base),
within 8 years.
3. Fujitsu COBOL, although an excellent implementation, doesn't have a
significant support base outside Japan. Unfortunately, they shot themselves
in the foot with abysmal support and stupidly restrictive licensing. I don't
know of any major players who are using Fujitsu COBOL (I think it's a pity;
it's a great product).
ASSESSMENT: Fujitsu will withdraw support for all COBOL except DotNET COBOL
for DotNet, within 4 years. As the realization starts to bite that even this
excellent implementation is not going to catch on, (people using the DotNET
platform are not going to pay substantial sums to do so in COBOL, when they
can do it in C# for free) we can expect Fujitsu to drop COBOL completely,
within 7 years. They are an immensly wealthy company and have been able to
afford to stay with COBOL for the long haul, even though their market
penetration has not been as effective as MicroFocus'. However, the Japanese
Management, even though they take a long term view, are not stupid. They
will not pour good money after bad, to flog a dead horse.
I have (recently) heard a
> RUMOR that Hitachi has implemented much (still not all) of the '02
> Standard, but there has been little or no interest from most vendors in
> implementing what was "new" in the '02 Standard. Even when it comes to
> "OO" - most of the vendors that had/have ANY OO support have used some
> pre-2000 version (or in IBM's case part of a version) and have shown
> little interest in the Standard.
>
This corroborates what I have said above.
> As far as "optional" goes, I think the COBOL user-base and the
> implementors have treated the entire '02 Standard as "optional" (and close
> to irrelevant).
>
Given the circumstances in the Real World, who can blame them?
> I, personally, will be VERY interested in what comes out of the WG4
> meeting in June and the J4 meeting in August. I don't personally see how
> (even through rose-colored glasses) they can claim that the Standard is
> pertinent or being enhanced, but I have been surprised before by what they
> can and do say - regardless of the facts that are facing them.
>
I don't think it really matters what they say any more. As far as COBOL
goes, they are an irrelevance, and as time goes by they simply continue to
reinforce this view. I cannot imagine any of the remaining vendors taking
them seriously. (Certainly not supporting them with cash or resources, over
a significant period of time. Look for more withdrawals in the next couple
of years...)
The marketplace is merciless and it sets direction as it always has. Vendors
will implement what is required by the marketplace, and gives them a
competitive edge, but even the long term customer base is coming to the
realization that COBOL is not viable for the future. (Vendors have known
this for some time now; but there is still money in the COBOL market. You
don't seriously think IBM would be promoting WebSphere, Java, and Linux, if
they believed the future was with COBOL....?)
I believe, (though I certainly can't prove it), that there are many IT
directors who would love to move away from their traditional Waterfall
development methodology and COBOL Language. The reason they don't, is
because it is by no means clear where they should go, or how they might get
there. (Object technology is a mystery to many of the old school; they
don't understand it and don't trust it, but, as I have noted before, once
the landslide starts, the pebbles don't get to vote.)
As the marketplace applies more pressure to the Business in general, and as
the cost of writing code line by line (and maintaining it) continues to rise
inexorably, they will HAVE to do something. Unless the vendors can show them
a pathway to something more viable, they (the vendors) will lose that
business.
Resources in the form of technical expertise are going to be at a premium
within vendor organizations. I can't see valuable resources being released
to sit on a committee that achieves nothing and has no relevance beyond
furthering its own existence.
Given that ANSI incompetence has played a considerable part in the demise of
COBOL, it seems to me that they are getting exactly what they deserve.
[color=darkred]
> --
> Bill Klein
> wmklein <at> ix.netcom.com
> "James J. Gavan" <jgavandeletethis@shaw.ca> wrote in message
> news:avnbi.16026$1i1.1606@pd7urf3no...
Compatible with what, Jimmy? If you mean interoperable across platforms,
that hasn't been true for many years now. If you mean that COBOL written for
one compiler should run unchanged on ANY compiler that has NEVER been true..
even when compilers complied with the standard.
Compatibility of source code is no longer important; compatibility of OBJECT
code (in the form of IL) is what matters. If C++, Java, COBOL, VB.Net and C#
(plus a few others) can all produce objects and Classes that can be invoked
from ANY of the other languages and run on ANY of the other platforms
(Windows, Mac, Linux, even z/OS), THEN you have something worth getting
excited about :-)
The DotNET framework delivers this, and that's why the only relevant
versions of COBOL for the next few years will be the ones that support this
(apart from mainframe COBOL which will remain aloof for the meantime...) The
message just isn't getting through here that Source Code is no longer King.
The source code of a given system could be in 4 different languages; each
one best suited to produce the required object class. It really doesn't
matter; all that is important is the encapsulated functionality.
Viewed from this perspective, the bloody COBOL standards committees are even
less relevant than I think they are...:-)
[color=darkred]
It is a Clydesdale suited to pulling a plough or carrying a knight in full
armour. A beautiful animal, worthy of respect. Unfortunately, today's world
uses tractors and missiles...
[color=darkred]
Absolutely. Spot on.
So why would anyone volunteer for the committee? Or even better, pay a
non-trivial sum of money to be on it? One thing if their company is backing
them and require them to be there protecting the corporate interest (and I
don't see that lasting too much longer...), but what about the one's whose
companies are NOT backing them?
Either they are dedicated people who should command our respect, but who are
being badly mismanaged, or bound up by red tape, to the point where they
achieve nothing; or it is simply an ego-trip... "Look Ma, I hold the future
of COBOL in my hands..."
Either way, it is an irrelevance.
[color=darkred]
It's a pity a few more on the committee weren't looking more carefully at
the real world...
Pete.
|
|
|
|
|