Cobol - J4 (INCITS aka ANSI COBOL) Committee update

Author

J4 (INCITS aka ANSI COBOL) Committee update

William M. Klein

2006-05-08, 9:55 pm

Subject should have had "J4" not "J$" (Freudian slip???)

--
Bill Klein
wmklein <at> ix.netcom.com
"Roger While" <simrw@sim-basis.de> wrote in message
news:e3p1im$6lu$00$1@news.t-online.com...
> Hmm, Link to INICITS/RD-3 within the doc doesn't appear to work.
>
> Roger While
>
> "William M. Klein" <wmklein@nospam.netcom.com> schrieb im Newsbeitrag
> news:YLS7g.431109$7i1.123759@fe06.news.easynews.com...
>
>

William M. Klein

2006-05-09, 6:55 pm

Actually, unlike some, I am ENCOURAGED by this. It seems to me (and I *know*
others will disagree - and some agree) that the current J4 process has totally
lost contact with what can/should be done with COBOL. Work on the '08 Standard
continues (see Chuck's separate note) with (as far as I can tell) ZERO attention
to the fact that there are no implementations of the '02 Standard. Furthermore,
the "quick and tentative" TR process has bogged down (just as the previous
amendment process did) and J4 seems to be spending more time (and resources) on
the XML TR than on the '08 Standard OR interpretations of the '02 Standard.

****

Once upon a time (and long ago) there were two groups:
CCC (CODASYL COBOL Committee) - that created a "futuristic" (not implemented)
JOD
and
X3J4 a "technical committee" that worked with real implementable standards and
its interpretation

When the 2 merged (under the X3J4 - soon after J4) title, the INTENT was to
bring these two closer together - with the thought this would be closer to what
was actually wanted and in use - and to provide "timely" enhancements.

What ended up was an "ivory tower" delayed specification that must be useful for
someone or some organization - but certainly doesn't (to me) seem to have any
relationship to current (or future) COBOL usage.

When (if) INCITS "closes down" J4, it would be my hope (not necessarily
expectation) that COBOL vendors (those remaining) will separately or together
support a programming language meeting the needs of their customers (current and
future).

--
Bill Klein
wmklein <at> ix.netcom.com
"Frank Swarbrick" <Frank.Swarbrick@efirstbank.com> wrote in message
news:4cbs69F14rcseU2@individual.net...
> William M. Klein<wmklein@nospam.netcom.com> 05/08/06 8:03 PM >>>
>
> consider
> most
>
> Wow, that certainly is discouraging, though not all that surprising. I'm
> guessing you aren't going to volunteer?
>
> Frank
>
>
> ---
> Frank Swarbrick
> Senior Developer/Analyst - Mainframe Applications
> FirstBank Data Corporation - Lakewood, CO USA

Pete Dashwood

2006-05-10, 6:55 pm

"William M. Klein" <wmklein@nospam.netcom.com> wrote in message
news:BB88g.367116$4P2.306208@fe03.news.easynews.com...
> Actually, unlike some, I am ENCOURAGED by this. It seems to me (and I
> *know* others will disagree - and some agree) that the current J4 process
> has totally lost contact with what can/should be done with COBOL.

Hear! Hear! They lost it years ago.

> Work on the '08 Standard continues (see Chuck's separate note) with (as
> far as I can tell) ZERO attention to the fact that there are no
> implementations of the '02 Standard. Furthermore, the "quick and
> tentative" TR process has bogged down (just as the previous amendment
> process did) and J4 seems to be spending more time (and resources) on the
> XML TR than on the '08 Standard OR interpretations of the '02 Standard.

I would expect nothing less. The XML facility is a stupid and pointless
addition which is simply unnecessary. Even if it were, there are many things
deserving of higher priority attention (like getting the 02 standard
implemented and ironed out).

No lessons have been learned; no efort to communicate with the tools
available in this day and age to find out what is needed and wanted from the
user base. Nanny knows best.

Oops... I'm getting myself

with someone who cares...

>
> ****
>
> Once upon a time (and long ago) there were two groups:
> CCC (CODASYL COBOL Committee) - that created a "futuristic" (not
> implemented) JOD
> and
> X3J4 a "technical committee" that worked with real implementable
> standards and its interpretation
>
> When the 2 merged (under the X3J4 - soon after J4) title, the INTENT was
> to bring these two closer together - with the thought this would be closer
> to what was actually wanted and in use - and to provide "timely"
> enhancements.
>
> What ended up was an "ivory tower" delayed specification that must be
> useful for someone or some organization - but certainly doesn't (to me)
> seem to have any relationship to current (or future) COBOL usage.

Beautifully expressed and right on the button.:-)
>
> When (if) INCITS "closes down" J4, it would be my hope (not necessarily
> expectation) that COBOL vendors (those remaining) will separately or
> together support a programming language meeting the needs of their
> customers (current and future).

The only reason you can still buy COBOL is because the vendors have done
that right through. Unfortunately, the high cost of producing the software,
coupled wih some short-sighted marketing in some quarters and dreadful
support in others, has simply nailed the coffin shut. Support for the legacy
will keep it going for a while, but it is no longer viable as a language of
choice for modern development. IBM are encouraging Java/Websphere, and they
can be considered the last bastion of Fortress COBOL. I believe the other
players will either withdraw or diversify, or go bust if they are a "one
trick COBOL pony" shop, within the next few years.

Yes, there were/are many factors that contributed to the death of COBOL, but
putting it into the hands of a totally useless and out-of-touch standards
committee that simply wrapped it in red tape while deciding which resort to
hold their next meeting at, has to have been a major one.

There were some good people who put in a lot of effort, but nothing valuable
was achieved. These thngs are judged by results. If you have no
answerability or responsibility, but simply set (and move) your own
schedule, it is hardly surprising that nothing gets done. It smacks of
appalling Management. There was no attempt to question or change the
antiquated procedures that had everybody hogtied, no communication with the
User Base, no reporting of progress. It was like:"You'll get it when we've
done it and be thankful..."

The demise of this committee is a breath of fresh air.

Pete.

James J. Gavan

2006-05-10, 6:55 pm

Pete Dashwood wrote:
> "William M. Klein" <wmklein@nospam.netcom.com> wrote in message
> news:BB88g.367116$4P2.306208@fe03.news.easynews.com...
>
>
>
> Hear! Hear! They lost it years ago.

And lost it even more recently ! I haven't commented before but take a
quick gander at :-

http://www.cobolportal.com/j4/index.asp?bhcp=1

The second group of documents "J4 - Standing Documents"

J4-02-0102 - Technical Report Styles - author, our buddy Thane

Pure bureaucratic bullshit !

Jimmy

Pete Dashwood

2006-05-11, 6:55 pm

"James J. Gavan" <jgavandeletethis@shaw.ca> wrote in message
news:wYp8g.146805$WI1.11147@pd7tw2no...
> Pete Dashwood wrote:
>
> And lost it even more recently ! I haven't commented before but take a
> quick gander at :-
>
> http://www.cobolportal.com/j4/index.asp?bhcp=1
>
> The second group of documents "J4 - Standing Documents"
>
> J4-02-0102 - Technical Report Styles - author, our buddy Thane
>
> Pure bureaucratic bullshit !
>
> Jimmy

Sorry Jimmy,

I didn't follow the link, but appreciate your posting it. If I go there I'll
just get wound up again and it really serves no useful purpose.

I've got to try and withold comment, and seeing more 'bureaucratic bullshit'
isn't going to assist that :-)

It is fair to say that I am heartsickened by the waste and the lost
opportunities. But it is water under the bridge.

Gotta get over it...:-)

Pete.

2006-05-14, 3:55 am

William M. Klein wrote:
> Actually, unlike some, I am ENCOURAGED by this. It seems to me (and I *know*
> others will disagree - and some agree) that the current J4 process has totally
> lost contact with what can/should be done with COBOL. Work on the '08 Standard
> continues (see Chuck's separate note) with (as far as I can tell) ZERO attention
> to the fact that there are no implementations of the '02 Standard. Furthermore,
> the "quick and tentative" TR process has bogged down (just as the previous
> amendment process did) and J4 seems to be spending more time (and resources) on
> the XML TR than on the '08 Standard OR interpretations of the '02 Standard.
>
> ****
>
> Once upon a time (and long ago) there were two groups:
> CCC (CODASYL COBOL Committee) - that created a "futuristic" (not implemented)
> JOD
> and
> X3J4 a "technical committee" that worked with real implementable standards and
> its interpretation
>
> When the 2 merged (under the X3J4 - soon after J4) title, the INTENT was to
> bring these two closer together - with the thought this would be closer to what
> was actually wanted and in use - and to provide "timely" enhancements.
>
> What ended up was an "ivory tower" delayed specification that must be useful for
> someone or some organization - but certainly doesn't (to me) seem to have any
> relationship to current (or future) COBOL usage.
>
> When (if) INCITS "closes down" J4, it would be my hope (not necessarily
> expectation) that COBOL vendors (those remaining) will separately or together
> support a programming language meeting the needs of their customers (current and
> future).
>
Bill, no one knows better than you that this is essentially what caused
the '1980 Standard' to become the '1985 Standard'... All the idealism
and attempts by those to have no code to upgrade, nor interest in the
practical use of the language almost killed the process 20 years ago.
This time, it looks like they were successful.

Your presentations at SHARE about the state of the standard were well
attended and well received. Users are interested in the future the
COBOL language. But, they also did not start yelling and demanding that
the entire standard be implemented as happened back in 1985. They saw a
few things that they liked, and I suspect that IBM will respond to those
needs. I would be very surprised to ever see an IBM compiler provide
the whole 2002 standard.

While I have not kept a close eye on the implementation of the latest
FORTRAN standard, I've not heard of a mass move to upgrade the FORTRAN
market place. It appears that 2002 COBOL has cloned the FORTRAN process
and the results. As with any other business, lose sight of the
consumers' needs and lose the consumer.

So, I am also encouraged by what may happen with the 'standardization'
effort. There may not be a new 'standard' but the users will get what
the need.

Just my US$0.02 worth... [That the price, the cost is zero! ;-) ]
Carl

Chuck Stevens

2006-05-19, 6:55 pm

I just returned from the most recent meeting of INCITS/J4, and rumors of its
(at least immediate) demise are premature. A member of the committee has
agreed to accept the responsibility of chairperson, and the appropriate
communications to this effect are in preparation.

I remain unconvinced that there is absolutely no value to a language
standard unless each and every implementation conforms to each and every
provision of that standard. I do agree that having some sort of mechanism
whereby vendors are required to provide fully-conforming compilers in some
context -- e.g., the US Defense Department up until sometime in the
1980's -- contributes mightily to the availability of conforming products.
So does the availability of some sort of mechanism whereby compilers can be
tested to verify that they do indeed conform to that standard.

And I do not believe that the fact that neither of these situations obtains
today is any sort of demonstration of a failure on the part of J4 to do its
job.

I agree that some sort of conformance enforcement would be appropriate. I
do not agree that it is the job of the half-dozen-or-so members of J4 (or
for that matter the National Bodies represented on WG4) to produce, out of
their goodness of their hearts, a complete conformance-validation suite. As
I recall it, J4 has gone on record as welcoming any effort in that
direction.

Given the fact that the current standard, as well as the draft for the next
one, is first and foremost an *international* standard, I am of the opinion
that such a validation suite ought to come from somewhere outside the US,
particularly given the fact that the details of the standard are delegated
to a primarily-American committee (INCITS/J4) by ISO/IEC.

I think the actual production of such a suite would make an excellent
project for an academic institution somewhere, particularly one that
maintains some interest in COBOL (e.g., the Free University of Amsterdam).
Whether the existing suite (which as I understand matters hasn't been
updated to include the Intrinsic Function amendment of 1989) could serve as
a reasonable and appropriate base for a validation suite for the 2002
standard is doubtful in my opinion (I have studied a number of the old "GSA"
tests -- both for the '74 and the '85 standards -- in some depth, and remain
unconvinced of their thoroughness, or of their value as examples of what
today passes for good COBOL!).

Finding someone to fund the management of such a project is probably the
sticking point here.

Any of you guys interested in writing a validation suite for the 2002
standard as a contribution to the world of computing?

-Chuck Stevens

Roger While

2006-05-19, 6:55 pm

Well, of course, as always was, and, it appears always will be,
the "other side of the ocean" is absolutely ignored/forgotten.
(Or perhaps "the rest of the world")

Roger

"Chuck Stevens" <charles.stevens@unisys.com> schrieb im Newsbeitrag
news:e4kmjt$29rh$1@si05.rsvl.unisys.com...
>I just returned from the most recent meeting of INCITS/J4, and rumors of
>its (at least immediate) demise are premature. A member of the committee
>has agreed to accept the responsibility of chairperson, and the appropriate
>communications to this effect are in preparation.
>
> I remain unconvinced that there is absolutely no value to a language
> standard unless each and every implementation conforms to each and every
> provision of that standard. I do agree that having some sort of mechanism
> whereby vendors are required to provide fully-conforming compilers in some
> context -- e.g., the US Defense Department up until sometime in the
> 1980's -- contributes mightily to the availability of conforming products.
> So does the availability of some sort of mechanism whereby compilers can
> be tested to verify that they do indeed conform to that standard.
>
> And I do not believe that the fact that neither of these situations
> obtains today is any sort of demonstration of a failure on the part of J4
> to do its job.
>
> I agree that some sort of conformance enforcement would be appropriate. I
> do not agree that it is the job of the half-dozen-or-so members of J4 (or
> for that matter the National Bodies represented on WG4) to produce, out of
> their goodness of their hearts, a complete conformance-validation suite.
> As I recall it, J4 has gone on record as welcoming any effort in that
> direction.
>
> Given the fact that the current standard, as well as the draft for the
> next one, is first and foremost an *international* standard, I am of the
> opinion that such a validation suite ought to come from somewhere outside
> the US, particularly given the fact that the details of the standard are
> delegated to a primarily-American committee (INCITS/J4) by ISO/IEC.
>
> I think the actual production of such a suite would make an excellent
> project for an academic institution somewhere, particularly one that
> maintains some interest in COBOL (e.g., the Free University of Amsterdam).
> Whether the existing suite (which as I understand matters hasn't been
> updated to include the Intrinsic Function amendment of 1989) could serve
> as a reasonable and appropriate base for a validation suite for the 2002
> standard is doubtful in my opinion (I have studied a number of the old
> "GSA" tests -- both for the '74 and the '85 standards -- in some depth,
> and remain unconvinced of their thoroughness, or of their value as
> examples of what today passes for good COBOL!).
>
> Finding someone to fund the management of such a project is probably the
> sticking point here.
>
> Any of you guys interested in writing a validation suite for the 2002
> standard as a contribution to the world of computing?
>
> -Chuck Stevens
>

2006-05-19, 6:55 pm

In article <e4kmjt$29rh$1@si05.rsvl.unisys.com>,
Chuck Stevens <charles.stevens@unisys.com> wrote:

[snip]

>Any of you guys interested in writing a validation suite for the 2002
>standard as a contribution to the world of computing?

Leaving aside, for the nonce, my utter inability to do such a thing...
must... resist... quoting... Drill... Sergeant... wisdom... about...
volunteering...

DD

2006-05-19, 9:55 pm

In article <e4kn1k$6ko$00$1@news.t-online.com>,
Roger While <simrw@sim-basis.de> wrote:
>Well, of course, as always was, and, it appears always will be,
>the "other side of the ocean" is absolutely ignored/forgotten.
>(Or perhaps "the rest of the world")

I'm sorry, maybe I missed something... where was it noted that
participation from *anyone* was ignored? Who said 'Sorry, Siemens, no
room for you' or 'Back off, Groupe Bull'? I, personally, may have my
own reservations about volunteering... but to say 'these people have
been ignored/forgotten' without there being any indication thus far that
'these people' attempted to participate seems, at least to me, to be
moderately deficient in a matter of fact or two... but I speak from the
standpoint of my own publicly-admitted ignorance.

Who was turned down, when and by whom, please?

DD

Pete Dashwood

2006-05-20, 3:55 am

"Chuck Stevens" <charles.stevens@unisys.com> wrote in message
news:e4kmjt$29rh$1@si05.rsvl.unisys.com...
>I just returned from the most recent meeting of INCITS/J4, and rumors of
>its (at least immediate) demise are premature. A member of the committee
>has agreed to accept the responsibility of chairperson, and the appropriate
>communications to this effect are in preparation.
>
I wish him/her well. It wouldn't be hard to do a better job than the
previous chair.

> I remain unconvinced that there is absolutely no value to a language
> standard unless each and every implementation conforms to each and every
> provision of that standard.

OK, but would you agree that the credibility of the standard is improved if
a full implementation of it CAN be shown?

>I do agree that having some sort of mechanism whereby vendors are required
>to provide fully-conforming compilers in some context -- e.g., the US
>Defense Department up until sometime in the 1980's -- contributes mightily
>to the availability of conforming products. So does the availability of
>some sort of mechanism whereby compilers can be tested to verify that they
>do indeed conform to that standard.
>
> And I do not believe that the fact that neither of these situations
> obtains today is any sort of demonstration of a failure on the part of J4
> to do its job.

I would agree that enforcement may not be within the scope of the committee.

>
> I agree that some sort of conformance enforcement would be appropriate. I
> do not agree that it is the job of the half-dozen-or-so members of J4 (or
> for that matter the National Bodies represented on WG4) to produce, out of
> their goodness of their hearts, a complete conformance-validation suite.
> As I recall it, J4 has gone on record as welcoming any effort in that
> direction.

Here we differ. I think they SHOULD have a validation suite. Anybody who
produces a standard should have an officially approved baseline that
determines whether an implementation is standard conforming or not. It has
nothing to do with the 'goodness of their hearts'; it has to do with a
yardstick that establishes credibility for the work they do and, as such, is
part of the job they have taken on.

>
> Given the fact that the current standard, as well as the draft for the
> next one, is first and foremost an *international* standard, I am of the
> opinion that such a validation suite ought to come from somewhere outside
> the US, particularly given the fact that the details of the standard are
> delegated to a primarily-American committee (INCITS/J4) by ISO/IEC.
>
They could certainly call on international affiliated bodies for help in
producing it, but at the end of the day it should be part of the package.
What right would anybody OTHER than the standards committee have to produce
a yardstick?

> I think the actual production of such a suite would make an excellent
> project for an academic institution somewhere, particularly one that
> maintains some interest in COBOL (e.g., the Free University of Amsterdam).
> Whether the existing suite (which as I understand matters hasn't been
> updated to include the Intrinsic Function amendment of 1989) could serve
> as a reasonable and appropriate base for a validation suite for the 2002
> standard is doubtful in my opinion (I have studied a number of the old
> "GSA" tests -- both for the '74 and the '85 standards -- in some depth,
> and remain unconvinced of their thoroughness, or of their value as
> examples of what today passes for good COBOL!).
>
> Finding someone to fund the management of such a project is probably the
> sticking point here.
>
As always. It needs to be sponsored by the same people who are requesting a
standard.

At least it would be money well spent for a change.

> Any of you guys interested in writing a validation suite for the 2002
> standard as a contribution to the world of computing?
>

Out of the goodness of our hearts, you mean :-)?

Spend a deal of time writing a benchmark for a language which is past its
sell-by (and being kept alive by companies who have an established user base
in it, and are not prepared to grasp the nettle and move on), for no gain
other than the doubtful kudos of having 'contributed to the world of
computing'?

Sure, I'll do that... about the same time I start giving my expertise away
for free... :-) Helping learners and interested parties is one thing;
working for free for corporations who aren't prepared to finanace what they
need done, is something else again.

Procedural programming has had it. Get over it. A more responsible approach,
(and one being gently supported by IBM) is to provide a pathway for the
existing cusromer base to refactor their investment and move off COBOL.

Pete.

Chuck Stevens

2006-05-22, 6:55 pm

"Pete Dashwood" <dashwood@enternet.co.nz> wrote in message
news:4d7nd9F19bpmsU1@individual.net...

> OK, but would you agree that the credibility of the standard is improved
> if a full implementation of it CAN be shown?

ABSOLUTELY!!!

But if it isn't, implementing features *from* the standard is, in my view, a
good reason for the standard to exist. *Pace* Bill, the fact that the
standard specifies "<>" instead of "/=" is an incentive for an implementor
to favor the former over the latter, whether or not that same vendor allows
very long nonnumeric literals.

> I would agree that enforcement may not be within the scope of the
> committee.

Thanks, I wasn't sure about that!

> Here we differ. I think they SHOULD have a validation suite.

Don't get me wrong, I think such a suite should exist.
> Anybody who produces a standard should have an officially approved
> baseline that determines whether an implementation is standard conforming
> or not. It has nothing to do with the 'goodness of their hearts'; it has
> to do with a yardstick that establishes credibility for the work they do
> and, as such, is part of the job they have taken on.

The validation suites for '74 and '85 were *not* produced or maintained by
the standards committee. I'm not sure that in and of itself was a Bad
Thing.

> They could certainly call on international affiliated bodies for help in
> producing it, but at the end of the day it should be part of the package.
> What right would anybody OTHER than the standards committee have to
> produce a yardstick?

Why would a standards committee NOT have the authority to *authorize* the
production of such a yardstick? Why would having the authority to
*authorize* the production of the yardstick necessarily preclude the ability
of the committee to retain the right to verify the correctness of the
yardstick?

> As always. It needs to be sponsored by the same people who are requesting
> a standard.

Not clear that it needs to be *financially* sponsored by, because the
financial resources of J4 are pretty much limited to the meeting fees
collected to pay for the meeting facilities with some in reserve to cover
the Project Editor's costs and the costs of hosting WG4 meetings in the US.
We don't have much to offer in that regard.

But I have no problem *at all* with the committees *authorizing* the
production of such a suite, regardless of who pays for it, so long as the
committees retain right of review.

> Out of the goodness of our hearts, you mean :-)?

So you have seemed to require of the other representatives on J4; why not
you? ;-)

> Procedural programming has had it. Get over it. A more responsible
> approach, (and one being gently supported by IBM) is to provide a pathway
> for the existing cusromer base to refactor their investment and move off
> COBOL.

There's also a significant movement to retain COBOL applications and make
them more accessible and more presentable in "modern" environments. No
doubt there's some significant movement away from COBOL, but there's also
been some backlash in that area (and I can't cite specific details).

-Chuck Stevens

Sponsored Links