Prev: COBOL Error Handling (was: What MF says about ROUNDED(was:Cobol Myth Busters
Next: Wildcat COBOL
From: Anonymous on 31 Oct 2007 19:01
In article <1193840492.880726.4060(a)z9g2000hsf.googlegroups.com>,
Alistair <alistair(a)ld50macca.demon.co.uk> wrote:
>On 31 Oct, 00:16, docdw...(a)panix.com () wrote:
>> In article <5opm2vFnnl6...(a)mid.individual.net>,
>>
>> Pete Dashwood <dashw...(a)removethis.enternet.co.nz> wrote:
>>
>> ><docdw...(a)panix.com> wrote in messagenews:fg6u2d$qvk$1(a)reader1.panix.com...
>>
>> [snip]
>>
>> >> All in all, it is usually a good thing to remember what Machiavelli had to
>> >> say about the introduction of new systems
>> >> <http://www.gutenberg.org/files/1232/1232.txt>
>>
>> >> --begin quoted text:
>>
>> >> And it ought to be remembered that there is nothing more difficult to take
>> >> in hand, more perilous to conduct, or more uncertain in its success, then
>> >> to take the lead in the introduction of a new order of things.
>>
>> >Only if you're a sissy. REAL Men embrace change and have no problem with
>> >being responsible for it. :-)
>>
>> Just like military officers have no problems leading their men over the
>> tops of the trenches... and the Gallipoli-like results which may ensue.
>>
>
>Gallipoli was a disaster because of the prevarication of the officers
>who failed to realise the dangers of being sniped upon by Turks on
>surrounding hills. So they leisurely organised cricket matches rather
>than marching off of the beachhead.

I was not there, of course, and rely on what I can recall reading in
various places... but what I recall is echoed by
http://en.wikipedia.org/wiki/Battle_of_Gallipoli#August_offensive :

--begin quoted text:

The landing at Suvla Bay was only lightly opposed but the British
commander, Lieutenant-General Sir Frederick Stopford, had so diluted his
early objectives that little more than the beach was seized. Once again
the Turks were able to win the race for the high ground of the Anafarta
Hills thereby rendering the Suvla front another case of static trench
warfare.

[snip]

The New Zealanders held out on Chunuk Bair for two days before relief was
provided by two New Army battalions from the Wiltshire and Loyal North
Lancashire Regiments. A massive Turkish counter-attack, led in person by
Mustafa Kemal, swept these two battalions from the heights.

Of the 760 men of the New Zealanders' Wellington Battalion who reached the
summit, 711 were casualties.

[snip]

Following the landing at Suvla Bay, casualties among the opposing armies
were particularly high, and the hot and humid weather made the stench of
bodies especially nauseating. A day's truce was arranged to facilitate the
removal of the dead and wounded; this momentary contact led to a strange
camaraderie between the armies much like the Christmas truce of 1914.

--end quoted text

An utter horror.

DD
From: Ranger on 4 Nov 2007 17:26
On Oct 30, 4:38 am, docdw...(a)panix.com () wrote:
> As an extension of an earlier thread regarding a COBOL-to-Java
> conversion... some meanderings.
>


I did want to provide a summary of a response that I provided to DD
privately, but maybe the group could benefit as well.

I own a technology company that provides an automated migraton for
COBOL based applications. Most of our clients have IDMS, Datacom, IMS,
VSAM and Adabas and are looking to migrate to DB2, Oracle or SQL
Server.

I am hearing from our clients that they need to migrate for
predominately two reasons - we getting off of the mainframe or we feel
at risk running some of the older databases/files. For those looking
to stay on the mainframe we have been migrating them to DB2 and rarely
do we hear a request for JAVA. Regarding clients that are migrating
off of the mainframe we are hearing an increased amount of requests to
convert part or all of the COBOL application to JAVA. We have one
client that said "I want to get off the mainframe, convert to JAVA and
outsource to India". Other client comments are "I had to sell JAVA to
ge the project funded".

What we are finding when clients actually migrate, they begin to see
the value in the COBOL based applications. They have years of
experience, and a temendous amount of business logic that no longer
has a subject matter expert.

Several of our projects started out with the intention to migrate the
COBOL applicatons to JAVA, and then they realize the cost and risk,
and usually determine the "customer facing, high value" functions will
be written in JAVA. At the end of the day, the application remains 95%
COBOL.

I am responding to a client RFP this month, the original conversation
was "We have to get off the maniframe, its a CIO directive. We thought
that this would be a good time to go to the board one time and ask for
the funding to get off the mainframe and convert to JAVA". My original
comment to them was to migrate the COBOL mainframe application to
COBOL distributed platform and take inventory. SOA and web services
can be initiated through COBOL and you can leverage your business
logic embedded in your existing programs. The training requirements
for coming off of the mainframe will be limited to environmental
issues, you can protect the productivity and experience of your staff
as well. At first they did not like that thought, and now, 60 days
later, they are asking for three approaches. The first is COBOL to
COBOL, the second is a hybrid, with all the heavy lifting being done
by COBOL, and the online/high-value programs being JAVA, and the third
option is all JAVA.

I hope this group benefits by some of the feedback that I have been
hearing. We have migrated over 50M lines of COBOL code the past few
years, with less than 2% going to JAVA. It is however, being requested
as an option on almost all of our "getting off the mainframe" RFP's.

Ranger


From: Anonymous on 5 Nov 2007 05:24
In article <1194215218.855930.75200(a)d55g2000hsg.googlegroups.com>,
Ranger <tarr(a)tiburontech.com> wrote:

>I am hearing from our clients that they need to migrate for
>predominately two reasons - we getting off of the mainframe or we feel
>at risk running some of the older databases/files. For those looking
>to stay on the mainframe we have been migrating them to DB2 and rarely
>do we hear a request for JAVA.

Outside of what is said in The Press about such matters... I wonder what
gives them this 'feeling'. Are their systems maxing out, are they having
trouble hiring folks for the pittance salaries they offer, are their own
people incapable of being trained in these technologies... all
possibilities
and others, I'm sure, but asking those questions might deny your own
organisation some business.

'Oh, you don't need to migrate anything... all you need is to train some
folks and then slap on some golden handcuffs to keep them on the job!'

>Regarding clients that are migrating
>off of the mainframe we are hearing an increased amount of requests to
>convert part or all of the COBOL application to JAVA. We have one
>client that said "I want to get off the mainframe, convert to JAVA and
>outsource to India".

To which one might respond 'And you believe that denying yourself control
of the resource of IT ownership would be of benefit... how?'

>Other client comments are "I had to sell JAVA to
>ge the project funded".

Welcome to the World of Business, aye... 'everyone's talking about
'longer, lower, wider and with port-holes, we can't get left behind!'

>
>What we are finding when clients actually migrate, they begin to see
>the value in the COBOL based applications. They have years of
>experience, and a temendous amount of business logic that no longer
>has a subject matter expert.

Oooooooooo... don't get me started on the Death of Business Analyst.
'COBOL is self-documenting so we only need programmers, not analysts...
but keep those programmers out of the meetings, they tend to be
distracting with all this 'logic' stuff.'

>
>Several of our projects started out with the intention to migrate the
>COBOL applicatons to JAVA, and then they realize the cost and risk,
>and usually determine the "customer facing, high value" functions will
>be written in JAVA. At the end of the day, the application remains 95%
>COBOL.

On the front end, nice screens, drop-down menus, easy-to-understand error
text ('1961 not a leap year' instead of 'XMB17355 - DATE ERROR')... on the
inside code that's been running for the past two decades. I have trouble
understanding why many folks consider this a problem... 'Well, all our
salesfolk drive cars to visit clients, we should have all our deliveries
made by car, too... no trucks!'

>
>I am responding to a client RFP this month, the original conversation
>was "We have to get off the maniframe, its a CIO directive. We thought
>that this would be a good time to go to the board one time and ask for
>the funding to get off the mainframe and convert to JAVA".

Who was the general who tried to demonstrate the loyalty of his troops by
making them march off a cliff?

[snip]

>I hope this group benefits by some of the feedback that I have been
>hearing. We have migrated over 50M lines of COBOL code the past few
>years, with less than 2% going to JAVA.

Thanks much for this view from Another Side of the Irregular Polygon of
Life.

DD

First  |  Prev  | 
Pages: 1 2 3 4 5
Prev: COBOL Error Handling (was: What MF says about ROUNDED(was:Cobol Myth Busters
Next: Wildcat COBOL