Re: Differences in data description in programming languages
From: Warren Simmons (wsimmons5_at_optonline.net)
Date: 12/01/04
- Next message: Michael Wojcik: "Re: Infinite Loops and Explicit Exits"
- Previous message: Warren Simmons: "Re: Differences in data description in programming languages"
- In reply to: Pete Dashwood: "Re: Differences in data description in programming languages"
- Next in thread: Richard: "Re: Differences in data description in programming languages"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 01 Dec 2004 10:45:26 -0500
Top Post only...
Peter, I understand. And perhaps I misled. This person knows JAVA and
OO. He uses them, but finds that for his work, a lot of which is
conversions, he can create the new programs faster with his system of
COBOL programs designed to create either COBOL, or with another
converter, JAVA. Someone has a conversion they want done, he will do it.
Using his programs. The output source may not look that much like you
might expect, but they are generated using his COBOL programs. They
work, and it apparently is cost effective. I believe he also creates
new applications with his COBOL generator. Of course there has to be
some beginning work to do that. Oh, and as I was born in Gary, IN.,
I know who you were talking about. There are some other Garyites that
I believe are more revered that the Jackson your referred to.
Warren
Pete Dashwood wrote:
> "Warren Simmons" <wsimmons5@optonline.net> wrote in message
> news:Nxard.3809$0H.3156@fe12.lga...
>
>>Top Post Only..
>>
>>One developer I know generates Java using a cobol system. I'm
>>not quoting him, but I believe he said Java is for the birds
>>or something like that, and that he can do more with cobol.
>>
>>Warren Simmons
>>
>
>
> Warren, you probably wouldn't be familiar with Jackson Structured
> Programming, but not long after Djkstra's famous paper on the elements of
> programming (where he shows there are only three requisites for a sequential
> process implemented on a Von Neumann processor), a hustler (oops...gotta
> watch for law suits here...) a "consultant" named Michael Jackson (no
> relation...and just as mentally sound) came out with a system that would
> enforce structured programming. Programmers indicated which of the three
> functions they required and the sequence they were to be applied in, and
> JSP generated a COBOL program that was then compiled.
>
> Now, this system had everything going for it. It could be proven to be
> mathematically "pure", it generated COBOL code, which everyone could read
> (this was in the days when Source and COBOL were still King). I worked on
> one site in North London were it was "mandated". I refused to use it for a
> particular job that required something more than batch processing and was
> told my job was on the line. (I was a contractor). I said fine, and started
> packing, (It was in the days when COBOL contracts where abundant), the
> senior manager heard about it and, after hearing both sides gave me
> dispensation for the specific programs I was writing. The manager who had
> enforced it on me vowed to see me off the premises. (A year later they had a
> party when he left and I signed his leaving card with the comment "I'm still
> here..." Small victories are often the sweetest, but I digress...
>
> One of the major criticisms about JSP was that it generated TERRIBLE COBOL.
> It was full of GO TOs (very anti-structured for a product that was
> Evangelical about structured programming...) and even used ALTER. (I liked
> it actually, but I never let on...and it was the concept I liked not the
> product...)
>
> So, anyone who used JSP could quite legitimately say (after looking at the
> generated code) : "Wow! This COBOL really sucks as a language. I could do
> things in RPG much better than this."
>
> In case I've lost you, the point was that your friend's experience of Java
> was as a COBOL programmer viewing Java that was generated from COBOL. This
> is no way to form an informed opinion about Java or anything else.
>
> Warren, the fact that Java is good, in no way dimiinishes COBOL, and it is
> silly to have a contest. They are both good for certain tasks.
>
> Java is more successful because the proponents of it are more dynamic in
> support of it (you don't find "infinite loop" threads in the Java forums
> <G>), it is virtually free (as opposed to several thousand dollars for a
> commercial COBOL compiler), has proliferated because it is pretty easy to
> learn (provided you don't approach it using "COBOL think"), implements OO
> (in my opinion, it does this even better than C++, but I don't profess to in
> depth knowledge of C++; this opinion is based on admittedly superficial
> reading and I will change it if evidence to the contrary becomes available),
> and is virtually platform independent. (Quite a lucrative market is emerging
> in writing JAVA games for WAP and 3G phones... I might even have a go at
> this next year <G>)
>
> So, irrespective of your friend's opinion, Warren, Java is successful and
> looks like being with us for some time to come.
>
> Procedural programming is on the way out (there are sound reasons for this)
> and COBOL is a procedural language. A fantastic effort was made to provide
> it with OO facilities, but these were not picked up by the COBOL community
> at large, and it has, largely, missed the boat.
>
> Even OO languages like Java do not have unlimited lifespan, but it will be
> longer than the languages which implement the procedural paradigm.
>
> I have to be honest here: the main reason I am still using COBOL is because
> it (well, I should really only speak for the Fujitsu implementation of it,
> as that is the only OO COBOL I have ever used) implements OO so well. (The
> other reason is because I am attached to it after making my living from it
> for 40 years...)
>
> Pete.
> <snip>
>
>
>
- Next message: Michael Wojcik: "Re: Infinite Loops and Explicit Exits"
- Previous message: Warren Simmons: "Re: Differences in data description in programming languages"
- In reply to: Pete Dashwood: "Re: Differences in data description in programming languages"
- Next in thread: Richard: "Re: Differences in data description in programming languages"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
