I both agree and disagree.  My experience is that there are many classes of
programmers.  However, for now, I just want to write about those who use
other people's code and tools and those who don't and those who will change
how they do things and those who won't.

In my opinion, one of the most compelling reason to use some of these new
programming paradigms centers around productivity.     To me the biggest
source of productivity gains are from the concepts of reusable code, use of
classes and class like things (i.e. COM objects, Java Beans, CORBA, etc.).

That said, if an existing programmer is willing to change how they do things
in order to be more productive, I agree that they can be an asset.  On the
other hand, all of my career I have seen people move into a new paradigm and
not change how they do things.  For example, code a 4th generation language
almost like assembler, or insist on writing all code from scratch.  I now
also work with some gray beards who can program circles around some younger
people because they combine their years of experience with use of improved
tools and techniques.

Charles Finley



> -----Original Message-----
> From: HP-3000 Systems Discussion [mailto:[log in to unmask]]On
> Behalf Of Joseph Rosenblatt
> Sent: Tuesday, February 13, 2001 6:56 AM
> To: [log in to unmask]
> Subject: Re: Cost of migrating COBOL Programmers to Java
>
>
> We have been hearing for years how COBOL/FORTRAN programmers
> could not learn
> the new languages. "OOP is different," has been the rallying cry
> of the new
> technologists. I believe and I think many of you would agree, that a coder
> is a coder. People that can write good code in one language can learn to
> write good code in another. Thought process may vary slightly and syntax
> greatly but coding is coding.
>
> The main thrust of the Gartner piece is that it is not cost effective to
> train a Cobol programmer. Let us keep in mind that Gartner is paid by and
> caters to executives not cubicle dwellers. I saw the article
> saying that it
> is cheaper to get a JAVA kid out of school than to train long
> time workers.
> This totally discounts the years of experience the older employee
> may have.
>
> If salary is the only criteria then definitely hire the
> non-experienced. If
> value is the issue then you may need to rethink that position. It's an old
> argument couched in a new paradigm.
>
> Just one old Cobol programmer's opinion.
>
> Joseph Rosenblatt
>