I had two reactions to Peter Relson's post.

The first was one of complete agreement.  I can think of few drearier
or more gratuitous tasks than "going through all of the manuals" to
change instances of 'load module' to instances of 'load module or
program object'.

I can imagine that as text is revised---for other, substantive
reasons---appropriate collaterial changes in text that refers to 'load
modules' could be made, but that is a very different matter.

The second is a little different.  In my view at least program objects
are still too little used.  There are circumstances, much discussed
here, in which they cannot be used and perhaps never will be usable;
but they are few.

What seems to be forgotten much too often is that the design and
implementation of programs objects were a response to the
deficiencies, many of them radical, of load modules.   We have just
had a discussion of the fact that the magic number 255 is more
problematic with load-module libraries than it is with program-object
ones, and other such discussions are frequent here.

For this reason it seems to me better to labor the distinctions
between load modules and program object, PDSs and PDSEs, than it is to
assume implicitly that they are much of a muchness, interchangeable in
use.

This community is full of instinctually risk averse, conservative
people who cherish and defend the old and are reluctant to master new
technology, often long after it has ceased to be at all new.  This
propensity is stubborn and hard to change, but it should be
discouraged where it can be, and terminology that obliterates
substantive distinctions is not helpful in doing so.

Portmanteau words like 'alphanumeric' are often useful.  Its
availability saves writing out the phrase 'alphabetic or numeric'
repeatedly.  Like Paul Gilmartin, I have sometimes found myself
lamenting the absence of a briefer locution for 'load module or
program object'; in this case, however, I suspect that making one
available would be counter-productive.

John Gilmore, Ashland, MA 01721 - USA

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to