Re: CL enhancements

[Mark Lazarus <mlazarus@xxxxxxxx>]

Chris,

At 11:15 PM 2/17/98 PDT, you wrote:
>** Reply to note from Mark Lazarus <mlazarus@ttec.com> Mon, 16 Feb 1998
21:09:54
>
>
>>  OK, maybe I was jumping to conclusions as to the *reason* CL hasn't seen
>> much in the way of enhancements, but the facts are there.  CL as a language
>> has hardly been touched.  We have many areas that we must kludge to work
>> properly.  A few examples: Mupltiple files opened, closing and opening a
>> file, working w/ various unsupported data types, loop control
structures, etc.
>
>Hey, I agree that there are plenty of features that could be added to or
>enhanced in CL. I think my point about is that I haven't ever been unable
>to implement something just because CL lacks a feature. If I really insist
>on multiple files being process in CL, I use nested CLs. 

 If a language requires you to kludge basic operations, then it's a
hinderance to programmer productivity.  If I have to create multiple
objects w/ their additional runtime, storage, source management and
maintenance overhead, then it s/b fixed.  It is considered crippled.  We
need CL. There is currently not a good replacement (including REXX).

>>  When a language is missing some basic elements (see above), no one should
>> have beat IBM over the head to get them to recognize its deficiencies.  The
>> IBM developers all agree that there are quite a few areas that need
>> improving.  We are not talking about niceties.  We are talking about basic
>> laguage constructs.
>
>I very much don't mean to be rude. But if these basic language constructs
>are so important how come I haven't noted the need for them?

 Just because you have found a way around deficiencies doesn't mean they
don't exist.  We think we are clever when we find a way around a
limitation, but it often makes for difficult to read code and costs your
company time and $$$.  Example:  Remember your RPGII or RPGIII routines to
center text using an array?  It's not easy to follow.  Contrast that to the
new RPG opcodes that can do the task in a clear, concise fashion.


>>  When IBM feels it's important, they will do it.  We've been asking for
>I think that's true. The key is to demonstrate it's importance.

 I mean strategic importance, not programming importance.  The latter is
clearly evident.

 -mark

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to "MIDRANGE-L@midrange.com".
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---