On Fri, 2 Nov 2007 09:08:02 -0500, Paul Gilmartin <[EMAIL PROTECTED]>
wrote:

>Now, I see that METAL C is LE independent.  But, I surmised,
>to be POSIX compliant it must have some RTL (well, I suppose
>it could all be done with macros, expanded at compile time.
>Unlikely.)  So, I wondered whether this new RTL would come
>with new license requirements.  Other plies in this thread
>have said, "No."
>

Ok, let's not let the train go too far off the tracks.  This thread diverged
on the discussion of PL/(S, AS, X) vs. Metal C.  

The PL/... variants:
- ARE NOT Posix compliant
- DO NOT have a runtime library
- DO NOT have standard services for input and output
- DO NOT have built in heap/stack/storage management (other than a primitive
to simply GETMAIN/STORAGE (OBTAIN) an area to accomodate the savearea and
variables declared by the module
- ARE insufficient for general purpose programming in many other ways, just
like assembler

Then, what good is PL/...?  System level programming, which implies a
fundamental binding to the underlying operating system and the hardware (and
in many cases, oddball environments - think SRB, system locks...).  Other
than the fact that it is a 3GL, it really doesn't purport to be a general
purpose language for anything.  This is role that Metal C chooses to play
for us unwashed masses.  It is for systems level activities on a particular
class of system.  That being said, it's up to the programmer to deal with
the above problems in their own way.  Do I get input from SYSIN?  Open a
DCB.  Do I use an operator REPLY for input?  Issue a WTOR.  Do I use MODIFY?
 Extract the CIB.  Just like assembler.

Additionally, I'd suspect that I could find a bright C programmer and get
them useful on METAL C faster than PL/X.  IBM doesn't do anything without
some level of informed self interest.  Perhaps they are having trouble
locating good PL/X programmers themselves?

Bait and switch?  I don't think so.  Unrealistic expectations?  More likely.

Scott Fagen
Enterprise Systems Management

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to