On Thursday, 1994-09-08  Murray Lesser wrote to David Noon about
"Pl/i" as follows:

ML>     I really don't care if the manuals don't handle multi-threading.  I
ML> doubt that I will be using multi-threading for REXX external
ML> functions.  However, it bothers me that you think I will have
ML> to buy a new Toolkit.  I was hoping to translate the few API calls
ML> that I might use to the PL/I format from either the C or ASM version
ML> that I already have.  All my REXX (and C) programming is for
ML> text-mode utilities, so I won't need the PM interface.  But, we
ML> shall see.
ML> 

[Text deleted]

ML> DN>If you need any help with PL/I syntax and semantics, I am more than
ML>   >willing to assist you or anybody else and can be reached here on
ML>   >FidoNet, on the IBM PCC BBS, or you can use Section 6 of OS2DF1 on
ML>   >Compuserve to address problems directly to IBM Santa Teresa Labs (the
ML>   >developers of the PL/I compiler).
ML> 
ML> Thanks.  I hope I won't have to bother you further.

Hi Murray,

It is certainly possible to define the API's in PL/I, by hand, as and
when you need them. But there are an awful lot of them; however, not
doing PM reduces the number significantly and removes most of the
necessity for multi-threading. That said, I think a full copy of the
compiler (the one on DevConn is a demo subset) and the PL/I Toolkit
will prove worthwhile if you are going to do a number of projects in
PL/I; if you are doing only this one then the demo copy might be
sufficient.

Incidentally, it is never a bother to me to help other programmers,
especially dealing with a moose-sized language like PL/I.

Regards

Dave

___
 X KWQ/2 1.2g X For five bucks I expect disembowelments.
--- Maximus/2 2.01wb