Re: members

Simon Coulter wrote:
> 
> Hello John,
> 
> Well, members _are_ objects (type *MEM) but they are not exposed.  IBM could 
>store object creation information
> in an associated space of the source member.  I like your idea although I 
>don't understand what you mean in
> your last paragraph.
> 
I think what I was trying to point out in the last paragraph is that
some programs JAVA, CGI, REXX, etc. can be or are already used as
interpreted languages where the final executable is created at run time
so they must be designed to have all information available to them at
run time.

> However, I suspect that IBMs preferred approach is for a third-party to 
>develop an application to manage the
> source creation process (or sell you Application Development Manager) rather 
>than build this into the OS.
> 
> Regards,
> Simon Coulter.
> 

Even if IBM makes it possible to store source/pgm as a true object you
will still need software to manage large projects so I don't think they
lose a market there.  This level of integration can only be built in at
the OS level.  Object type *MEM could be expanded to *RPG *RPGIV *C etc.

Once they allow this capability (src/pgm) then adding a new capability
to the object can be done transparently to the applications that use
that object.  This design could also allow them to improve on the ILE
program model.  It's great that IBM has finally caught on to this but it
is still not up to par with "class libraries" used in c and other
languages.  Look at the requests for additions to RPGLE that have been
showing up on this list and then think about how you would handle them
in c. 

John Hall
Home Sales Co.
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@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
+---