John D. Alleyn-Day wrote:
> Jim Phillips writes:
> >John Alleyn-Day writes:
> >>Some time ago I switched for a period to programming on an IBM
> >>mainframe using CICS.

> >Uh-oh!  I've worked with CICS and DLP (Unisys) screen handlers and
> >they are totally different from MPE.  Not so much the screen handler
> >itself (field mapping, display attributes, etc) but the methodology
> >used is psuedo-conversational as opposed to HP's conversational mode.

Well, while we are showing our age...  :-)

CICS was largely transaction oriented - you had one program servicing
multiple users with little or no user-local storage.  Not unlike web cgi
where you have very little state information, maybe the process has gone
full circle?  But the MPE concept of addressing ONE user is completely
foreign to someone with a CICS background.

That was the beauty of MPE to me, back in 197x.  Your code was
explicitly re-entrant, shareable; and user save areas (data stacks) were
inherent in the OS.  You write it for one-at-a-time, it works for
many-at-a-time.  It was a totally alien concept at the time for that
mindset.

Similarly, the single-threaded concepts that carried over into MPE are
foreign, like the old shared record pointers in CM KSAM.

The more things change, the more they stay the same :-)

Jeff Kell <[log in to unmask]>