Much of the speculation about possible futures for MPE seems to ignore the
fact that MPE is undoubtedly very hardware-specific in its design.  To port
it to a platform with a different architecture would be a huge effort.  This
is acknowledged in HP's decision to not port MPE to IA-64, which in turn is
probably a major factor in HP's decision to drop MPE.  Since HP is the only
manufacturer of PA-RISC, and the sales volume of MPE-based systems is
evidently insufficient to make a port an attractive investment, it seems
unlikely that IBM or other hardware manufacturers would be interested.  It
is even more unlikely given the fact that IBM undoubtedly tells its
customers that OS/400 is the best choice for that market segment.

It seems to me that the most likely scenario for a post-HP life for MPE is
for a company that has experience in MPE support, such as Allegro or
Beechglen, to take over the OS, and the hardware would continue to be sold
by HP.  If that model generates enough sales to justify it, the new owners
of MPE could do a port to IA-64 or whatever.  HP would be likely to find
such an arrangement attractive, because it would continue to generate
hardware sales revenue.  The open-source approach is certainly possible, but
at the expense of some of MPE's famous stability and reliability.

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *