1)  The number of MPE users continues to whither away, and with it any
source of money, direct or indirect to finance it.  Also, as the market
shrinks, so does the number of customers affected by the lack of support.
HP knows this very well.

2)  Expertise necessary to support MPE continues to whither away also.
There are few left here on this list, but most have left, and most of the
few that are left are just here to pick over MPE bones looking for
opportunities to migrate the weakened die-hard homesteaders to their
particular solutions and to squeeze a few more dollars out of what is left
of the HP3000 user base.

3)  I have some expertise in operating systems and their internals, and MPE
in particular having patched MPE before acquiring even the source code, and
cracking its security easily by finding backdoors lurking in SL.PUB.SYS,
also before acquisition of source code.  I don't see how the current code
could be maintained by the few remaining people left capable of doing it,
especially maintaining hardware drivers.

4)  Reliable hardware drivers are very hard to create and maintain as nearly
all hardware deviates from published standards in some way and/or have
hidden timing issues that are difficult to duplicate.  People, this is
"rocket science"!  Current hardware drivers in MPE become obsolete as the
hardware they support becomes obsolete.  New hardware require new drivers.
Who will be able and willing to create them?

5)  As far as I can tell, the remaining HP3000/OpenMPE community rely on
prayers to the HP gods to save them from oblivion and are unwilling to lift
a finger to save themselves.

On Sun, Oct 26, 2008 at 6:44 AM, Jim Phillips <[log in to unmask]> wrote:

> On Sat, 25 Oct 2008 16:50:13 -0400, Wirt Atmar <[log in to unmask]>
> writes:
>
> > As I've written before, having the OpenMPE committee agree to sign non-
> > disclosure agreements was one of the dumbest mistakes that the
> > committee could have ever made, and in the end, could well be attributed
> > to be the single greatest contributor to the death of MPE, beyond HP's
> > own mishandlings of the product.
>

The NDAs only provide a cover for HP and fodder for prayers to the HP gods
for deliverence.  If everyone on this list were part of the NDAs, what
little hope of prayers to the HP gods would disappear, and there would be
much wailing and gnashing of teeth as all would see the futility of their
prayers ever being answered, and HP would experience some unneeded
embarrassment and attention.  Currently, MPE is dying a quiet death
unnoticed by the vast majority of HP customers and potential customers.  HP
would like to keep the slow death of MPE and the HP3000 quiet and out of the
media.  It is working very well for them.

I, for one, have all but given up hope that OpenMPE will ever do anything
> that will lead to MPE living beyond HP.  Almost seven years of haggling,
> pleading, begging, moaning, complaining, and griping (BMC) have done nothing
> to HP's steadfast refusal to part with anything even remotely related to
> MPE.


This is their optimum, least cost strategy.  It is working nicely.

OpenMPE tied their own hands with the NDA and if anyone wishes to see MPE
> survive to fight another day, we need to get an emulator going so that the
> O/S is not beholden to some arcane, HP-controlled architecture.  It is my
> personal opinion that even an emulator will not "save" MPE (in any business
> sense of the word), but only allow it to live as a hobbyist toy in the hands
> of a dedicated few.
>

I agree.  The only feasible way to save MPE now is to create an MPE or
MPE-like environment that runs atop Linux.  But, even that seems to be
wishful thinking as their doesn't seem to be enough people left capable to
make it happen technically and financially.  Let the prayers to the HP gods
continue!


> It was to this end that Peter Eggers and I talked about writing an MPE
> emulator to run on Linux.
>

I was thinking more of a business application development and operational
environment based on the Classic MPE concepts of simplicity and reliability
that would run atop a minimized (or full, if desired) version of Linux that
would only be accessed through MPE's Privileged Mode, but still within Linux
user space.  Could be done by a very small group with time and motivation to
do so.  Unfortunately, they don't appear to exist anymore.

The addition of Unix/POSIX and file system(s) was a knife in the heart of
MPE.  It added complexity and decreased reliability.  It did provide a path
for MPE people to become familiar with the Unix/Linux world, so that most
would have less of a barrier to migration off of MPE to HP-UX/Linux, and
more incentive to do so, due to increased complexity and unreliability.
Good for HP who wants MPE to quietly pass away into oblivion.

From HP's point-of-view, may hope and prayers of the HP3000/MPE community
never die until the last HP3000 is retired.

Here to eventually pay my respects, possibly add to MPE's eulogy, and be
entertained by the OT threads, Pete.

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