Chuck Ryan after Jeff:

>> > Perhaps HP would consider releasing the source to an
>> > earlier version of MPE?  Version 5.5 or 6.0 would be
>> > a great starting point for future development and
>> > would take care of eliminating STM and some of the
>> > other "enhancements" we have seen of late.
>>
>> At this time HP has no plans to release the source code
>> for any version of MPE to the general public. There are
>> many reasons, and others on this list are aware of at
>> least some of the reasons.
>>
>> But, even if we did offer MPE source to the public, how
>> easy do you think it would be to make a change in MPE
>> and verify that the change is correct and does not have
>> any negative side effects?
>
> Um Jeff, you do know that some of us are actually
> programmers that do have some small idea of what is
> involved in maintaining and enhancing old code?

I would bet that many if not most people on this list
consider themselves at least part-time programmers...  and
I would dearly love to be able to toss CSTM and replace it
with an updated / expanded version of SYSDIAG...  BUT:

I would be skeptical about trying to "go back" to older
versions of MPE and using that as a base for independent
development.  Remember that in 6.5, 7.0, and now 7.5 there
are a large number of enhancements and limit expansions.
As the keeper of the SIGImage/SQL "Now / Soon Available"
list I see that my "counter" has added 22 new enhancements
just to IMAGE in 2000, 2001, and 2002.  Some may be
backward compatible;  some are not...  And start to think
of all the new / expanded features in MPE from 6.5 and
going forward...  Sites have started using those features;
"going back" would break many running systems.  Also keep
in mind that "MPE" (in the OpenMPE context: The OS and ALL
subsystems, regression test suites, etc., etc.) is very
much more than just the OS itself "(MPE/iX, in OpenMPE
parlance).  In the aggregate that is millions and millions
of lines of complex inter-related code; a very non-trivial
project to attack.

I still hope that OpenMPE and the rest of the MPE
community will eventually get to where there might indeed
be a possibility that at least some low-risk, "localized"
enhancements could be made...  But if and when that is
done, we better have a rigorous quality control and test
process in place like HP does now (and even with that (as
we all know) bugs STILL get out).  To attempt that overall
process after HP exits will be major effort.

Also...:  SYSDIAG doesn't even run on or support the new
A-and-N class boxes, so (at least for now) they have to
have a later version of the OS...  and actually now that I
think of it:  6.0 doesn't support the A-and-N class at all.
While there are still a lot of 9X9 and 9X8 and even 9X7
boxes out there (hi, Kiwi Ken...), if we try and fast
forward to end of HP support at end of 2006 I think it is
reasonable to assume that A and N class will make up a
much larger percentage of the "up and running" installed
base than they do now...

And...:  While 6.5 is now going to be supported until the
end of 2006, if there are any further enhancements and
even for bug fixes, in many cases I would guess that the
7.5 "parking release" is going to get by far the most
attention.  So for a lot of reasons looking beyond end of
HP support in '06, IMO the 7.5 EX/PP1 "HP parking release"
is the one to look at for potential further work, IF that
should become possible...  I really don't think there is
much choice about that:  If it ends up being possible to
mount an effort to ever put out, say, a 7.5 PP2, it is
likely going to take just about every non-HP MPE expert in
the world to pull it off...  better focus effort on that.

SIDEBAR:  If last above ever becomes possible, my first
enhancement request:  Save all the other good enhancements
that were done / are being done in 6.5, 7.0, and 7.5;  but
junk CSTM and replace it with something better (almost
ANYTHING would be better).

Now if you are talking about one or more true "hobbyists"
who might like to fool around with MPE versions older than
6.5, hey:  As far as I'm concerned, I'd be happy to let
them have it...  But I respectfully have to doubt whether
there would be many serious production customers who would
be interested in paying any money / running their business
on that AFTER 2006;  or would even be ABLE to do so even
if they wanted to (I realize right now may be somewhat of
a different story, but I'm focusing on >= 2006).


> The repeated statements made by you and others at CSY
> have led me to expect a mass of undocumented hacks and
> quick fixes that will be no small challenge to work with.
> Which is why I think rolling back to a version before
> the large number of appeasement changes might provide
> the most stable code base.

I have seen or heard nothing that makes me believe "CSY"
has thrown their previous methods of quality control and
care in production out the window.  A major amount of work
has been done beyond 6.0, and for the most part it has
turned out well as far as I can see.  We have been running
6.5 for nearly two years, and (other than cursing CSTM) I
can't think of any significant problem we have had in all
that time;  not counting network outages that were not the
fault of our 3000, we have had essentially ZERO down time.
...  and remember:  CSTM (STM on the HPUX side) was for
the most part "foisted" on "CSY" by their unixen breathern.

But yes:  I would DEARLY love to get rid of CSTM...  I
would pay money NOW, while still under HP support, if I
could get even roughly equivalent functionality with an
updated version of good old SYSDIAG (if my managers ever
get any more money to give me, that is)...

Ken Sletten