[Info-vax] Reimplementing VMS, was: Re: HP adds OpenVMS Mature Product Support beyond the end of Standard Support

[JF Mezei]

On 14-02-02 07:03, Jan-Erik Soderholm wrote:

> Note the difference between "VMS" n the first part and "BCS OS"
> in the second part. That was HPUX and that has been verified several
> times. Last I heard of it was at some HP seminar.

My sources told me that all 3 were evaluated for porting costs.
The Oracle documents only spoke of HP-UX because the HP lawsuit was
about Oracle hurting HP-UX sales.  (hard to claim damages on VMS when
you don't market the product, expect no sales and tell everyone at
conferences you only expect the remainiing installed base to use VMS).

Remember that this was during a period where Intel signalled it wanted
to pull the plug on IA64. HP had to look at its options. In negotiating
with Intel, HP would have to know what was involved in porting all BCS
OS to the 8086, versus paying Intel to keep IA64 on life support.

Before talking to Intel, HP would need to know what special features (if
any)  would have to be added to 8086 to support its 3 BCS OS.  HP-UX is
the problem child due to endianess, so HP would have had to negotiate
the costs of adding big endian to 8086 vs costs of converting HP-UX to
little endian.

So yes, HP would have had to take a serious look at the cost of porting
all 3 OS, and technical feasability so it could go to Intel with "if you
don't continue IA64, you will have to pay us $x to port the 3 OS to the
8086".  For such negotiations, you don't want to forget the less
important OS, you want to show the costs are going to be as high as
possible in order to negotiate a good deal with Intel.


Consider when Compaq bought Digital. It quickly announced that it would
port NSK to Alpha staring with EV7 once it got lockstep. This means that
NSK engineers had done an evaluation of porting to Alpha well before the
merger was announced, would have met with Alpha engineering and the
costs of this endeavour calculated on both sides.

Same thing when Capellas announced the Alphacide. key people within VMS
engineering would have been involved in evaluating this to look at costs
and technical feasability so that Capellas could negotiate the big
suitcase full of money Compaq would get to help fund the port.

People will deny this ever happened because they either didn't know who
was involved in that evaluation, or they were involved and signed an NDA
where they woudl deny this ever happened. But it happened.

Note that while Palmer's deal with Intel did involve porting
Digital-Unix to IA64, it didn't touch VMS, but in that time frame, I
suspect VMS was looked at from a feasability point of view. So perhaps
it only needed a refresh for the Alphacide to update it according to
final specs for IA64 and manpower costs,