[snip]
>   My point:  I would find it very hard to plop down $100+ K on a
platform
> with an uncertain future.

It it time to throw in that all platforms have uncertain futures. Just
having an O.S. with the same name doesn't even mean much with some
vendors. What will IA-64 and NT X.x bring? How backward compatible will
it be? Is what your saying is to just buy cheap PC hardware (I've heard
that argument from others) and replace it with new stuff when it
misbehaves? If so, many of use would rather have hardware that lasts 5
years without having to bother with it, but, to me, it is important to
know that you can scale MPE/iX systems to keep up with growth without
re-engineering all your software - that also isn't common.

If the MPE application software you have fits your needs and provides
for your near-term future, how could you do better than to upgrade to a
new (or used, but higher performing) MPE/iX platform? Migration from a
Classic is easier than any other O.S. "jump" I'm aware of by any vendor
- it is certainly easier than going from HPUX 9.x to 10.x

If you buy commercial software, then much depends on you software
vendor. The vendor can hide a lot of the troubles with O.S. updates and
smooth the path to using the hardware they use, or they can leave it up
to you.

>
> My $0.02.
>
> Dan Kosack
>
Mine, too.
Richard Gambrell