I received this reply to my last post from Duane and he has kindly allowed
me to post it here along with a few further comments of my own:

> I think all this discussion boils down to the commodity small/mid-range
> server business being worth bothering with or not. You seem to argue
> for it not being worth the effort and for them to concentrate on higher
> end stuff.

Actually, I agreed that HP needs to cover this business opportunity as a
way to provide an all HP solution and as a way to keep competitors out of a
customers shop.  My point was that commodity computer products are not very
profitable.  Overall corporate strategy needs to concentrate on sources of
higher profitable or slow death will result.

> I would argue that there is lots of money to be made in this space if
> you are the leading vendor.  If you can work it out to be the leading
> vendor you will have the unit-volume to make the business work.
> If not then you will be in trouble and not ever able to catch up.
> I would also argue that the leading vendor has an opportunity to
> capture significant mind-share such that they become almost a defacto
> standard.  This is why I think the merger is important to hp/compaq.

HP is losing money on low-end commodity products now.  Compaq is losing
money on low-end commodity products now.  So how is HP + Compaq with higher
volume (doubtful if they can keep it), multiple part number systems, short
term higher costs, etc. going to make money?  Therefore, I disagree with
the premise that the merger is important (although likely to be a
significant disaster).  It will not improve profits in this area.

> I would also argue that this is a play to squeeze out SUN and to a lesser
> degree SGI.  HP has invested significantly in ia-64 to get a high-volume
> server chip produced by the king of chip-making.  If the chip goes
> high-volume this will cause SGI and Sun to become extra-niche players
> and render them ineffective given the price/performance hp will have.

I agree with the above paragraph.  I have thought that Sun is headed for a
dead-end for a while.  Their CPU's are designed by Fijitsu, I believe, and
have got to be higher cost than the Intel IA-64 based on relative volumes.
SGI is a niche player and not on my radar.

> BTW - my prediction is that if the merger is approved and ia-64 rolls out
> that over time Sun will falter and be bought by IBM. HP will acquire SGI.
> Or maybe Apple will get Sun :-)

The merger is not required for this to become a reality.  HP and Compaq,
among others, have already committed to IA-64 and would buy it regardless
of the merger.  Sun is stuck either way.  HP buying SGI would go the way
the Apollo purchase went  --  badly.

> But, as a customer, I am more interested in the lower-cost equation all
> this brings. I have no allegiance to hp. Never really have. They stopped
> being interested in small fish about 1989 (IMHO).

You are right about HP's small fish interests.  If these small fish should
grow into big fish, HP will have lost contact with the customer and the
higher-end business could easily go somewhere else.

> We buy our ia-32 servers from Dell.

> duane

As a buyer/planner for your company you should certainly do what is best
for them.  If using low-end commodity products at the lowest price where
price overrides all other considerations, then that is what you should use.

It sounds like you are expecting this merger to drop prices.  Usually the
elimination of a competitor does the opposite.  But, it is also possible
that one of these companies was not a serious competitor, and so the loss
of one of the product lines will not be significant.

It is interesting that you are in the position of moving from the HP3000 to
these low-cost servers, but that you have not chosen HP or even Compaq
products.  This means that the only value of the merger to you is any
competitive effect it might have on Dell, for example.  This is a great
example of how HP not only threw away the HP3000 business, but has blown
the migration opportunity as well, and the merger won't fix that.

If HP were still offering you the supported solution of the HP3000 you
wouldn't have to waste your time and your company's money on the Phase II
of your migration plan.  I suspect that this will cost you more than any
savings in low-end server hardware.  Instead you would be doing something
more useful for your company and be grateful that an HP3000 allowed you to
have that option.

Best regards,

John R. Wolff

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