In article <[log in to unmask]>,
  Jeff Kell <[log in to unmask]> wrote (heavily snipped):
] Subject:      Soliciting HP3000 support
 
] If you were faced with the question of "Why stay with the 3000 when the
] entire media universe is touting Unix client/server?"  With the SCT product
 
] If you were in a similar situation, supporting an "Open" application, why
] would you stay with the 3000?  (I say reliability and management simplicity,
] plus no staff retraining).  I would like to solicit "testimonials" from any
 
] wishes.  I would prefer non-HP responses (unbiased) but will take all replies.
 
] [\] Jeff Kell, [log in to unmask]
 
Ahem.  I selected the first HP-3000 at UTC.  At the time it was quite the
fashion to have IBM - UTC had a 360/30 on lease from Greyhound.  The UC
foundation had put in serial # 1 of the HP2000 Access.  Interesting times.
 
Dr. Drinnon decided to move UTC ahead - radically.  During one of the most
hideous (yet enjoyable) 6 months, an IBM 360/30 with "punched-cards in the
gym for two days" software was replaced with an on-line interactive system.
 
HP wrote it up.  First in the southeastern US.  Without quoting actual
numbers, the HP3000 solution cost $600 for every $1000 an IBM solution
(370/115) would have cost. It met the bid specs: on-line, interactive,
spooled printing, batch capability, etc.  There were lower bidders: one
proposal was a DEC 11/34 (an 11/70 would have been rational) just to be
the low bidder.
 
I selected HP (the second _highest_ bidder) because it could do the job
that needed to be done.  We first determined what we needed to do.  My
question is, then, what additional business goals need to be met now?
 
After specific goals are identified, then (and only then) you can examine
alternatives to the current system.  Pin them down: "we don't know exactly
what we want to do" is not a good answer.  Make them explore the other
options, travel and see alternatives, go to user meetings, visit the
factory - and write reports for others with _their_ conclusions.
 
A lot of thinking, time and hard work went into the initial HP selection,
and the decision has proven over the last 18+ years to be correct.  It may
indeed be time to change - but the analysis deserves, if anything, more
care these days.  There is a significantly larger status quo to lose.
 
Your job may depend on their selection.  Make sure those participating in
the analysis are equally involved in the project.
 
And you may quote me.  I'm proud of the part I played in UTC's success.
 
--
Regards  --  Tony B. Shepherd  --  [log in to unmask]