Re: FW: OS/400 upgrades for CISC

Dave,

In a message dated 97-04-20 17:07:13 EDT, you write:

> > That's irrelevant.  How many old Corvettes are there out there compared
>  > to the new model ?  Do you think if you ask GM nicely they'll build a
>  > new 1996 Corvette with the new engine for you ?
>  > 
>  It is EXTREMELY RELEVANT.  If there are 800,000 CISC boxes out there and
>  200,000 RISC boxes, IBM should get darn worried about losing market
>  share (very important for a business).  What is irrelevant is whether GM
>  will build a 1996 Corvette.
>  
>  People forget that if a business decides that the cost outweighs the
>  benefit in retaining a old machine, they will replace it.  IBM does not
>  have to tell them.  For example, when S/36 came out, S/34 people in
>  droves converted to S/36.  When S/38 came out in 1978, S/34 people
>  yawned, but S/3 people were forced due to the aforementioned cost
>  benefit considerations to move to S/38.  IBM made a big error in
>  assuming S/36 people will move to AS/400 when they introduced their
>  S/36E in an AS/400 (ever tried DELETE X,F1 in R2? Can finish your lunch
>  by the time it was done).  The point I am trying to make to all the
>  people with the shiny black boxes is the gray boxes will remain there
>  and if the manufacturer doesnt feel like supporting it, they will do a
>  cost benefit analysis and may move (could be RISC or nowadays it points
>  to NT !).
>  
>  > On Sat, 19 Apr 1997, Dave Mahadevan. wrote:
>  > 
>  > > You seem to be very sure CISC is dead.  I am curious, as to how many
>  > > CISC is out there vs RISC.  Anybody have these numbers?

Please see my earlier post, and I believe that Neil's analogy is correct.  I
believe that you are missing mine and Neil's points.  Would you petition IBM
to build you a B70?  I think not.  Like I mentioned earlier, keeping 800,000
customers happy is NOT IBM's goal.  EARNING (with the emphasis on EARN) the
NEXT 800,000 customers is IBM's intention.  They (IBM) have come up with a
box competitive with it's major adversaries that will take it into the next
millenium against so-called "open" boxes.  The latest performance
announcement proves that...

Granted, the /36 outlived IBM's predictions (it's still the bain of my
existence because I happen to know a little about it), but it doesn't fit the
profile of what you're asking for.  The System/36 survived because it did
what it's owners needed.  It was (and still is) easy to use, rarely fails,
and replacements/upgrades are still fairly cheap.  You want the CISC system
to do MORE than it does, to stay current with technology, and incorporate
features that it does not have today.  These positions are diametrically
opposed.  You CANNOT stay current on old hardware!  Would you try to run
Office '97 on a 286?  I don't think so (beside the point that you couldn't
run a 32 bit application on a 16 bit machine, or maybe that IS the point).

If CBR (cost-benefit ratio) is all that you're looking for, then YOU need to
find a new job.  Your employer isn't keeping you current, and you don't stand
a chance in today's job market.  I would stand an eight year old AS/400 up
against an eight year old equivalent (DEC, HP, DG) ANY time.  It's been to my
detriment in the past, but I've recommended competitive systems where they
fit.  If your company can ACTUALLY justify training technical people on an
entirely different platform as well as all of your users and still save
money, more power to them.

As I'm fond of recommending, "the fit is the fit".  I don't care if it's on
an ES/9000, a DEC, a Honeywell, a Tandem, or any other.  If a CISC AS/400 is
holding you back, then you didn't have the right "fit" in the first place.
 You NEED to move to another platform.  You buy what your company NEEDS --
the hardware is inconsequential.

CISC is GONE.  The number of machines doesn't matter, the number of
applications doesn't matter.  The number of us that "wish it weren't so"
doesn't matter.  We can want it with "all our hearts" and it doesn't matter.
 IBM wants to sell more hardware, and they're not going to do it with CISC.
 RISC is the future, and YOUR only choice is to decide when you're going to
implement it.  You don't need RISC to be Y2K compliant, and you don't need it
to take advantage of most of the new internet technologies.  If you're SO
DARN HOT on using the latest stuff, why not upgrade to the latest hardware?
 You probably need it anyway...

<<end soapbox mode>>

Dean Asmussen
Enterprise Systems Consulting, Inc.
Fuquay-Varina, NC  USA
E-Mail:  DAsmussen@AOL.COM

"Always take a job that is too big for you." -- Harry Emerson Fosdick
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* This is the Midrange System Mailing List!  To submit a new message,   *
* send your mail to "MIDRANGE-L@midrange.com".  To unsubscribe from     *
* this list send email to MAJORDOMO@midrange.com and specify            *
* 'unsubscribe MIDRANGE-L' in the body of your message.  Questions      *
* should be directed to the list owner / operator: david@midrange.com   *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *