Re: Linux (see smaller server Iseries) -- MIDRANGE-L

----- Original Message -----
From: <thomas@inorbit.com>
> I realize this is a devil's advocate position, but I don't want
it to go unspoken,
> especially if someone can contribute a solid fact.

Since you label everything a guess, I have to wonder what isn't a
guess. What would you accept as absolute, incontrovertible fact?
What is your standard of proof?

> Sure, CFINTxx exists. But that in itself says absolutely nothing
about the cost to continue supporting interactive.

Well, I've offered what I think is a reasoned hypothesis that
CFINT exists to prop up the price of AS/400 interactive systems in
the absence of competition. I've offered several reasons for it.
It seems to me that we need to make our deciusions about it in the
absence of what you call a solid fact, since you haven't offered
any likely way to get those facts.

You are saying more about your own preferences for a foundation
for decisions making, rather than anything about the situation
itself. If someone came out of IBM with a piece of paper that
says, "we are doing this to soak our Iseries users because they
have no other choice", how would you know it was a 'solid fact'and
not a forgery, a disgruntled emplooyee?

In the real world it is sensible to take the behavior,
particularly the planned and continually executed behavior, of
persons and organizations as 'solid fact' and as evidence of their
intentions.

>  Technically, I'd guess it even increases the cost. I mean,
SOMEbody has to maintain and test it.

Sagittarian?

That's like allocating the cost of the gate back to the sheep who
are penned in by it.

Think of it this way - if IBM stopped maintaining and testing
CFINT, how much 'cost' would they save? It's a negative number,
running to many millions of dollars. If CFINT goes away, customer
dollars flow to third party software houses, or never leave
customer's pockets at all. they don't flow to IBM. The business
partners who earn significant revenue from software must be
howling mad. But many get a big chunk from the hardware markup.
Customers pay for hardware without hardly thinking about it.
Software dollars get squeezed.

When you communicate with me on this topic, I can never know your
intention absolutely. You might be an IBM apologist or IBM
loyalist. You might think IBM is doing exactly what I infer that
they are doing, but you think they are justified. You might not
like me and not want to agree with me. You  might be confused. You
might not understand what I am saying. You might just like to
leave things as uncertain as possible. Or there may be some other
reason I don't know.

The only solid fact I have is your behavior. In the real world,
with limited time and a lack of omniscience, that's what I have to
make real world decisions on.

So if you point is that we can never 'REALLY' know, I agree with
you. Even when someone comes from IBM and tells us, even when the
judge makes a ruling, we can never 'REALLY' know. Epistemology.

Meanwhile I think it would be nice if the customers would explain
to IBM that they will have to find another way to make their
money. Right now the net economic effect is that they are taking
money out of the pockets of iSeries users and are giving it to
their Web and Linux users (assuming Linux does not get affected by
CFINT), to finance IBM's competitiveness in that arena.

It's great fun as long as they can get away with it. It's like a
tax on every RPG programmer, decreasing your wage compositeness
and reducing the future market for your talents.

Maybe someone can create an AS/400 clone.

Brad Jensen

Brad Jensen