Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!tank!shamash!raspail!bga
From: bga@raspail.cdcnet.cdc.com (Bruce Albrecht)
Newsgroups: comp.misc
Subject: Re: Ethics of crippler circuitry
Message-ID: <1379@raspail.cdcnet.cdc.com>
Date: 4 Mar 89 01:22:11 GMT
References: <176@ucl-cs.UUCP> <1989Mar2.193443.17196@sq.com>
Organization: Control Data Corporation, Arden Hills, MN
Lines: 73

In article <1989Mar2.193443.17196@sq.com>, msb@sq.com (Mark Brader) writes:
> > Do ethics really enter into this?  Nobody actually gets stung,
> > and the only reason people feel dissatisfied seems to be some
> > kind of resentment because "I could have done that".
> 
> If the fast version of the machine sells for $B (B for Big Number), and
> the crippled version sells for $S, and the crippling really is done
> simply by adding something (something non-functional and of negligible
> cost) to the fast version, then the manufacturer must be making $(B-X)
> profit on each fast machine and $(S-X) on each crippled one, for some $X.
> Unless they're engaging on "loss leader" or similar strategies, both numbers
> are positive.

Suppose the situation is this: the company expects to sell N fast machines
for $B, and the R&D for developing the fast machine is $R, or $R/N per machine.
X is the cost to manufacture and sell the machine.  The profit for the big 
machine is therefore $(B-X-R/N).  If, in addition to the fast machine, there 
is a market for M "cripplied" machines, and it would cost .1 R to design 
cripple circuitry, but .8 R to design a separate, slower machine.  Therefore, 
if it costs roughly the same to build the fast machine or the slow machine, 
the profit for the slow machine is either $(S-X-.1R/M) or $(S-X-.8R/M).  As 
you can see, it is far more profitable for a company to build a system that 
is as fast as it can build, and slow it down, than it is to design two 
separate systems.  So the company is better off selling a cheap "crippled" 
system, even though the manufacturing costs aren't cheaper, if the R&D costs 
are high.

When the computer system has proprietary OS and applications, the buyer is 
mostly buying the applications, but also buying price performance.  Even if 
the customer buys the original system for price-performance reasons, any 
subsequent purchases are probably for the applications.  It makes a lot of
business sense, especially when you're selling proprietary software, to
sell wimpy machines to capture new customers that are on a low budget and
buying applications for the first time, because repeat sales are easier than
finding new customers.  After all, there are a lot of businesses out there
that buy a cheap system, wait until they are exceeding their computing
resources, and then upgrade to a larger system, rather than spending the
big bucks to buy the more powerful system in the first place.

> The ethical issue arises in two ways:  if several companies are in
> the same market and many or all of them are using cripplers, there is
> an appearance of price-fixing.  And if only one company is in a market
> and is using a crippler, they appear to be taking advantage of a type
> of monopoly.  But for these purposes "a market" may mean one segment of a
> highly segmented market; computers of different brands sometimes don't
> really compete against each other; so these situations do happen.  Now, is
> price-fixing ethical?  Is this price-fixing?  There's your ethical question.

It's not price fixing unless the companies agree to set prices together.  In 
reality, the computer companies set their prices based on their older
equipment, other companies current prices, manufacturing costs, R&D costs,
etc.  It's no more price fixing than when your local gas stations all sell
gas at the same price, or within 2-3 cents.

> The resentment arises from the perception that this may be happening.
> It also arises from the fact that the existence of the crippler circuit
> makes the buyer feel that the profit increase of $(B-S) is *unreasonable*;
> after all, the vendor is earning that increase for nothing.  So the buyer
> feels taken advantage of.  Probably if the vendor sold the fast machine
> only, the price would be between $B and $S; but the buyer sees only that the
> vendor could have sold for $S and still made a profit, and resents that.

I doubt that most corporate buyers feel cheated.  They are, after all, buying
a product with a certain performance.  How it obtains that performance is
irrelevant.  I suspect that the people that are the most outraged at "crippled"
machines don't even have any financial stake in it, don't understand the basis
for pricing, and are mostly upset because they really want a supercomputer for
the price of a PC, and they think that building a "crippled" machine is
preventing them from getting one.

> I have tried to present this neutrally, but my emotion is with the buyer.

It shows.