Re: More details on the NT Workstation TCP limit

Robert S. Thau (rst@ai.mit.edu)
Wed, 7 Aug 1996 13:39:47 -0400


[Deliberately crippling products to manipulate the marketplace is a
strong sign of monopoly power... dee3]

Hmmm... there's an ongoing thread on Lusenet (widely crossposted)
discussing tricks like this by various firms over the course of time.
Many of these come from people with first-hand knowledge, including
some from firms which were nowhere near a monopoly position.

One instance was a Honeywell minicomputer from the 1960s, all of which
were built with a fully functional interrupt system, which was always
tested as part of the ship process. This test was a little more
awkward to perform on the cheap ones, since the CE performing the test
had to remove the "cost-reducing" card which was shorting the
interrupt request line to ground. Data General did something similar
with the power-fail detect option on some of their machines in the
'70s, clipping the leads on an IC to disable the feature if the
customer hadn't paid for it. (In both cases, the people reporting
this stuff had first-hand knowledge --- the one posting the Honeywell
story actually was a CE who performed those tests).

There is of course, a long history of nasty stories and nastier rumors
about DG's past marketing practices, but neither they nor Honeywell
were ever near a monopoly in any sector of the market. In fact, the
Honeywell machine was going directly up against IBM, which had near-
total market dominance at the time (it was nearly 1401-compatible, and
a translator to convert your 1401 assembly programs --- "the
Liberator" --- was a heavily touted feature). So, ugly as these
stories are, I wouldn't call them a sure, or even strong, sign of
monopolistic behavior.

rst