Rash's Judgment: Industry Barometer Is Stable
 
Microbytes Daily News Service
Copyright (c) 1989, McGraw-Hill, Inc.
Comdex is always a good barometer of the microcomputer industry.
Sometimes it serves to tell the direction of new technology,
sometimes it shows the latest rage in applications. More often,
though, it shows the great changes that grip the industry and
mold it for the long run. The hype and excitement obscure this,
but deep within the movement of Comdex is the direction you will
find yourself taking over the next months.
 
This month's Comdex showed very clearly that the movement is
toward consolidation, stability, and a period of catch up. It will be
a period in which software finally begins to reflect the
capabilities of the hardware that it is expected to support. It
will also be a period in which products are allowed to become
more fully mature, more thoroughly developed, and a better deal
for the buyer.
 
This is not to mean that the relentless move to more speed,
greater memory, and higher data densities will cease, but
rather that they will become more predictable. The evolution of
the Intel 80486 illustrates that point. For the first time since
the company began to dominate the central processor market,
Intel's latest CPU is completely compatible with its predecessor.
Software that runs on the 80486 will also work with an 80386
equipped with a math coprocessor and a cache controller.
 
This means that the hardware environment for IBM-compatible
computers is stable. Developers know what to expect, for a while
at least. Instead of being intimidated by the headlong rush of
change, they now have the time to create software and peripherals
that will fill real needs. It's equally clear that the operating
system wars in the IBM market have calmed down also.
 
While there is some doubt about exactly what caused it, IBM and
Microsoft have made it clear that they agree on the future
direction for MS-DOS and OS/2. It's clear that there will be a
place for Windows, and it's equally clear that IBM is serious
about supporting OS/2. Software developers will have to decide
whether to support either or both, but at least they will know
what to expect.
 
More important, though, is the effect that this stability will
have on the customer. Prior to this, products often were not
developed as completely as they could have been because of the
short life cycle they were expected to have in which to earn
their keep. Often, this time was only about 18 months, of which 6
months were spent in development. Products were obsolete before
they were out of warranty. As Zenith Data Systems president John
Frank said to me during Comdex, "We were devouring our young."
He explained that the customers were getting less than they
should have because no company could afford to support adequate
development. There simply wasn't time.
 
Now that we know the direction of the next few years, there will
be time. We will see our goals years in advance instead of only
weeks away. Products will benefit, and we as customers will
benefit.
 
                              --- Wayne Rash
 
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
