According to several reports, for the fist time
in history sales of personal computers are decreasing
(especially in the United States).
In a way... thats pretty obvious. It had to happen. All
markets reach saturation: a time when most of the
people who want something have it, and the replacement rate
is necessarily slower than new purchase. If computer
manufacturers thought that the same growth rate could
continue forever, they were out of their minds.
For several years the growth rate was forced by phony
innovation. Unnecessary changes in technology (especially
software) were developed for the purpose of making perfectly
viable machines obsolete. They even invented a
law that doesnt exist: so-called Moores
Law is based on a warped interpretation of an analysis
published by Gordon Moore in 1964 and not confirmed by
facts in the following 37 years. Its quite surprising that
these manipulations could continue for so long. But sooner or
later, in one way or another, they must come to an end.
On the other hand... there is huge market, unsatisfied
potential demand, for new computers (as well as properly
re-engineered old ones). Billions of people, millions of
organizations worldwide that dont have any electronic
equipment at all or much less than they need. But to
satisfy that demand we need machines that are very different
from those on the market now. Not only much cheaper (which of
course is possible, considering the decreasing cost of
artificially obsolete components and the enormous
economies of scale). They must also be more reliable, have
low energy consumption and above all they must be stable
over time. With software that is simple, fully tested,
thoroughly debugged and not subject to unwanted
and whimsical innovation. And of course
opensource operating systems. What is needed is a sturdy machine
that can work efficiently without any changes (and limited maintenance)
for ten years or more. With peripherals
and accessories of the same sort. Can it be done? Yes, of course.
But we still dont know how, where and when.