Re: [Gnu-arch-users] [OT] funding free software R&D

[Stephen J. Turnbull]

>>>>> "Andrew" == Andrew Suffield <address@hidden> writes:

    Andrew> And that's not even considering all the new design
    Andrew> techniques and workarounds that engineers keep coming up
    Andrew> with. I've seen more stuff with significant potential in
    Andrew> the past couple of years than I can count.

They're irrelevant to the argument.  The basic pessimistic argument
says that Heisenberg's uncertainty principle puts a limit on the
number of angels that can dance on the head of the pin, and then
extrapolates the exponential increase in the number of angels until it
bumps into that limit.  So even if we conservatively say "the best we
can do is 1 million times the Heisenberg limit" to figure that the
worst case is 20 years, we don't buy much with optimism.  A factor of
100 improvement by "workarounds" postpones the inevitable by (say) a
year.  Another factor of 100 gives you another year.  Now you've run
out of factors of 100, and Moore's Law is dead.

That's the way exponential growth works.  It's hard to buy order of
magnitude improvements when you're working against exponential growth.

So you need a fundamental change in either the architecture (eg,
quantum computing, as you suggest) or the definition of the nature of
the problem.

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.