[extropy-chat] Re: A slow down in Moore's law?

[Acy James Stapp]

I promise this is my last post on the subject, unless it comes up again
:)

Tuomi addresses many variants of Moore's law, all leading toward the
metric Kurzweil proposes, which is in essence, computation per dollar.
So Kurzweil's analysis is applicable in the context of the Tuomi paper.

Kurzweil sidesteps the whole "Moore's Law" issue by restaging the
advance of information processing capability into his own "law of
accelerating returns". The introduction of the phrase "Moore's Law" to
describe Kurzweil's ideas on the future of computation was my doing. 

There is a fundamental difference of opinion between the two on the
rate, shape, and limits of information processing capability which
drives their argument. Tuomi does seem to be advocating that the advance
of computing technology is in fact slowing (or rather, the acceleration
is decelerating). 

I will finally note that one can be fervent about something without
being religious. Fervor is a a strong expression of passion for and
excitement about some thing. Fervor in service of an ideal serves to
direct and motivate one to work toward that ideal. Sometimes, even
often, it is to the detriment of rationality but it can drive one to
action in support of a rationally determined goal. (Of course, for most
humans, this goal will still serve some basic genetically determined ned
or desire.)

Acy

-----Original Message-----
From: Christian Weisgerber

Acy James Stapp <astapp at fizzfactorgames.com> wrote:

> > http://www.firstmonday.dk/issues/issue7_11/tuomi/index.html
> And, to counter:
> http://www.kurzweilai.net/meme/frame.html?main=memelist.html?m=1%23593

To counter what?  People are arguing at cross purposes.  The Tuomi
paper shows that "Moore's Law" is ill-defined and the usual attempts
to phrase it are counterfactual.  The thrust of the argument is not
that computers aren't getting faster.

> Kurzweil argues that the relevant metric for Moore's law is price per
> transistor cycle and provides much more convincing evidence (IMHO)
that
> Tuomi.

Kurzweil comes up with yet another metric that he tries to shoehorn
into "Moore's Law".  All this does is to reinforce that people can't
agree on what "Moore's Law" actually is.  The term is meaningless.
I know people on this list like to reference it with religous fervor,
making them sound all the more naive.