Re: Algorithmic Revolution?

From: Tim May <tcmay.domain.name.hidden>
Date: Tue, 19 Nov 2002 18:47:12 -0800

On Tuesday, November 19, 2002, at 05:12 PM, vznuri.domain.name.hidden wrote:
> I would take all of TCMs own citations and turn them around in my
> favor. I would classify all the following as occurring under
> the heading "algorithmic revolution" (not the greatest moniker I
> admit..
> a provisional one for me right now)
>
> - advent of cyberspace, web, email, browsers etcetera
> - advent of mass software
> - PC revolution
> - microsoft & intel from zero to billion dollar companies in short
> decades
> - quantum computing is on the way
> - fractals. could not be discovered without algorithms. a new metaphor
> for not only nature but all reality.
> - complexity theory. again, not possible before the algorithmic
> metaphor
> and mass computational capabilities
> - simulation, "in silico science"
> - moore's law
> - photorealistic rendering
> - (relational) databases
> - mass economic shift into information technology as driving force..
> "bits versus atoms".. (negroponte)
> - video games
> - etcetera!!

I agree that these are all huge changes.

I interpreted your "algorithmic revolution," in the context of this
list and the Kevin Kelly article and the Wolfram brouhaha, to be about
a revolution in terms of thinking of the universe (or multiverse) as
being primarily computational.

My point is that the verdict on the
Zuse/Fredkin/Wheeler/Lloyd/Wolfram/Tegmark/Schmidhuber/etc. views of
reality is still way, way out. I stand by this point.

If by "algorithmic revolution" you meant that computers are
increasingly important, then of course I agree.
...
>
> ahem!!! what is the relevance to a TOE??? well historically it is clear
> our perception of reality is based on our favorite metaphor of the
> times.
> in recent ages it was (a) the clock, "clockwork universe", (b) the
> steam engine. and now it is (c) computer/algorithm/information.
> clearly
> it is no coincidence whatsoever that new TOEs are essentially
> algorithmic.
> its the human race's latest-and-greatest metaphor for reality.

Yes, and our past experience in going through all of these metaphors or
"mathematical fictions" has made many of us wary of saying things like
"The universe is like a hologram" or "The universe is about connection"
or even "The universe is a gigantic Game of Life."

The issue of an ontology being a metaphor is an interesting one.

I currently have no view of any particular metaphor for what the
universe "is." It may have computational aspects, and mathematics
(superset of computer science, of course) may be woven throughout the
structure of reality. It may even have "holographic" or "clockwork" or
"cellular"-like aspects. But aspects are not the same thing as
equivalence.

Greg Egan makes a good point in "Diaspora" about the limitation of the
mathematical models we sometimes use as metaphors for reality. A mass
falling through a borehole through the Earth acts exactly as if it's a
mass on a spring tethered at one end. Same precise equation of motion.
Yet a spring is not at all what the Earth is, and confusing the
mathematical model with reality is dangerous.

Of course, at this point we have much, much less reason to speculate
that the universe "is" a cellular automaton of some sort.
>

> scientists have been slow to adopt to this shift, and I would argue
> they
> are still underutilizing simulation to some extent. science & physics
> is still yet to be influenced fully by the algorithmic revolution. one
> striking example I think will happen-- I believe billion
> dollar particle accelerators
> may be downgraded in importance in favor of extremely effective
> simulations.

The reason experiments are still done is because they are the real
proof of the pudding about what the universe really _is_.

> (besides-- does anyone fully realize how much software plays already
> such a
> crucial, foremost role in existing accelerators??)

Not sure what you mean. I did some coding of Monte Carlo simulations in
my physics days, and I hired some of the coders from SLAC to work on
some of the stuff we were doing at Intel. Software is used to design
the accelerators, the detectors, the experiments, etc. As with the rest
of the world, computers and software are undeniably important.

I'm not doubting the importance of computers. Nor the importance of
clocks and wristwatches. But just as we know "the universe as a
clockwork mechanism" was not the whole picture, I think "the universe
as a computer" is not, without a lot more evidence, very compelling. To
me, at least.

(I am interested in being convinced otherwise. And I have my own
interests. Today I ordered the Peter Johnstone 2-volume set "Sketches
of an Elephant: A Topos Theory Compendium." Not that I am saying "the
universe is a topos.")


--Tim May
Received on Tue Nov 19 2002 - 21:52:47 PST