John F. Sowa wrote:
Azamat
et al.,
I received an offline note with a pointer to a web site
of "dangerous questions" -- one of which is a point by
the physicist Lawrence Krauss that perhaps a "Theory of
Everything" is impossible, and the only thing that is
possible is an open-ended collection of "phenomenological"
theories about the various ways of perceiving the universe.
(Copy below).
Another voice speaking out against a "Theory of Everything,"
and advocating the replacement of "reductionist" approaches to
natural science (based on deduction from first principles) with
empirically-based approaches of discovering "principles of
collective organization" is Nobel-prize winning physicist
Robert Laughlin. He has a recent popular book:
A Different Universe: Reinventing Physics from the
Bottom Down
that develops a thesis he presented (with David Pines) in a
PNAS essay:
p01apr99
-
R. B. Laughlin and D. Pines, "The Theory of Everything",
Proc. Nat'l. Acad. Sci., 97, 28 (2000).
For a quick overview, see his slides on "Self-Organization of Matter"
http://large.stanford.edu/rbl/lectures/index.htm
He makes a comment in a footnote of his book about at technical
sense of "protection" that he uses, which is a popular stand-in for
some idea involving of fixed points or renormalization groups. I
suspect this may connect to the uses of category theory being
advocated in Robert Kent's Information Flow Framework, which
SUO has accepted as a starter document, and applied in the
work of Kalfoglou and Schorlemmer to coming up with practical
operations for mapping and merging ontologies:
"Using Formal Concept
Analysis and Information Flow for modelling and sharing common
semantics: lessons learnt and emergent issues"
Y.Kalfoglou, Marco Schorlemmer
Abstract -
draft
copy (PDF) - BiBTex entry
"A Channel-Theoretic
foundation for ontology coordination"
M.Schorlemmer, Y.Kalfoglou
This EUMAS'04 paper is
similar to that of MCN'04, but we were invited to present it to a
different audience: agents' community.
In Proceedings of the 2nd
European Workshop on Multi-Agent Systems (EUMAS'04), Barcelona,
Spain, December 2004
Abstract
- draft
copy (PDF) - BiBTex entry
"IF-Map: an ontology
mapping method based on Information Flow theory"
Y.Kalfoglou, M.Schorlemmer,
In this JDS article we
extend the work presented in the ODBASE'02 paper. We present IF-Map, an
ontology mapping method based on Information Flow theory. This the
definitive IF-Map resource and we present the theoretical background
with the channel theory preliminaries and their interpretations in
ontology mapping, the methodological approach to ontology mapping, an
operational architecture for executing the IF-Map engine, and an
exemplar case from the academic departments organisational ontologies
domain.
Journal
on Data Semantics 1, LNCS 2800, pp.:98-127, Springer, ISBN:
3-540-20407-5, October 2003
Abstract
- draft
copy (PDF) - BiBTex
entry
Cheers,
Fred
--------------
Frederick B. Kintanar
NEC Telecom Software Philippines
Cebu City
The word
"phenomenological" is important because all the
theories that we have been calling "ontology" should more
properly be called "phenomenological". They don't really
explain the ultimate categories of Being (Aristotle's
most general type, "to on"), but the range of possible
ways that people (or cats or chimpanzees or computers or
extraterrestrials) perceive the phenomena of the universe.
Krauss states the hypothesis that perhaps there is no
ultimate TOE. Penrose makes the point that there may be
an ultimate TOE, but it certainly won't resemble any of
our current theories of physics. Just consider how far
the relativistic or the quantum mechanical views diverge
from the Newtonian view. Any ultimate TOE will certainly
diverge from our current theories at least as much.
Summary: Forget the elusive search for an ultimate unified
ontology because (a) it might not exist, (b) even if it
did exist, it's not likely to be found for at least another
century or so, and (c) even if it were found today, the most
important things we need in our ontology are descriptions
of the phenomena as we and our fellow earthlings (of all
species) perceive them. And those phenomena are as varied
and variable as there are earthlings, species of earthlings,
and instruments (such as telescopes, microscopes, MRI
scanners, etc.) for perceiving and recording them.
John Sowa
__________________________________________________________
Source: http://www.edge.org/q2006/q06_8.html#krauss
The world may fundamentally be inexplicable
LAWRENCE KRAUSS
Physicist, Case Western Reserve University
Science has progressed for 400 years by ultimately explaining observed
phenomena in terms of fundamental theories that are rigid. Even minor
deviations from predicted behavior are not allowed by the theory, so
that if such deviations are observed, these provide evidence that the
theory must be modified, usually being replaced by a yet more
comprehensive theory that fixes a wider range of phenomena.
The ultimate goal of physics, as it is often described, is to have a
"theory of everything", in which all the fundamental laws that describe
nature can neatly be written down on the front of a T-shirt (even if
the T-shirt can only exist in 10 dimensions!). However, with the
recognition that the dominant energy in the universe resides in empty
space — something that is so peculiar that it appears very difficult to
understand within the context of any theoretical ideas we now possess
— more physicists have been exploring the idea that perhaps physics is
an 'environmental science', that the laws of physics we observe are
merely accidents of our circumstances, and that an infinite number of
different universe could exist with different laws of physics.
This is true even if there does exist some fundamental candidate
mathematical physical theory. For example, as is currently in vogue in
an idea related to string theory, perhaps the fundamental theory
allows an infinite number of different 'ground state' solutions, each
of which describes a different possible universe with a consistent set
of physical laws and physical dimensions.
It might be that the only way to understand why the laws of nature we
observe in our universe are the way they are is to understand that if
they were any different, then life could not have arisen in our
universe, and we would thus not be here to measure them today.
This is one version of the infamous "anthropic principle". But it could
actually be worse — it is equally likely that many different
combinations of laws would allow life to form, and that it is a pure
accident that the constants of nature result in the combinations we
experience in our universe. Or, it could be that the mathematical
formalism is actually so complex so that the ground states of the
theory, i.e. the set of possible states that might describe our
universe, actually might not be determinable.
In this case, the end of "fundamental" theoretical physics (i.e. the
search for fundamental microphysical laws...there will still be lots of
work for physicists who try to understand the host of complex phenomena
occurring at a variety of larger scales) might occur not via a theory
of everything, but rather with the recognition that all so-called
fundamental theories that might describe nature would be purely
"phenomenological", that is, they would be derivable from observational
phenomena, but would not reflect any underlying grand mathematical
structure of the universe that would allow a basic understanding of
why the universe is the way it is.
|