Theories all the way down (Was: Uploading)

From: Anders Sandberg (asa@nada.kth.se)
Date: Tue Jun 25 2002 - 14:04:15 MDT


<grandiose speculation mode on>

The popular idea that quantum fuzziness/discreteness is a sign of the
universe being run on a computer or romething similar (being an
"embedded system" so to speak) is based on the assumption that such
implementations have to be discrete. It makes a lot of sense in our
universe, since stuff generally is discrete, good information processing
seems to require some discreteness (error correction at least) and
finite resources doesn't allow arbitrarily detailled simulations. But
these are common sense observations based on our world, and do not
necessarily carry over to the hypothetical surrounding universe.

Maybe the best way of approaching this issue is to look at what systems
can sustain complex systems such as observers (if such are assumed to
exist in the surrounding system) and whatever systems are needed to
emulate our kind of universe. Given that most rule sets do not produce
anything like this, there might be just a few classes of systems that
work (this is BTW where I think Wolfram is actually saying something
worthwhile; but in fairness Marvin Minsky made the same point about
"interesting systems" in his essay on alien communication; I also think
we need to look at simple continuous systems more than Wolfram does). If
there are just a few classes, then it might be possible to say something
about the relative likeliehood of us being embedded in a discrete
or continuous surrounding system.

On Tue, Jun 25, 2002 at 03:05:35PM -0400, Mike Lorrey wrote:
>
> I think an interesting thought experiment would be to devise a set of
> physical laws which are analog all the way down to the most discrete
> subatomic level. Such could represent a really 'real' universe from
> whence all simuverses derive.

Why stop there? Why not construct a set of laws that are analog all the
way down. Just imagine a soft universe made of fields, where particles
act as solitons.

Another interesting possibility (and hence the title of this post) is
whether there exist laws that are self-similar or periodic in some way
all the way down (and presumably up). I personally hate "maybe atoms are
universes" speculations for some odd reason (maybe too much bad sf when
I was young, or a dislike of idle speculation based on loose analogies),
but the hypothesis that physical laws are scale invariant seems to be
possible to refute or do some real thinking with.

Most likely it would imply that as we go on larger and larger scales new
forces (very weak to compared to the others) would appear
(quintessence?) and on small scales certain forces would actually become
less powerful (hence the experiments on short-range gravitation would be
very useful to test this idea) and other, extremely-short range forces
would show up.

The distribution of stuff would have the same statistical properties at
different scales (although spacetime messiness at the Planck length
could make this rather strange; it could well be that in a system like
this spacetime as we know it is a large-scale phenomenon akin to the
entire inflationary universe, while below certain length-scales another
*kind* of spacetime exists (akin to the bubbles). It could be something
related to our spacetime as a dual or in the way wave functions relate
to quanta, i.e. very abstract. In any case it seems that having the
periodicity or self-similarity constraint as well as the demand that
there can exist complex systems could be enough to constrain such
theories to a non-trivial set.

Just some idle speculations instead of working. Infinite Fun Space is so
tempting.

-- 
-----------------------------------------------------------------------
Anders Sandberg                                      Towards Ascension!
asa@nada.kth.se                            http://www.nada.kth.se/~asa/
GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y


This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:15:01 MST