Re: Fwd: Implementation/Relativity

From: Hans Moravec <hpm.domain.name.hidden>
Date: Fri, 30 Jul 1999 16:22:07 -0400

Christopher Maloney <chris.domain.name.hidden>:
> the problem, which makes such a point of view [finding
> interpretations that contain self-aware observers in non-traditional
> objects] useless, was pointed out by Hal. Fictional scenarios are
> lawless. In our world, we invariably see that some sort of order
> and persistence manifests itself. But in fiction, I, or any author,
> is free to make up whatever he or she wants, almost without
> constraint.

Universes without constraint are know as chaos. They're interesting
in their own right, but fiction stories don't fall among them.

To be enjoyable, or even readable, fiction has to have overwhelming
structure and self-consistency, enough to be recognizable as physical
life in this, or some nearby, universe. Many authors complain they
lose their freedom to the demands of recurring characters they
created: facts and character traits introduced in past stories begin
to narrow the possible behavior of the character in subsequent
stories. The character becomes more and more clearly its own person.
Created universes more and more define their own possibilities. The
young Isaac Asimov said he couldn't do a fourth Foundation book
because the work of keeping track of everything had become too
large. (i.e. the complexity of the Foundation galaxy had outgrown
his capacity to simulate it! Many decades later, the old Isaac
Asimov did write a couple of additional books connected to Foundation
in a semi-detached sort of way, but anyone could see they weren't
really about the same universe.)

If you refuse to interpret (say) detective stories as containing
self-aware characters, then there is no point to reading them.
Outside of a certain interpretation they are indeed just strings
of meaningless ink blobs on paper. A useful point of view for a
paper recycler.

By the same token, if I don't interpret you as a self-aware
being, then there is no point listening to you. Without the
proper interpretation of your sputterings you are just an noisy
bag of meat. A useful point of view for a hungry tiger.

> I also believe that it must be possible to establish some kind of
> measure over the universes, that gives rise to the physical laws we
> witness. So, in some sense, they are "real". If you disagree with
> this, kindly define what you mean by "abstractions"

A universe I see depicted on the screen of a simulator in front of me
has enormous measure of existence (1.0) for me. As do a universes
recorded in books that issued from the wet simulator we called
Asimov's brain.

> my own existence is the only thing I can be really sure of
> ... trapped inside some sort of universe that I never made, and that
> obeys physical laws. How do you explain that?

You interpret the universe as real. The universe returns the favor.

Sherlock Holmes interprets his fictional London as real. His London
treats him as existing in it.

Same deal.

> If you work with simulators, you must know that the lower the
> fidelity of the simulation, the less useful it is

Leaving out inessential details is how mathematics generalizes. It is
often immensely useful, because with details gone, your results apply
to a larger universe of examples.


> As I said in my earlier post, I could cut open a persons head, and
> see all kinds of gunk in there that I could use to explain that
> person's behavior. But how would I explain the behavior of a
> fictional character?

By cutting open his head in the story and finding all sorts of gunk
that explains his behavior. I know lots of stories where that
happens. I guess you could group them under the term "neurosurgery
fiction". (Of course, most stories don't involve cutting open
characters' heads. I hope the same holds for most of your physical
world relationships.)


> [The Sherlock story in an AI VR implementation]
> Then, and only then, does he become an implementation of a
> conscious structure.
> ...
> [interpreting a human being, predictively, in purely mechanical
> terms]
> And I would say that they are wrong.
> ...
> [Mapping psychological traits onto non-traditional things]
> Then they become useless.
> ...
> (Many additional tediously predictable pronouncements)

Says you. You're entitled to your opinion, but I wouldn't give you
two cents for it.


> Are you saying I'm wrong, and my self- interpretation is
> questionable?

You self-interpretation is fine, but it's not unique. It is one of an
astronomical number of possible alternatives, and some of the others
(like the mechanical one, for biologists) are useful to us even now.
Others may be significant to other observers of whom we are not
presently aware, because they are interpretively too distant from us.
But as our computing power grows, each time we build a new simulator,
or a gadget that looks at our surroundings in a new way, our
interpretive horizon expands.
Received on Fri Jul 30 1999 - 13:31:44 PDT