Re: Uploading
Mon, 4 Nov 1996 16:53:23 +0000

> >Information itself cannot be moved because it has no
> >physical properties.
> It seems to me you could make exactly the same argument if you said
> information always moves and can never be brought to a halt, which leads me
> to think this is not a fruitful avenue to pursue.

I was saying that you cannot produce a physical or spatial translation of
information without also producing an equal physical or spatial
translation of the encoded substrate.

> >The physical translation of information from one point in
> >space to another is meaningless in an information context.
> I think it's very meaningful. After Turing machine A produces an output
> there will be a delay before Turing machine B can use it, this delay is
> proportional to the distance between A and B.

This is a physical context, not an information context. You are
generating a copy of the output of Turing machine A on Turing machine
B. The information that is output by Turing machine A exists as soon
as it is generated. The problem of copy latency is a physical one,
not an informational one.

> >If I am not mistaken, the "Information can not be moved,
> >only copied" theorem has been proven mathematically.
> I very much doubt such a murky idea can even be expressed mathematically,
> much less be proved or disproved.

Nothing murky about this at all. This is a very basic theorem that
has been proven in very concrete terms. There are many similar
theorems. A large areas of information theory is counter-intuitive.

The science as a whole is very practical and is generally used in the
design of complex, distributed information systems. Large-scale
distributed databases and distributed operation systems use this
area of mathematics extensively. In its full scope it also includes
topics such as data compression and cryptography.

Another interesting mathematically proven theorem of information
science: It is mathematically impossible for an information agent
(such as humans or computers) to prove that their information is
synchronized with that of another information agent.

The above theorem is also counter-intuitive (with many implications)
but true nonetheless. In the real world, distributed systems assume
synchronization because they can't really know. Humans also "assume"
synchronization, often called "common knowledge".

> >I am not saying it is impossible to upload; it is impossible
> >only in the sense of uploading a single, existing stream of
> >consciousness as opposed to creating a new stream of
> >consciousness.
> Every time the telephone rings my stream of consciousness is interrupted,
> when you read this post your consciousness is different than it would be if
> you were reading another post. I don't see why that is such a big deal that
> it makes us different people.

Your consciousness is interrupted, but not terminated. Our minds,
like most modern operating systems, are event-driven and capable of
running more than one process at the same time, some in the
background, some in the foreground. If this was not the case, you
would have to think about breathing to make it happen, and a
phone call would make you forget to breathe.

> >If consciousness is based upon information and information
> >processing, my above statements hold.
> Unless the religious people are right, consciousness must be based on
> information processing, that's why I don't understand your theory. You're
> saying, I think, that information exists outside our physical universe,
> if so then the information in my head and the information in my upload a
> thousand miles away must be the same, after all you also say " a physical
> translation in space of the substrate is irrelevant". Also, I don't see why
> you keep distinguishing between "originals" and "copies", if your theory is
> right then everything in the physical universe is a "copy".

Information exists in the physical universe, but it has no physical
properties. There is a big difference. There is no
difference between originals and copies. The problem starts when
consciousness comes into play. If you think of your mind as an
instance of a very complex block of information, you create a
situation where a block of information contains the ability to
recognize the difference between itself and all other copies of
itself via the mechanism of self-awareness. Self-awareness allows
a block of information to uniquely recognize itself without necessarily having
any differences between itself and it's copies.

Yes, the copies may be identical, but to the self-aware, they are not
the same as itself. Other self-aware, exact copies would be
identical to each other from its own viewpoint, though.

> >If every piece of substrate containing a specific piece of
> >information is destroyed, the information is destroyed
> >permanently.
> Not necessarily. It would be difficult to destroy the information that
> 2 + 2 = 4 because nobody knows where or how it's encoded. Even Shakespeare's
> plays could be rediscovered by a monkey banging on a typewriter, it would
> take a long time, but not an infinitely long time.

2 + 2 = 4 is hard to destroy for a number of reasons. Mostly,
though, it is because it is thoroughly encoded on many, many
substrates throughout the universe.

As for Shakespeare, you wouldn't know who Shakespeare was if every
bit of information on Shakespeare was destroyed, and therefore, you
couldn't "rediscover" the information. Rediscovery requires prior
knowledge hence the information could not have been completely
destroyed. The monkey might
bang out plays, but they would no longer be Shakespeare's plays
because there would be no evidence in the universe that Shakespeare
ever wrote a play. They would be random noise generated by a monkey
pounding on a typewriter.

> To my mind this does indeed give some support to the idea that information
> can exist external to the physical universe, but certainly not to the idea
> that there is a fundamental difference between a copy and an original or that
> exactly the same information can't be encoded in 2 very different ways.

How information is encoded is irrelevant to the information itself.
Information *can* be encoded in two different ways, and usually is.
The way your mind stores a block of text is very different than the
way it is stored on a page, but the information content is the same
either way.

-James Rogers