[ExI] Uploads
Ben Zaiboc
ben at zaiboc.net
Sat Mar 28 17:27:12 UTC 2020
On 28/03/2020 15:53, billw wrote:
> Imitating biochemistry: the brain initiates a lot of behavior that
> other organs carry out, such as endocrine glands. So - will an upload
> contain or create virtual endocrine glands? Virtual leg muscles for
> virtual running? Liver etc. for virtual digestion? And so on.
> Will they disconnect links to how to curl your toes?
>
> It seems that you all want a brain that is pure thinking, and there is
> so much of it that is devoted to other things that maybe you will
> upload far less than what is in your full brain. What good will
> uploading your medulla do? No lungs, no heart,etc. Or the cerebellum
> - no motor control needed.
>
> Will all emotions be included? Could we just not upload fear? Or
> whatever?
Personally, I think the sensible approach would be to start by emulating
everything that you can. Including the stupid things, the things we
don't know why they exist, the things that seem irrelevant, etc. So yes,
endocrine glands, leg muscles, liver, toe-curling reflexes, the lot. And
that includes all the emotional responses (which you probably couldn't
avoid anyway).
Initially.
Then, we're in a good position to start fine-tuning things. Think the
recurrent laryngeal nerve is a nonsensical by-product of blind
evolution? (Yes, of course it is). Right, then re-route it. Bear in mind
this is all an emulation, so such things would be easy, once the basic
emulation is working. Did that work? Any unexpected side-effects? No?
Fine, carry on. Tweak one thing after another, go cautiously, maintain
rollback points, undo things that didn't go as expected. Imagine how
much we will learn! It will take a long time, yes, even at software
speeds, but you can probably see that gradually, an improved virtual
organism would emerge.
"It seems that you all want a brain that is pure thinking"
This is *far* from the truth. Anyone who gives any serious, informed
thought to the matter of uploading will realise that a sense of
embodiment will be essential. The aim, initially, would be to emulate as
much as possible everything that we can sense, plus the things that we
can't directly sense, that contribute to our selves. Only after we have
mastered that would we start to tinker with the design, and who knows
where we might go after that.
Here's how I would describe it:
Imagine that you woke up one morning, and someone told you that you'd
been uploaded (or you remember making the decision to upload, the
previous day). But /nothing has changed/, subjectively.
You feel exactly the same. Things look, feel, taste, smell, exactly the
same. The world is the same. Your emotions are the same, your body is
the same.
Then someone shows you that by thinking certain things, or even by using
a control panel in your environment, you can change things selectively.
Press this button, and you get 2cm taller. rotate that knob and your
apparent age goes up and down (with all the attendant internal changes
as well as your appearance). You realise that your body doesn't hurt in
the small ways that biological ageing has gradually brought on. Then you
learn that you can make much bigger changes, including to your
environment. Over a period of subjective months or years (which may or
may not be real-world months or years), you learn more and more how to
change things, control yourself and your environment, and how to
communicate in many different ways with other people, including other
uploads. At some point, you learn to use an interface with the real
world, and perhaps inhabit the body of an avatar, or a non-human robot,
or, well, or just about anything really. A hummingbird. A dinosaur,
whatever.
At no point do you even remotely feel like a disembodied mind (unless,
that is, you normally used to!). At no point do you notice anything
/missing/. Except the things that you deliberately change, like that
arthritic pain you always got in your left hand. Maybe at some point you
decide to find out what it's like to do without the feelings of hunger.
Maybe that doesn't go so well, so you reverse the decision. And so on,
and so on, for many many centuries. Or more.
--
Ben Zaiboc
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20200328/38db7dc3/attachment.htm>
More information about the extropy-chat
mailing list