[extropy-chat] Identity Transfer (Re: A view on cryonics)

[Brett Paatsch]

Slawomir Paliwoda wrote:

> Brett wrote:
> >You are currently located in
> >your brain just as I am. Your aspirations for cryonics will have
> >to deal with that.  A cryonics procedure that goes hunting
> >about the room to pick up astral travelling selves is unlike to
> >be of interest to either of us.
>
> That is exactly right. If we assume that protecting our identities must
> necessarily involve preserving our original mind processes instead of
> our mind patterns, then we must expect the future cryonic revival
> procedure to offer a workable process-preserving solution before
> people like me can sign up.

I'd also be a lot more interested then. I'd still want to know more before
handing over money, but your definitions of identity don't rub the wrong
way right off the bat.

> Let me offer such a solution, I've been sitting on for few years
> now, which is a transfer mechanism that relocates an original mind
> process to various substrates. Because preserving identity-carrying
> mind activity goes few steps further than what is required for
> protecting a mind pattern, this new transfer postulates additional
> constraints on the cryonic revival procedure.
>
> Unfortunately, the bad news for potential cryonic patients who
> subscribe to a "true identity is defined by the uniqueness of a mind
> process in space-time" view is that cryonics is almost a hopeless
> technology as far as identity preservation is concerned. This is
> because any reliable mind relocation method prefers not only as
> little loss of mind pattern data as possible, but also, more
> importantly, a mind that is able to function within the original
> substrate.

Ah but where some of John Clarkes objections were off base in
my opinion before, they would not be in your new scenario if I
understand it correctly. I am much more willing to transfer off
my substrate in stages while alive, including theoretically, all the
way of it altogether, in stages provided I can stay aware the
whole time. After all, that IS what I have been doing all my life.
New neurons growing etc. New memories forming etc. Sometimes
while I slept I'd grant John but never while I was totally offline.

An ANALOGY might be that I am a lot more comfortable
conceptually having a new hard disk appended onto my existing
substrate forming a sort of RAID array, provided the process
can be done while I am still "online". I'd be willing to take some
calculated risk for that sort of opportunity. But the tech is likely
to be much harder and my timeframe of potential interest is
someone less than my current life expectancy.  Cybernetics
is perhaps risky but in principle its a clear good. I'd want the
option of it.

> While cryonics is very much vulnerable to the loss of mind
> pattern data, it is probably even more so when it's expected
> to  repair the original hardware to the point when the mind
> process can resume its operation within that substrate. Or
> maybe it's up to nanotech to fix this problem. In any case,
> it seems like we're in deep trouble.

Yup. But of course John C and co are right in that biology
isn't exactly "hardware" anyway.

>
> The good news is that cryonic revival methods are not
> entirely hopeless. Even though a task of reviving a mind
> process within an original hardware will be incredibly hard,
> it doesn't seem impossible.
>
> However, from a theoretical point of view, one might observe
> potentially big problem. Could a mind process survive the
> break in its continuity?

This doesn't bother me excessively. Its a factor but I'd be willing
to risk a little mental illness if I could manage the process. Heck
worse case scenario I die a little sooner and/or the world would
have another nut for a short time. Mental illness is not a fate worse
than death.  I'd dare so long as the daring was done on my terms
and my timeframe. I'd also have a battery of tests I could think
of to incrementally explore and manage and minimse risk.  That
sort of risk taking is fun and rational I reckon.

> After  all, a mind cannot emerge if its
> hardware cannot function. It would be  tempting to reason
> that freezing evaporates identity, which is a property grounded
> in a brain activity that would be unquestionably absent while in
> a deep freeze, but the reason why this would be wrong is that,
> instead of  being absent, the process is merely pausing its
> execution. The process  survives because the hardware and
> software that enable the emergence of a mind are still there
> and so is the identity.
>
> With that out of the way, let me move on to the details of the
> transfer itself. In order to successfully transplant a mind onto
> a different substrate, I impose two constraints on the procedure.
>
> 1) Mind process must function within the original substrate. (It
> doesn't have to be conscious. It just has to execute in some form).
>
> 2) Unity of mind must be preserved at all times.
>
> The first constraint might be obvious. The transfer cannot happen
> without an existing process ready to be relocated.
>
> The second requirement constrains the way in which the shift is
> to be performed to avoid creating additional copies of the mind.
>
> Even though cryonics deals with biological-to-artificial transfers,
> for the sake of clarity, I'll assume artificial-to-artificial transfer to
> better  illustrate the mechanism.


---------

> Two separate machines are located on different planets, and are
> connected by some interplanetary, and sufficiently fast communication
> link. Machine A  executes some person's mind process, while
> machine B consists of hardware suited to receive that mind process,
> but is not running one at the moment. Then, the mind is divided into
> N number of imaginary mind units. At first, machine A carries 100%
> of the units, while machine B has 0%. Transfer begins by reading one
> unit on A at a time and writing it to B. Then, the unit now sitting on B
> gets integrated into the overall mind process. This makes the original
> unit fetched from A redundant, so it is deleted. At this point, the mind
> operates using N-1 units on A and 1 unit on B. The operation
> progresses until 100% of mind units that used to occupy A now inhabit B.

---------
Ok as an analogy.


> In this way, any mind process could potentially be relocated onto any
> suitable medium without the loss of identity, grounded in the mind
process.
> It is that grounding that allows us complete disregard for the substance
of
> the substrate as long as the unity of mind is maintained.

Yes unity is key.

> If the focus was
> only on mind pattern, which so many people are happy to ground their
> identity in, the second constraint would be meaningless here, (what's the
> point of maintaining unity of mind if all we care for is the data itself),
> and that would lead only to logical confusion and situations where two
> copies of the same data might be assigned the same identity.

In the interests of communication clarity I would not agree to using
identity to describe any outcome where people could talk of two
contemporaneous and disconnected instances of an "identity". If there
are two it is not an identity. For identity there can be only one, period.

New words for types of near identity might be coined.

> Obviously, the procedure present here is very strict because it shifts one
> mind unit at a time and allows coexistence of only single unit across the
> machines. I imagine that, in practice, whole chunks of mind units might be
> allowed to coexist during the transfer because they alone wouldn't
> contribute that much to the emergence of a complete and independent mind.
> The second constraint prevents the procedure from creating any additional
> copies of the original process because that would inevitably lead to
> creation of a new mind process and, automatically, a new identity. In that
> case, the procedure wouldn't result in a successful identity transfer, but
> merely a birth of a perfect clone.
>
> The second constraint also requires the mechanism to maintain
functionality
> of a single mind process stretched across different platforms. It's
crucial
> that the one and only copy of the mind subjected to the transfer does not
> degenerate into a non-mind process at any point during the relocation.

> Ideally, a person shouldn't sustain any degradation in his subjective
> experience during the operation.

Ideally, but a little "sleepiness" might be tolerable :-)

> Ultimately, it's all about subjective experience. No transfer can claim to
> be identity-preserving unless it guarantees the integrity and continuation
> of subjective experience during a substrate switch.

Yes !

>  Recognizing
> identity as uniqueness of mind process in space-time helps us define the
> conditions  necessary for subjective experience to flow between different
> computational mediums. This allows us to view the mind as a physical
> phenomenon and a tangible container for the seemingly intangible
> subjective experience.

Yes.

> This, in turn gives us license to define transferring
> procedures in scientific  terms, i.e., measurable parameters of physical
> objects. Whatever subjective experience is, mind process enables it and
> that's the only thing we need to be paying attention to, confident in the
> truth that as long as we preserve the physical integrity of that process,
> we'll keep subjective experience  intact.
>
> Slawomir Paliwoda

Good post. Imo, you are raising the bar in terms of how you are thinking
about this.

Interestingly, I can see how Rafal's view (or rather his communication of
it) could be reconciled with yours to some potential benefit of both your
aspirations. He'd just need to acknowledge that you are using the word
identity in a tighter more strict way than he is. Like a person with a
rectangle
recognizes that they have a rectangle and you have a square and they
should not try an call it a "square". If a rectangle is all you want thats
fine.
If a rectangle can be achieved easier thats fine too. If something less than
full identity works for you then don't pretend that it is full identity when
you know it isn't going to be seen as full identity for others. To do so
dumbs stuff down.

Its about clarity, and politics, and communication, and truth in advertising
that identity continues to mean only one.  The is the sort of identity most
people will relate too. That some people will settle for less than full
identity preservation rather than nothing is ok but beside the point when
the point when what both you and Rafal would want would be to
increase the chances in the real world of satisfying both your criteria
and his will be easier than yours.

If Rafs vision is held up by cryonics folk as the one to try and attracts
others than it will repel folk like me at the same time as it attracts folk
like him and John.  On the other hand if yours is held up, Raf's lower
criteria is likely to fall out of a technological pipeline faster as a sort
of easier prototype perhaps as more people and resources may go
in pursuit of yours (assuming of course yours is not impossible - a
question I've not considered in this post).

Regards,
Brett