[ExI] Alcor's new statement on ASC

[John Clark]

On Tue, May 15, 2018 at 10:17 PM, Stuart LaForge <avant at sollegro.com> wrote:

> *> If you consider your connectome to be the essential you, then your
> argument does  follow in a narrow sense.*

Narrow sense? It follows unless you assume particular atoms confer
individuality to us even though science tells us atoms have no
individuality, and I don’t see hot atoms can confer that property to us
when they don’t have it themselves. And if that was the key I also don’t
see how we seem to remain the same person even though atoms are constantly
shifting in and out of our bodies from birth to death.

And I think it is beyond debate that the connectome is essential, there may
or may not be other things that are essential too we don’t know.

> *> But keep in mind that your connectome is a malleable ever evolving
> thing. Every second of every day, you are making new memories and also
> forgetting old ones. In a continual state of becoming.*

We're trying to decide if ASC or Alcor's method should be used, so how is
the above relevant?

> *> Maybe by placing so much stock in a frozen snapshot of your connectome
> you would lose the information of who you were becoming.*

Alcor's method is just as much a frozen snapshot as ASC is, and I don't
know if the connectome information would be enough to bring somebody back
but ASC would preserve more than that, I can't think of any sort
of information that Alcor's method would preserve but ASC wouldn’t, and
almost certainly scramble it less too.

> *> Kind of like a Heisenberg uncertainty principle of identity.*

If the Heisenberg uncertainty principle is relevant then we become a
different person a billion times a second.

> Which memories were just starting to form? Which were on the verge of
being forgotten.

Those would be the memories of just before I died when my body was breaking
down and I was probably in great pain, and I don't mind forgetting them so
much. And anyway the same thing would happen with Alcor’s method so the
issue has no relevance in deciding between the two.

> *> Life and consciousness are not just nouns, they are also verbs i.e.
> dynamic processes.*

I am an adjective, I am the way generic atoms behave when they are
organized in a johnkclarkian way.

> > > I think the warming up process almost certainly causes even more
>> damage than the freezing process.
>
>

*> From a biomedical standpoint, absolutely. Especially if the warming
> comes too fast. Also from ischemia considerations, it would be better to
> warm the person from the inside out.*

It would be even better not to warm the brain at all, instead have
nano-machines note the position of all the atoms in the outer layer of the
frozen brain by feel, that is to say by using the same basic principle that
Scanning Tunneling Microscopes use, then they would remove that layer and
do the same thing with the next layer and then the next until the position
of every atom in the brain is known,  Ralph Merkle figures this would take
about 100 days; the original brain would be gone by then but that would be
OK because if you've seen one atom you've seen them all. Its not the
particular atoms in your brain that makes you be you its the particular
information.  Of course to be in a usable form that can be used to make
another brain (or more likely an upload) that raw data is going to need a
lot of processing. In the current issue of Cryonics Ralph Merkle divides up
the brain into 10^27 little boxes,  each box would then only contain a very
few molecules and most of them would be water molecules whose reaction with
aldehyde is simple and well known and and whose exact position is less
critical than for proteins. If we figure each tiny little box would need a
billion floating point operations to sharpen up the raw data, which a think
is a generous estimate, then the entire brain would require 10^27 billion
floating point operations.

If we have Drexler style nanotechnology (and if we don't nobody is coming
back) one watt for one second could power about 10^12 billion floating
point operations, there are 3.6*10^3 seconds in a hour and 10^3 watts in a
kilowatt, so one kilowatt-hour of energy could make 3.6*10^18
floating point operations, thus to do the entire brain you'd need
280,000,000 kilowatt-hours of energy. Right now in the USA a kilowatt-hour
costs about 12 cents so the energy cost would be about 34 million dollars,
much too high for most people, but if we have nanotechnology it is
certainly reasonable to expect energy costs to be very dramatically lower
than what they are now.

> *> But I am sure you had more reductive reason so let's hear it. I am
> curious. :-)*

During freezing if a piece breaks off it won't be able to diffuse very far
away so you can figure out where it came from but with rewarming that piece
could end up anywhere and any damage produced will continue to evolve and
get worse unless you immediately step in with very sophisticated technology
to stop it. But with freezing the damage automatically stops when
things become solid, there are no time constraints and we can leave the
problem of repairing the damage that has occurred to future technology. Or
at least we can provided the brain information has not been so scrambled
that even Nanotechnology can't unscramble it, and that could happen if
turbulence sets in.

So the key question is " will the micro-currents in my brain be in a
turbulent state when it is frozen or will the flow be laminar ?". If it's
turbulent then very small changes in initial conditions will result in
large changes in outcome and I'm dead meat, even nanotechnology couldn't
put Humpty Dumpty back together again; but if the flow is laminar figuring
out what things were like before they were frozen would be pretty
straightforward.

Fluid flow stops being smoothly Laminar and starts to become chaotically
turbulent when a system has a Reynolds number between 2300 and 4000,
although you might get some non chaotic vortices if it is bigger than 30.
When chaotic turbulence starts a very small change in initial conditions
will result in a huge difference in outcome and that is exactly what we
want to avoid because we want to be able to figure out what the brain was
like before it was frozen.

We can find the approximate Reynolds number by using the formula LDV/N.  L
is the characteristic size we're interested in, we're interested in cells
so L is about 10^-6 meter. D is the density of water, 10^3 kilograms/cubic
meter.  V is the velocity of the flow, during freezing it's probably less
than 10^-3 meters per second but let's be conservative, I'll give you 3
orders of magnitude and call V 1 meter per second.  N is the viscosity of
water and at room temperature N is 0.001 newton-second/meter^2, it would be
less than that when things get cold and even less when water is mixed with
glycerol as it is in cryonics but let's be conservative again and ignore
those factors. If you plug these numbers into the formula you get a
Reynolds number of about 1. 1 is a lot less than 2300 so it looks like any
mixing caused by freezing would probably be laminar not turbulent, so you
can still deduce the position where things are were from the position of
where things are now, you can figure our how the parts of the puzzle are
supposed to fit together.

> >> And more important the current method leads to the shrinkage of the
>> brain by 50%.
>
>

> *All cryobiotic organisms that survive freezing undergo a certain amount
> of dehydration in the process.*

That works fine for microorganisms like tardigrades but as far as I know no
adult mammal has ever survived being frozen solid, much less brought back
from liquid nitrogen temperatures, there is always some liquid water
remaining.

> > *If you are that worried about your connectome, just get a picture or
> better yet a movie of it while you are still alive. The data will be
> fareasier to stably store than a biologically useless head.*
> http://www.humanconnectomeproject.org/gallery/

I know of no non-destructive method that could produce the connectome
information of my brain right now, unfortunately MRI, PET scans and X rays
don't provide nearly enough detailed information. However a human brain has
been chemically fixed and sliced into 5000 thin sheets and
detailed microscopic images taken of every square nanometer of all 5000
sheets, and that might do the trick, but for obvious reasons I'm not quite
ready for that.

> > Your DNA won't survive covalently bonding with every nitrate group in
> its vicinity. But your DNA can survive boiling water. Think on that before
> you pickle yourself in glutaraldehyde for posterity.

Why on earth should I think about that? Talk about redundancy, my entire
genome is repeated in every cell of my body, my hairbrush alone is enough
to guarantee it will survive intact, but that is not nearly good enough
because there is a lot more to me than just my genome.

> *> Aldehyde is one of the most toxic substance known.*

Then why can I buy it over the counter at my local drugstore?

> *  > Tardigrades can survive outer space but cant survive aldhehyde.*

I know, and the same is true for the closely related chemical formaldehyde
, but neither would make for a very good weapon of war.

> *> It inactivates *viruses* which is why it works on that wart.*

I know but that’s not a bug it's a feature, inactivating the entire
biochemical metabolism is exactly what makes aldehyde so good, we want to
shut it down and keep things in place, its the entire point of Cryonics.

> *> you are not just poisoning your cells with aldehyde,*

Yes aldehyde is a poison in that it renders cells non-viable, but I don't
care about that, I only care about saving the essential must have part of
them, the information.

> *> you completely cross-linking all your biomolecules into one gigantic
> covalently bonded molecule.*

Right.

> *> You will be will still be able to see the forest, you just won't be
> able to see where one tree ends and another one begins.*

How do you figure that? We know what the chemical properties that a
molecule of aldehyde has and we know what proteins are found in brain cells
and we know what shape they have and we have Nanotechnology which means we
have one hell of a lot of processing power. And there is a hell of a lot of
redundancy in the brain and that will help too.

> *> Again, take a picture of your connectome before you die and let Alcor
> do its thing.*

I'd love to get a picture of the connectome of my brain but please tell me
how to cut my brain into 5000 slices far thinner than tissue paper without
killing me. If Alcor has the ability to do that right now then they've been
holding out on me.

> *> Why would you want your connectome preserved at the moment of your
> death anyway? So you can live forever with PTSD?*

First of all there is no reason to think ASC would preserve connectome
information and nothing else, and second of all why is ASC a snapshot but
Alcor's method is not?

> *> Dozens of examples of proof of principle of vitrification already exist
> in nature. I can't show you a mouse at liquid nitrogen temps but I can show
> you a frog at -16 C:*

I admit the frog is impressive but 3F (-16C) isn’t very cold, that’s why
its metabolic process never stops entirely and 1/3 of it remains unfrozen,
and it can only remain in that semi-frozen state for a few months. And −16C
is not nearly cold enough for long term storage. Alcor uses liquid nitrogen
at −196C and that is cold enough to keep things unchanged for centuries, in
fact −140C would probably be cold enough and it is above the glass
transition temperature so the rate of cracking would be less, but I don’t
consider cracking to be a very serious problem, it should be pretty simple
to figure out what things looked like before the crack happened. And
storing things at −140C would be more expensive than storing them at −196C,
and it would be far more complicated than simply dumping in liquid nitrogen
and that would increase the likelihood of catastrophic failure. The less
maintenance required the better.

> > *I was pleasantly surprised to discover Natasha Vita-More did it in the
> lab with nematodes.*
> https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4620520/

In a way Natasha’s work is more impressive than the frog even though the
worm is so small it can barely be seen by the naked eye because she got
down to −80C and because she showed that memory is retained. But −80C is
STILL not cold enough for long term storage, the chemical reaction rate is
just too high,  the worms were only frozen for 2 weeks.

> *> I can't show you a single example of *anything* surviving getting
> dipped in aldehyde at *any* temperature.*

Because aldehyde destroys cell viability, but it doesn't destroy
information and therefore it doesn't produce information theoretical death:

https://en.wikipedia.org/wiki/Information-theoretic_death
<https://en.wikipedia.org/wiki/Information-theoretic_death>

> >> to restore biological viability you are going to need information about
>> what atom goes where, and from everything I have seen ASC does a better job
>> preserving that information than Alcor’s current method. Electron
>> Microscopes don't lie.
>
>
> *> What goes where is not enough. You need to be able to distinguish
> between hydrogen bonds and covalent bonds.*


Nobody knows that you're just guessing. But it doesn't matter because I
don't insist that Electron Microscope pictures provide enough information
to deduce the connectome or that the connectome information is enough to
bring a individual back; but I do insist those Electron Microscope pictures
are excellent evidence that ASC distorts information less than Alcor's
current method, and not just information about the connectome.

*> Scanning electron microscopes don't have the resolution to distinguish
> hydrogen bonding from covalent bonding. You need an STM for that.*

I always assumed the trillions of nano-machines that examine the layers of
the brain would gain that information by useing STM technology for two
reasons:

1) They could take better pictures.

2) Its far easier to design a STM microscope a few nanometers across than
electron microscope the same size because I don't know how such a tiny
machine could produce the very high speed electrons needed.

> *> I think what you see would be a lot messier with ASC under an STM.*

You think that with Alcor's method things would look neater than ASC at
that high resolution even though we know they look much messier than ASC at
a lower resolution??

> *> Nanosanta will thank you for not cross-linking your molecules into a
> Gordian knot.*

But I don't want to untie the Gordian knot, I just want to take a picture
of it.

 John K Clark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20180517/8b8c617f/attachment.html>