[ExI] Zombie glutamate
avant at sollegro.com
Thu Feb 19 06:18:40 UTC 2015
Quoting John Clark:
> Message: 2
> Date: Mon, 16 Feb 2015 11:09:55 -0500
> From: John Clark <johnkclark at gmail.com>
> To: ExI chat list <extropy-chat at lists.extropy.org>
> Subject: Re: [ExI] Zombie glutamate
> <CAJPayv0r+GF+_LaT2KqrQeAzMa8nXvG0Mv5uOsQ4tAKO0uL0zA at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
> On Sat, Feb 14, 2015 at 4:44 PM, <avant at sollegro.com> wrote:
>> While Brent is not completely wrong because susbtrates do have very
>> specific structures that enable their function, the structural
>> considerations outweigh the simple identity of the substrate. For example
>> a hemoglobin molecule denatured by heat would still chemically be
>> hemoglobin, but it will have lost its delicate folded structure and
>> thereby all of its biological function.
> Denatured hemoglobin chemically reacts very differently than non-denatured
> hemoglobin does, and the logical structure of a brain fed by denatured
> hemoglobin would be quite different from your brain, the neurons would
> respond to signals differently because they were dead, killed by lack of
> oxygen. But if done competently they logical schematic of your uploaded
> brain in a electronic computer would be identical to the logical schematic
> of your biological brain.
I might be in the minority on the list, but I think you underestimate
the Kolmogorov complexity and information content of the brain by
orders of magnitude.
All schematics are, by necessity, simplifications of the real deal.
Have you ever driven the schematics of a car? Yeah sure, you probably
played a racing video game, but do you think the developers programmed
your virtual car to have virtual cylinders burning virtual fuel? Now
you want to talk about simulating your brain. How do you know
something as crucial as your imagination, for an arbitrary example, is
like the spark plugs of the virtual race car which is *not* simulated.
Would you as an upload even be aware that that component of your mind
>>> If you want to simulate the mind, you would have to
>> simulate the human brain from the atoms up along with any attendant
>> chemistry and physics. You might even have to simulate the rest of the
>> body as well, after all, I wouldn't feel quite like myself without my
>> adrenal glands or my testicles subtly influencing my thinking.
> I see nothing sacred in hormones, I don't see the slightest reason why they
> or any neurotransmitter would be especially difficult to simulate through
> computation, because chemical messengers are not a sign of sophisticated
> design on nature's part, rather it's an example of Evolution's bungling. If
> you need to inhibit a nearby neuron there are better ways of sending that
> signal then launching a GABA molecule like a message in a bottle thrown
> into the sea and waiting ages for it to diffuse to its random target.
I never claimed hormones are sacred but they do have utility. Your
example shows you don't understand how chemical messengers like
hormones function. The target is not random at at all but actually the
40% or so of neurons that are expressing the GABA receptor at any one
time. And each one of those neurons can modulate the expression levels
of GABA receptor based on the reaction of those cells to the
concentrations various chemical messengers including GABA itself. If
GABA is a message in a bottle, it is a bottle that only the intended
recipients can open. But really the bottle itself *is* the message,
and it only means anything to a select subset of neurons that are
receptive. Those neurons intentionally express the genes to be
> I'm not interested in chemicals only the information they contain, I want
> the information to get transmitted from cell to cell by the best method and
> few would send smoke signals if they had a fiber optic cable.
> The information content in each molecular message must be tiny, just a
> few bits because only about 60 neurotransmitters such as
> acetylcholine, norepinephrine and GABA are known, even if the true number
> is 100 times greater (or a million times for that matter) the information
> content of
> each signal must be tiny. Also, for the long range stuff, exactly which
> neuron receives the signal can not be specified because it relies on a
> random process, diffusion. The fact that it's slow as molasses in February
> does not add to its charm.
I think you and Tomaz both severely underestimate the complexity of
living systems. Adding together the Beckenstein limits, calculated
using atomic masses and covalent radii of the individual atoms in
GABA, yields an information storage capacity of approximately 35 MB
and that's just for one molecule of GABA. Of course Beckestein bounds
are upper limits, so the relevant information content is likely much
lower. But the relevance of information is context dependent. So even
spam can be relevant to those who generate it. So just how much of the
maximal information capacity of our brain (approximately 10^41 bytes)
nature uses to generate John Clark is a tough open question. Oooh I
just had an insight on how might be able to estimate but I don't have
> If your job is delivering packages and all the packages are very small and
> your boss doesn't care who you give them to as long as it's on the correct
> continent and you have until the next ice age to get the work done, then
> you don't have a very difficult profession. I see no reason why simulating
> that anachronism would present the slightest difficulty. Artificial
> neurons could be made to release neurotransmitters as inefficiently as
> natural ones if anybody really wanted to, but it would be pointless when
> there are much faster ways.
Again, you don't understand how cell signaling functions. These
signals are not indiscriminate at all but instead highly specific and
targeted at the cells that have the appropriate receptors.
Furthermore, the messenger *is* the package and its function is
encoded by its shape. It is analogous to the copies of a key that only
open certain locks. These keys can only initiate a response from cells
that have the correct lock.
Yes simulations certainly could eventually get much faster but the
first few are liable to be much slower. The difficulty in simulating
signaling molecules and other biochemicals is that the all important
shape of the molecules is determined by the distribution of electron
densities, and thereby electric charge, over the molecules. The
distribution of these electrons are a quantum mechanical phenomenon
and as the Beckenstein bound example I gave above illustrates, quantum
mechanics is an information processing rabbit hole that has the
potential to go quite deep.
> The great strength biology has over present day electronics is in the
> ability of one neuron to make thousands of connections of various strengths
> with other neurons. However, I see absolutely nothing in the fundamental
> laws of physics that prevents nano machines from doing the same thing, or
> better and MUCH faster.
But biomolecules *are* nano machines that collaborate to replicate and
they have had billions of years to optimize themselves. We are just
now getting started. If general AI was easy, everybody would be doing
it. And the Fermi paradox could imply that nobody is doing it because
it is very hard.
More information about the extropy-chat