A thread that has nothing to do with memes or the Singularity

Lyle Burkhead (LYBRHED@delphi.com)
Mon, 09 Dec 1996 16:44:22 -0500 (EST)

Hal Finney and Eugene Leitl write,

>> I have never heard of this phenomenon.
> Nor me.

I can't help it if you guys are a decade or two behind the times in
protein science. Go to a meeting of the Molecular Graphics Society,
and look at some animations of proteins. They do in fact behave
as I described.

James Rogers writes,

> There is a type of "active" composite that has the properties you are
> describing. It is a laminate composite consisting of piezo-electric
> ceramics and shape-memory alloys from the nickel-titanium family.

Beautiful. Sounds like something that could be scaled down.

> When vibrations hit the ceramic outer laminate, the ceramic acts as a
> transducer, converting the vibration into an electrical signal.


> The signal is inverted ... and fed into the shape-memory alloy.
> The alloy flexes and reproduces the inverted signal, similar to a
> speaker, through the ceramic.

In other words the alloy turns the electrical signal back into sound, and
sends it back through the ceramic, but it is inverted, and so cancels out
the original sound. Neat. But this is going to defeat our purpose,
because --

> The net effect is that the material nulls the vibration, but produces heat
> with the excess energy.

So the big, regular vibrations are replaced by small, random vibrations,
i.e. heat. The noise is attenuated, and the material becomes warmer.
The thing is, we are trying to design a refrigerator. When a piece of
this material absorbs the sound in a room, does it lower the temperature
of the air? No! The material itself becomes warmer, and this heat is
radiated right back into the air.

This composite is wonderful stuff, and I'd like to have a house made of
it. Walls that absorb all noise!! I'd like to see entire cities made of it.
Silent cities. That would be heaven. But this composite doesn't solve
the problem we were trying to solve in the cell.

We want to take the energy from the protein's vibrations, and send that
energy somewhere else, far away from the protein (or far away from
whatever we are trying to refrigerate).

Let's go back to the first step. We put the ceramic next to the material
to be refrigerated. The ceramic acts as a transducer, and converts the
sound into an electrical signal. I'm not sure how this would work at the
atomic level, but suppose it does.

Now, what you want to do is send that signal elsewhere, to a heat sink,
and *then* have it pass through a resistance, thus generating heat --
far enough away so the heat stays away from the material we are trying
to refrigerate.

Some part of this process has to involve a net input of energy.
I guess the energy would be required to maintain the heat sink at a
lower potential than the ceramic material.

Anders Sandberg writes,

> What about the microtubuli? They are sturdy, good at self-organization
> and if we could figure out how to use the dyneines and kinesines,
> we could move molecules along them as we want. If we also could
> fixate the tubuli, they might make a good scaffolding.

I agree. Microtubuli are just sitting there begging to be made into
nanocomputers. The question is how to get a handle on these things.
How does one design a cell in which microtubuli form a scaffold or
some other useful structure?

One possibility is to have a vat with a large number of cells, filter out
the ones that have microtubuli in some approximation to the desired
configuration, and only let those cells reproduce (discard the other
cells). After n generations, it should be possible to breed cells to fairly
exact specifications.

Design problem #1 is the first of a long series. There is also the inverse
problem, i.e. how to imbed cells in machines. This discussion of
combining cells and machines should take place among the employees
of a company, and the upshot of the discussion should be a decision
about which product to work on next.