Re: Alcor on KRON

Date: Sat May 05 2001 - 03:13:14 MDT

Spike Jones wrote:

> Heres one I tried today: if you took the computer that is
> sitting in front of you right this minute and put it in a time
> machine, sending it back one hundred years, you could
> collect the *very best minds* in the world of that time, show
> them the machine and what it can do, and *not one* of
> them would have the foggiest clue how it works. They
> could take it apart, examine every piece, and the best
> scientists of 1901 would be quite profoundly baffled.

Would they? It would seem to depend on how many of
the machines you would send them, so they would be able
to break them for analysis, and how much time
they would have to play with it. Would they have juice,
at at least know what kind of juice it takes? Handbooks?
Okay, no handbooks.

Assuming they were motivated, what they probably would be,
if you tell them the device is basic part of economy's
infrastructure, and a valuable scientific instrument on
its own right. The mechanical parts and the macroscopic
stuff would be comparatively easy to figure out. Keyboard?
A remote descendant of the typewriter. Hmm. Wires. Hey,
this thing uses electricity. Cool stuff! (Electricity
in 1901 was something like nooklear in the 1950s) Mouse,
largely mechanical (unless you're evil enough to give
them one of these fancy purely optical ones), wires again.
Hard drive. Just unscrew it and look at it in operation. Look
at the surface with a microscope. Once you realize it's
magnetic, you win a lot. CRT (unless you're evil enough to
give them a TFT). The color capabilities would astonish them
without measure, but the basic shape of the device is almost
immediately recognizable if you look under the hood. Or apply
a magnet. See the color triples with a loupe. See the color
distortions with a magnet.

Now figuring out the semiconductors would be hard, because
they are sealed, have tiny leads, switch very fast and
are a micropatterned substrate, which would exceed their
analytics. They would do much better with a transistor radio,
or a vacuum tube computer. In fact the CRT would act as a veritable
Rosetta stone, because it's based on an ancient technology, and
uses macroscopic high-power devices, and simple signals.

Of course this doesn't give them the science behind it,
and the context in which it was produced. But as a hint and
as a data point that thing would have been very valuable.

Things would have been infinitely difficult if you gave the
same machine to scientists in 1800s, or in 1700s. They'd be
most likely stymied. Or it'd took them almost forever to
figure it out.

> Likewise 100 years from now there will be machines
> that even the most insightful among us will declare
> indistinguishable from magic.

Except that we'd knew they're not magic, but artifacts
based on scientific principles. To stick with computers:
though we're not seeing 99% of the future, we probably
could resolve the elementary cell of computronium, paradoxically
omega-grade computronium would be easier to figure out,
because it would likely have a very small elementary cells.
Of course if it's not orthogonal, i.e. if you have to figure out a
macroscale artifact which is heterogenous at molecular scale,
we'd be out of luck. And figuring out the other stuff, we'd
be in a situation worse than 12th century alchemist trying to
figure out a modern multimedia machine.

Hey, someone send us a piece of computronium. Preferably a fully
functional device, so that it would explain itself, and assist in
its own reconstruction, provided we're treating it not too shabbily.
Okay, allright, it would be magic. Artifacts which are also persons
are a classical requisite of fairy tales. Mirror, mirror on the wall,
give me the dirt on blue chips today.

This archive was generated by hypermail 2b30 : Mon May 28 2001 - 10:00:02 MDT