[extropy-chat] Alternative to Cryo was The Amazing CellularRepairdevice
eugen at leitl.org
Sat Oct 15 09:07:29 UTC 2005
On Sat, Oct 15, 2005 at 02:01:44PM +1000, Brett Paatsch wrote:
> But how do you compare the *original* analog store with digital
> copy #1 ? How would you do a baseline MD5 checksum on
> the frozen or vitrified brain?
A cryptographic checksum is a wrong metaphor to use on a
physical system. Measurements are almost always precise
in a discrete system, so single bit errors almost never
occur (and if they do, you usually notice them quite soon).
Cryptographic checksums are deliberately designed to be
brittle enough to detect single bit mutation. Moreover, it
is a digest (tiny destillate of a large mass) by definition.
Of what value is a brittle digest on a very large, very
dynamic, nonlinear but also self-homeostating system?
There are many ways to describe a complex physical system,
the space between our ears included. Shape and property descriptors,
and similiarity metrics upon it are not that hard, actually.
A really dumb brute-force approach would just destructively scan the bulk
at atomic resolution, and store atomic properties along with
the position. You could even compute a hash on that vector,
though I don't see why one would want to do that. It would
also be a pretty idiotic representation level for whole
body/brain emulation similiarity metrics, as the relevant
function occurs quite a few storeys above, at single cell
and cell compartment level.
A good metric would be spatiotemporal activity patterns,
in which case picking out a few relevant ones for a fingerprint
(not hash) would do. A crude fingerprint of a physical system
would be high-resolution fMRI, MEG and EEG. A less crude one
would be instrumenting each 100th cell or so with a realtime
recording probe and extract operational signatures from the
trajectory, and correlates with internal and external stimuli.
For obvious reasons, this won't be practical for quite a long
time. It is, however, the only way to make sure your process
is accurate enough and doesn't confabulate similiar but not
In ALife simulations individuals are characterized by behaviour,
which we are well equipped to abstract.
Eugen* Leitl <a href="http://leitl.org">leitl</a>
ICBM: 48.07100, 11.36820 http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 189 bytes
Desc: Digital signature
More information about the extropy-chat