[MURG] Neuropeptide Function?

[Yan King Yin]

From: Eugen Leitl <eugen at leitl.org>

>Unfortunately, a complete model of information processing in biological
>tissue is equivalent to a complete model of biological tissue. It's a dirty
>computer. Theories tackle only few select aspects. It's a caricature of the
>real thing.

Complete modeling is *NOT* necessary. We can apply statistical
methods to modeling and make the error / deviations small enough
to be insignificant. In fact, a lot of modifications are bound
to happen after uploading, we'll all want some or other forms of
that.

>I'm having my doubts about our future. There's a considerable potential
>locked in the raw switch count of a 300 mm wafer, or a near-future molecular
>computer which is not available due to architecture issues. A true machine
>intelligence can achive very high improvement rates by breaking through these
>barriers.
>
>It will take a long time to creat a supercritical seed, but it will go from
>zero to hero in record time, leaving us far behind. It is very important that
>we will survive that bootstrap process.

I have thought about it recently quite in depth. I think the
future will be a balance of power between and amongst uploads,
AIs, humans riding AIs, humans, other organisms, and computer
viruses etc. A single AI taking over is highly unlikely, so I
don't worry too much about that.

YKY


____________________________________________________________
Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail!
http://login.mail.lycos.com/r/referral?aid=27005