Ben Goertzel wrote:
About self: you don't like Metzinger's neurophilosophy I presume?
(Being No One is a masterwork in my view)
I agree that integrative biology is the way to go for understanding
brain function ... and I was talking to Walter Freeman about his work
in the early 90's when we both showed up at the Society for Chaos
Theory in Psychology conferences ... however, I am wholly unconvinced
that this work implies anything about the noncomputationality of
consciousness.
You mention QED, and I note that the only functions computable
according to QED are the Turing-computable ones. I wonder how you
square this with your view of QED-based brain dynamics as
noncomputable? Or do you follow the Penrose path and posit
as-yet-undiscovered, mysteriously-noncomputable quantum-gravity
phenomena in brain dynamics (which, I note, requires not only radical
unknown neuroscience but also radical unknown physics and mathematics)
-- Ben G
The comment is of the kind "when did you stop kicking your dog". You
assume that dog kicking was an issue and any answer in some way
verifies/validates my involvement in dog-kicking! No way! :-)
Turing computable or Xthing-computable...is irrelevant. I am not
'following' anyone except the example of the natural world.....There's
no inventions of mysterious anything... this is in-your-face good old
natural matter doing what it does. I have spent an entire career being
beaten to a pulp by the natural world of electromagnetism....This is
really really simple.
Nature managed to make a human capable of arguing about Turing
computability and Godellian incompleteness without any 'functions' or
abstractions or any 'model' of anything! I am following the same natural
path of actual biology and real electrodynamics of real matter. I have a
brilliant working prototype: /the human brain/. I am implementing the
minimal subset of what it actually does, not a model of what it does. I
have the skills to make an inorganic version of it. I don't need the ATP
cycle, the full endocrine or inflammatory response and/or other
immunochemistry systems or any of the genetic overheads. All the
self-configuration and adaptation/tuning is easy to replicate in
hardware. When you delete all those overheads what's left is really
simple. Hooking it to I/O is easy - been doing it for decades...
Of course - like a good little engineer I am scoping out electromagnetic
effects using computational models. Computational chemistry, in fact.
Appalling stuff! However, as a result my understanding of the
electromagnetics of brain material will improve. That will result in
appropriately engineered real electromagnetics running in my AGI, not a
model of electromagnetics running in my AGI. Quantum mechanics will be
doing its bit without me lifting a finger - because i am using natural
matter as it is used in brain material.
Brilliant tho it was, and as beautiful a piece of science that it was,
Hodgkin and Huxley threw out the fields in 1952ish and there they
languish, ignored until now. Putting back in the 50% that was thrown
away 50 years ago can hardly be considered 'radical' neuroscience.
Ignoring it for any more than 50 years when you can show it operating
there for everyone to see...now that'd be radically stupid in anyone's book.
There's also a clinical side: the electrodynamics/field structure can be
used in explanation of developmental chemistry/cellular transport cues
and it also sorts out the actual origins of EEG, both of which are
currently open problems.
It's a little brain-bending to get your head around.. but it'll sink in.
cheers
colin
-------------------------------------------
agi
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
https://www.listbox.com/member/?member_id=8660244&id_secret=117534816-b15a34
Powered by Listbox: http://www.listbox.com
