DZ:I just don't even know where to begin, except to say that image schemas
are only "made of lines" in the most superficial way; it's like saying
language is made of lines because that is how we write it down.
Language is indeed made of lines - written language - and that is not a
superficial but a profound feature. We have to process it first as lines.
Computer programs don't have to - but then they're still dumb without any
real contact with the world. (Ask yourself why everything we centrally
depend on here - language, logic, numbers, geometry - all depend on lines).
And image schemas are profoundly not superficially made of lines. Their
lines reflect the lines and outlines of objects and the world - including
the lines of letters. That is an absolutely fundamental and essential
connection.
Image schemas connect us to the world while isolated, brain-in-a-box
computer programs, that cannot delineate the world with our fluidity, have
no connection at all.
And the point of thinking in terms of a robot rather than just a computer
program, is that - as cognitive embodied science provides massive evidence
of - real world computation/problemsolving is a process that does and must
engage not just a brain-in-a-box but an entire body simulating the
alternative lines of action that may solve problems.
You seem to have a Platonic, abstract idea of intelligence and
problemsolving. Real world intelligence is embodied through and through -
every step of the way.
-----Original Message-----
From: Derek Zahn
Sent: Saturday, January 26, 2013 6:01 PM
To: AGI
Subject: RE: [agi] The many different types of embodiment
Mike,
I'm going to have to bow out of this conversation; there is just too large a
distance between how we think about these things, and I don't get anything
from it. I don't necessarily disagree with your basic judgment about which
parts of cognitive science are the most fruitful, but you interpret them and
apply them in such strange and useless ways... When you write:
What do you think image schemas, on which cognitive
embodied science is founded, are made of ... but lines?
I just don't even know where to begin, except to say that image schemas are
only "made of lines" in the most superficial way; it's like saying language
is made of lines because that is how we write it down.
I was happier with the idea of maps because maps are abstract
correspondences (of an interesting type) between an internal representation
and the domain being mapped; that correspondence is the important thing.
"Line", especially the way you use the word, is not nearly as useful a
concept. "Path" or "Link" would be better than "Line" as parts of a
framework for thinking about thinking.
Just one last thing, which I really hope you manage to get someday:
The key difficulty here is that you seem to be still thinking of *computer
programs* - rather than robots.
A robot is still controlled by a computer program. You don't get away from
the difficult issues of AGI by giving a computer a body. I believe it is a
good idea because of the richness of the connection between program and
world that is provided by the sensory and and motor capacities of such
embodiment, but it still runs a program which is what interprets the sensory
data and tells the motors what to do (provided that it is not some radically
new design -- which would be fine, but needs to explained).
The task of AGI is to write a computer program (or build a device out of
some other tangible specifiable concrete "stuff"). Yes, it is a challenge
to get programs to do what is needed, but that is exactly the challenge that
defines AGI as an area of study.
Derek Zahn
-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/6952829-59a2eca5
Modify Your Subscription:
https://www.listbox.com/member/?&;
Powered by Listbox: http://www.listbox.com
-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com
