On Sun, Jul 25, 2010 at 5:05 PM, Mike Tintner <tint...@blueyonder.co.uk>wrote:

>  I think it's v. useful - although I was really extending his idea.
>
> Correct me - but almost no matter what you guys do, (or anyone in AI does)
> , you think in terms of spaces, or frames. Spaces of options. Whether you're
> doing logic, maths, or programs, spaces in one form or other are
> fundamental.
>
> But you won't find anyone - or show me to the contrary - applying spaces to
> creative problems (or AGI problems). T
>


I guess we may somehow be familiar with different and non-overlapping
literature, but it seems to me that most or at least many approaches to
modeling creativity involve a notion of spaces of some kind.  I won't make a
case to back that up but I will list a few examples: Koestler's bisociation
is spacial, D. T. Campbell, the Fogels, Finke et al, and William Calvin's
evolutionary notion of creativity involve a behavioral or conceptual fitness
landscape, Gilles Fauconnier & Mark Turner's theory of conceptual blending
on mental space, etc. etc.

The idea of the website you posted is very lacking in any kind of
explanatory power in my opinion.  To me any theory of creativity should be
able to show how a system is able to generate "novel and good" results.
 Creativity is more than just outside what is known, created, or working.
 That is a description of novelty, and with no suggestions for the why or
how of generating novelty.  Creativity also requires the semantic potential
to reflect on and direct the focusing in on the stream of playful novelty to
that which is desired or considered good.

I would disagree that creativity is outside the established/known.  A better
characterization would be that it resides on the complex boundary of the
novel and the established, which is what make it interesting instead just a
copy, or just total gobbledygook randomness.



-------------------------------------------
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=8660244-6e7fb59c
Powered by Listbox: http://www.listbox.com

Reply via email to