Replying to your questions helped me put this in perspective. I now
feel that I have a very simple place to start. This is really
important. I don't need to solve problems that I can't solve just to
get started. It all looks very reasonable to me now. At first the
program will be a simulation but it will be based on simplifications
of how a computer program might actually derive information to use in
a subsequent analysis of simple text. (It won't be me creating stock
answers to a few variations of questions.)

Because I can start with extreme simplifications of my ideas, I can
examine how these ideas might be implemented in an actual AGI program
using prototypes. I keep emphasizing the simplification of these
strategies because that is a necessary jumping off point. However, my
program is not going to be an implementation of old AI or NLP ideas.
My ideas are built on a foundation of old ideas but the problem
requires a lot of innovation and the only way you can check how good a
conjecture is by trying it in an actual program. I am really looking
forward to getting back to work on my program now.


-------------------------------------------
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

Reply via email to