I really did not understand what John was talking about at first. The
methods that he is working with are evidently very different from
mine. Then when I tried to think about how I might write a program to
have a sense-of-self when being creative I could not really figure it
out.  But I finally got it.  A program cannot notice every detail of
what happens in the IO data environment, but when it does it should
tag the observation with some sort of simple mark to signify that the
event was noticed in the IO environment.  But, when it takes some
action of its own, especially when it tries to use imagination to
solve a problem, the imaginative action should be tagged differently.
The imagined event or the conjectured relation should be tagged that
it was something that the program created.  These simple meta-tags
would not be as important in a habitual response and they could be
discarded as soon as they were used.  The important thing in a
unconscious or semi-conscious animal habit is that it is effective.
However, when a person first starts to think about what he is doing
when he reacts with a well practiced habit he starts trying to
identify the steps he takes.  That is why (in this conjecture) why you
have to actually try doing something that you do so well that you
don't even have to think about in order to understand what it is that
you are doing in a way that will allow you to explain it to another
person.
So any internal action, or at least any internal action that involves
some creativity, should be tagged differently than an event that was
observed in the data environment. A well practiced habitual response
would be tagged at a higher level, but the individual steps taken
might be discarded as soon as they were used.
Jim Bromer


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