I missed the original post that started this thread ;
I hope you are still reading it.
The original poster wrote :
> > Now the way I want to use this to involve the program
> > in conversation is to have it chat about a topic to
> > the extent of it's knowledge and then propose a
> > similar topic, much like a real life conversation.
> > AKA - If you are talking about sailing, even if you
> > never mention water or ocean or anything, the program
> > will realize that sailing is done on the ocean
> > and ask you what kind of fish you like to eat.
> >
> > I am very confident that I can do this ...
I understand how you can get a program to take in
information as you type sentences into it.
This part sounds relatively straightforward.
But how do you propose to get it to "converse"?
It has to somehow decide what to say, to get the user
to keep typing. What it says should preferably be
related to what the user is typing in.
What are your plans for this?
Do you have a simple model of what a conversation is?
So that the program can hold up its end of the conversation.
Perhaps you are just hoping that the program can get by
with the use of built-in statements like Eliza used,
such as "Tell me more about your <mother>.".
Or do you have something different in mind?
Someone commented :
> ATM: You are making a start (bravo!), but you may need
> to develop a much deeper theoretical framework ...
I agree that its good to make a real start.
You are starting the right way by writing a program.
I think that you are taking the right approach in one sense.
You are trying to write a real program, and get it working,
instead of talking about A.I. or the human mind theoretically.
I think this is definately the right approach generally.
I belong to a couple of mailing lists where people discuss A.I.
vaguely and theoretically, without ever writing a program.
Believe me, they never make any progress in their
thinking ! They don't advance at all.
--
Martin Sondergaard,
London.
