[Heartlogic-dev] Re: 3 ratings at a time

[Joshua N Pritikin]

On Wed, 2005-05-18 at 00:01 -0500, William L. Jarrold wrote:
> On Tue, 17 May 2005, Joshua N Pritikin wrote:
> > Are you with me?
> 
> I don't know.
> 
> Suppose, Jill wants water for herself.  Therfore she is goal wrt
> water for herself.  She is no goal with respect to water for Jack.

OK

> New scenario.  Suppose Jill wants water for Jack.  Therefore
> she is no goal wrt water for herself.  She is goal with respect
> to water for Jack.
>
> Is this correct?  In what way is it corret/incorrect.

Correct.  That's why I ask the questions in a funny way.  I'm trying to
isolate an agent's proper goals from an agent's concern-for-others
goals.

> >> maybe what you are trying to do is figure out a theory of what goals
> >> and anti goals people have/do not have?
> >
> > Yes, a taxonomy.
> 
> A taxnomny?  To me a taxonomy is a hierarchical tree shaped data struture. 
> What might the tree look like?
>
> > This is the taxonomy of the goal
> > pairs collected so far.
> 
> It looks like a 3 X 3 matrix or table to me.  Not a taxonmy.

Sure, this is just the root of the taxonomy.  Think of it this way -- my
taxonomy for individual goals (Goal, AntiGoal, No Goal) is trivial.
However, when we look at situations involving just two people, there is
a 3 x 3 matrix of possibilities.  As you mentioned a tree, depending on
which stories fit into which categories, we might add branches to some
of the goal pairs later when the data supports it.

> > What is not obvious is which stories will fill
> > in the rest of the boxes.  There are nine boxes and only two have
> > examples.
> 
> What examples?  What are the 9 boxes. What does Goal/Goal = 0.21 mean?

Goal/Goal = 0.21 means the most believable example of a story with a
Goal/Goal interpretation has a believability of 0.21 on a scale of
[-1,1].

> What does Goal/No Goal = 0.36 mean?

Similarly, the most believable example of a story with a Goal/No Goal
interpretation has a believability of 0.36 on a scale of [-1,1].

> I clicked on 0.21 and got
> the same story as when I clicked on 0.36.

Sure, because there is only one story in the database.  As soon as
someone adds another story then there will be some competition for which
story is most believable for each goal-pair category.

> Oh Joshua, that would be terrible feedback.  Feedback should be 
> constuctive.  It should say what is good and what is bad and say
> how to make the bad better.  I always strive to give feedback
> that meets those requirements.

Dunno, it seems like we are making good progress just going though the
web pages.

> Just send me a three sentence blurb.

To empirically induce a taxonomy of goal-pairs.  Does that help?

-- 
If you are an American then support http://fairtax.org
 (Permanently replace 50,000+ pages of tax law with about 200 pages.)

signature.asc
Description: This is a digitally signed message part