RE: Profiling in the News: Was: RE: Taxonomies, URN's etc..

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

The profile is the point.  We 
never claimed we could model humans.  We claimed 
we could create systems that enable humans 
to model humans which if you think about it, 
is what profiling is.   We told the world 
this was about reducing miscommunication. 
Your public safety systems do this kind of 
thing every day and they use these with other 
information to make decisions about how, 
where and when to deploy agency resources. 
Without episodic memory, for example, 
gang rivals are put into the same cells 
with bad results.  Child molesters are 
allowed into the general prison population 
with bad results.  Every time a prisoner 
is booked, a psychological profile is created 
that is used to determine how resources will 
be applied (for example, a suicide watch). 
There are numerous cases where such profiling 
has enabled serial killers to be located and 
more when they were not and continued killing 
because the information systems did not share 
the profiles and this is true on both sides 
of the pond.  As a matter of fact, the British 
systems are working very hard to solve this 
problem *today*.  I read their RFPs.

Hypotheticals are a way to find in a situation 
a matching theory so it can be tested.  In this case, similar to 
the kind of deduction Holmes might employ, the 
art is one of recording and assessing observations. 
In the case of the girl, she does have episodic 
memory and that is beyond the observation but not 
her ability to select.  We have to account for 
that or otherwise, we can't use this for some 
applications such as autonomous agents.  They 
depend on episodic memory for negotiation.  Unless 
limited by design, they can also screw up badly 
using it just as people can with the exception 
that they may not have a "this doesn't feel right" 
sense.  I note this in my Golem paper as part 
of the limiting scope of authority for certain 
decisions, ultimately, who chooses choices.

I provided an example in which the miscommunication 
was possible but did not occur and asked if 
we could construct a plausible description 
of what kept that from happening.  It is just 
as likely that based on the context, she would 
have looked at the Dog:dog and said, "yes that 
absolutely is an ugly dog."  Now they would be 
in agreement but failed to communicate.  They 
might go on through the night talking about the 
"ugly dog" with each communication getting ever 
more hilarious.  We also have a show on the 
comedy cable about George W. Bush and his wife 
in which he is talking about getting rid of their old 
cat and she thinks he is talking about her genitals 
based on a word for both.  Comedy works by 
stretching the plausible into the implausible 
to the point of pain.

In the beginning, I claimed the central metaphor 
of HumanML would be "awareness".  This was not 
without thought.

Len 
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Sean B. Palmer [mailto:sean@mysterylights.com]
Sent: Thursday, August 30, 2001 3:23 PM
To: Bullard, Claude L (Len); Rex Brooks
Cc: humanmarkup-comment@lists.oasis-open.org
Subject: Re: Profiling in the News: Was: RE: Taxonomies, URN's etc..


> Her reply is based on the conjunction of her (emotional)
> reaction to her interpretation of the statement and the
> emotional reaction is based on her past history and the
> (culture) she is a part of that considers calling a woman
> a dog an (insult:symbol)

Ah, but now you're building up a hypothetical psycological profile of her
that I am not aware of, and that makes the entire situation unrealistic.
What if she doesn't consider that event an (insult:symbol)? There could be
any number of reasons for one person ascribing the label of
lamppost-attracted-object to (which is actually hypothetical in this case
too... anyone who's read "Blindless" could not do such a thing!) and
another countering with a suggestion of "beautiful" (even considering that
that two classes are not disjoint).

But all I'm really saying is that you should have modelled this on two
entirely abstract people, rather than trying to inflict YAUEC (unrealistic
event-case) on the world. I agree with the rest of your theorized model.

> She counters with beautiful because
> she has a goal and it is now up to
> you to figure out what it is. If she
> slaps you (haptic) try again.

We have a show in the U.K. which may or may not air over the pond, and
which you may or may not be aware of, called "The Fast Show". It has a
number of excellent actors and comedians in it, and has met universal
acclaim.

One of the charaters from one of the sketches, played by Arabella Weir, has
the catchphrase, "does my bum look big in this?". It's interesting that in
all of the shows that I can remember this sketch being done, no one has
replied to that question...

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .