[Robotgroup] (no subject)

[Betty]

Wow, Glenn, 

You covered a lot, there. I'm still avoiding the roads, and have a touch of
cabin fever, but let me see if I can respond coherently.

 

About models and not just using humans: 

The LEGO kids I work with seem to think a robot is a little car model. A car
that doesn't know where it is going.

 

 
"Anthropology can be defined as the study of humanity":
Funny to think that when I was getting my degree back in the Eighties it was
still called the "study of man."
 
My anthropology is pretty hazy. I remember lots of studying kinship systems,
trading practices (mostly cattle), religious beliefs, warfare, and language.
Robots, not having cousin or marriage networks, possessions, religion,
(much) speech, settlements or warfare (okay, they fight when we make them
to), wouldn't yet make for a good Field Study yet. Although, field Studies
sort of failed as really useful study, anyway; my professors admitted
Anthropology didn't live up to its early promise. Have psychology or
sociology done any better?
 
I did take an interesting class in Crossing Cultures, about culture shock
and all our unseen assumptions. Trying to understand how Others think and
why they behave the way they do. That would apply to robot interaction. I
was reading yesterday about programming and the "conceptual gap between the
representations people use in their minds when thinking about a problem and
the representations that computers will accept during programming." Bridging
the gap would mean "either teaching people the computer's epistemology
(i.e., teaching them to think like computers) or teaching computers to
accept representations that people find convenient." The purpose of
programming classes is to teach people to build a mental model of computers
[so now we're forced to learn their model, not base them on us] which people
don't enjoy and if they're successful they don't like where they end up:
"They don't want to think like computers; they want to control computers."
The writer then proposes some programming changes I didn't quite grasp (not
being a programmer) that help bridge the gap. This is all from a chapter
called Making Programming Easier for Children in the book The Design of
Children's Technology. Adults could excel more easily than little kids, of
course, but the fact remains that most adults are just end users using
someone else's application. In order for this to change, new programming
styles are needed - and robots, along with game building, would be an
enticing way to teach them. And not just teach them to promote clear and
logical thinking, but to enable new kinds of learning, of assisted thinking
that Alan Kay writes about. I'm all into Alan Kay this week.  
 
 
 
But back to modeling robots on us - so much is still to come, regardless of
what model is chosen. I'm teaching a class starting next week and I'm afraid
the kids will be disappointed we won't have movie-style robots to play with.
But since they may outlive me by thirty years they'll see more interesting
robots than I ever will. Hurry up already, I'm middle aged! And why aren't
the robot toys more FUN? I see Robosapien 2 and the Chimpanzee head are both
reduced to $50 clearance now. They don't keep the kids attention any longer
than mine, and are amusing mostly when they "mess up."