[Edu-sig] Re: Alice

[Jim Harrison]

on 1/6/01 12:31 PM, Arthur_Siegel@rsmi.com at Arthur_Siegel@rsmi.com wrote:

Re: Alice
>>> You have more of a specific objection to this project.  You consider
>>> it opportunistic and bad science.
> Keyword - "opportunistic" -- etc. etc. etc.

I hesitate to add to this thread, partly because it's difficult to know
exactly what the various participants are really thinking from what's been
written. I hope I'm not severely missing the point when I offer these
several thoughts:

It's true that Randy Pausch sometimes comes across like a motivational
speaker when he talks about Alice, and his presentations for general
audiences are mostly gee-whiz and don't really focus on the science. In my
experience, most successful leaders of university R&D groups have this
characteristic--marketing yourself and your work energetically is almost a
requirement in a setting where external funding defines a person's worth.
The incentives are strongly in the opportunistic direction, but this doesn't
mean that the science isn't also good. It's just that you have to sell your
work as hard as you can using anything you can think of (including your
personality or your work's gee-whiz aspects) in addition to the science. To
evaluate the science, observers must ignore the hype and go to the original
published work.

Useful information on Alice and it's development group is at
http://www.alice.org/stage3/projects.html and a list of peer-reviewed
publications is at http://www.alice.org/stage3/pubs.html. Arthur, you may be
very familiar with the studies that have been carried out in conjunction
with Alice but your references to opportunism and bad science--which could
be correct--are difficult to evaluate as appropriate for this list or to
respond to because you haven't mentioned specific examples of bad science or
critiqued specific published hypotheses or results (forgive me if you've
done this and I've just missed it).

I'm not an expert on Alice, but I have heard Randy Pausch speak and I have
read a bit about Alice. I understand it to be a limited but interesting
environment that provides opportunities to study various kinds of people's
interactions with computer-generated three-dimensional representations.
Using an environment like this, experimental user interfaces for creating
and manipulating the environments can be created and tested. One interesting
issue that has seen some investigation within the Alice group is the
perception of 3D computer environments by visual artists, writers and
engineers. These perceptions bear on whether a unified interface for
manipulating 3D objects might be optimal for humans or whether the optimal
3D interface might vary based on the point of view and background of the
user. In a sense, this use of Alice is similar to the studies that Xerox
PARC drew upon to create the Star and strongly influence the Mac (note that
Alice is not analogous to the Star, but to the laboratories where the key
issues were identified and tested that influenced the Star design).
Ultimately the work done with Alice will combine with a large number of
other studies to influence 3D interface design.

An environment like Alice might be re-purposed to be useful in other areas.
Once could conceive of a number of settings where creation of 3D
environments might be useful in education. As a simple example, Alice might
be useful in teaching programming. The manipulation of 3D objects provides
an outcome that is easy to grasp cognitively and logic errors are quickly
perceived. Manipulation of 3D graphics is also strongly reinforcing to many
people. On the other hand, aspects of Alice's design may be suboptimal for
this use. Enthusiasm from Alice proponents in this direction is, I think,
understandable opportunism. In any case, it doesn't invalidate Alice's
usefulness as a 3D interface-design laboratory. Alice will be used for
secondary purposes to the extent that it is truly beneficial.

It's reasonable for Python to point to projects with a certain coolness
factor like Alice as examples of the capabilities of the language. In the
end, though, Alice itself plays to a limited audience. Python will
incorporate lessions from Alice and other research, but it's real success
will depend on projects like Zope, optimal relational and object-oriented
database access libraries, porting up-to-date XML processing software into
the Python libraries, Kirby's math curriculum examples, Jeff Elkner's
students' demonstration of excellence, and the like.

Jim Harrison
Univ. of Pittsburgh