Saturday, March 22, 2008

Doing VR Wrong

I recently attended a presentation on what universities are doing with Second Life, a freeform virtual reality environment. A lot of it seemed to involve requiring a class of students to spend an hour or so learning to get around in the virtual world in order to then hold a class there instead of in an ordinary classroom. Since a class in a virtual world has lower fidelity video, lower fidelity audio, and less bandwidth in the form of facial expressions and the like than a class in realspace, it seemed a bit pointless.

It reminded me of my experiences about twenty years ago with educational software. I had written a price theory text and some computer programs to go with it, and gave demonstrations of the programs at economics meetings where my publisher was trying to sell the book. One of the standard questions I got was "how many chapters of the book are on the disk?" My response was that the chapters of the book were in the book, where they belonged. What was on the disk were not chapters of the book but computer programs designed to teach ideas in ways that could be done by a program better than they could be done with text and pictures.

My conclusion at the time was that most "educational software" was bogus—doing things on the computer that could be done just about as well in a book. The motivation was that computers back then were supposed to be exciting, sexy, exotic, so the same student who would be bored reading an explanation of supply and demand in a book would be riveted to the same explanation on a computer screen. I have the feeling that the same thing is happening now with University involvement in Second Life.

Two other points struck me. One was that our university had apparently spent a fair amount of money hiring people to construct its virtual campus on its island in Second Life. But part of the beauty of Second Life is that you don't have to be a professional to do stuff in it. It's a decentralized system where, provided you have access to space to work on--your university's island if you are a student--you can build things. The right way to use it is not to try to replicate a real world classroom with a bunch of student avatars but to give students access to what they need to do things in the virtual world. Such as create a virtual campus.

The other was a comment by one of the presenters that someone in physics wanted to set up physics experiments in the virtual world. It struck me as an oddly perverted idea. What is exciting about doing a physics experiment is discovering that the real world, physical reality, actually obeys the equations physicists use to describe it. Doing the experiment in virtual reality, where the physics professor has programmed the pendulum, billiard balls, or whatever, only demonstrates that the equations obey the equations.


Anonymous said...

Heh, so the education industry is still totally missing the point of computers... just like in the '80s. And the '90s.

Kat said...

Worse than that, I wonder how many students would see the physics experiments in SL and take that as convincing proof that it works in real life.

Anonymous said...

I can imagine the "physics experiments in Second Life" being pedagogically valuable if done well. One could design several virtual universes with different laws of physics, then have students perform the same experiment in each, figure out what laws apply in which, and perhaps design followup experiments. They could practice thinking like real scientists who don't "already know the answers."

Or it could be a pointless gimmick.

Daniel A. Nagy said...

As someone who has run your programs in an emulator in order to learn basic economic concepts, I obviously agree with most of what you write.

However, I would like to respectfully disagree with the assertion that running simulated experiments is pointless; actually, your software did the same. The point is to demonstrate not that equations obey equations but that equations, when "brought to life" behave similarly to what we can observe in reality. Also, it helps understanding what equations are responsible for what phenomena.

One of my most successful forays into education was being a TA for Mike Roth's Vector Calculus class at Queen's University (Canada). For explaining the wave equation, I put together this webpage, complete with a java applet solving the 2D wave equation numerically. It went quite a long way in explaining how the wave equation relates to the observable phenomena of wave propagation.

Mike Huben said...

I worked in educational software 30 years ago on the old PLATO system, for almost a decade. Among other things, I programmed Law and Economics lessons, including animations of train/fire commonlaw burdens and simulations of self-segregation.

Most educational software is (and always has been) crap. David is right on that text and pictures do not need to be reproduced on the computer: as a matter of fact, to the extent that they're less convenient than printed material, stick to dead trees. As time goes on and working on computer gets cheaper and even more convenient, then perhaps switch to the computer.

But I'd disagree with the other two points. Having students build the second life campus is not important unless you want them to learn the skills required to use second life. Instead, build the basic campus so that it fills the requirements immediately, and then let students extend it instead if they discover problems with the original design. Placing obstacles to usage is not the way to encourage usage: it is a different educational endeavor that you may not want.

The second point is correct if the objective is belief that the equations model reality. And confirmation of predictions in the real world can be exciting and thus valuable. But I'd think that far more often the problem is really understanding of what the equations represent. That usually requires equipment or apparatus which may or may not be convenient or affordable: a simulation is essentially a public good. And a simulation can model really expensive and sophisticated apparatus, rather than junk, for the same price.

I think hudebnik's idea of simulations providing new opportunities for theorizing and discovery is a really good one. Modeling real principles has the problem that students can look up the answers rather than think through the evidence. Simulation in imaginary worlds can provide a near-endless supply of novel problems.

DerekL said...

""But part of the beauty of Second Life is that you don't have to be a professional to do stuff in it."

Well, not really. Not if you expect your virtual environment to not look like crap. It's takes a pretty serious skill level and a fair amount of experience to make anything in Second Life look decent, the tools available inworld are primitive and (to say the least) idiosyncratic.

Robbie said...

A decent physics simulator would be very useful for helping students get a feel for some of the strange phenomena of quantum mechanics or relativity. In these cases the experiments are often too difficult (if not impossible) for students to perform in real space.

Unknown said...

I like to have the text of books in digital form so that I can quickly search for some specific content I need later. And I like to be able to copy and paste excerpts for sharing/criticism.

But I would rather read it originally in dead tree form.