Notice how the wheels roll down the track. On each axle, the wheels are rigidly attached, and the wheels are slightly tapered. If an axle gets a tiny bit off center, the wheels roll on different circumferences, which steers that pair of wheels back to the center. In a turn, the axle steers itself off-center by the same mechanism. Ideally, the wheels don't slip in a turn any more than they do on straight track. The flanges are a backup system.

If you watch a train, tidbits of science and engineering are in view. You can smell the cars marked Molten Sulfur. When locomotives pull a heavy train uphill, watch for the nozzles squirting small amounts of sand for traction. Listen to the horns and hear the Doppler effect.

I use Pilot EasyTouch. Look on http://www.staples.com and search on pilot easytouch . I buy medium retractable, a dozen each of black, red, blue, and give some as gifts. I never notice skipping or blobs. Fine point is available. The simple retractable ones (NOT "Pro") have a clear barrel so that you can observe the mechanism, a great nerdy touch.

VRML graphs are used for non-trivial purposes on my web site, http://www.jimworthey.com . The general topics of the web site are lighting and color.

It depends on your overall plan whether you need new dead-tree books. But the Schaum's outline books are good, with plenty of worked problems. Look in a college bookstore or do a web search on Schaum's outline .

I was not aware of time cube until now, but of course I found it and the related Wikipedia article. Let me go at it one more time. I have a BSEE, an MS in Physics, and a PhD in Physiological Optics (human vision).<br>

If you look at the world around you, you'll see, for example, that shading of 3D objects gives you important shape information. Look at a black-white photo, where shading is the primary cue. Highlights are even better at giving high contrast and defining regions of high curvature. Color helps identify things, but also to outline them and even gives shape cues. All such cues depend on the optical interaction of objects with lighting.<br>

Now if you can rewind 50 years to the age of slide rules, nobody was prepared to think seriously about the details of optics and color. Illumination engineering developed on a very simple basis, considering light to be a liquid. Since then, lights evolved, but the words and theory have evolved little, except in my work.<br>

A ceiling full of fluorescent lights does everything wrong, but the usual discussion never distinguishes bad from good. Again, see my web site. A good place to start is the article on light source size.

The problem is that the whole lighting discussion is in vague terms. Lighting in fact controls the stimulus to vision. The usual discussion (books, research) throws out the direct effects of lighting on the stimulus to vision. They assume that light is a liquid (minimizing any reference to optics). Designers praise daylight, but there is little discussion (except my articles) about how daylight is different from artificial sources. Lighting could be done like other engineering, with a cause-and-effect discussion. But that's not what is done. Clearly some existing sources are better than others. Much better. But if the discussion fades into vague statements, change will be hit and miss. Results of years of research are on my web site.

I made an entry last year when the contest was open. It concerned a neglected problem that is ripe for better research. I am disappointed that the voting will choose among rather bland and general ideas. My original entry is available from my homepage, http://www.jimworthey.com/ .

This page: http://www.jimworthey.com/jimtalk2004nov.html is the graphics that I used for a talk last year. As you read through, you'll see 3D pictures and animated graphics. When you see a 3D graph with a border, that links to a VRML pic that you can zoom and rotate. For free VRML viewer see http://www.parallelgraphics.com/ for example.