I'll second the .28 mm uni-ball signo. It's the best thin-line pen I've ever used. It's great for doing neat, precision writing, but I can't write very fast with it. In my opinion the .18 is a little too small. I haven't tried the .38 version.

I've also tried the Pentel Slicci .25 mm pens, as mentioned by other posters, but I don't like them as well.

For a thicker-writing pen, I highly recommend the Uni-ball vision elite micro.

Rucker's "the fourth dimension: toward a geometry of higher reality" is a very good read, or at least i really enjoyed it when i was in high school. Much of the book is about Flatland, with a lot of commentary and additional ideas about A. Square. I'd recommend it.

Download it while you're at the university.

That first picture reminds me of watching cinemax when I was a teenager, minus the naked women.

I'd suggest a $100 or so USB microscope. You can use it to look at opaque objects, and you can have the picture on your computer screen. That would be a big help when trying to point out what the kid is seeing.

Wikipedia says that the boxer linked to above is 6'3", and the other link says that "In the modern era, middleweight means that the fighter's official weight does not exceed 160 pounds."

But all of that is beside the point. Using height and weight to determine how well someone can fight is like using clock speed to determine how fast a computer is.

And 160 lbs at 6'3 is so skinny as to be a total non-threat.

Henry Maske and I'm sure plenty of other middleweight boxers would disagree.

Mr. Soupy Sales

They have 13$ arduinos.

If you're making more than one device you can go even cheaper if you just buy a bare ATMega328 + a couple of cheap components. Granted, you also need a USB to TTL cable which is another $20 or so, but you only need one of those.

Did it come from IKEA?

You wouldn't need the A/D converters. The atmega328 (main chip on the arduino) has them built-in.

Indeed. The development of this ability in hardcore Scrabble players is similar to the Tetris effect.

This is not quite equivalent to the TSP. TSP tries to find a minumum weight hamilton cycle, which does not allow repeated vertices. In the problem here, they are allowing vertices to be repeated, but they are trying to minimize the number of repeats. Also, the graph obtained by overlaying the triangular grid might not be hamiltonian.

That said, I suppose you could translate an instance of this problem to the TSP by doing something like adding weighted edges between nonadjacent vertices, and letting the weight of each new edge uv be the distance between u and v in the original triangular graph.

... from wayne's world.

"Was it Kierkegaard or Dick Van Patten who said, 'If you label me, you negate me'?"

I seem to remember a guy named Steve saying that there's better money selling magazines door-to-door.