Every once in a while, you encounter a person who collects something truly odd. This guy has a website dedicated to the insulators that are used on telegraph poles along British railways. That’s… just… Wow. How niche can you get?

Teleramics – Insulator collecting UK

The new WordPress 2.1 Ella was released about 17 hours ago, and since this coincided with a few minutes of boredom, I upgraded. Check out the link for details on what’s improved: most notable to me is that it does appear to be significantly faster. I noted a couple of strange things though: the “Press It!” bookmarklet that I often use must have changed, because the old one doesn’t work. I got rid of it, and dragged the new one back to my toolbar, and it works fine.

Had no problems upgrading. Seems like a no-brainer.

[tags]Wordpress[/tags]

When I was probably ten or twelve years old, I remember that the book Flatland by Edwin Abbott somehow came to my attention, probably through something related to Martin Gardner’s Mathematical Recreations column in Scientific American. It is an interesting book that tries to present an easy-to-understand example of how we might visualize four dimensional space by hypothesizing a two dimensional world, inhabited by beings who can only perceive the two dimensional world around them. The narrator, who is a simple square, dreams that he visits a one dimensional world, and tries to convince its inhabitants of the existance of two dimensions. He’s then visited by a Sphere, a 3 dimensional being who helps him perceive the three and more dimensions that he inhabits.

It’s kind of a fun book, and being written in 1884, it’s in the public domain and available in a wide variety of formats and sites which you can find on Wikipedia. Interestingly, it’s going to be made into an animated educational film, with Martin Sheen as the vocal talent. Check out the trailer here.

[tags]Mathematics,Flatland[/tags]

We woke up and got in line for the Best Buy in hopes that we might get a Wii. They had fifty. We were fifty second in line. Sigh.

[tags]Nintendo,Wii,Disappointment[/tags]

I’ve mentioned it before: when you are interested in as many strange things as I am, web surfing can be dangerous to your time. Lately I’ve been going through my blogroll with two purposes in mind:

• Delete the stuff that I don’t read anymore, or consider boring.
• Tag the remaining links with slightly more meaningful tags in Google Reader.
• Lastly, find more cool stuff by mining the links of blogs I already have.

Courtesy of one of the mathematics blogs (I can’t remember which one), I located The Fitful Flog, a fascinating blog which had many interesting bits of paperfolding. Not the traditional crane stuff mind you, but interesting stuff that used a combination of straight and curved folds to make things that I have known as “developable surfaces”. Don’t know what that means? It means that is has zero Gaussian curvature. That didn’t help? It means that you can flatten them onto a plain without distortion (stretching, compressing, tearing). In other words, it is the class of surfaces that you can get by folding paper.

The thing that piqued my interest was this Champagne Flute, which I folded up…

It’s a very simple pattern, using a series of straight and curved folds. Very neat.

I meantioned this to Tom, who reminded me that the person that probably brought the term “developable surface” into my vocabulary, Paul Haeberli (one of the many notable computer graphics people with the initials PH) had some interest in the topic, and, in fact, had recently received a patent on the topic as well as apparently developing some commercial software for the design of developable surfaces. Very cool. It reminded me of this cool paper from SIGGRAPH 2004, as well as perhaps this successor. It’s all very cool.

Isn’t it amazing where the stream of the web can carry you on a Friday?

[tags]Origami, Mathematics, Developable Surface[/tags]

Addendum: Here are some more papers that could be relevent.

Addendum²: I was pondering this paper pattern, and realized I knew how it was created. To prove it to myself, I wrote a little C program that spit the model out in the form that my homebrew raytracing program could render. Voila.

Let’s face it: I’m not Canadian, so I don’t really care about hockey. But still, I found the following story to be both amusing and thought provoking, and i suspect that similar situations can occur in baseball (a sport that non-Canadians enjoy) so I’ll comment briefly.

All-Star games are odd things because they take place between players who the fans select by voting. In the past, ballots were usually distributed by the league, and collected. In recent years, it has been popular to use the Internet to allow fans to vote.

So, we now have a popularity contest decided by vote, where essentially everyone can vote as often as they like, where the marginal cost in terms of times and resources is exceedingly low, and where automation of voting is a very real problem.

What could possibly go wrong?

[tags]NHL All-Star Game,Rorygate[/tags]

Addendum: Check out the entry under “Stuffing the ballot box” to see how the 1957 Major League All-Star voting worked out.

During a stop to the new Coppola winery in Healdsburg, I noticed that they had a kit for a pinhole camera on sale for a mere $4.99. I’ve goofed around with pinhole photography a bit before, so the allure was too much to resist. A couple of hours later, I had the camera on the right assembled. I’m hoping to run a roll or two of film through it this weekend.

Stay tuned for updates!

[tags]Photography,Pinhole Camera,Pinhole Photography[/tags]

You only need to make one mistake when it snows in Portland: that’s trying to drive at all. Here’s a clue: walk to your car. If you can push it sideways away from the curb, you shouldn’t attempt to drive it. Just stay wherever you are at. If you do decide to venture out, at least have the common decency to not insure with the same company as me.

[tags]Snow,Portland,Idiocy[/tags]

Thanks to Josh for bringing this one to my attention.

This took me a few minutes, but I figured it out, and archived it here so I wouldn’t need to spend that time again.

mplayer -cache 128 -tv driver=v4l:width=640:height=480:outfmt=i420 -vc rawi420 -vo xv tv://

It works.

[tags]My Projects,Webcam,Mplayer[/tags]

Addendum: You can specify which video device by adding it to the tv options:

mplayer -cache 128 -tv driver=v4l:width=640:height=480:outfmt=i420:device=/dev/video1 -vc rawi420 -vo xv tv://

It’s sometimes odd to think of the ways that the eddies and currents buffet you around while reading on the Internet. Frequently, you encounter odd bits that resonate with stuff you have seen before. On such current lead me to return to a topic which I’ve dabbled in from time to time: gravitational n-body simulation.

Some of my first computer experiences were in solving versions of this problem, beginning perhaps with the classic lunar lander written in basic. I recall writing simple solar system simulators based upon the math presented in A.K. Dewdney’s Computer Recreations column in Scientific American in an article entitled A Cosmic Ballet, and later reprinted in his book The Armchair Universe. Later, when I gained some mathematical sophistication, I read about work by Carrier, Greengard and Rokhlin and Appel’s An Efficient Program for Many-Body Simulation” (I haven’t found a link online). Later, I would revisit this stuff when Pat Hanrahan introduced fast hierarchal radiosity.

Today’s surfing lead me to a website I hadn’t seen before. The Art of Computational Science visits this stuff for me again, but in a way that is actually designed to help you learn about how to write code to do this kind of stuff. Pretty neat.

[tags]Computational Science,N-body simulation[/tags]

When I dropped into Tom’s office earlier today, he mentioned that he had found a copy of Albrecht Durer’s Of the Just Shaping of Letters available as a printable PDF. This version was scanned from the 1969 Dover reprint of the 1917 edition which is in the public domain. Cool! A bit of googling turned up the original scanner, one Sean Gleeson, as well as the PDF itself. Cool. This has the constructions for each of the 23 capitals of the Roman alphabet. Very, very neat.

[tags]Public Domain,Typography,Albrecht Durer[/tags]

I should have mentioned this before, but in case anyone didn’t know, there’s been a naked eye comet visible in the west as the sun is setting (at least, for people in the northern hemisphere), and for the last couple of nights, I’ve had binoculars out and let my fellow Pixarians have a glimpse at the fun. I snapped a bunch of photos in the most ghetto way imagineable, simply holding the Panasonic point-n-shoot to the eyepiece of my binoculars, and snapping away, hoping to get one that I like.

Most were deplorable, but a couple aren’t too bad. Here’s one of the better ones from the bad lot.

It’s rapidly sinking as the sun is setting, so you might not get a chance to see it, but give it a shot.

[tags]Astronomy,Comet McNaught[/tags]

Addendum: I was going through the pictures that I snapped, and realized that you could see the comet actually moving (well, in reality, the earth rotating) in real time in the sequence that I shot). I combined them all into an animated GIF which you can see here or here (first is larger, and about 2 megabytes, the second smaller and 1 megabyte). I just thought it was kind of neat.

Another link stashed away for safe keeping:
A Paint-Brush Laser Range Scanner

It’s really quite clever in the way it uses a double frame to provide a reference coordinate system and allow for free motion of the camera and the laser.

Despite having done quite a bit of computer graphics over the years (even getting paid to do it), I haven’t really done much in the way of computer vision. I’m currently trying to work on a project for the upcoming Maker’s Faire that has something to do with computer vision (he says slyly, teasing his reader(s) mercilessly), so the first step is to try to calibrate a camera. This quantifies the distortions in the camera, and allows you to locate the camera in 3 dimensional space relative to a base coordinate system. Toward that end, I wrote a simple ellipse detector, and it works really well. On the fastest machine at my disposal, it processes this 640×480 frame (reduced here to 400×300) in about 3.5 milliseconds, and locates all of the ellipses in the scene to subpixel accuracy. Or at least, that’s the ultimate idea. But I have the feature detector working.

As I get more of the ultimate project working, i’ll post more info here, and it is my intention that eventually the entire project be available for download.

[tags]Computer Vision,Camera Calibration[/tags]

Just thought I’d stash a reference to this paper on low cost motion capture. It uses two webcams to track the lower half of someone’s body. Just keeping this for inspiration in a related project.