2,3 Turing Machine Type 1,2

The above image comes from Stephen Wolfram’s blog announcing proof of the simplest universal Turing machine. Believe me, I’m not about to describe a Turing machine, but I will happily describe my utter confusion when seeing the image. It initially reminded me of a Sierpinski Triangle, but obviously, something else is going on, and there’s no explanation to help! In my typical, touchy way, I got annoyed at this silly little picture, lacking any caption or descriptive text, presented as some kind of straightforward statement of the problem and its solution. Grrr…

Then I started trying to figure it out. I’ve previously blogged about the Mathematica Player and Wolfram Demonstration Project, and indeed, there’s a similar (actually, mathematically identical) example of a Turing machine in the collection. That discovery clarifies things somewhat, and yet another actually makes it quite comprehensible. In fact, reading the “New Kind of Science” prize from Wolfram’s blog, it became obvious to me that Wolfram (at least, maybe others, for all I know) has a visual shorthand that he uses in describing Turing machines—and that Mathematica evidently uses in displaying them. And I then realized that I had gotten in a huff a little hastily.

Allow me to extol a bit of personal pedagogy. In general (as I’ve mentioned in my “What Is Viz?” presentation), I divide science visualization into three basic types: 1) communication with oneself, 2) communication with a peer group, and 3) communication with public audiences. In typical astronomer parlance, I will refer to these as Type 1, Type 2, and Type 3 visualizations. In this blog, I most often comment on Type 3 visualizations (those addressed to a broad audience), and I initially mistook the above image for that type.

But it’s not. Instead, I think Wolfram has devised a Type 1 visualization of the Turing machine that, thanks to Wolfram’s influence, has transitioned to a Type 2. Presumably, there’s some small audience of “peers” out there for whom the above makes plenty of sense—perhaps they can even extract useful information from it.

Back to Jupiter

I know that I just blogged about the Jovian magnetosphere, but here I go again. And it’s another press release from SwRI, of all things. There’s a lot going on in this diagram! First off, kudos on getting the dipole to look right, but then things get a little confusing…

The profusion of orbit lines and magnetic field lines (or tubes, I guess, if I look at the high-resolution version of the image) might make sense to a well-informed viewer, but they seem confusing for the uninitiated. I’m also wondering why the faint structure that connects the moon Io to Jupiter, which indicates ionized gases trapped in Jupiter’s magnetic field, doesn’t actually follow a magnetic field line. Yeah, they got the dipole bit correct, but then garbled the message! And the Io torus, which also looks somewhat tubular in this depiction, doesn’t seem to lie in the same plane as the orbit lines. All very odd.

Here’s the image caption, BTW: “About [one] ton of volcanic gases are spewed out by Jupiter’s moon Io every second. When ionized, these gases become trapped in Jupiter’s strong magnetic field (shown in blue) and form a vast ring (shown in red) around the planet with Jupiter’s 10-hour spin period. Jupiter’s strong magnetic, rapid rotation and Io’s prodigious source of material result in a giant magnetosphere whose dynamics are very different from the Earth.” Not such a bad explanation, really, although it helps to know that the ionized material rotates along with Jupiter’s 10-hour period, whereas Io orbits more slowly, so the stuff gets smeared out along the length of the moon’s orbit.

If you’re interested in a bit more on the topic, you can also check out an actual image of the Io torus and even see its rotation with Jupiter (the latter page actually has a much better description of the torus than the above as well). Um, did I mention I almost did a Master’s project looking at the Io torus…?

Brightness and Darkness

A brief note. I just ran across “Brightness and Darkness as Perceptual Dimensions” on PLoS. I can’t say I fully grok the contents, but the authors suggest that brightness and darkness behave not simply as polar opposites, rather as axes of a brightness-darkness space (i.e., the brain processes the two differently). Hrm.

As the authors state, “Vision scientists have long adhered to the classic opponent-coding theory of vision, which states that bright–dark, red–green, and blue–yellow form mutually exclusive color pairs.” But, “Here we provide direct evidence that brightness and darkness form the dimensions of a two-dimensional (2-D) achromatic color space. […] Our 2-D model generalizes to the chromatic dimensions of color perception, indicating that redness and greenness (blueness and yellowness) also form perceptual dimensions. Collectively, these findings suggest that human color space is composed of six dimensions, rather than the conventional three.”

Definitely merits a closer reading…

Preaching to the Choir

A press release from the Southwest Research Institute describes observations made of Jupiter’s magnetosphere by the New Horizons spacecraft. The above image (sorry, it’s quite low-res, and to take a closer look, you’ll need to open up the huge version linked from the above) summarizes some of the results. To summarize my response: it would work quite well in a scientific publication, but it just doesn’t cut it for public use.

I admit that it’s nice to see actual data represented—and nice to see an attempt at providing context for them—but the context in which the data fails to help much; furthermore, it really only conveys the context for an expert viewer—one who knows about the solar wind, magnetic fields, and such. In a previous post, I complained about depictions of Earth’s magnetosphere; I won’t bother reiterating my gripes, but they can be applied to the top portion of the above image. Honestly, some version of the schematic portion of the image would probably have sufficed for a press release, but it would have required significant work to be made more comprehensible.

Also, we’re given no hint as to how to read the spectrograms below the schematic diagram, and furthermore, they utilize opaque units such as “Energy/Q [eV/q]” and “DOY 2007 [UT].” Oh, yeah, and pseudocolor. ’Nuff said.

Making matters worse, the picture’s caption incorporates a trult impressive quantity of jargon. To call it “incomprehensible,” at least for public audiences, would be kind. The press release is better, but not by much. The only audience I can imagine picking up on this story is a quite sophistication publication such as Scientific American. I guess that’s all well and good (better than nothing), but a little more effort could make this result more accessible to broader audiences.

(I’ll just add that the New Horizons folks actually produced a spiffy press kit that describes the fly-by, with some decent diagrams, too.)

BTW, I’m in Athens attending the Communicating Astronomy with the Public conference. Fun stuff! And I finally achieved my goal of presenting a PowerPoint using no bullet point slides. A personal victory.

XOX, XO

An article in today’s San Francisco Chronicle (yes, I do read my new hometown’s daily paper) features the above image of the XO laptop designed as low-cost hardware to be supplied to children around the world. What attracted me to the image is the manner in which it anthropomorphizes the technology: the screen swiveled at an angle as if the XO had just turned to face us, the ear-like antennae perked up as if it were listening, and of course, the images of children to bring home the point. It would be difficult to make technology appear any more friendly.

And indeed, the technology seems quite friendly! Not just in terms of its use, but also in terms of its creators’ goals. One Laptop per Child (OLPC) intends “to provide children around the world with new opportunities to explore, experiment and express themselves,” through an innovative combination of hardware, software, and economic strategy. I’ve had an opportunity to see several XO laptops, but I haven’t played with one much. Very clever, in more ways than one.

BTW, the OLPC website features an interactive showing how the wee laptops communicate: the ear metaphor for the antennae actually suggests how the hardware daisy-chains together connecting one another. Humanizing and illustrative!

More Abstraction

Okay, I give up.

No, not with the blog, in spite of my lousy track record posting lately. I give up trying to figure out the image above…

I mean, it’s pretty and all, but what does it mean? I’m so baffled that I won’t even complain about the pseudocolor (indeed, I’m quite fond of orange). I read through the press release and the accompanying caption (which seems to have been removed recently), but… Huh?

Here’s the caption, BTW: “Spectroscopic image showing the microwave-frequency magnetic resonances of an array of parallel, metallic thin film nanowires (‘stripes’). The peak in the center is due to resonances occurring at the stripe edges while the strong horizontal bar is due to resonances in the body of the stripes.”

Since I’m trained in astronomy, my tendency is to read frequency along the horizontal axis, which would imply a peak of some sort at a particular frequency, but that doesn’t feel right, somehow. Maybe it’s actually a spectrogram of some sort, with the horizontal axis representing the spatial extent of the nanowires?

Whatever the image tries to show, the real question is: why confuse people with it?

The NSF Nose Best

It’s that time of year again! The National Science Foundation has announced the winners of the 2007 Science and Engineering Visualization Challenge.

What I find clever about the image above is exactly what is remarked upon in the caption from Science magazine’s Ben Lester. “Normally, CT renderings meld slices together into smooth surfaces, but, in what he terms the ‘Rainbow Technique,’ Fung instead broke them apart, creating a topographical map of the airspaces described by the contour lines of individual slices, and colored according to the density of the tissues that border them.”

I question whether that’s a completely accurate description of the technique: the contour-like color variation suggests that there’s more than tissue density informing the color selection. Regardless, the technique draws attention to the asymmetries in the image, which would be far less apparent if the same data were rendered in a photorealistic fashion. As always, I wonder how the uninitiated interpret images such as this, but overall, I rather like it. Even the Moiré patterns I manage to find both engaging and distracting at the same time.

(A much less appealing—in fact, presumably inadvertant—appearance of contours shows up in an image associated with an ESO press release that came out today. I would recommend a Gaussian blur, kiddos!)

Anyway, take a look at the other winners. Interesting stuff. You can check them out via the link above or by going to the corresponding page on the Science magazine site.

One short year ago, I blogged about a 2006 winner (while I was visiting Chicago and listening to Wolf Parade, evidently), which also happened to be a CT scan. And that reminds me! This blog is just a little over a year old. Sadly, I’ve been unable to pay as much attention to it of late, but so it goes. I won’t give up just yet (although I will cringe when people make reference to it at conferences or before I give a talk, since I’m embarrassed at how rarely I post nowadays).

Mathematica Mojo

Visiting Chicago for the annual meeting of the Astronomical Society of the Pacific, I attended a one-day session on astronomy visualization (the usual suspects). Lots of things caught my attention, but I thought I’d highlight the Wolfram Demonstrations Project, a collection of visual (non-Java) applets that utilize the free Mathematica Player.

The image above is a snapshot of the “day and night world clock” from the collection. Nothing too special about it, particularly as a snapshot, but within the player, you can fiddle with the time of year and time of day to see how the dividing line between day and night changes over the course of the year. As a kid, I was always fascinated by the Geochron clock at my local planetarium, so perhaps I’m just being nostalgic. Still, I enjoy watching the terminator’s projection on the flat surface, bending through the course of a year.

Among the other demonstrations, I like the Cepheid light curve, too, as well as the hydrogen orbitals and the inverse Hilbert matrices.

Ball and Stick, Apple and Orange

Today,
Science Daily reported on research from Rice University that had actually appeared in a press release from Rice last week. Go figure. The new article includes the above image, however, which could be perceived as an improvement (or not) over the text-only copy from Rice.

A quick glance at the image caused a sudden nag, and I started to browse on before I figured out what was bothering me.

The nanotube should be made of atoms, right? Presumably those little grey shiny balls in the molecular model above. But interior to the nanotube, we see brightly-colored (one might be tempted to call them radioactive-looking) blobs that look like a scanning-electron micrograph of something-or-other. But these are supposed to be atoms! Specifically, astatine atoms, which should be a fair bit bigger than shown here.

This isn’t a big deal, I suppose, but it’s oddly distracting. First off, they use different visual vocabulary to represent the same kind of thing: atoms are shown in two distinctly different ways in the above image. Secondly (and I know I’m going out on a limb here), the image they choose perhaps even vaguely suggests cancerous cells… And given that the press release concerns using nanotubes to treat cancer, that’s potentially problematic.

Holy CMB, Batman!

A press release from NRAO announces, “Astronomers Find Enormous Hole in the Universe.” Hmmm. I’ll refrain from commenting on the overzealous word choice (except insofar as I just did) and focus on the image above.

I have to admit that the first thing I thought of when I saw the diagram was a poorly-rendered traffic cone—with a circular base, executed with an acute lack of graphical perspective.

The caption reads: “Illustration of the effect of intervening matter in the cosmos on the cosmic microwave background (CMB). On the right, the CMB is released shortly after the Big Bang, with tiny ripples in temperature due to fluctuations in the early Universe. As this radiation traverses the Universe, filled with a web of galaxies, clusters, superclusters and voids, it experiences slight perturbations. In the direction of the giant newly-discovered void, the WMAP satellite (top left) sees a cold spot, while the VLA (bottom left) sees fewer radio galaxies.”

The viewer (i.e., astronomers with their WMAP satellite and radio telescopes) is off to the left of the image, and it would probably be worth continuing the sides of the traffic cone until they meet—at Earth! Otherwise, it really doesn’t make much sense. Given its opacity and apparent solidity, the traffic cone looks like a structure, and truncating it simply exacerbates the problem.

Plus, the pictures of the two telescopes distract from what’s going on and further confuse things. They hover there by the tip of the cone, as if they belong there. But the radio telescope wasn’t even part of the observation depicted by the diagram: radio observations supplied confirming evidence.

I admit that I don’t have an immedite solution on how to depict the observations better, although the above image could be improved by making the cone appear more transparent, more a part of some continuous medium affecting the observations, and more connected to an observation point to the left of the image. Oh, and more appropriate in its perspective.

Curiously, the image is offered as a 73KB JPEG, a 278KB JPEG, and… a 34.3 MB TIFF! Now, I’m all about lossless compression of images, and I noticed that the giant TIFF had no compression whatsoever. So, just for kicks, I saved it out with LZW compression and it shrunk to 9.1MB. Yeah, disk space is cheap, but c’mon, let’s be sensible.