bioephemera

Okay, I knew that planets are big, intellectually, but a well-done graphic is worth a thousand words, and a pretty HD video is even better. Brad Goodspeed made this video to suggest what other planets would look like, if they orbited Earth at the same distance as the Moon does. I’ve embedded it, but you should seriously watch it in HD, full-screen for maximum effect.

Scale from Brad Goodspeed on Vimeo.

I have nightmares like that. Seriously. But is the video accurate?


In addition to being full-on creepy, Brad’s video produced a fascinating discussion in the comments and on various sites linking to his blog. The reason is that the original version of the video had foreground elements, which were misleading scale-wise. Brad deleted them. Then people thought the video represented the entire night sky – as in, if you went outside you’d see nothing but Jupiter wrapping horizon to horizon, which obviously isn’t possible, unless Jupiter is now concave or the Earth is concave or something else creepy and Inception-esque.

Brad explained that what he was actually trying to capture was a sector of the night sky – what we tend to “see” in our field of view when we look at the moon:

Had I represented my moon as being only half a degree on the screen, it would have appeared as a tiny circle of light, which is not how the moon ‘feels’ to us when we look at it. That’s because, I think, when we look at the moon we tend to shut out all the empty space around it and narrow in on it. Not with our eyes but with our brains.

That should make perfect sense to anyone who’s ever tried to take a photo of a stunning sunset or cloudy moon, only to discover that the camera doesn’t zoom in enough to replicate what you “see” so easily. Anyway, the debate about whether Brad’s animation was or wasn’t misleading, inaccurate, etc. feeds into interesting issues of art vs. science. Brad readily admits that he was not striving for technical accuracy, but rather for a subjective experience consistent with having the planets looming above us:

I’m no Astronomer. I’m a guy who makes mistakes and gets away with it because I’m ‘creative’. Believe you me, anyone in the arts will tell you how cool it is to be considered the ‘artsy’ one, and how many rules it gets you around. Hell, I haven’t worn anything dressier than blue jeans and a collarless shirt to a meeting in my entire lifetime.
And I don’t really feel that bad about the mistake, (although I would certainly have preferred if I had gotten it right) because I was intending on creating a piece of art and not a scientific document. I’d argue that while I certainly got the geometry wrong, I hope I got the subjective feeling of impact right. Through the same process that causes the moon to seem larger than just half a degree on the sky, Jupiter above our heads would seem to fill the heavens. To show that in a little Quicktime is way tough. That sort of subjective perception of the world is not easily portrayed by science, and must rely on art if it is to be communicated.

(I added the emphases).

I think Brad has a very good point. Yes, I can see why experts in this area would take exception to a video that seems to claim accuracy. But the fact is, if Brad had made the video window = an entire night sky, the result wouldn’t have looked anything like what we “see” when we look at the sky at night. So it wouldn’t have been “accurate” as a visual experience, either. This is a fundamental challenge with all scientific visualization: it has to take into account the interpretations our brains superimpose on our sensory inputs – like lightness constancy. A scientific visualization can be completely accurate, yet not reflect how we would “see” the data if it were before us – and vice versa. Figuring out how to maximize the accuracy of the data and the experience is one of the hardest things about scientific illustration.

Take, for example, videos portraying cellular processes, like the classic Biovisions/Xvivo Inner Life of the Cell. In those videos, in addition to size and scaling issues, you also have color, transparency, and temporal issues. Cytoplasm is chunky (as is blood), but animators generally pretend it’s clear to let you see the structures around and in it. Then, they add false colors to help you resolve different structures and molecules. These simple adjustments should be pretty obvious to most people, if they think about it. But what may not be so obvious is that the time scale has to be adjusted, too. If you think about the speed with which DNA is translated or replicated to enable bacteria to divide as fast as they do, then watch a video of tRNAs slotting on and off an mRNA, or a replication fork zipping and unzipping nucleotide by nucleotide, you know you have to be watching some seriously time-dilated animations. But that’s not obvious if you already don’t know how fast these things happen, and most people watching cellular animations are getting an inaccurate impression of how cellular processes work. Is that a problem? I don’t think so – the animations are so helpful on other valences, the timing thing is minor. Personally, I have more of a problem with the smoothness, nonrandomness, and apparent purposeful motion displayed by molecules in so many animations. (Molecules bounce around, they don’t glide straight toward their binding sites like homing missiles!)

Anyway, I’ve talked quite a bit already in previous posts about whether or not sciartists have a duty to be scientifically “accurate” – and what that even means – and how every portrayal of data, no matter how close to the original, is going to be the product of interpretive choices. Brad’s “Scalegate” once again raises these issues: what do we expect of, or value about, sciart when we run across it in a gallery or on the internet, etc.?

Brad’s response to the criticism of his work is the best I can imagine: he doesn’t presume to know the answer, even though he created the artwork:

It appears I messed it up. Well, kinda. I guess it depends on what you consider the animation to be, art or science? This blog was always meant to promote art and science, but which one is ‘Scale’?

I love that Brad is willing to say hey, this is both science and art, and depending on the context and your expectations, it may or may not be “wrong.” I think that’s exactly right – there is no “right” or “wrong” when it comes to sciart. There’s only opening the doors to more questions about how the world works, how we relate to it, and how we can understand it better given our limited tools and senses.

And being scared of Jupiter, of course. It’s so BIG. (As is the traffic when your video goes VIRAL. Poor Brad. . . ).

Read more at Brad’s blog, which deserved to go viral – it’s very good.