Bad maps

People had some peculiar ideas in 1932. Try reading this wonderfully detailed diagram of evolution (if you’ve got the bandwidth, download the 3212×8748 pixel version).

The vertical axis makes sense: it’s a logarithmic scale of geologic time. It’s not quite right, since it has life arising about 1.6 billion years ago, when we now have good evidence that that occurred more than twice that long ago. I’m not going to complain about that — science does march onwards, and it probably represents the best estimates of that time.

The horizontal scale is a real problem, and is revealing something about early attitudes towards evolution. It’s completely unlabeled and poorly explained. Each of the bands of color is, apparently, a lineage; in the excerpt from the beginning of the histomap below, the light green are the bacteria, the dark green are the chlorophyllic plants, the yellow are the porifera, etc.

i-d69a25495954a7fb43567bb55ce776c5-histomap_early.jpg

The extent of each lineage along the horizontal axis is drawn with some care, but it’s meaningless. The legend says, “The horizontal width of any strip at any time suggests in a general way the relative dominance of that type of life at that time.” It’s got the Porifera as the bulk of chart 900 million years ago, for example, with plants and bacteria as a narrow strip, tiny in proportion. It doesn’t even make sense to talk about “dominance” in these terms, and obviously there is no way to measure this.

Another problem with this particular rendering of evolutionary history is that almost every band arises independently from a spot on the left border. Read it literally, and this is a chart of sequential creation, with no detectable relationship between most lineages.

The absurdity grows increasingly apparent as we read down the chart. Here’s the bottom of the histomap; the modern era is completely dominated by the bloated bands of humans to a ridiculous degree.

i-fdcaa56352c5d80a2d31af55cf6c8722-histomap_late.jpg

The yellow (of course!) bands on the left are the Chinese and Japanese — the map is broken down by human races and nationalities by this point — the orange are the Russians, and that broad pink grouping near the middle are the Americans on the left, the English on the right, and a narrower lighter pink swath separating them representing Canadians, Australians, New Zealanders, and South Africans. Then the Germans in blue, the Latins in light green, etc. If you look way, way over on the right, you’ll find those unimportant bacteria, plants, insects, and mammals as teeny-tiny little ribbons which together compose about as much relative importance as the Greeks and Italians.

The focus of the entire chart is on two trivial and poorly defined entities: human “races” and “dominance”, with evolution as only a badly represented premise. All that impressed me is how badly it is done.

The creator, John B. Sparks, seems to have done a lot of this kind of visualization work. He also did histomaps of world history, religion, and who knows what else. The world history map seems to be uncritically praised, but what little I’ve seen of it looks like more of the same — charts with the author’s subjective impressions of importance illustrated over time. I don’t get it. They look utterly useless to me, except as pointless accumulations of words and a false representation of a few events in time. Oh, and as a historical curiosity. I hope no one is trying to learn history or science from these things.

A celestial omen!

Here I am, rooming with that space case, Phil Plait, and what should appear on the astronomy photo of the day but a cosmic cephalopod, a picture of Comet Holmes that has a resemblance to a cartoon octopus.

i-25000e197330296c3808cde2d9e5bf05-comet_octopus.jpg

Perhaps this is a sign of reconciliation? That the savage enmity between two science blogs shall be soothed? That the ferocious competition between the best science blog on the web and the blog that tied with a junk science site shall be at an end? That disparate disciplines can find common ground in the beauty of the natural universe?

Naaah, I hope not. The rivalry is too much fun.

You, too, can be an agent of selection

Here’s a website of mutating pictures, a collection of images made with a splatter of scattered triangles. Your job is to browse through them and score them for how much they resemble a face — which isn’t easy. If I stare at any random pile of symmetrical shadings, they all start to look like faces to me.

Anyway, pictures that get higher scores produce more progeny, with slight mutations, in the next round of picture generation. You can see where this is going…