Monday, October 5, 2015

Monday, September 28, 2015

For Better Nutrition

From Visual News, a revamped design of a boring nutrition label. Using bar charts and a pie chart, nutritional data is presented in a better and visually more informative way. Also shown at Visual News are redesigns for the myriad of parking signs and also a wine menu. These are from the folks at Column Five Media who produce lots of infographics.

Monday, September 21, 2015

YADDA New York's Penn Station

Yet Another Door Distribution Again. Doorway to a train track at New York's Penn Station showing the two halves of nearly circular bell-shaped curves of wear. But from what? Most likely the edge of Amtrak luggage carts pushing the door open. The carts seem to push equally on each door, with perhaps a slight bias for the left side. Compare this to the bookstore doors or the Dulles airport doors from earlier posts.

Monday, September 14, 2015

Fall Foliage Map

Here is an interactive 2015 USA Fall Foliage Map from website
Here it is set for the coming week of September 19. Perfect for planning any Fall leaf peeping trips. the site also explains why and how the leaves change color. Enjoy.

Monday, September 7, 2015


From Oliver Roeder at the blog FiveThirtyEight this is a 'scatter plot' of the size of all 2,229 paintings in the Museum of Modern Art. The size of each painting is plotted with Width on the x-axis and Height on the y-axis. This data point serves as the top right corner of the painting's rectangle. There is so much variability among so many smaller paintings that they get over plotted on the graph, so much that only a purple blob is produced.

They go on to consider only the smaller paintings with only the top right corner's data point. This resolves the purple blog into data points for each painting, but some detail is still lost in the over-plotting of these points.
More at FiveThirtyEight.

Monday, August 31, 2015

Puzzling World

Here is a busy graphic representing the composition of the world's economy. Countries are given polygonal sections proportional to the relative size of their economies. Each sector is divided into three smaller regions: Most lightly colored is Agriculture, somewhat darker is Industry, darker yet, the sub-segment that is greatest for most countries, is Services. Starting with the largest (US) the polygons seem to be arranged in a rough Pareto-chart fashion from largest to smallest in a counter-clockwise fashion. Although, individual countries are hard to locate, those of similar size contributions are somewhat easy to compare. I found it fun to puzzle out comparisons that I found surprising, for example, US industry has about the same area as China's. Mathew Yglesias from Vox points out others:
You can see some cool things here.
  • For example, compare the US and China. Our economy is much larger than theirs, but our industrial sectors are comparable in size, and China’s agriculture sector looks to be a little bit larger. Services are what drive the entire gap.
  • The UK and France have similarly sized overall economies, but agriculture is a much bigger slice of the French pie.
  • For all that Russia gets played up as some kind of global menace, its economy produces less than Italy. Put all the different European countries together, and Russia looks pathetic.
  • You often hear the phrase “China and India,” but you can see here that the two Asian giants are in very different shape economically.
  • The only African nation on this list, South Africa, has a smaller economy than Colombia.
This graphic has gotten criticism as being too busy, too unorganized, and too difficult to make comparison, among others. But I do like the puzzle-it-out quality of the graphic. It makes us work a little to understand and make comparisons. I especially like the quote from Andrew Gelman at Statistical Modeling, Causal Inference, and Social Science:
Paradoxically, the best thing about this graph may also be its worst: Its tricky, puzzle-like characteristic (it even looks like some sort of hi-tech jigsaw puzzle) makes it hard to read, hard to follow, but at the same time gratifying for the reader who goes to the trouble of figuring it out.
From howmuch, via Vox, via SM,CS and SS

Monday, August 24, 2015

Carport Oil Drips

This is a stain on carport floor of my mother's home.  It comes from the the persistent drip of oil from her parked car and the variability of her parking location. When the car enters the carport from the bottom of this picture, the front of the car (and its motor) stops over the developing stain. The stain pattern running from the bottom to the top in the picture is from parking less or more of the car, respectively, in the carport. The left to right stain is from the car entering the carport more to the left or right.

Notice the prominent lines of oil drops on the right side of the stain. It could be from repetitive and quite accurate placement of the car more on the right of the carport. But it could also be a re-positioning of the car more forward or backward after it stopped on an established an entry line. This illustrates the conditional distribution of bottom to top oil placement given a fixed left to right car approach. Notice that it's likely that the floor is not level, indicated by the lighter, more liquid stain running towards the entry of the carport.

What results is a joint, two-dimensional scatterplot of automobile placement in the carport. The overall pattern of the oil is consistent with a targeted, ideal car placement. With a few other assumptions, such as independence of bottom to top and left to right placement, it is possible to prove that the resulting joint distribution is that of a bivariate normal distribution. More on the details in a future post.