
Just as the GoPro camera and YouTube have enabled us regular viewers to see things from a perspective we've never seen before, aerial photography, satellite imagery, number-crunching computers and GPS trackers can help us understand data flows too complicated to easily imagine.
The fantastic PBS miniseries America Revealed, which "explores the hidden patterns and rhythms that make America work," makes stunning use of data-viz techniques to stimulate the eye-candy part of your brain while teaching you something. Pictured up top is what our internet access looks like (image rotated to fit). Below is "the route of a family-run combine harvesting business as they zig zag across the U.S."

These pinpricks of light represent U.S. job losses (rotated to fit).

Most fascinating to me are these blue lines showing the Manhattan travel routes of a pizza delivery guy, taken from the episode looking at how food moves across the country.

This wicked image shows Domino's Pizza's national supply chain.

Here's the trailer for the Food episode, which I'll say is your must-see video of the week:
The full episode is available for free viewing online, at least for U.S. residents, here. (UK readers, you should be able to catch the program through Sky. Everyone else outside the U.S., hit the link and let us know in the comments if the player is allowed in your country.)





Comments
Dominos in NYC, really?
Tested the link in France, and it's working over here
Video works fine here in South Korea.
I love how it looks. It's really cool to see everything in the ccontext of location and mapping. The Job losses map is a little scary though. It's interesting to see the states with the largest job losses are the progressive/democrat states; California, Illinois, Ohio, Pennsylvania, New York, etc. I just hope the entire country doesn't go in the dirrection they're sending their states and citizens.
Worked fine here in Sweden.
We have the same here in the Netherlands. If you like a little different view, more based on the UK series, you should check out Nederland van Boven (Netherlands for above). I find the graphics and narration of the US series to be a bit of sometimes.
But still data visualisation on all of the series is brilliant and gives a totally differant view on the country.
Daniel Huffman has written a critique of the errors (and poor design choices) in these visualizations:
http://cartastrophe.wordpress.com/2012/06/22/beautiful-errors/
These are complex visualizations, and Daniel himself made some errors in his first interpretation of the images that he later corrected.
In response to Jack's comment above that "the largest job losses are the progressive/democrat states: California, Illinois, Ohio, Pennsylvania, New York, etc." I will point out that Jack is misinterpreting the image (following the misleading statement of the original author of this post that the "pinpricks of light represent U.S. job losses"). According to Daniel Huffman's critique that I cited above: "This is actually a map showing the distribution of manufacturing jobs in the 1990s. In the video, the dots wink out -- turn black -- to show the decline of the manufacturing sector. Seen at about 12 min, 50 sec in the video..." So the pinpricks of light represent manufacturing jobs more than a decade ago, not recent job losses. Moreover, since the dots represent the aggregated quantity of jobs per state that have been evenly distributed over the area of each state, a state with larger surface area will seem dimmer (less luminous) than a smaller state with an equal number of jobs. (The distribution of dots does not represent actual geographic distribution, which would be more heavily concentrated in urban areas and not evenly distributed over rural areas and even wilderness areas where human population is very low!) Hence Texas may have had nearly the same number of manufacturing jobs as Ohio in the 1990s, but Ohio appears brighter (more luminous) because its pinpricks of light are evenly distributed over a smaller area.
Finally, to prove a correlation between aggregate number of manufacturing jobs per state and aggregate political party affiliation of each state's voters, as Jack does in his comment above, would require a great deal of investigation and analysis and cannot be assumed in such a facile way.