The Financial Times has created a giant videographic project in NYC’s Grand Central Station. Check out details about the installation and watch some of the videos (on business and the global economy) at http://ftgraphicworld.ft.com. Has anyone seen it yet in person?
Artist Gary Simpson created a series of frescos in 2006 based on global indicators from the CIA’s factbook. A bit stylized, to say the least, but I applaud the effort. Below are my favorites:
How dare they! Well, actually, it’s a fun exercise. Declared by Tufte to be one of the best statistical graphics ever drawn, Joseph Minard’s graph of Napoleon’s march on Russia is definitely a classic (a copy hangs in my bathroom).
John Boykin recently took a crack at redesigning the classic, and goes into quite a bit of detail on his website about the choices he made:
John links to a series of other re-creations and re-interpretations of Minard’s dataset, as collected by Michael Friendly:
I particularly like this googlemap version:
And then there’s the executive summary version. Bwahahahahaha!
This is an example of why you keep checking back on mediocre data visualization tools. The last time I looked at the OECD’s explorer, it was slow, kinda clunky, and not very innovative. This morning I took another look. Wow! It has interactive choropleth maps, motion scatter plots, profile plots, time graphs, and cool histogram tools – and all of them have excellent filters and fine tuning controls, can be viewed over time, are smoothly animated and you’re allowed to load your own data.
But wait! There’s more! MUCH more! It turns out the explorer is just one tool created by the Swedish National Center for Visual Analytics (NCVA), who have constructed a set of Geovisual Analytics Visualization (GAV) Flash tools, including what you need to create your own statistics explorer. The NCVA also has a spin-off company that sells a desktop version of the explorer, a Flow Map explorer that draws proportionate arrows on maps to track flows, and a multi-dimensional explorer (which I only played with a little – but is very very cool).
Check out the scatter tables in the MDIM as a way to select data in the other two panels:
I’m almost embarrassed I haven’t seen these before. On the other hand, I love that there is such innovation going on – all the time.
The White House released a video of last week’s State of the Union address, with a split screen showing supporting charts, diagrams, and talking points. It’s not a brilliant model of visualization best practices by any stretch of the imagination, but it is a step in the right direction, and represents a recovery from the ridicule that Ross Perot received for using charts on the national political stage in 1992.
For further insight (on both design and economic issues), I HIGHLY RECOMMEND watching this in conjunction with Jodi Beggs critique over at Economists Do it With Models.
An interesting (ok, quick poll: should I stop calling things interesting? I only post stuff I think is interesting – seems a bit redundant, no?) long term look at new technology. At first look, it’s a bit boring and geeky – but the interactive popup descriptions really flesh it out. It would be easy to quibble about where things fall on the timeline, but overall I really enjoyed reading about all the hypothetical tech. Now that we have realized most of the science fiction from the 1900s, it was nice to see that we still have ways to dream and imagine. (via)
Here’s an example of how more isn’t always better. Compare the interactive presentation of retail sales data below to the static version – both from the Wall Street Journal. In my opinion the static version presents the information in a much clearer and usable format. The only thing it’s missing is a chart for “total retail sales”. Regarding the interactive version, I’ve never liked stacked bar charts over time unless they illustrate very clear trends – with this many similar segments I think they are pretty silly (though the ability to drill down does alleviate this a smidgeon). (related article)
This is an interesting example because they really tried to have every part of the graphic convey information: color, size, and line type. (via The Big Picture)
If you look at the 2011 update of the report, you can see they have used a similar, but somewhat cleaner design. The labels are all clearer, and the use of a lighter background map and grey text improves the contrast considerably. Finally, I think using the color of the arrows instead of width for the value was a good call – as that’s easier for the eye to distinguish.
Unfortunately, I don’t know of any software that let’s you produce these quickly – you generally have to draw them by hand.
Based on a survey of 1000 British adults, this interactive display lets you select questions and filter answers in a variety of interesting charts, where each dot represents a person in the survey. The live animations are really cool, though some of the presentations are more effective than others.
Harvard has released an interesting new index of “economic complexity” which is the productive knowledge of the economy, based on analysis of its output composition.
… the Economic Complexity Index (ECI) is based on the number and the complexity of the products that a country exports with comparative advantage. Empirically, countries that do well in this index, given their income level, tend to achieve higher levels of economic growth. The ability to successfully export new products is a reflection of the fact that the country has acquired new productive knowledge that will then open up further opportunities for progress.
The index is then used to make detailed growth projections, and identify export opportunities on a country-by-country basis.
There are also interactive versions of most of these visualizations that you can explore and filter:
Dark sky is an interesting short-term (<60 minutes) weather app. The creators interpolate weather radar information and present it based on your GPS location in the form of a smooth animation, along with a precipitation estimate. So instead of knowing that there’s a 50% chance of rain in your region that afternoon (typical weather forecast), you’d know if there was a very likely chance of rain falling on your head in the next 15 minutes. The app is currently in development, and there’s a Kickstarter campaign if you want to contribute to the cause. There’s also a blog post that explains in more detail how they are using the standard NOAA data.
Eric Fischer analyzed twitter meta data for location and language and ended up with a pretty good approximation of the normal geopolitical borders. He chose grey for English, since they speak that everywhere – it would be interesting to see a map of just English. There’s also a world version. (via FlowingData)
An addictive collection of beautiful charts, graphs, maps, and interactive data visualization toys -- on topics from around the world.