This picture has been doing the rounds, and it’s certainly an interesting explanation of the Economist’s house style. I particularly like the idea of a red rectangle as a way of keeping the chart identified as coming from the Economist, even when it moves onto Twitter, Tumblr, etc. But, why on earth would you want to “move the chart as far into the background as possible”? There’s an inferiority complex, here, with respect to “the surrounding article”, which is unhealthy.
Most interesting, however, is the squib at the bottom, saying that the chart “has been created for educational purposes only” and “has not been created by The Economist”. Which means that this chart shows what a *designer* thinks an Economist chart should look like. As opposed to a data visualization professional, who understands numbers and how to convey information using charts. Read More…
… and a “warm welcome”? You have approx 3 seconds to suck me in. There’s no time for cuddling in visualizing data.
Engineer & applied physicists, Jean-luc Doumon, spinning in his grave. Just kidding, he’s alive and well, selling a fantastic (necessary/expensive) book for anyone interested in visualizing & communicating data…or, for those of us who are punished by eyesore presentations. …& that’s everyone, innit?
John’s picture page, though well-meaning, is terrible. There certainly are universal rules for charting. Here’s one: include all the info you need to communicate the data, but no more. Which really means, you need very, very little if you know what you’re doing. Title should be the headline (not an attempt at wit). What does the graph tell you? There’s the title. He has Google and Apple, twice..wastin’ my time. Always bin up the x&y, and truncate even if you do have the space, which you don’t & no one likes clutter. Yes, that background advice is BS, and if that noisy space between the x dashes and the year are representative of that year (it looks sloppy), I mean, look at all that’s going on in there. I’d either ask a more appropriate question (perhaps the one offered by felixsalmon), or a better by month/by year metric, or a different visual all together. Also, he’s loosing an opportunity to tell us more on the left y. Maybe the capped price could run along there. Let the data speak, translate it clearly while making it effortless to read without mimicking what we are used to seeing… because it’s probably wrong. Long story longer: Trees, Maps and Theorems.