Bella consults

I visited my friend Namib. He lives at the Epupa falls. They are in the North of Namibia, on the border to Angola. At day he tends goats. In the evenings he thinks about visualization.

Bella, symbols are for worse, you say. That is why we create them as analog as possible. You cannot misinterpret:

Our design is pure, organic and as close to the object as possible. Left: marks for road workers. Middle: a signal, that this Himba village does not welcome visitors. Right: sign for a tire repair station.

We adhere to your Bella Reporting Standards and label directly, without decoration:

Understanding outplays beauty:

Sincerely, your Namib

German newspaper “Die Welt” doesn’t know about tickers. Correct: information passes by your eyes. Very comfortable. Especially, if information otherwise had not enough room for display.

Wrong: The reader passes by the information. Perhaps you know a dumb dog. Who can pass the newspaper by your eyes.

Source: Die Welt, 2008–08–01, p. 21, click for a longer version

Here another broad jump for your eyes.

At November 18th I received this email. Unasked. They cannot blame Excel’s diagram wizard for that. It likes to chop axes. But not always. And not with this data.

Legends are mostly unnecessary. Instead: right to the data. Look, here it’s for Dollar and percentage signs. And for scale infos.

Source: Wall Street Journal, 2006–11–21, p. C4

Also unnecessary: repetitions. Once a Dollar sign and once a percentage sign is enough.

You can reduce skillfully: 2000, 02, 03 etc.

Reference lines may actually help. If they visually increase the relative deviations. That’s what they do in the middle graph.

On November 18th msn news asks me: „How good do you score in the PISA-Test?“ I click and answer questions on caries, evolution and earth rotation. My brain hurts. Poor students.

After wind turbines and marble splints in vinegar: question 6 on Windows, question 7 on Word. At the end antibiotics and dancing bees. 2 out of 9 questions on Microsoft. Peesa in Amerieca? Dear Microsoft, did you cheat?

Question 6 in the German msn PISA test: Does PISA ask for Windows?

A German one without Microsoft here. I got 16.

Obama times: Yes, we draw. Bars as long as they just are. No problem at all in Obama country:

The Wall Street Journal paints them over the whole page (click for complete view).

In the German newspaper Die Welt, flashes hit the charts (left: original, right: me). The flash makes the chart go absurd. A table would be better. On the right I drew a correct one.

Wall Street Journal, 2006–11–21, p. C4; Die Welt, 2008–08–13, p. 26.

Symbols are difficult. In most cases they don’t work out. Some of them have to be learnt. It takes longer than reading something you already know – e.g. a word.

Awkward symbols make me sad. The arrow is at the beautiful river Saone, in beautiful Burgundy:

Source: me

Should you follow the arrow?
Do you have to follow the arrow?
Are you allowed to, if you are careful?

And here from the magazine “Der Spiegel” – oh boy:

Source: Der Spiegel 29/2008, 2008–07–21, p. 65

Oh boy, because: Americans are rigorous with everything concerning their flag.
For instance: whoever wants to be president needs to wear it on the revers.
Always.
I don’t think that the editors of Der Spiegel are still allowed to enter the US.
Even my flag looks best non-shrinked and non-stretched.

And that:

Source: Wirtschaftswoche 27, 2008–06–30, p. 104

From now on: whenever we see three stars we know they will be part of the stock index DAX, soon.
One means: probably not.
Says the German magazine “Wirtschaftswoche”.
I am not sure.
Will that be accepted?

I believe for most symbols the rule is:
Entrance prohibited.

Today I have to give advice to Barack. He takes a dog to the White House. He doesn’t know yet which. Best choice: a Labrador.

Barney, the Terrier of George W., had its own website. The most intelligent thing he said so far: „Bark, bark, bark. Woof, woof. Rowf. Arf.“

Well: no Terrier.

Labradors have been successfully in charge of the White House before: Bill Clinton’s dog Buddy chased away Socks the cat, ate a reporter’s donuts and threw over the President. Buddy was run over later. His successor, Seamus: a Labrador.

Buddy saves the Oval Office.

Labradors like to go for a walk. Barack saves the world until next spring. Then he’s got enough time.

The Wall Street Journal (2008–10–23) does it right. Nothing chopped off. The Royal Bank of Scotland in free fall. Nearly 14 %. You see it. The disaster as long as the column wide.

In the German newspaper Süddeutsche Zeitung (SZ, 2008–10–23) 29 out of 50 bars for the Stoxx index end trembling in haze.

The reason: To spare nervous investors a lot of stress the SZ cuts everything above 5 %. Consequently, the 15.25 % lost by Repsol, too. Soothing. But nonsense.

Dear SZ, now that you are using beautiful graphical tables, why don’t you use them correctly?

Intention of a data graph: review values. Register differences. Identify patterns. With this here only reviewing values is possible. You don’t see differences. You don’t see patterns.

Source: Welt am Sonntag, No. 36, 2008–09–07, p.39

I have to calculate differences myself: all (”gesamt”) minus each value. 42 times.
And transpose all. Because the types belong into columns.
Patterns are best identified with graphs. Therefore: bars.

Reading newspapers can be exhausting.

We have got a sense for numbers. Naturally. Or acquired. Doesn’t matter. Anyway: it brings space and numbers together. Linearly and proportionally. Intuitively. Automatically. Because of that my law of proportions: Graphical changes have to be proportional to the changes in values depicted by a graph. This sense for numbers can be verified. With an experiment. I have done it. With the people in the office here. Everyone had to draw an image. That’s what I asked them:

“Take a piece of paper. No matter what paper. Draw a straight line on it. As long as your hand is wide. Approximately. Not too short. Not too long. Write zero on one end. Write 100 on the other. Draw marks for 25, 50, 70, 80 on your line.”

That’s what they drew:

The people did quite an accurate job. Here are the deviations from an exact scale:

Consequently, in every graph: Everything must be proportional to the depicted values. Column lengths, for instance. Without proportion is manipulation.

Right: all data within eye’s sight.

Wrong: Data as broad as eye’s sight. As on the finance pages of the German newspaper Die ZEIT:

Die ZEIT No. 25, 2008–06–12, p. 30.

Such few values fit onto less newspaper. Below each other and you could have compared them. Sorted descendingly would be great, too. Then it would be brain jogging and not broad jump for my eyes. Just like that:

More than 70 percent of German citizens are satisfied with Angie. With Guido Westerwelle 39. With Oskar Lafontaine 19. Helmut of Focus magazine likes Angie. A lot. That’s why Focus calculates 71 divided by 19 as 6.7 and not 3.7. The 19 of Oskar is 1 cm from the bottom line. If drawn correctly, Angie’s 71 percent would have been at 3.7 cm of the scale. They are not. They are at 6.7 cm. Exaggeration factor 1.8. Nearly double wrong. If the paper is too short, you can scale down. You must not chop.

Source: Focus 20/2008, May 10th, 2008, p. 16.

I prefer that over Sudoku: Take any newspaper and estimate the exaggeration, then measure it.

The labels are odd, too. Guido is not labeled. Kurt is labeled Guido. Erwin is labeled Oskar. Oskar is labeled Erwin. (Hover over the portraits to see who’s who.)

I am going to ask Focus what they think they are doing. Dear Focus, what do you think you are doing?

Again. My law of proportion. The idea of a chart is: Display proportions between values with proportions of length. Proportional. Proportional. Proportional. You can ignore it. You can also lick out an empty bowl. If you are a rather dumb dog.

I yowled on cheating grids for time axis already. They manipulate. This one is even more elaborate. At first glance you think: a nice man, admits he hasn’t got the data. Wrong. He says: “I tease you but I admit it”. Remove 2001. It suggests that 2002, 2003, 2004, 2005 have been the same as 2001. Were they, really?

Source: Euro 06/07, p. 42.

This is Howard’s rule „Ignore the visual metaphor altogether” in action. The data doesn’t fit on paper but I show it. A graph doesn’t fit on your paper? Use a table.

Source: Wirtschaftswoche 27, 2008–06–30, p. 82.

I like „Die Süddeutsche“. Its the first newspaper with graphic tables in the stock market section. And now this. The most interesting information – the outliers, the hotshots and the losers, that what you need to know: hidden. I’ll be on vacation.

Süddeutsche Zeitung, 2008–07–16.

Everyone starts into their vacation. We take a map with us. Maps show wherefrom, whereto, how long, how far, how mountainous, next to what, where can you swim, sleep, refuel. Where Lothar Mathäus is playing football. The Tour de France doesn’t go in circles. Where Robert, Nick, Nico, Timo and Sebastian drive at 320km/h. Where new oil is found. A classic: Netanya? New Caledonia? Where was it?

Source: WAMS no. 27, 2008–07–06, p. 19 and 77.

Impossible without a map: where to where? Chic: the „when“ is there, too.

Source: WAMS no. 27, 2008–07–06, p. 21.

As many elements as variables, perfect.

Source: Motorsport Total

As soon as numbers are in, its difficult. Area and value don’t get along well. It’s merely bearable. We had this issue before. Chic, but superficial. How many states? How many electoral votes?

Source: WAMS 23, 2008–06–08, p. 2.

My Tipp: Show where in the map. Show how much in a table. That’s safe. Like that. Old, new, where, who else. You compare and analyze. Nice.

Source: Die ZEIT 25, 2008–06–12, p. 21.

Under no circumstances: decoration in a map. It ruins everything. Until I recognized what the yellow things meant. Oh dear! Cornfields? Deserts? No. The symbol for radio activity. Nuclear desasters? Contaminated areas? No. Just a little bit of deco …

Source: Die ZEIT 29, 2008–07–10, p. 6.