Can someone please stop the infographic madness? | Hacker News Reader

Can someone please stop the infographic madness?

om.co

48 pointstaylorbuley15 comments

Even the people who are supposedly masters at this type of data presentation mess it up big time: http://blog.jgc.org/2009/11/how-to-fail-at-data-visualizatio...

> He uses circles with dots in the middle to represent the size. So is the value being plotted proportional to the radius of these circles or the area? Not clear

It's worse than that. Is the size of the circle proportional to the number of incidents, or to the density of incidents? Large circles look like they should the most 'dangerous', but that would imply a high density, which would imply a small circle.

The entire visual representation serves no purpose except to contradict the data it presents.

He spent a year creating over 200 visualisations. So, 4.5 graphics a week, for a 46 week year, makes 207 graphics.

I'm not convinced that people can gather suitable data, analyse it, turn it into an attractive graphic, and then check the graphic for statistical or numerical or logical or whatnot errors.

I find there are roughly two clusters of infographics people.

One side is the statisticians who are adept at handling large data and making reasonable inference from it. Their primary goal is to tease out a story from the data. Unfortunately, they often don't care if it's ugly or impenetrable so long as it's "correct" or at least strongly defensible.

The other side are the designers. Their primary goal is to explore the attractive nature of patterns in data and use those patterns to tell people an interesting story. Unfortunately, if the data doesn't seem to fit the story they want to tell then they're likely to fall prey to confirmation bias and other errors of detail that can completely change or pervert the support for their story.

Neither side is successful in isolation.

Somewhat parallel to what I see as the two legitimate uses of infographics (with no claim that there are only two):

1. presenting/supporting analytical results

2. providing the viewer/reader with a way to explore data for new conclusions (cf Tufte)

The first one is easier. E.g. I've identified that there were successive waves of support, in the Republican primaries, for Bachmann, Cain, Gingrich, etc. and I can show a chart drawn from the Nate Silver do support this - http://fivethirtyeight.blogs.nytimes.com/2011/12/25/reads-re...

The chart both shows and supports the conclusion. Another "type 1" infographic in a poster/narrative style (via Tufte himself): http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0...

The second use is more tricky. The famous "Napoleon's campaign against Russia" chart does it well - http://upload.wikimedia.org/wikipedia/commons/2/29/Minard.pn...

You can explore lots of different points. (And it works in the first sense as well - hitting you between the eyes with how many French soldiers died).

My test for the first purpose is "what's the story?" If it's just "ooh, look at the pretty", the infographic has failed. And then of course you verify that the story is presented fairly.

My test for the second purpose is whether the infographic rewards more examination, and, even more so, whether I can find patterns or links that the author/graphic artist might now have anticipated. E.g. in the Minard/Napoleon graphic, you could check how losses en route to Moscow correlate with losses on the way back.

Given how often infographics fail both these tests, I'd say we're in an infographic bubble. Can't wait for it to burst.

I agree with all you've said here. Infographics is fashionable today in the same way that every hour there's a new article on how Big Data Will Change Everything. As an article of fashion, it's easy to see that people have a real desire to understand their world.

So the trick is just to prove that you're actually better at it than everyone else. Tufte has done this, I think. In some circles.

Very generally speaking, any successful idea accumulates adulteration over time as it is overused and over-exploited. The number of high quality instances do not decrease but have to be sifted out from a larger group. If the author hasn't realized this yet, he need not look far for numerous examples in writing, music, movies, startups; appealing to make a u-turn on the "madness" is futile.

The problem with ending the "madness" is that you need to convince the web spammers that create these solely for increasing pagehits to stop doing so. That doesn't seem likely to be successful as long as it is still a workable strategy.

I got two solicitations yesterday to place infographics on some sites I own. Here's the pitch:

+++++++++++++++++

I came across __________ while searching for resources around education and quality and was wondering if this is the correct contact in regards to the content on the site. My team just created a graphic on these topics, would you be interested in taking a look? I'd appreciate any feedback.

Thanks in advance.

+++++++++++++++++

My site ranks highly for several keywords related to online education, which is why I got this pitch. The other pitch came from someone who saw that I once posted a gag infographic on another site.

I stopped looking at Infographic as they've become cheesy these days. The purpose of an Infographic is to simply data consumption and allow a consumer to get as much data with just a glance or a cursory browse.

Many seem to have taken to the Infographic craze with bizarre and meaningless meandering, that makes things lot more complex than the data being conveyed. Well, there are Infographics where a simple bulleted list would have done a much better justice.

From an online marketing perspective: We'll stop producing them when you stop linking to them.

This is like asking Om not to take money for GigaOM..

I will admit that there are lots of "spammy" infographics running around on the web. A lot of people use them without good reason.

However, I feel that they are a great way way to visualize data and communicate to an impatient audience.

Ending the infographic plague: http://www.theatlantic.com/business/archive/2011/12/ending-t...

I don't understand what you're trying to say... if only you could tell it to me in some kind of visual, easy to understand format...

Well with visual.ly's launch on HN - one of the worst launches I've ever seen by the way - I think it's almost over.

Reply on news.ycombinator.com