How David McCandless makes beautiful visualizations that go viral | Hacker News Reader

How David McCandless makes beautiful visualizations that go viral

vox.com

28 pointsgk17 comments

The information is Beautiful visualizations usually frustrate me.

They often hide the data's insights inside a fireworks-show of other design choices. So, they wind up being abstract art that may, upon close study, reveal some fragments of information. (The Tuftean "data-ink ratio" runs very low.)

Though they're obviously quite popular, I prefer more-usable comprehension-aids, that draw attention to underlying trends and relationships.

I think you're right to be frustrated. It feels like the main motivation for creating the infographics is to fill the books, not to communicate previously opaque information. He's not a very convincing person: https://www.youtube.com/watch?v=RnpJQyDXjQ4

I agree, I think one of the take-aways from Tufte is that everything you do to a visualization should have meaning.

Don't just change the colors of different objects to make them pretty, use colors to reveal another dimension of data.

I think more dangerously, and something that Tufte also warns about, is using visualizations to skew the data towards a biased point, and all the clutter in infographics seems to be more towards this end than actually discovering and showing patterns in data, and then using your visualization to inform the viewer of those patterns.

Infographics are so edited and so commercialized, you'll notice many of them aren't even available in readable resolutions, but are helpfully for sale as $29.99 posters on the author's webpage.

You know what sells? Shocking or contrary items. So why wouldn't an infographic author skew what they're showing? Infographics almost seem like an advertisement selling themselves and not sources of information.

I'm almost at the point of thinking that infographics are what not to do in infoviz.

edit on the other hand, here's an infographic series he likes

http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0...

I loved the rhetological fallacies one... (direct link)

https://cdn2.vox-cdn.com/thumbor/_vHiY0Rp2T_SPwJMvJdK94awg0s...

There is something powerful about David's work. I remember stumbling onto it some time ago and thinking "there is a way of using this kind of communication to make information so much more accessible". Projects like the Google Public Data explorer [1] are a step in this direction but by no means are the final solution.

I find that when I am researching something I have to spend so much time wading through people's attempted (and sometimes good) prose in order to find the conclusion I need.

In an era of overwhelming data I feel like we should have a way to make common data much more accessible and David's work (despite it's potential flaws in regards to empiricism) seems to be a step in the right direction.

[1] http://www.google.com/publicdata/directory

Just noticed one of the infographics mention a pacemaker has on average 0.5-1 million LOC. Does this look right?

I think you're reading the chart wrong, it's listed as just short of 100k. (As read from the top scale. - It's not a very clear infographic at all, xkcd ones using area squares tend to be a lot more clear!)

Reply on news.ycombinator.com