word clouds as visualisations
Jacob Harris of the New York Times offers a useful critique of word clouds, which as he says are often used indiscriminately, providing little insight, or worse, making the actual data less understandable. His examples are excellent. Look at this meaningless wordcloud from Fast Company made from the Wikileak war logs:
— and compare it to the NY Times’ (of course) visualisation of the same data:
Of course, it’s easy to argue that a genre sucks by showing a terrible example of that genre and an excellent example of whatever genre you prefer. One of the commenters to the article, Hardy Leung, linked to his blog which consists of nothing but word clouds, most of which most certainly “tell a story”, as Harris insists that data visualisation should do. Here, for instance, is his word cloud for Ada Lovelace day last week:
Obviously a word cloud like this is far more carefully (and time-consumingly) crafted than a standard wordle.com cloud. (Though you can make your own, too – I haven’t tried that yet.)
I do like Harris’ emphasis on story, though. He writes that “visualization is reporting, with many of the same elements that would make a traditional story effective: a narrative that pares away extraneous information to find a story in the data; context to help the reader understand the basics of the subject; interviewing the data to find its flaws and be sure of our conclusions.”
Do you have really good examples of the use of word clouds?