This is a beautiful representation of Twitter’s public timeline, similar to twittervision, but with goofy avatars in 3D. It reminds me a lot of the work stamen design has done with Digg, e.g. arc. The visualization shows how many people are active in the Digg community at any given time. Tools like TwittEarth and twittervision do the same for Twitter.
Google recently released the Google Visualization API, based on software written by Hans Rosling. Check out his software in action at TED a few years ago. It’s both phenomenal and convincing, showing the true power of good data visualization.
One of the few Facebook apps that I still use is Friend Wheel, which shows how my network is intertwined and especially the clustered areas of friends.
Aside from being cool eyecandy, these representations give substance to data. Visual representations of data are something we take for granted. Remember the last time you built a slide deck with some data? I’ll bet you’ve used fonts, colors, pie charts, graphs to emphasize the data you had to present.
People have a much easier time relating to abstract concepts when they are represented visually. Frequently, the metadata collected provide an equally or more interesting representation. For example, TwittEarth includes the tweet (data), but what makes it really appealing is the metadata (location mapped to the globe). At a glance, you can tell that the US and Europe are active now, probably due to the time of day.
We’ve been trying to build visualization into Mix and Connect for a while. It’s easier said than done. I do think that a great visualization leaves an impression, e.g. I saw Hans’ presentation months ago, and when Google finally released the Visualization API, I spent a few hours trying to figure out ways to use it. His presentation was made more compelling by the eyecandy.
What do you think makes a good data visualization? What type of visualization could make Mix and/or Connect more compelling or interesting?