As we get access to more and more data, we need more, and more interesting, ways of looking at it--and I keep tripping over some of these interesting visualizations. Sometimes they're even worth using!
- One of the nice things about online bookstores like Amazon is that you can find just exactly what you want, instantly. Of course, that eliminates part of the charm of going into a bricks-and-mortar shop and discovering something you weren't expecting. So, consider Zoomii, which is a more real-world like bookstore experience. Working like the love-child of Amazon and Google Maps, you browse books visually, zooming in on "shelves" and looking at book covers. As a front-end, you can buy the books you find on Amazon; Zoomii makes their money through the Amazon Associates program.
- Want to explore the Linux kernel? Try the Interactive Linux Kernel Map. It's a lot like the Zoomii (or Google Maps, for that matter): scroll-wheel to zoom in and out, click and drag to move around, and click to open up source code. Not as slick, perhaps, as other visualizations, but certainly useful if you're interested in understanding how the Linux kernel works.
- It's one thing to remap huge databases into user-friendly interfaces; it's another to demonstrate relationships among different entities. That's what these Maps of science show: how different scientific fields are related to each other.
- Not exactly data visualization, but it's a neat application, Asirra (Animal Species Image Recognition for Restricting Access), from Microsoft Research, is a friendlier and possibly easier alternative to CAPTCHA for sorting out bots from humans. The idea is to require completion of a task that's pretty easy for humans, but almost impossible for computers: differentiate between photos of cats and dogs. I played with it for a while, and it seems as if you don't have to get all the cats perfectly, but as long as you only miss one (I think) and don't misidentify any dogs as cats, you can pass for human.
Posted July 7, 2008 6:00 AM
Permalink | No Comments |



