I've been trying, unsuccessfully, to get my hands on a demo model of Lytro's new and purportedly revolutionary light-field camera. This is a camera whose sensor records the direction of incoming light in addition to its intensity and color, allowing the camera to "see" over a broad focal depth. Users take a single picture and focus after the fact.
Not clear? Here's an insect-themed Lytrograph (is that a word?) by Adam Gould. Click on different parts of the image to refocus:
Call me a luddite, but I don't think Lytro will upend photography as we know it. I find the Lytro is more a solution in search of a problem than an upheaval of imaging technology. I don't wish to downplay the technological achievement- I am duly impressed that Lytro has been able to build, inexpensively, a truly novel way to take pictures. Refocusing still images is indeed fun. More than fun, even. I'd happily use Lytro's images to teach principles of focus and composition.
The fundamental problem for a light-field revolution, though, is that focus is not the limiting factor for most photographers. Fixing focus issues doesn't add much to a photographer's tool kit. Light sensitivity, magnification, dynamic range, stabilization, resolution- these are all areas where technological improvements solve immediate hurdles for different genres of photography. Focusing? Not so much.
There is, however, at least one area of science photography where Lytro has great potential: wildlife camera-trapping. That's the technique where a camera is placed out with some sort of trigger that trips the shutter when an animal walks by. Camera traps are useful for gathering data about wildlife with minimal disturbance.
Since the trapper is not present to compose an ideal shot, a camera that focuses everywhere at once might be rather useful. With a light-field camera, the trap can be set much more broadly than a traditional trap, covering a larger area of forest and casting a wider net. Pictures can be refocused after the fact to bring out the subject's relevant detail.
If Lytro ever does send me a demo, I hope they don't mind if I set it out in the woods overnight.