The Beauty of Bones (and Skinput)

I've always been a fan of anything that uses the concept of bone conduction. A friend who worked as a field medic for public protests years ago told me that he'd often diagnose and locate bone fractures by taking a tuning fork, striking it, and holding it against the limb in question--the sound would travel up and down the bone and cause a stronger 'sensation' (ouch!) wherever there was any sort of a break in continuity. Now, thanks to research being done at Carnegie Mellon and Microsoft, you can use this same basic technology to play tetris!

The video has a more in-depth demonstration, but the idea is based on the fact that our bodies are pretty effective conductors of minute acoustical information, so vibrations from something like a tap on the forearm or fingertips can be picked up by a bio-acoustic sensor positioned somewhere else along the arm. Because every part of the body is composed of specific combinations of different kinds of tissues with various densities, every location hypothetically has a signature resonance that can be tracked.

While it's still in development, they're already teaming the technology up with wearable pico projectors. I think it's really interesting for the future of AR, in terms of creating the ultimate ephemeral user-interface--Tablets are SO 2010.

12

  1. Reminds me of the short lived “Bone Fone” AM/FM radio that you wore like a scarf (or yoke) and that played music by conducting it into your body via . . . some method I never understood. In fact I think BB had a post about it a few years ago.

  2. But if, as it says, “every part of the body is composed of specific combinations of different kinds of tissues with various densities, every location hypothetically has a signature resonance that can be tracked”–wouldn’t these resonances change if I gained weight, or even got dehydrated one day?

    I guess it wouldn’t be too hard to re-calibrate, but I wonder if the fidelity needed to discriminate two spots on my arm an inch or so apart would be a fidelity finer than the range of temporal variation in a single location.

    1. I suspect that’s a predictable enough effect that two or three taps would suffice to recalibrate, if recalibration is needed. Or, depending on what kind of resolution you’re trying to achieve, it might just be a matter of designing large enough buttons — which is one of the factors that already goes into robust touchscreen UI design.

      I like it. It’s not exactly high-res, but for simple devices with limited controls/menus it makes sense to me. Might be possible to extend it by tapping the monitored arm on things rather than vice versa — or maybe snap your fingers to open/close the interaction?

      The question is going to be whether it can be made comfortable enough and cheap enough and unobtrusive enough. And I suspect their prototype would have trouble with long-sleeved shirts, so it may wind up being a special-purpose tool. But more modalities is generally a Good Thing.

  3. Yes, also reminded me of the Bone Fone. Unfortunately for the Bone Fone makers the Walkman came out a few short months later.

  4. Huh, I read this about two weeks ago and now find it on Boing Boing. It used to be the other way around…

Comments are closed.