3 ms·
Excellent, we get a “faster horse”. When do we get the future we imagined 20 years ago? https://youtu.be/PJqbivkm0Ms https://youtu.be/PJqbivkm0Ms I thought S
by melling 5y ago
Excellent, we get a “faster horse”.
When do we get the future we imagined 20 years ago?
https://youtu.be/PJqbivkm0Ms https://youtu.be/PJqbivkm0Ms
I thought Soli was a step in the right direction.
https://atap.google.com/soli/ https://atap.google.com/soli/
- bitwize 5y agoJust because it looks cool in a movie does not mean it's a practical interface for everyday use. That Minority Report scene just screams "gorilla arm". I get shoulder cramps looking at it.
- sleepybrett 5y agoIndeed it looks awesome in the characters enormous office where he is alone. Now imagine a software engineer 'pit' at a typical corporation, rows and rows of engineers standing waving their arms to and fro. Hell they'd need more square footage per person, that's never going to fly.
- melling 5y agoYes we’ve had that “gorilla arm” discussion before. That’s why I tried to sidestep that discussion by mentioning Soli, which has subtle movements. When “inventing the future”, feel free to skip what the “movies get wrong “
- bitwize 5y agoIt's really hard to beat the ergonomics of a keyboard for things like writing, coding, and data entry. These are still important applications, they're not really amenable to literal handwaving, and a keyboard presents a dead simple solution. Dismissing improvements on the keyboard as a "faster horse" when gesture interfaces still have enormous issues to work out is silly, especially if you mention Soli -- so groundbreaking, Google left it out of phones after the Pixel 4! And yes, the gorilla arm is but one of these issues. Gestures are vague enough to be challenging to recognize with a higher than acceptable rate of false positives and false negatives. They're also not very discoverable. Certainly there's room for research in this area, but there's also room for making working systems better.
- urtie 5y agoThe problem with space sensing interfaces is one of intentionality. The best example of how this is hard that I can think of is that of a passive IR sensor wired to trigger opening a door. If that door is parallel to a hallway, the door will probably be opened a lot by people that have no intention whatsoever of going through the door. It is that particular problem that is well captured by buttons: by decreasing the actual sensing area you increase the intentionality of the measurement. Furthermore, requiring specific directional movement increases it further. Another example: the capacitive sensing 'buttons' on my washing machines can be accidentally activated by touching with the side of the arm, which can occur for instance when I'm loading detergent. That accidental activation is harder to achieve and easier to detect before accidental activation for physical buttons that have some travel.