4 ms·

At first I thought this was just a gimmick without any real-world usage, the last several demos in the video changed my mind. Having a button raise itself unti

At first I thought this was just a gimmick without any real-world usage, the last several demos in the video changed my mind. Having a button raise itself until pressed would be a great feature for something like a car touchscreen. The argument I always hear against automotive touchscreens is that you can't feel them without looking at the screen (and thereby taking your eyes off the road). However, with this screen it should be possible to adjust the fan speed, or turn on the seat heater, without even looking at the screen.

neuralRiot 3y ago
The main feature of physical buttons is that they’re always there. The fan speed knob on my car is always on the same position and does always the same thing, adjust the fan speed. Now if a “magic button” is sometimes there, sometimes not, sometimes adjusts the fan speed and others the temperature (land rover I’m looking at you) the user will have to visually check what is doing.
- blincoln 3y ago
  It could still be a good way to differentiate between different UI elements, especially when UI style skews toward the "is this an integrative element or not?" end of the spectrum, as it has currently. Are there any phone/tablet display technologies that are flexible enough to work with something like this and not wear out almost immediately? The general concept isn't new, although this is the first implementation I've seen. It seems like the big challenge would be combining it with a working display that would last long enough to be useful. I'd always imagined a MEMS device where each pixel was on an independent pillar that could be raised or lowered (magnetically, pneumatically, etc.). Basically the original vision of the Optimus keyboard combined with the heightmap table display from the first X-Men film. It would be trickier to get an element there to feel like a single button instead of pushing your finger into one of those "box of pins" desk toys, but not impossible.
- Bjartr 3y ago
  > the user will have to visually check what is doing. In theory at least, you could have an indicator for every configuration in a fixed position of the screen. The screen's name in braille would work. Then you could at least check without looking and you could build muscle memory for the sequence of inputs. Not ideal, but it could help.
  - throwthrowuknow 3y ago
    The display could also have a time out function where it resets to a beginning state if you don’t use it for a while. That way you learn to use the different options as a sequence that begins from the same configuration.
ncr100 3y ago
Making touch screens operate like mechanical controls seems like it would need to satisfy hurdles. PROBLEMS: - sensing the "amount" of input, based upon tactile response. Touchscreens respond to varying amounts of pressure, sometime responding if a finger-tip hovers 1mm. Touchscreens may respond only after 100ms of contact. Touchscreens, in my experience, respond the same regardless of contact pressure. - providing tactile evidence describing the dimensions of the control. Touchscreens are flat and smooth. Touchscreens lack bumps and ridges to outline the non-interactive bezel areas, or the semi-interactive boundaries between a GUI component and its external margin. So, I think it would be nice for puffy bubbles to SOLVE by: - depress (un-puff / flatten) or change texture, after a fixed touch-duration, when activated - provide an outer "ring" to guide the finger to the inner "sensitive" area

neuralRiot 3y ago

The main feature of physical buttons is that they’re always there. The fan speed knob on my car is always on the same position and does always the same thing, adjust the fan speed. Now if a “magic button” is sometimes there, sometimes not, sometimes adjusts the fan speed and others the temperature (land rover I’m looking at you) the user will have to visually check what is doing.

blincoln 3y ago
It could still be a good way to differentiate between different UI elements, especially when UI style skews toward the "is this an integrative element or not?" end of the spectrum, as it has currently. Are there any phone/tablet display technologies that are flexible enough to work with something like this and not wear out almost immediately? The general concept isn't new, although this is the first implementation I've seen. It seems like the big challenge would be combining it with a working display that would last long enough to be useful. I'd always imagined a MEMS device where each pixel was on an independent pillar that could be raised or lowered (magnetically, pneumatically, etc.). Basically the original vision of the Optimus keyboard combined with the heightmap table display from the first X-Men film. It would be trickier to get an element there to feel like a single button instead of pushing your finger into one of those "box of pins" desk toys, but not impossible.
Bjartr 3y ago
> the user will have to visually check what is doing. In theory at least, you could have an indicator for every configuration in a fixed position of the screen. The screen's name in braille would work. Then you could at least check without looking and you could build muscle memory for the sequence of inputs. Not ideal, but it could help.
- throwthrowuknow 3y ago
  The display could also have a time out function where it resets to a beginning state if you don’t use it for a while. That way you learn to use the different options as a sequence that begins from the same configuration.

ncr100 3y ago

Making touch screens operate like mechanical controls seems like it would need to satisfy hurdles. PROBLEMS: - sensing the "amount" of input, based upon tactile response. Touchscreens respond to varying amounts of pressure, sometime responding if a finger-tip hovers 1mm. Touchscreens may respond only after 100ms of contact. Touchscreens, in my experience, respond the same regardless of contact pressure. - providing tactile evidence describing the dimensions of the control. Touchscreens are flat and smooth. Touchscreens lack bumps and ridges to outline the non-interactive bezel areas, or the semi-interactive boundaries between a GUI component and its external margin. So, I think it would be nice for puffy bubbles to SOLVE by: - depress (un-puff / flatten) or change texture, after a fixed touch-duration, when activated - provide an outer "ring" to guide the finger to the inner "sensitive" area