3 ms·

So, with the voice UI, eye-tracking, and lots of subtle Soli gestures (https://atap.google.com/soli/ https://atap.google.com/soli/), think I could work for 20-3

So, with the voice UI, eye-tracking, and lots of subtle Soli gestures (https://atap.google.com/soli/ https://atap.google.com/soli/), think I could work for 20-30 minutes? -->"But, IIRC, people whose job it is to talk all day have been shown to have a higher rate of vocal cord cancer." Because nowhere did I say how long I wanted to work. For some reason you're taking your lack of imagination and trying to extrapolate it into an all day affair where I'm holding up my arms. Btw, I wouldn't mind that holographic display either but I'm willing to settle for an 8k display for now. After 17 years, and zero "Minority Report" gestures, I've learned that these things take time.

teddyh 8y ago
> think I could work for 20-30 minutes? Barely, maybe. Gesturing with your arms for that long is hard work. I don’t know what the practical limits would be, but even lecturers who hold long speeches with lots of gestures rest their arms on the podium for a reason, and can continue speaking while resting their arms. > you're taking your lack of imagination and trying to extrapolate it into an all day affair On the contrary, you’re taking my writing “for any length of time”, and somehow assuming I meant “all day”. Which I didn’t mean. I only used the phrase “all day” in an unrelated aside about voice interfaces, and only then while referring to a study which (IIRC) specifically was about people who talked all day. But this was unrelated to gesture interfaces, which is why I did not follow this line of inquiry. My point is that gesture interfaces are very hard work to use for longer periods of time (I’m unsure of the exact number), and if they can only be used for shorter periods, are they worth the hassle to learn? If you have work you need to do which takes longer than your arms can handle, you need an alternate interface anyway, (presumably some keyboard/mouse variation which have proven themselves to be reasonably effective) and if you already have that, why do you want to have to learn the gesture one? This assumes, of course, that no gesture interface can solve the gorilla arm problem. But there might be one. Case in point: iPads were successful in avoiding the gorilla arm problem even though they are touch interfaces which the “gorilla arm” phrase originally refers to, and iPads avoid the problem by being small enough that people can use one while holding their arms in their lap – essentially while mostly resting their arms. There might be some solution to gesture interfaces which I’ve not yet seen. > your lack of imagination That’s twice you’ve now insulted me without provocation. Are you upset about the fact that I question the premise on which gesture-based interfaces are based?
- melling 8y ago
  I’m only moving my hands. I can still move my hands while resting my arms. I think you should watch the Soli video again. I’m getting tired of repeating myself. No affiliation with Google or Soli. I keep repeating it because you seem to to be hung up with the extended arm motion and “gorilla arm” MIT has done some gesture UI work. Don’t know much about it: http://www.pranavmistry.com/projects/sixthsense/ http://www.pranavmistry.com/projects/sixthsense/
  - teddyh 8y ago
    From what I could see from the Soli page, all the video showed people with arms in the air not resting on anything. OK, there’s one quick shot of a resting arm at 3min 36sec on the longer video. But all this is irrelevant to the original issue, which is specifically about large-scale arm-based movements as exemplified by Minority Report. I can’t see how those, specifically, can ever be practical. > I’m getting tired of repeating myself. Repeating what? I asked you upthread if you had seen any gesture-based interfaces which avoided the gorilla arm problem, and you did not answer. > No affiliation with Google or Soli. (Yes, I can see that. I apologize for asking, implying you might have been, when I could easily have looked it up. I deleted the question from my text.) > I keep repeating it because you seem to to be hung up with the extended arm motion and “gorilla arm” Well, you’re the one who specifically wanted “a Minority Report interface”. Which I can’t see ever being practical. Google Soli, perhaps, if used in the small, like you describe, with armrests. Maybe. But not Minority Report or Jarvis. > MIT has done some gesture UI work From what I can tell, that work looks like it’s all 10 years old. Also, from what I have seen in the past, all efforts like those are essentially about creating nifty-looking models for interaction (with stunning visuals and huge wow-factors), and not ergonomic interactions made to be used for any length of time. Indeed, the entire “gorilla arm” jargon term was entirely about how people simply forgot about how different something is if you’re using it for doing real work for 4 hours as opposed to doing a 5 minute demo (or, as it were, a two-minute movie scene).
    - melling 8y ago
      No, I didn't specifically want "Minority Report". I clarified that immediately in my first followup. And it's ok if you don't see the value in a gesture UI. I'm not trying to convince you. I'm not really sure why you feel the repeated need to tell me I don't need one.

teddyh 8y ago

> think I could work for 20-30 minutes? Barely, maybe. Gesturing with your arms for that long is hard work. I don’t know what the practical limits would be, but even lecturers who hold long speeches with lots of gestures rest their arms on the podium for a reason, and can continue speaking while resting their arms. > you're taking your lack of imagination and trying to extrapolate it into an all day affair On the contrary, you’re taking my writing “for any length of time”, and somehow assuming I meant “all day”. Which I didn’t mean. I only used the phrase “all day” in an unrelated aside about voice interfaces, and only then while referring to a study which (IIRC) specifically was about people who talked all day. But this was unrelated to gesture interfaces, which is why I did not follow this line of inquiry. My point is that gesture interfaces are very hard work to use for longer periods of time (I’m unsure of the exact number), and if they can only be used for shorter periods, are they worth the hassle to learn? If you have work you need to do which takes longer than your arms can handle, you need an alternate interface anyway, (presumably some keyboard/mouse variation which have proven themselves to be reasonably effective) and if you already have that, why do you want to have to learn the gesture one? This assumes, of course, that no gesture interface can solve the gorilla arm problem. But there might be one. Case in point: iPads were successful in avoiding the gorilla arm problem even though they are touch interfaces which the “gorilla arm” phrase originally refers to, and iPads avoid the problem by being small enough that people can use one while holding their arms in their lap – essentially while mostly resting their arms. There might be some solution to gesture interfaces which I’ve not yet seen. > your lack of imagination That’s twice you’ve now insulted me without provocation. Are you upset about the fact that I question the premise on which gesture-based interfaces are based?

melling 8y ago
I’m only moving my hands. I can still move my hands while resting my arms. I think you should watch the Soli video again. I’m getting tired of repeating myself. No affiliation with Google or Soli. I keep repeating it because you seem to to be hung up with the extended arm motion and “gorilla arm” MIT has done some gesture UI work. Don’t know much about it: http://www.pranavmistry.com/projects/sixthsense/ http://www.pranavmistry.com/projects/sixthsense/
- teddyh 8y ago
  From what I could see from the Soli page, all the video showed people with arms in the air not resting on anything. OK, there’s one quick shot of a resting arm at 3min 36sec on the longer video. But all this is irrelevant to the original issue, which is specifically about large-scale arm-based movements as exemplified by Minority Report. I can’t see how those, specifically, can ever be practical. > I’m getting tired of repeating myself. Repeating what? I asked you upthread if you had seen any gesture-based interfaces which avoided the gorilla arm problem, and you did not answer. > No affiliation with Google or Soli. (Yes, I can see that. I apologize for asking, implying you might have been, when I could easily have looked it up. I deleted the question from my text.) > I keep repeating it because you seem to to be hung up with the extended arm motion and “gorilla arm” Well, you’re the one who specifically wanted “a Minority Report interface”. Which I can’t see ever being practical. Google Soli, perhaps, if used in the small, like you describe, with armrests. Maybe. But not Minority Report or Jarvis. > MIT has done some gesture UI work From what I can tell, that work looks like it’s all 10 years old. Also, from what I have seen in the past, all efforts like those are essentially about creating nifty-looking models for interaction (with stunning visuals and huge wow-factors), and not ergonomic interactions made to be used for any length of time. Indeed, the entire “gorilla arm” jargon term was entirely about how people simply forgot about how different something is if you’re using it for doing real work for 4 hours as opposed to doing a 5 minute demo (or, as it were, a two-minute movie scene).
  - melling 8y ago
    No, I didn't specifically want "Minority Report". I clarified that immediately in my first followup. And it's ok if you don't see the value in a gesture UI. I'm not trying to convince you. I'm not really sure why you feel the repeated need to tell me I don't need one.