info-mac@uw-beaver (info-mac) (12/03/84)
From: TIM%MIT-OZ@MIT-MC.ARPA It is difficult to takes sides in this ongoing keyboard vs. mouse debate, but we all seem to agree that there are sometimes when you want both, but we only have two hands. Has anyone done research on the ergonomics of a foot operated mouse which would roll around on the floor, or combining a keyboard and a mouse in one piece which would slide around on the table? For a foot mouse to work best, it might be desirabled to have a special curved surface to roll the mouse on (perhaps with the dimensions and curvature of organ pedals). Of course if the surface were curved you might want to tie the mice to your feet so they don't slide down hill. While we're at it, how about one mouse for each foot? This way you could easily specify a rectangle by having the left foot specify the upper left corner and the right foot specify the lower right.
info-mac@uw-beaver (info-mac) (12/03/84)
From: John W. Peterson <JW-Peterson@UTAH-20.ARPA> Yes, people have done research on foot mice. This is reprinted from the WorkS list. It's by Doug Engelbart, who invented the hand mouse at SRI. Date: 14-May-84 11:43 PDT From: DCE.TYM@OFFICE-2.ARPA Subject: Foot mouse, knee controller, nose pointer; more research Cc: WBD.TYM@OFFICE-2.ARPA, KIRK.TYM@OFFICE-2.ARPA Re: WORKS Digest V4 #22 At SRI in the 60's we did some experimenting with a foot mouse. I found that it was workable, but my control wasn't very fine and my leg tended to cramp from the unusual posture and task. I assume that these would be overcome eventually by practice. I got to thinking about skill development with fine foot control, and realized that most of us developed a very high degree of fine control with the accelerator pedal. I tried controlling vertical cursor position with such a pedal, and it worked quite well. Thinking about concurrent horizontal control, I realized that I can swing my knee from side to side with fairly good control (in terms of fraction of total range of swing). That worked fairly well, better I found than with the foot mouse. Then I migrated the vertical actuator from the floor pedal to an up-down bar actuated by the knee -- which, I reasoned, was controlled by foot/ankle actions against the immovable floor (to raise or lower the knee) that would be essentially the same as when using a pedal. So we had a compound pair of actuators operated by a knee -- up and down for vertical cursor control, right and left for horizontal. It was natural to learn, and gave better control immediately than did the foot mouse (I believed we called the latter our "skate"), but I tended to get leg cramps from this as well as the vskate. About that time I also rigged up a mechanism that utilized a light-weight helmet for the user to wear: turning his head from side to side would move the cursor horizontally, and nodding the head up and down would move the cursor vertically. This looks a bit strange, but it worked. AND this also gave me cramps, in the neck, after ten minutes or so. I don't think that our experiences would be enough to discard any of these possibilities from a list of research candidates. The cramping I think would pass as skill developed, and the degree of control achievable has to be judged after more practice than we gave ourselves. The hand mouse didn't have these early problems, and gave good results with considerably less practice than these devices seemed to require to get past the cramping stage. We thought about eye control, but at the time didn't want to spend the money to implement it. We had many things, at other levels and domains of an "augmentation system" that I wanted to explore, besides the best means for concurrent screen-selection and character-entry operation. We finally settled for the mouse in one hand and the chord keyset in the other -- and went on with the other system pursuits. I have no particular conviction that the hand-controlled mouse will be the best screen-select control means that will emerge; and I applaud any pursuit of better means. I assume that the end objective is for best over-all, concurrent control & input means. Perhaps a combination of eye pointing, hand controls (including character input), and coded-voice input might be the winner -- or even more parallel motor channels? Any such system would have to be evaluated for a specific working context, though. For instance, what is best would depend upon such things as the level of user training assumed, the responsiveness of the system, how much concurrency can be harnessed in simultaneous task pursuit by the system, the range of media being handled (e.g., text, graphics, color, dynamics, voice, video frames, people-intercom control,) etc.. Doug Engelbart, Tymshare ------------------------------ End of WORKS Digest ******************* ------- -------
dcmartin@ucbvax.ARPA (David C. Martin) (12/11/84)
In article <2398@uw-beaver> info-mac@uw-beaver (info-mac) writes: >From: TIM%MIT-OZ@MIT-MC.ARPA >It is difficult to takes sides in this ongoing keyboard vs. mouse >debate, but we all seem to agree that there are sometimes when you >want both, but we only have two hands. Has anyone done research on >the ergonomics of a foot operated mouse which would roll around on the >floor, or combining a keyboard and a mouse in one piece which would >slide around on the table? > >For a foot mouse to work best, it might be desirabled to have a >special curved surface to roll the mouse on (perhaps with the >dimensions and curvature of organ pedals). Of course if the surface >were curved you might want to tie the mice to your feet so they don't >slide down hill. > >While we're at it, how about one mouse for each foot? This way you >could easily specify a rectangle by having the left foot specify the >upper left corner and the right foot specify the lower right. Xerox has done a significant amount of research based on mice controlled by other extremities. Both feet and knees were considered, but discarded as either too strenous or not exact enough. Now the real trick would be an optically controlled ``mouse''. Airlines use vision tracing systems to lay out cockpits, why not do it for a $2500.00 computer :-).