Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!think.com!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!decwrl!ads.com!potomac.ads.com!jclose
From: jclose@potomac.ads.com (Jeff Close)
Newsgroups: comp.human-factors
Subject: Re: Track ball on a keyboard? vs Eye Trackers
Message-ID: <1991Jun12.193251.13921@potomac.ads.com>
Date: 12 Jun 91 19:32:51 GMT
References: <91163.083144CJS@psuvm.psu.edu> <1991Jun12.142307.210@ux1.cso.uiuc.edu> <1991Jun12.155623.15673@wpi.WPI.EDU>
Organization: Advanced Decision Systems, Arlington, VA
Lines: 70

In article <1991Jun12.155623.15673@wpi.WPI.EDU> dmorin@wpi.WPI.EDU (Duane D Morin) writes:
>In article <1991Jun12.142307.210@ux1.cso.uiuc.edu> klefstad@ux1.cso.uiuc.edu (Sue Klefstad) writes:
>>CJS@psuvm.psu.edu writes:
>>
>>>I understand that users of track balls find them superior to mice.
>>>They are now readily available as mice replacements.
>>
>>>But I would like to see keyboards with track balls built-in.  I'll bet
>>>there are some now or rsn.  Comments?
>>
>>>It seems to me that it would be nice to reach the ball without moving
>>>your fingers from the home row of the keyboard (unlike most cursor keys
>>>and function keys).  I wonder how a location right below the space bar,
>>>reachable by either thumb, would work.  Comments?
>>
>
>As a user of a trackball (more specifically, a "thumb ball") I have to agree
>with an earlier post that says they're not as accurate or as fluid as a 
>mouse.  Then again, my use of the ball is limited to rare occasions when I
>forget a function key sequence in Wordperfect, or whatever.  In cases like 
>that, I can take the time to line up my cursor where I want it to be, since
>Im only dealing in text modes.  I wouldn't wish it on anyone doing anything
>remotely graphic, like CAD.  Although I've heard that it can be used rather
>well in a Windows 3.0 environment.
>
>I've seen keyboards with trackballs built in, but, as you point out, they are
>little more than extensions to the right hand side of the keyboard, past the
>numeric keypad, making an even longer reach for your hand.  (Well, my track
>ball sits on my desk next to my keyboard anyway.  Six of one, half dozen of
>another...)
>
>A location right under the spacebar seems feasible, provided that:
>	1) Its NOT in a location that can easily be bumped by your hands.  
>	   Some applications have auto activating windows or menus that 
>	   simply need to be pointed at to come alive, and a rogue pointer
>	   wouldn't be good here.
>
>	2) It should have some sort of design that makes the use of the thumbs
>	   comfortable and natural.  Look at the placement of your hands on the
>	   keyboard right now.  Move your thumb around a little bit.  Now try
>	   and move it, fairly accurately, in each of the four compass 
>	   directions without getting a cramp or moving the rest of your hand.
>	   The joint simply doesn't want to move like that.
>
>Take the idea of having your hands never leave the home row one step further,
> and equip people with a simple headset that they can don while at the 
> terminal.  Just look at the screen and move the cursor wherever you want it
> to go.  Unfortunately, this is in direct violation of rule (1) that I just
> listed.  How do you know when to turn it on or off?  A hot key sequence?
> That defeats the purpose, somewhat.  It could be argued that a "keyboardist"
> as opposed to a "typist" can manipulate the function keys just as rapidly
> as a traditional typewriter, and therefore could hit F10 or something without
> much problem.  The technology to do this certainly isn't unrealistic - if
> Nintendo can mass produce such a monster, anyone can.
>
>dmorin@wpi.wpi.edu

There were several examples of work with such a device at CHI this year.  One
company, LC Technologies, in Fairfax, VA, makes a relatively (!) cheap
version, and Bob Jacobs at Naval Research Lab gave a presentation on their
research in developing applications using them.  What they've discovered is
that there are certain input "tokens" that really don't lend themselves to
this technique, because of the timing required or the ambiguity involved.
However, they have had success in some cases with tracking the eye movements
and interpreting an appropriately long delay as a desire to shift focus.
-- 
---------------------------------v------------------------------------
InterNet: jclose@ads.com         | VoiceNet: 703-243-1611
ADS, 1500 Wilson Blvd #512       | "Now it's time for something you'll
Arlington, VA 22209              | REALLY like.." - Rocky the Squirrel