Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!ux1.cso.uiuc.edu!ux1.cso.uiuc.edu!m.cs.uiuc.edu!p.cs.uiuc.edu!bentrup
From: bentrup@p.cs.uiuc.edu
Newsgroups: comp.cog-eng
Subject: Re: one-finger keyboard
Message-ID: <78600001@p.cs.uiuc.edu>
Date: 17 Oct 89 16:40:15 GMT
References: <221013@<1989Oct6>
Lines: 59
Nf-ID: #R:<1989Oct6:221013:p.cs.uiuc.edu:78600001:000:3164
Nf-From: p.cs.uiuc.edu!bentrup    Oct 16 19:19:00 1989


/* Oct 15, 1989 by adams@math.berkeley.edu in p.cs.uiuc.edu:comp.cog-eng */
>I can envision a couple of applications for a one-fingered keyboard:
>1.  Handicapped people may not have the
>    use of individual fingers, or they may need to type with a stick
>    held in their teeth or attached around their heads, or whatever.
>    They, then, are performing "one-fingered" typing.

I don't believe a universal one-fingered keyboard is the solution.  For
many individuals with a motor impairment, the input device is more passive
than active.  I would consider a morse code keypad an active device (i.e.
the user must repeatedly manipulate the key to communicate).  However I
would view as passive, a scanning device which the user merely indicates a 
word/phrase choice as the cursor automatically moves through a list. (E.g.
Gibler & Childress "Language anticipation with a computer based scanning
communication aid", Proc IEEE Computer Society Workshop on 'Computing to
Aid the Handicapped', 11-16, 1982).

>2.  Hand-held computers / memory aids / phone dialers.  In all of the
>    examples of these I have seen, the size of the device and the size
>    of the keys are such that the user typically holds the device in
>    one hand, and pokes at the keys with the index finger of the other
>    hand.  A particular drawback to these (potentially very useful)
>    pocket-sized devices is that the QWERTY keyboard, while well-known,
>    is not at all suited to the task.  Some of the most common letter
>    combinations require jumping across the entire keyboard.  Also, the
>    size of the QWERTY keyboard is effectively 3 rows of 10 keys, making
>    it difficult to arrange a keyboard on anything that will fit in a
>    pocket.

Scott Minneman did some work while he was at Tufts (?) using a 12 key phone
pad as a fairly effective keyboard.  Some follow up work (which would
provide a pointer to Scott's original work) is reported in Foulds et al,
"Lexical Prediction Techniques Applied to Reduce Motor Requirements for
Augmentative Communication", Proc of the 1987 RESNA Conference.  Basically
they exploited the transitional probabilities for quadgrams (4 letter
clusters) in English to predict what the words should be.  Improvements
could have been made using this in conjunction with a context sensitive 
dictionary.

>3.  I imagine that there would be situations where someone with restricted
>    finger dexterity (in particular, if they're wearing heavy gloves or
>    the like) may need to enter data into a keyboard.  An astronaut on
>    a spacewalk, for example, may need to type something while wearing
>    a spacesuit.

Presuming that the vocabulary in such situations is fairly limited and ease
of communication is essential, how about a palette containing
pictures/stock phrases from which the user could select? 

>          Jeff Adams                 adams@math.berkeley.edu    

John Bentrup
University of Illinois at Urbana-Champaign
Department of Computer Science

        arpa/internet	    bentrup@CS.UIUC.EDU 
        bitnet	            bentrup%uiucdcs.BITNET  
        usenet/UUCP net	    {pur-ee,convex,ihnp4}!uiucdcs!bentrup