Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uwm.edu!rpi!ssimmons
From: ssimmons@unix.cie.rpi.edu (Stephen Simmons)
Newsgroups: comp.sys.mac.hardware
Subject: Re: Chords (was Re: a plea to Apple -- something for the offhand
Message-ID: <~Y5$FJ*@rpi.edu>
Date: 31 Jul 90 01:25:34 GMT
References: <1396@idunno.Princeton.EDU> <1990Jul26.024139.11905@csrd.uiuc.edu> <10911@spool.cs.wisc.edu>
Organization: CIE, Rensselaer Polytechnic Institute, Troy, NY
Lines: 63

In article <10911@spool.cs.wisc.edu> tonyrich@titanic.cs.wisc.edu (Anthony Rich) writes:
>Anyway, about chord keyboards:
>
>Chords are only more efficient than other ways of accomplishing the same
>thing if you can recall them instantly.  You lose BIG every time you have
>to look one up in an online or offline manual, so they introduce a learning
>curve inefficiency that's based on
>
>    1) How fast you can memorize the chords,
>    2) How many chords there are, and
>    3) How many different meanings the same chord has (in different
>       programs, for example).

Another important factor is what kind of chord keyboard; court
reporters use them for text entry; some people here seem to want to
place the modifier keys on them.  As to #3, you need, basically two
standards: one for modifer keys and one for text entry.  And you would
need some kind of mode chord to switch between the two.

As to #2, in the modifier mode, the answer is clear: one key for each
modifier and chords are common-sense combinations of these modifier
keys.  But there is a more serious problem with the text entry mode.
With five keys, we are limited to 32 chords.  It's still possible to
fit the entire ASCII set into this small set by using mode-chords
(like shift-lock or caps-lock, etc.) and chord sequences.  However,
this type of encoding slows the chordist down which defeats the purpose.

Another possibility is a sixth button on the keyboard (how about the
base of the palm) which would immediately double the number of chords
available.

Another consideration that chord designers need to look at is this:
are they designing a chord keyboard for primary entry (i.e. two hands)
or to use in conjunction with a mouse?  Or both?

Could we have a two-handed chord device with wheels on the bottom --
that is could we combine a keyboard with a mouse? :)

Could we have a two-handed chord device that could be easily switched
into yet another mode that allowed the above one-handed entry?

Yes, I know that the above sounds complicated.

I know that the average user would not want such a device (because you
have to learn 80 or so chords before you can do text entry).

>Another factor in recall speed (after they're "learned") is how frequently
>you USE the chords (and your Mac, for that matter).  Take a long summer
>vacation, and whoops!  Let's see, was that command-click or option-click?

More importantly, how much speed do you gain by having the modifier
keys readily available?  Average user=.001%, power user < 2%?  This
capability seems worth paying for only as an added (i.e. free) feature
of a chord keyboard designed for text entry.

The real problem is, that the device will only be commercially
developed if there are enough people interested in it.  Quite a number
of people seem to have some interest in it here.  Anybody want to take
a net poll?  Something like "how much would you pay for such a
device?"  Anybody want to volunteer to be the pollster?  (i.e., a
manufacturer...)

-- Stephen Simmons