Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!henry
From: henry@utzoo.UUCP (Henry Spencer)
Newsgroups: net.database
Subject: Re: a word to the wise
Message-ID: <5894@utzoo.UUCP>
Date: Wed, 21-Aug-85 15:05:03 EDT
Article-I.D.: utzoo.5894
Posted: Wed Aug 21 15:05:03 1985
Date-Received: Wed, 21-Aug-85 15:05:03 EDT
References: <578@bu-cs.UUCP>, <29771@lanl.ARPA>
Organization: U of Toronto Zoology
Lines: 52

> >                          Now the layout of the typewriter keyboard was
> >designed to facilitate the even use of both hands.
> 
> Just a little note Edward to remind you that the typewriter keyboard was
> "designed" to slow typist down so they wouldn't jam the keys together.
> The first typewriters used gravity to return the keys and any typist who
> typed too fast had to have the boss disassemble the platten and unjam
> the keys.

Sigh, this old misconception comes up again.  It's *almost* true, but not
quite.  The keyboard was designed to put frequently-used keys far apart,
because fast typing tended to jam the machine if *adjacent* keys were hit
in fast succession.  This does *NOT* necessarily slow the typist down;
there is some evidence that it speeds things up, by promoting alternation
between hands.

> ...  Back in the 40's when
> Devorak was attempting to persuade the military to specify DSK keyboards,
> the government held contests.  After the Devorak typists broke 300 words
> per minute and it became obvious that the qwerty typists weren't going
> to break 110 wpm ( at that time ), the government discontinued the
> contests because "they don't prove anything".

It is worth remembering that these were *manual* typewriters.  One of the
biggest warts of the qwerty keyboard is that it loads the little fingers
much too heavily.  But this is much more serious for manual typewriters
than for electronic keyboards.

> Anton Devorak did a lot of research and proved that good touch typists can
> retrain to ~double their speed in < six weeks...

On *manual* typewriters.  More skeptical people, with more modern keyboards,
have been unable to replicate these results.  I am told that no unbiased
test with modern keyboards has shown more than a 10% improvement.

Note also that retraining for a Dvorak (there's no e in the name) keyboard
ruins your qwerty touch-typing; there is a severe compatibility issue here.

> The Devorak keyboard was designed with the comfort of the typist in mind.

No, it was designed to be as good as it could within the severe physical
constraints imposed by mechanical typewriters.  If you put your hands
loosely on the keyboard, you will notice that they come in at about 45
degree angles from each side.  This is a rather poor match to horizontal
rows of keys.  There has been some work with radically-redesigned keyboards
with a little cluster of keys for each hand, set at a better angle.  There
are claims of significantly better results this way.

This does not have very much to do with databases, by the way.
-- 
				Henry Spencer @ U of Toronto Zoology
				{allegra,ihnp4,linus,decvax}!utzoo!henry