OT: keyboard layouts

[Peter Hiscocks]

Since this comes up from time to time (like most religious wars ;), there
might be a web reference to the tests that Henry is referring to,
done in 1950s by the US government's General Services Administration. Has
anyone ever seen such a thing?

> The Dvorak typists never did substantially better than the Qwerty group.

> Similar tests, on a smaller scale, have been done by modern ergonomics
> researchers.  Same uninspiring results -- at most, a small advantage. 

Any reference for this?  I can get most papers, however obscure, through our
inter-library loan service at Ryerson.

One criterion the GSA would apply to this test would be 'is it worthwhile to
change our training materials and our typewriters to obtain this marginal
advantage'? (I assume we're talking typewriters rather than computer
keyboards in the 50's). So although there might be some benefit, it was not
deemed worth the costs. For someone learning a keyboard on their own, the
cost-benefit might be different.

Further, I would assume this performance metric is 'speed'. It would be
interesting to test for 'effort', but that's a lot more difficult to
quantify. The other metric would be 'ease of learning', which would be
relatively easy to measure.

The nice thing about this issue in 2003 is that people are relatively free
to choose. That wasn't true in the era of manual typewriters, when
manufacturers generally produced only the Querty keyboard.

Peter


On Sat, Dec 13, 2003 at 01:03:05AM -0500, Henry Spencer wrote:
> On Sat, 13 Dec 2003, Anton Markov wrote:
> > I think the best test for Dvorak vs. Qwerty would be to have two people 
> > who have never used a keyboard (yes, there are such people) of about the 
> > same age, background, command of the english language, etc. to learn 
> > touch-typing (in say 1 month) and have periodic speed contests 
> > throughout the period.  That would be a real un-bias (if there is such a 
> > thing) test.
> 
> More or less this test was done, with careful controls, in the 1950s by
> the US government's General Services Administration.  They retrained
> existing typists rather than picking keyboard-naive people, but they were
> careful to keep it an equal competition -- for example, when the Dvorak
> typists got up to their previous speed level, all further training they
> got was matched by extra training for the non-retrained Qwerty control
> group.  (Training and practice will improve typing speed, so you have to
> make sure both groups get similar levels of it.  And you have to control
> carefully for individual differences in ability, which is one reason why
> it's convenient to choose people whose performance as typists is already
> somewhat calibrated.  This test was generally much more carefully and
> fairly run than Dvorak's badly-slanted tests.)
> 
> The Dvorak typists never did substantially better than the Qwerty group.
> 
> That test largely ended interest in the Dvorak layout.
> 
> Similar tests, on a smaller scale, have been done by modern ergonomics
> researchers.  Same uninspiring results -- at most, a small advantage.
> 
> Folks, this is not some unexplored mystery; such careful, fair trials have
> been done a number of times in the last 50 years.  The only reason the
> issue keeps coming up is that True Believers in the Dvorak layout insist
> that all those negative results must be wrong.
> 
> By the way, although the early typewriter layouts were not done with the
> benefit of modern human-factors knowledge, there *were* quite a variety of
> them -- Qwerty was far from the only choice -- and organized speed
> competitions were common and got serious attention from customers.  One
> reason why Qwerty survived is that it consistently did fairly well in
> those.  It *is* a reasonably good layout, if far from perfect.
> 
> > Any other keyboard layouts out there?
> 
> The Maltron keyboard really is what the Dvorak keyboard pretends to be --
> a new keyboard layout designed for electronic keyboards and based on
> modern ergonomics.  It is, alas, hard to find and expensive; it's mostly
> marketed to people with RSI and similar medical problems. 
> 
>                                                           Henry Spencer
>                                                        henry-lqW1N6Cllo0sV2N9l4h3zg at public.gmane.org
> 
> --
> The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
> TLUG requests: Linux topics, No HTML, wrap text below 80 columns
> How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

-- 
Peter D. Hiscocks                         	   
Department of Electrical and Computer Engineering    
Ryerson University,                    
350 Victoria Street,
Toronto, Ontario, M5B 2K3, Canada

Phone:   (416) 979-5000 Ext 6109
Fax:     (416) 979-5280
Email:   phiscock-g851W1bGYuGnS0EtXVNi6w at public.gmane.org
URL:     http://www.ee.ryerson.ca/~phiscock
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml