NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/comp/misc/4797

Internet Message Format | 1993-01-07 | 3.6 KB

Path: sparky!uunet!elroy.jpl.nasa.gov!usc!rpi!usenet.coe.montana.edu!news.u.washington.edu!news.u.washington.edu!news
From: [e-mail] (Lloyd P. Goldwasser)
Newsgroups: comp.misc
Subject: Re: Really better? (Re: DVORAK keyboard?)
Date: 7 Jan 1993 20:06:12 GMT
Organization: University of Washington
Lines: 78
Message-ID: <[e-mail]>
References: <[e-mail]>
NNTP-Posting-Host: zooid.zoology.washington.edu

In article <[e-mail]> [e-mail]
(Carl Johnson) writes:

> I remember an article that concluded the advantage of dvorak keyboards
> was greatly overstated.  Some studies have stated that dvorak has a 2-3
> times advantage based on finger movement.  This article mentioned that
> they all neglected to figure the motion of the finger pressing the key,
> which is the same in either case.  I believe that they concluded that
> dvorak had about a 15% advantage still; not enough for a touch typist to
> change over.


I'm not sure how you can quantify the way that the Dvorak layout
feels nicer and easier to use than Qwerty, but it certainly does, and
being able to type with one's fingers relaxed counts for something.
Regardless, some numbers about letter frequencies may hint at some
differences:

(2 caveats:
  1. The frequencies and rankings of the letter depend on the choice
     of text, and what I cite here is calculated from what I have written.
  2. The actual layout I use is a personal variant of the Dvorak layout,
     which I describe below.)

The two most frequent letters, E and T, account for about 20% of
one's keypresses; in the Dvorak layout, they are under the two middle
fingers in "resting position"; in Qwerty, they both require movement
and create an awkward combination.

Question: of the 9 most frequently occurring letters, how many do
your fingers touch when they are in resting position?
Answer: Dvorak: 7;  Qwerty: 2.
These letters account for more than half of one's keypresses, and if
you can do them without moving a finger away from resting position
then you *have* saved finger movement.  This comparison alone should
make one skeptical of the means by which they calculated that 15%.

These numbers don't address the issue of *combinations* of keys,
which is the real strength of the Dvorak layout: many consecutive
letters lie comfortably under alternate fingers (3rd and 5th; 2nd and
4th) rather than, say, under different positions of the same finger
(e.g., Qwerty's E and D, or T and R).



Here is my variant of the Dvorak layout, which I recommend to present
and potential users:

   '  ,  .  Y  J  F  G  C  R  L
    A  O  E  I  U  D  H  T  N  S
     ;  X  Z  P  K  B  M  W  Q  V

The major differences are my switching of I and U (putting the more
frequently used I under resting position) and the shifting of Y and
P.  Q seems to make more sense on the right hand, since it is always
followed by vowels with the left.



In my experience, switching over from Qwerty to Dvorak was *not*
difficult or confusing.  It took basically an afternoon of
concentration.  Although it took a while to get my speed up, the new
layout felt better almost immediately, and eventually my speed with
Dvorak ended up much faster than that with Qwerty.



Finally, it may be worth noting that on a NeXT, each user can specify
his or her own preferred keyboard layout, which is automatically
invoked upon login, and which applies to *all* typing one does,
within *any* program.  I can send my keymapping file to any NeXT user
who is interested.


Cheers,

Lloyd Goldwasser
[e-mail]