Aucb.114
fa.editor-p
utzoo!decvax!ucbvax!C70:editor-p
Sat Dec  5 13:17:37 1981
experiments
>From JWALKER@BBNA Sat Dec  5 13:15:42 1981
   [ This should be fairly easy to measure, though of course will vary
     with keyboard and training of the typist.  David Rumelhart, at San
     Diego, or Moran at Xerox, may have some data.  Does anyone volunteer
     to do the experiments if they do not?  (ADMIN.JQJ@SU-SCORE) ]
This might be "fairly easy to measure" in the intrumentation
sense.  However, it is not fairly easy to do the experiment in
such a way that the results can be given a useful interpretation.
I would urge anyone who gets interested in doing this kind of an
experiment (who is not already a psychologist) to PLEASE talk to
a psychologist about the design of the experiments.

    I'd be interested in seeing some data on average time to type
    <esc> <ctrl><char> vs. <char1> <char2> vs. <ctrl><char1>
    <ctrl><char2>

I agree with Ellis' intuition that control-meta-d is more like
one and a third keystrokes than two or three.  In my own
experience, C-X C-x is faster than ESC x (note: for EMACS on
standard keyboards, ESC x is the same command as ESC C-x).  This
I found after moving my EMACS metizer from ESC to a control key.
I don't have measurements to prove it was actually faster but it
feels faster, therefore seems easier.

(Or should that be "feels easier, therefore is faster"?  One
would have to do careful research before claiming that things
that are actually faster always feel easier subjectively.)