[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Why does the tutorial talk about C-n/C-p etc?

From: Chad Brown
Subject: Re: Why does the tutorial talk about C-n/C-p etc?
Date: Sat, 12 Mar 2016 11:14:38 -0800

There are some extensive UI studies on these topics, and the results
are basically (paraphrased from memory):

  * Using the mouse for nearby positional editing is measurably slower
    than using 1-key or 2-key keyboard commands, for people used to

  * Using arrow keys in a typical IBM 101 or similar layout is *very
    slightly* slower than using home-row keys (I believe both WASD and
    HJKL were compared).

  * Similarly, using 2-key combinations (like emacs’) is *very
    slightly* slower than using home-row keys.

  * Using the mouse for distant positional editing is often faster and
    rarely slower than using the keyboard. The theory I recall is that
    tying scrolling to analog physical movements enables people to use
    the spatial-reckoning hardware in our brains.

  * Interestingly, emacs has seen a recent flowering of new navigation
    modes that are largely based on searching rather than positioning
    (ace-jump, avy, swiper, etc).

  * For many users, using the mouse for positioning *feels* faster or
    as fast as using the keyboard, even if it is not (mostly for
    nearby positioning).

  * Similarly, many users *feel* like moving their hand to the mouse
    for any positioning task is slower than using the keyboard, even
    if it is not (mostly for non-local positioning).

  * Interestingly, showing users the stopwatch doesn't change the feel

  * Whether it is worthwhile or not to break what users feel in favor
    of lower stopwatch numbers depends on your domain. Breaking feel
    creates an irritation point that slows down both editing and
    reading/writing, so it's definitely not free.

The practical upshot of all of this is that there really is no “one
best answer” for whether to use dedicated keys, key-chords, or mouse
for an editing environment; the answer is still "it depends".

It’s been a while since I followed this space, and there’s probably
been some new research since I looked. If memory serves, the sources
for these come from a combination of Apple’s/SIGCHI’s research, Bruce
Tognazinni’s work at Apple and Sun, some stuff from the MIT Media Lab,
and the work on Plan 9’s 8½, rio, and acme user interfaces. Much (but
not all) of it predates the widespread adoption of advanced scroll
wheels and multi-touch scrolling gestures, but those are unlikely to
suggest a stronger insistence on emacs's control-key style of

I hope this helps. Apologies if it should have been sent to emacs-tangents 


reply via email to

[Prev in Thread] Current Thread [Next in Thread]