Re: Keyboard usage on some Gnome windows not working

[Matthew Thomas <mpt myrealbox com>]

On 20 Oct, 2005, at 1:21 PM, Eric Larson wrote:
> On Thu, 2005-10-20 at 08:49 -0200, Matthew Thomas wrote:
> ...
> > I see there is some research showing that the keyboard is faster for
> > common commands <http://ad-astra.ro/research/view_publication.php?
> > publication_id=1508&lang=en>, but that wouldn't include access keys
> > unless you were encountering particular dialogs or alerts very often.
>
> I think this screams for a little personal user testing! I use emacs 
> and I feel as though never having to switch to the mouse is very 
> helpful.

That may well be true. Emacs has keyboard equivalents based around the 
home row of the keyboard, and it has incremental search for positioning 
the caret. Neither of these are standard in GNOME.

> Yet, I have wondered if I do lose productivity by having to remember 
> so many commands and not having as many facilities for refactoring.
>
> So in the interest of usability (and a little fun), I propose trying a
> short hiatus on using short cut keys and see what happens. I will 
> report my findings on my blog (http://ionrock.org).

Unfortunately, that won't be at all useful. To measure the keyboard 
against the mouse usefully you'll need to test multiple people, with a 
real-world task, using a stopwatch, on an app following the GNOME HIGs.

If you test only one person (such as yourself), learning the required 
actions during the first approach (whichever it is) will make the 
second approach faster than it otherwise would have been. If you don't 
test with a real-world task that includes a variety of commands (e.g. 
"add keywords to these photos for each person in them", or "open this 
file and resolve the merge conflicts in it"), the test subjects will be 
able to keep individual keyboard equivalents and/or mouse actions in 
short-term memory, which will be unrealistic. If you don't use a 
stopwatch, you'll undoubtedly conclude that the keyboard is faster, 
just as Tognazzini's test subjects did even when they were wrong 
<http://asktog.com/SunWorldColumns/S02KeyboardVMouse3.html>. And if 
you're testing an app like Emacs that doesn't use the keyboard 
equivalents specified in the GNOME HIGs, there's little point in doing 
it at all, unless you want to change the HIGs, and even then you'd need 
to compare the results with those of an otherwise identical app that 
does use the HIG keybindings.

> One thing to consider is that Tog reported this in 1989! I think it is
> safe to say that users would not anywhere nearly as accustomed to 
> typing as we are today.
> ...

It's even safer to say that people weren't anywhere near as accustomed 
to using a mouse as they are today. Most computers before 1989 (or 
maybe a couple of years earlier) didn't even *have* a mouse.

--
Matthew Paul Thomas
http://mpt.net.nz/