Re: Enter the build sheriff: Jacob.
- From: Gregory Merchan <merchan phys lsu edu>
- To: gnome-hackers gnome org, desktop-devel-list gnome org
- Subject: Re: Enter the build sheriff: Jacob.
- Date: Thu, 14 Mar 2002 19:16:34 -0600
On Thu, Mar 14, 2002 at 03:37:59PM +0100, Liam R. E. Quin wrote:
> On Thu, 2002-03-14 at 18:08, Gregory Merchan wrote:
> > Last night, as part of a favor, I had to use MS Word on WindowsME.
> [description of pressing tab too many times]
> You can press shift-tab to go backwards, of course.
> So if you overshoot, no big deal. Although not everyone knows this,
> and not everyone knows you can press tab either.
I know and that's what I did, but awkwardly. It's still an unnecessary delay
of about 2 seconds by GOMS.
MPT is Mental Preparation Time
n is the number of tabs required to get to the target.
MPT + n * Tab + activation
= 1.35s + n * 0.2s + 0.2s
MPT + (n + 1) * Tab + MPT + Shift + Tab + activation
= 1.35s + (n + 1) * 0.2s + 1.35s + 0.2s + 0.2s + 0.2s
(The second MPT is here because you're basically starting anew.)
The difference is:
3 * 0.2s + 1.35s
If the default button were OK and clearly indicated, then the time
for the operation would be about 1.55s. (MPT + Enter) The combination
of bad UI factors more than doubles the time required to complete
I'm by no means an expert at GOMS analysis and this is probably
abuse or misapplication of it, but even without numbers it should
be obvious that you need time to recognize and correct the error
as well as take the extra keystrokes. Someone with a copy of
Jef Raskin's The Humane Interface can probably provide a correct
analysis. (There are other sources, but I don't know them.)
] [Thread Prev