Re: [Usability] "Finish" vs. "Close" in gnome-control-center dialogs

["Elijah Newren" <newren gmail com>]

On 3/23/06, Alan Horkan <horkana maths tcd ie> wrote:
> I dont recall the earlier versions including a Finish button either.

Right.  Depending on what you mean by 'earlier versions', but making
my best guess at it, those versions didn't include a finish button. 
Rodney changed the dialog to not be instant apply during 2.13 in order
to 'fix' problems seen in user testing.  Then he also added explicit
OK & Apply buttons.  Then the backlash and those two things were
reverted so that it became an instant apply dialog again.  Then, as an
alternative fix to the problem, he changed the wording from close to
finish.

> > When dobey mentioned the user testing and its results, it rang a
> > strong bell with me--I had felt the exact same way about that dialog
> > (and a few others) when I first used them.  Personally, I strongly
> > feel that "close" sucks for that dialog and that "finish" is much
> > better.
>
> The button label is a much smaller issue compared to changing the dialog
> to an instant apply dialog, which puts "better" and the results of
> usability testing in a bigger context.

That wasn't really a change...  The dialog was an instant apply dialog
when the testing was done; it was changed to not be one as an attempt
to 'fix' the problem seen, then reverted back to an instant apply
dialog a little under a month later due to the outcry.

> > (Which is why I gave one approval to the UI change despite
> > the inconsistency we knew it would introduce; others objected strongly
> > on the consistency basis, but that's all water under the bridge now.)
>
> I'm a little suprised or should I say disappointed you deliberately broke
> consistency.  (Well tough shit boo hoo, I should get over it, yeah I
> know.)
>
> How consistent can Gnome ever be if it is not something developers
> consider important?

When did you stop beating your wife?  ;-)

To be more clear, how do you jump to the conclusion that it isn't
considered it important?  I disagree with that claim (it's definitely
false for me, in particular).  The problem is that there are often
times multiple important issues in conflict and one must be chosen
(yeah, it'd be nice to get all the important items covered, but it's
not always possible.)  Now, if you want to claim that it isn't always
the #1 priority, then you have a strong claim.  :)

(Also, to prevent or at least correct any possible confusion, let me
note that I'm not a control center maintainer)

> Always come down to not enough time.

Yeah, speaking of which, there's some pretty nasty Metacity bugs that
I need to get around to fixing and a bunch of patches I need to
review...  ;)

Cheers,
Elijah