Re: [Usability] Control Center Appearance Capplet

[Thomas Wood <thos gnome org>]

On 19/04/07 13:57, Calum Benson wrote:
> On Wed, 2007-04-18 at 19:21 +0100, Thomas Wood wrote:
>
> > Maybe we get some of these things with the new Xrandr? I haven't been 
> > following it too closely. The problem I can see here though, is that 
> > Display may not be an obvious place to look for when wanting to set font 
> > options.
>
> I agree, especially now that we're all used to looking in the Fonts
> capplet for them.
>
> The trick if we did that would probably be to re-cast those settings as
> something affecting the overall 'display quality', rather than something
> specific to fonts.  For that matter, it would probably be worth
> revisiting whether we really need to offer such an exuberant choice of
> font rendering settings anyway; Mac and Windows seem to manage fine with
> much less.

Definitely. I've never used the Details window of the font capplet. 
Ideally I think we'd have a single checkbox marked "Enable font 
smoothing", but unfortunately I don't think there is a way to detect LCD 
screens yet, so we ought to have at least three choices.

I wonder if there is a reliable way of determining what percentage of 
people actually use these settings.

> > > Well, IIRC, the sort of concerns raised at the time were:
> > >
> > > - Without the Close button, a blind user has to infer that the dialog is
> > > instant apply by the absence of any action buttons.  The presence of the
> > > Close button gives them a more positive indication.
> > I think this goes against what the Novell studies show, and what I 
> > suggested in the beginning. Does a Close button really indicate explicit 
> > apply? I would suggest that for the many people who don't realise the 
> > difference between Close and Cancel, it really only serves to confuse 
> > matters further.
>
> That may be true for sighted users, but I don't think Novell tested with
> any users who were relying on screenreaders for their information (I'd
> be delighted to hear otherwise, though), so I'm not sure it would be
> wise to draw any conclusions about their needs from that particular
> study.

The point I was trying to make should apply equally to partially sighted 
users as to sighted ones. The problems associated with the Escape key 
and the instant apply confusion can be traced to the confusion between 
Close with Cancel. I suspect this is largely because many people are 
used to the explicit apply model in Windows, and just do not understand 
that the controls in preference windows are instant apply. If there was 
a better way of demonstrating to users that the settings are instant 
apply, I think it would solve a lot of our problems.

However, I don't have any data to back up my hypothesis about this, so I 
am not suggesting we change the status quo unless just yet.

Regards,

Thomas