Re: the keyboard accessibility capplet

[Havoc Pennington <hp redhat com>]

Jody Goldberg <jody gnome org> writes:
> On Wed, Sep 25, 2002 at 08:14:06PM -0400, Havoc Pennington wrote:
> > 
> > Presumably this refers to the global enable/disable toggle. But that
> > doesn't help anyone; because enabling it the first time doesn't _do_
> > anything, as all the sub-checks are disabled. The global toggle here
> > just slows things down.
> 
> It does do something.  It enables key handling to enable disable the
> other features in ways that do not require the capplet.

A nicer approach to that might be:
"[  ] Pressing <whatever the key sequence is> enables bounce keys"

I think Windows just has this always enabled, and chooses sort of
obscure key sequences so no one will do it accidentally - I could be
wrong.

> > The global toggle isn't even clearly associated with the
> > group of things it affects.
> It used to be in a frame containing the other features which if I
> recall you previously objected to.

I should have been objecting to the double-nested enable/disable
groups, rather than the layout, if I wasn't.

> > How about losing some of the complexity from the dialog so you have
> > room to spell it out. ;-)
> which of the complexity.

Anything. My argument is that we have requirement #1: less
complexity. Something else should give.

The fact that it's hard to choose what needs to give just means UI is
hard. It doesn't mean we get to avoid choosing.

> - Remove the sliders ?  Seems like a bad idea in that most of us
>   prefer sliders to spinners.

But almost no one prefers _both_. That's the key point.

> - Remove the global button ?  See above.

No rationale given so far outweighs the complexity of the overall
dialog IMO.

> Spinners are useful for some users and sliders for others.  This is
> not a 1 size fits all area.

1 size works at least adequately for all. 2 sizes at once doesn't make
sense.
 
> Seems reasonable.  Do we have consensus on this one ?  If so I'll
> make the change.  It will involve schema and string changes, so this
> would be a 2.2 change.

Yes, I'm talking about 2.2 for all of this, to be clear.

Havoc