Re: Review of accelerator changes

[Tim Janik <timj gtk org>]

On Sun, 18 Nov 2001, Bill Haneman wrote:

> Havoc Pennington wrote:
> > 
> > Owen Taylor <otaylor redhat com> writes:
> > > Should we kick the question of what the default should be to the
> > > useability project? since I don't think we'll really get anywhere
> > > with the:
> > >
> > >  "It's confusing for users"
> > >
> > > vs.
> > >
> > >  "It's a cool feature that everybody who has been using GTK+ since
> > >   before 1.0 wants"
> > >
> > > argument. Just much not data there.
> > 
> > I can remember at least a few mails to user lists where users thought
> > this was a bug, and at least a few mails to developer lists where
> > developers wanted to turn it off due to user complaints/confusion. So
> > there is at least some data. Whether it's conclusive is another
> > issue. ;-)
> 
> Really, really we should turn off on-the-fly editing of keybinding stuff
> by default.  It's too easy to do this by accident, and for blind users
> it's mind-bogglingly confusing.

i don't see how blind users can be used as argument for the
general case here. they need a bunch of atk code run in the
first place to get to use GUIs, so you can also turn off accel
changes there, similar to how you're going to tweak other
code portions accordingly.

also, changing accels is not so easy if they need to be combined
with modifiers by default, and accidental changes can be reverted
just as easy as the wee made.

>  Unexpected for everybody the first time
> it happens, I'd dare say.

well, pretty much _everything_ new is going to be unexpected, it wouldn't
be something new other wise (if you come from a different toolkit), that
doesn't autmatically make it bad though.

> 
> -Bill
> 

---
ciaoTJ