behaviour of F-keys on XO HS

Tiago Marques tiagomnm at gmail.com
Sat Jul 17 00:39:38 EDT 2010


On Thu, Jul 15, 2010 at 11:50 PM, Daniel Drake <dsd at laptop.org> wrote:

> Hi,
>
> On the XO HS (highschool edition, the one with a more "normal"
> keyboard) we're facing some questions about how the F keys should
> function, under sugar and GNOME.
>
> The technicalities are in http://dev.laptop.org/ticket/10213 and here
> is a picture of the keyboard:
> http://wiki.laptop.org/go/OLPC_Spanish_Non-membrane_Keyboard
>
>
What's up with the layout? The lowercase letters are gone and that was a
fine addition.


> Under non-sugar environments (e.g. GNOME), myself and Paul are in
> agreement that in order to change brightness and volume, you should
> press e.g. Fn+F9 (to decrease brightness).
>
> This matches behaviour of "normal" laptops, including the Dell that
> I'm writing on. Linux already has mechanisms (once through hal, now
> through udev) so that when I press Fn+F8 on my Dell, X receives the
> "volume down" key press (instead of the Fn+F8 key press), matching
> what is printed on the keyboard.
>
>
> We want all of the unmodified F-keys to send the normal F-key events
> (we don't want to map them especially). This is for maximum
> compatibility with non-sugar environments.
>
>
> For Sugar we have an open question.
> The F1-F4 keys have the zoom levels printed on them, and pressing the
> keys unmodified will cause the zoom levels to change (because this is
> how Sugar is already coded, it responds to the literal F1 keypress).
>
> The other keys are:
>  - F5: search/journal
>  - F6: frame
>  - F9: brightness down
>  - F10: brightness up
>  - F11: volume down
>  - F12: volume up
>
> For these other keys, when using Sugar, should the user have to press
> the Fn modifier while pressing the key in order to reach the named
> function?
>
> The advantage of not having to press Fn (i.e. the keys would work
> unmodified) is that sugar works the way it always has on XO (we retain
> consistency with XO-1).
> As for the other option, the advantage of requiring Fn is that we gain
> consistency between Sugar and GNOME, and between the XO HS and
> "normal" laptops (where Fn *is* necessary to reach those alternate
> functions). (but we do end up with some confusion with the zoom level
> keys,which will continue to work unmodified)
>
>
Since on the XO-1 & 1.5 the keys don't have the F numbering print, it seems
to me a good idea to have *all* OLPCs running the modified keys by default,
even in GNOME, and having Fn change it to F keys. This bring some
advantages:

   - XO 1.5 currently has no way of accessing F keys that are mapped for
   functions(at least that I know of), which is a problem.
   - XO 1 would now be able to run other desktop environments and have the F
   keys available
   - Consistent "by design" - although it has become the norm for the Fn key
   to modify to non F keys, it makes more sense to have to press the Fn key to
   get an F key output, just like we press shift to get the alternate keycode.
   - Consistent with Apple and newer laptops

Although this may break usability at first, new users may quickly realize
that while the zoom keys don't do anything, the volume and brightness ones
do and that may suggest them to press Fn to check the F key functions on the
zoom keys when the F1 keycode will expose the GNOME help, suggesting a
normal behavior.

If we decide to make the keys available unmodified in Sugar we would
> have to change sugar (worldwide, not interested in downstream forked
> patches) so that F5 on any system opens the journal, F6 on any system
> opens the frame, etc. How would people feel about that?
>

Ideally if you could check the keyboard at boot and make the changes, it
would be better. Can you?

Best regards,
Tiago


>
> cheers,
> Daniel
> _______________________________________________
> Devel mailing list
> Devel at lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.laptop.org/pipermail/devel/attachments/20100717/b692dd1c/attachment.html>


More information about the Devel mailing list