On Fri, 2012-05-11 at 10:41 +0300, Pekka Paalanen wrote:
> On Thu, 10 May 2012 09:45:44 -0700
> Christopher James Halse Rogers <christopher.halse.rog...@canonical.com>
> wrote:
> 
> > This brings up something that I've been wondering: what (if anything) is
> > the long-term plan for input in Weston - and Wayland compositors in
> > general?
> > 
> > Is Weston going to essentially fold in all the interesting bits from all
> > the input DDXs to the core (and require this to be duplicated in all
> > other compositors), or should these conceptually be modules, eventually
> > crystallising as an input-module ABI compositors could share?
> > 
> > I think I'd like the latter to happen, but what are others thinking
> > about this (if anything ☺)?
> 
> There is also one approach no-one mentioned here yet.
> 
> All compositors that will be interesting to the user, that are session
> compositors, could have a Wayland backend only. Therefore they
> would not care about the hardware driving at all.
> 
> Session compositors would run on top of a system compositor, that
> interfaces with the kernel input devices, and does the basic
> interpretation into Wayland events: normal and raw input events.

I agree completely; session compositors should just speak wayland to the
system compositor.  I was thinking of system compositors - it's not
clear to me - I need to go through and check - that we'll end up with
Weston being the sole distro system compositor, and in that case it
would be useful for system compositors to be able to share that input
code.

If Weston is suitable to be a system compositor for everyone who cares
about generic hardware then I care much less ☺.

> Also this approach won't fit for all Wayland uses, but I expect it to
> become a dominant one on the desktop and similar high-power systems,
> especially ones that may have more than one real user account.

Right.  Phones and tablets may well not want to bother with multiple
compositors and just go the one.  They also tend to be environments
carefully composed to work on just that hardware, so it's easier.

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
wayland-devel mailing list
wayland-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/wayland-devel

Reply via email to