[PATCH wayland] protocol: Add set_xwayland request for new shell surface type

[Pekka Paalanen]

On Thu, 13 Dec 2012 15:10:25 -0800
Bill Spitzak <spitzak at gmail.com> wrote:

> Tiago Vignatti wrote:
> 
> > About the availability of it to regular Wayland clients, I agree, it's a 
> > problem. Bill Spitzak pointed this out on the previous set. It sets a 
> > bad example because it exposes global position to all regular clients 
> > and we don't want this.
> 
> Actually I prefer this. I think if you put it into some xwayland api, 
> then all that is going to happen is that clients are going to use the 
> xwayland api, rather than the shell api. So moving it to X makes client 
> behavior worse, not better.

I don't think so. Arbitrary clients cannot get to the xserver nor wm
Wayland interfaces, because we can make them privileged, just like the
screenshooter interface and desktop-shell private interface are.

So when we move the problematic part to the xserver or wm interfaces,
no ordinary Wayland clients can access it, since they cannot bind to
the global objects.

> You have to realize that any commercial multi-platform software is going 
> to get this information, no matter how hard it is, because it is 
> difficult to imagine any scheme were adding #ifdefs and a new 
> implementation to software that positions windows correctly on X, 
> Windows, and OS/X, is a smaller amount of work. The fact that the X api 
> needs this information means it is there, and the hackers are going to 
> get it, no matter how much you wish they did not.

We're not giving up on this battle yet.


Thanks,
pq