On 4/2/07, David Zeuthen <[EMAIL PROTECTED]> wrote:
On Mon, 2007-04-02 at 23:17 +0100, Richard Hughes wrote:
> On Mon, 2007-04-02 at 18:10 -0400, David Zeuthen wrote:
> > I agree, and I'd even say that gpm and others shouldn't
>
> Shouldn't or should?
Shouldn't....
> > start providing the SM interface as it may be a bit more complicated
> > than the proposed Shutdown()/Reboot()/Logout() methods (e.g. Lubos
> > Lunak's comments earlier today).
... because of the reasons above.
> Jon, what have you spec'ed out so far with GDM?
But gdm can never provide an interface on the D-Bus session bus in a
desktop session. And it shouldn't - gdm is highly sensitive code so we
want as few attack vectors as possible. However, gdm can (and already
does) provide a mechanism that e.g. gnome-session (which should provide
org.fd.SessionManagement) can use. Hope this clarifies.
Right. The interface that Dan and I were suggesting would be on the
system bus and essentially just a nicer replacement for the current
socket protocol. And... offtopic for this thread - sorry.
Jon
_______________________________________________
xdg mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/xdg
