Re: Geoclue 2.1.8

[Peter Robinson <pbrobinson gmail com>]

On Wed, Apr 16, 2014 at 2:43 PM, Zeeshan Ali (Khattak)
<zeeshanak gnome org> wrote:
> On Wed, Apr 16, 2014 at 9:43 AM, Peter Robinson <pbrobinson gmail com> wrote:
> > > - Switch to Mozilla Location Service for geoip as well.
> > > - Allow specifying nickname to submit network data with.
> > > - Report actual available accuracy level and keep it up2date.
> > > - Add 'Active' property to indicate client state.
> > > - Expose altitude info (if known).
> > > - Timeout on inactivity, rather than no clients.
> > > - Automatically:
> > >   - stop the client if max accuracy level is set to 0.
> > >   - start the client if max accuracy level is set to non-0 and it was
> > >     automatically stopped.
> > > - Don't disable modem after using it. Disabling the modem, disables it for
> > >   everyone so if there was an active connection going on through the modem,
> > >   geoclue will cause it to disconnect. While we need to have some centralized
> > >   means to enable/disable modem and ideally geoclue shouldn't do any
> > >   enabling/disabling of modems at all, for now lets just only enable the
> > >   modem.
> >
> > Maybe I'm missing the context here but shouldn't both the
> > enable/disable and use of the modem be dictated by RF kill switch
>
> Whats that? :)

Basically the same as the airplane mode (Radio Frequency Kill switch).
Generally done by the HW kill switch like on most laptops but
sometimes is implemented in SW, ultimately if the RF Kill is triggered
in the kernel you shouldn't override it.

> > airplane mode as well?
>
> Yeah, it actually should.Thanks for pointing out.

Yea, we should, basically it shouldn't override those sorts of user
decisions :-)

Peter