Re: [gpsd-users] setting/gettin almanac data trough gpsd

[Eric S. Raymond]

Tomalak Geret'kal <address@hidden>:
> K, then I guess we'd need to know more details about the mechanism
> that his receiver is exposing to allow [almanacs]. As far as I'm aware,
> there's no "standard", other than for those receivers that have
> built-in network comms of some kind that can reach a proper assist
> server.
> 
> As for whether the gpsd bods have been seriously considering this
> anyway, of course I couldn't comment...

I'd certainly like to support something like this. It's fraught with 
difficulties, though.

First, as you note, there's no standard.  We'd have to store a pair of
capabilities for each device - how to pull the almanac/ephemeris from
it, how to push one to it.  I'd be OK with extending the device-driver
structure to carry these.

The larger problem may be that very few devices ID themselves
properly.  It doesn't help gpsd to know that the Foonly GPSomatic 2317
has these push/pull capabilities if the Foonly GPSomatic 2317 behaves
indistinguishably from a dumb generic NMEA device and can't be provoked.
into revealing its identity.
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>