Future of location services on OM

Ludwig ludwigbrinckmann at gmail.com
Sat Jan 10 12:08:27 CET 2009


>
>
>  When checked Android, they have a few interesting features in the
> positioning area. I think they have their own GPS hardware interfacing
> and interpretation, which is then supplied to the userspace. Also,
> they have a number of different location providers, like the gps
> satellites, GSM cell information, Wifi hotspots.... [3]


I have just been toying with the Android location services on OM. At least
in the Koolu version, the Android GPS interface to GPS hardware as it exists
on the G1 phone with a close-source libgps.so library is replaced by the
qemu code which parses (only two) NMEA sentences, but which are sufficient
to supply location fixes to the Android stack. (The restriction to two NMEA
message is AFAIK because Android does not currently make anything but
location fixes available to Android application code, so there was no need
to parse anything else).

This worked, together with the koolu kernel, out of the box to supply
location information. (Of course Android MapActivities will not work as this
is closed-source, but that is an entirely different issue).

I am in the process of adding parsing of some more NMEA messages to
hardware/libhardware/gps/gps_qemu.cpp, so that additionally satellite status
information is supplied on OM machines. I posted some of that code on the
koolu forum at
http://forum.koolu.org/viewtopic.php?f=10&t=58#p363

I agree that ideally there should be one code-base for parsing NMEA
messages, but then there will be a need to parse whatever intermediate
format there is, to propagate it into the Android stack.
So for any Android work the question is whether there is a better way than
to hijack gps_qemu.cpp for this.

Ludwig
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.openmoko.org/pipermail/devel/attachments/20090110/7d661590/attachment.htm 


More information about the devel mailing list