new development, release plans?
jku at linux.intel.com
Thu Feb 18 03:51:45 PST 2010
I've pushed the ofono-using gsmloc provider, the new nominatim geocoder
and some provider fixes to jku-branch. The gsmloc provider is still
fairly experimental (my modem doesn't work 100%, preventing proper
testing), but in my opinion it's already far more useful than the
previous one. Nominatim is the best geocode/rev-geocode source we have,
I think. If anyone objects to the gsmloc changes or has other comments,
please say so.
In other news I'd really like to get this project moving again. If you
have anything in mind that should be done before a new release, let me
know -- I do intend to go through bugzilla but I might have forgotten
things discussed elsewhere... So please comment if there's something we
should have done already.
I'll start: I propose adding freeform address input to Geocode interface
Rob Bradford asked about this, and it's something I've been thinking as
well. Many address sources are freeform: think user inputting an
address, twitter location, etc. Currently the best you can do is fill
one of the existing fields with the freeform data. That usually works,
but can fail in some cases...
I see two better options:
1. add another method:
<arg name="address" type="s" direction="in" />
<arg name="fields" type="i" direction="out" />
<arg name="latitude" type="d" direction="out" />
<arg name="longitude" type="d" direction="out" />
<arg name="altitude" type="d" direction="out" />
<arg name="accuracy" type="(idd)" direction="out" />
2. add a "freeform" field into current AddressToPosition(): if that
field is present, do geocoding based on that field only.
I don't see much benefit in option #1 so the second one seems more
attractive because it's less work. Opinions on the idea in general or
the different options?
More information about the GeoClue