W3C home > Mailing lists > Public > public-geolocation@w3.org > June 2008

Re: GPS recommended minimum

From: Erik Wilde <dret@berkeley.edu>
Date: Thu, 26 Jun 2008 12:27:30 -0700
Message-ID: <4863EDA2.80204@berkeley.edu>
To: public-geolocation@w3.org
CC: Kai Hendry <hendry@aplixcorp.com>

hello kai.

Kai Hendry wrote:
> Most  GPS devices embedded in mobiles like the Nokia N810 output
> something called the Recommended Minimum sentence C ($GPRMC).
> http://www.gpsinformation.org/dale/nmea.htm#RMC
> I think GGA is also very common, though most device drivers filter GGA
> (and other lines) and expose only RMC.
> http://www.gpsinformation.org/dale/nmea.htm#GGA
> RMC is the minimum standard defined by
> http://en.wikipedia.org/wiki/NMEA which a device must support to be
> considered 'a GPS device'.
> http://aprs.gids.nl/nmea/#rmc
> http://answers.yahoo.com/question/index?qid=20080109031033AAkHMFQ

correct me if i am wrong, but all of this is on a slightly lower level 
that we are talking about here, right? it is still relevant, of course, 
because it allows us to better understand what information to expect 
from an average *GPS device*, but in the end, the API developed here 
does not really care whether the location is derived from a GPS, Wi-Fi 
locations, cell phone towers, or a user claiming at being a location by 
simply marking it on a map. all the API must be able to do is define a 
way how this location information can be accessed by scripting code, and 
there must be well-defined semantics for the retrieved values. which i 
think still leaves us with the basic questions:

1) which location types are supported? coordinates are pretty much set 
here, but i think we also should consider URIs to support locations not 
just looking at spatial concepts, but also supporting place concepts.

2) for each of the types defined in 1), there must be well-defined 
semantics. for coordinates, this means the reference system (including 
the question whether there should be support for more than one reference 
system, and if there is such support, whether there are mandatory and 
optional reference systems). for URIs, this can either mean to just 
defer the semantics to the URI scheme and/or some discovery mechanism, 
or it could also mean to restrict URIs to only allow URIs using 
location-oriented URI schemes.


Received on Thursday, 26 June 2008 19:28:27 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:33:49 UTC