Re: The Geolocation API must provide location data in terms of a pair of latitude and longitude coordinates.

Hello, 
  I'm new to the mailing list, but do have some experience in this area. I 
was the co-chair of the OGC Open Location Services (OpenLS) specification: 
http://www.opengeospatial.org/standards/olscore and also worked with both 
LIF and subsequently  OMA on the Mobile Location protocol (MLP): 
http://www.openmobilealliance.org/Technical/release_program/mlp_v31.aspx 

  We decided to allow the default to be WGS 84. However, we also allowed 
an extension where a coordinate reference system could be specified. As 
pointed out earlier, many street databases are in other coordinate 
systems, so errors will occur with street matching if the two systems do 
not match.
  I contacted one of the staff members at OGC (Carl Reed) and asked for 
his input: 

  In my experience with OASIS, IETF, OMA etc, every standard I have been 
involved in has specified WGS-84 2d as the default. That said, both OASIS 
and the IETF go into greater detail so that the reader has the proper 
context.  The same is true for GeoRSS. The following are the typical words 
used in standards by other standards organizations that are using GML 
schema (or at least the ISO 19107 geometry model).
 
Implementations shall support the following coordinate reference systems 
based on WGS 84. These are identified using the European Petroleum Survey 
Group (EPSG) Geodetic Parameter Dataset, as formalized by the following 
Open Geospatial Consortium (OGC) URNs: 
?       3D: WGS 84 (latitude, longitude, altitude), as identified by the 
URN urn:ogc:def:crs:EPSG::4979. This is a three dimensional CRS. 
?       2D: WGS 84 (latitude, longitude), as identified by the URN 
urn:ogc:def:crs:EPSG::4326. This is a two dimensional CRS. 
The most recent version of the EPSG Geodetic Parameter Dataset should be 
used. The CRS shall be specified using the above unversioned URN notation 
only; implementations do not need to support user-defined CRSs. 

Implementations MUST specify the CRS using the srsName attribute on the 
outermost geometry element. The CRS must not be respecified or changed for 
any sub-elements. The srsDimension attribute should be omitted, since the 
number of dimensions in these CRSs is known. 

The reason for a very clear expression of the CRS, even if WGS84 is the 
default, is to insure that there is interoperability across jurisdictional 
boundaries (including international). 
Regards,
Will Wilbrink 
Pitney Bowes MapInfo



Matt Womer <mdw@w3.org> 
Sent by: public-geolocation-request@w3.org
06/25/2008 01:40 PM

To
public-geolocation@w3.org
cc

Subject
Re: The Geolocation API must provide location data in terms of a  pair of 
latitude and longitude coordinates.







Hi all,

On Jun 25, 2008, at 1:09 PM, Erik Wilde wrote:

> ...and please forgive me if i am not entirely correct in what i am 
> saying, i am not a GIS person. over the last year i have at least 
> learned that location is a more complex concept than i first thought 
> it would be.

I'm glad to see these issues coming up.  I'm hoping some of our 
geospatial technology experts will chime in here and give us some 
guidance.  From past experience, things like the grid systems used in 
some countries, and the ellipsoidal coordinate systems used in others 
make interchange difficult when not knowing what each part of the 
system expects for a datum.

(Similarly for timestamps, it should be noted that GPS time is not 
quite UTC.)

-Matt