W3C home > Mailing lists > Public > semantic-web@w3.org > December 2008

Re: Ontology for points in a three-dimensional space

From: Norman Gray <norman@astro.gla.ac.uk>
Date: Tue, 9 Dec 2008 14:05:37 +0000
To: Toby A Inkster <tai@g5n.co.uk>
Message-Id: <58097EE5-7F65-434C-8368-CD2588A77C8C@astro.gla.ac.uk>
Cc: Semantic Web <semantic-web@w3.org>


Toby, hello.

On 2008 Dec 9, at 12:49, Toby A Inkster wrote:

> For the time being, I've set up this:
>
> 	http://buzzword.org.uk/rdf/ungeo#
>
> But will keep reviewing alternatives.

In this context, x, y, z would usually (or most conventionally) be  
linear coordinates with an origin at the centre of the body, such as  
the earth.  On the other hand, latitude and longitude would be angular  
coordinates, and altitude a radial coordinate above the earth's  
reference surface, also centred on the earth.  Thus coordinate (0,0,0)  
would be the centre of the earth in (x,y,z) coordinates, and the point  
on the equator south of Greenwich in (long,lat,alt) coordinates.  Thus  
they can't really be subProperties.

Unless you're talking about (x,y) being coordinates on a plane, which  
is to say a map, which involves you in mappings.

As a basic set of properties, you could skip (x,y,z), define longitude  
as east-longitude (so that (long,lat,alt) defines a right-handed  
coordinate system -- http://www.w3.org/2003/01/geo/wgs84_pos# still  
fails to distinguish east- and west-longitude), define altitude as  
height above some reference datum to be specified in the :system  
comments, and give units -- (rad,rad,m)? -- for each of them.  That  
should be enough for simple cases.

That boils down to something which is almost the same as <http://www.w3.org/2003/01/geo/wgs84_pos# 
 >, but neutral as to the reference system, and specifically without  
the restriction to WGS84.  The terms in that ontology would be  
subclasses of the terms you want to define.

>> if you're interested, I could roll you an RDFS version of it).
>
>
> Thanks. It's a big spec, so don't do so on my account, but if you  
> were planning on doing it anyway, then I'd certainly be interested  
> in the result.


I can send you a mechanically-generated version of an earlier STC  
release off-list, if you like.

All the best,

Norman


-- 
Norman Gray  :  http://nxg.me.uk
Dept Physics and Astronomy, University of Leicester
Received on Tuesday, 9 December 2008 14:06:17 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 21:45:26 GMT