- From: Frans Knibbe | Geodan <frans.knibbe@geodan.nl>
- Date: Tue, 21 Jan 2014 16:56:43 +0100
- To: public-locadd@w3.org
- Message-ID: <52DE98BB.6020302@geodan.nl>
On 2014-01-21 15:44, Andrea Perego wrote: > Hi, Frans. > > Comments inline. > > On Tue, Jan 21, 2014 at 11:48 AM, Frans Knibbe | Geodan > <frans.knibbe@geodan.nl <mailto:frans.knibbe@geodan.nl>> wrote: > > [snip] > > It could very well be that I am overlooking something. But let me > try to explain my thinking: The original suggestion was to model > centroid, MBR, etc. as subproperties of locn:geometry. But the > range of locn:geometry is locn:Geometry. So I think you can't say > something like: > > ex:London > a locn:Location; > locn:centroid ex:aCentroid; > locn:mbr ex:aMbr; > locn:geometry ex:goemetry1; > locn:geometry ex:geometry2; > locn:geometry ex:geometry3. > > > Sorry, Frans, probably I'm missing something here. Do you mean that > MBRs, centroids, etc. cannot be considered as types of geometries? No. It makes sense to consider them special kinds of geometry, so to model them as subproperties of locn:geometry. At this moment I can't recollect what I was thinking. Perhaps I confused rdfs:range and rdfs:domain, or it was something else... Anyway, I am sorry for the disturbance and I think I fully agree with the proposal :-D Regards, Frans
Received on Tuesday, 21 January 2014 15:57:13 UTC