[GeoSciML] Vancouver meeting notes
Boisvert2, Eric (NRCan/RNCan)
eric.boisvert2 at canada.ca
Fri Jun 15 07:24:40 EDT 2018
>- whole geometries can be expressed as RDF literals
But is this supported by any client application ?
From: Simon.Cox at csiro.au <Simon.Cox at csiro.au>
Sent: June 12, 2018 21:45
To: geosciml at lists.opengeospatial.org
Cc: Boisvert2, Eric (NRCan/RNCan) <eric.boisvert2 at canada.ca>
Subject: RE: Vancouver meeting notes
I notice you have re-discovered the JSON-LD gotcha - no nested arrays.
I'm not convinced this is a big problem. The RDF model is for logic and reasoning, not for math operations. Thus I don't think RDF is relevant to coordinates and geometries - whole geometries can be expressed as RDF literals, but not individual coordinate values. So there is no need for nested arrays. GeoSPARQL recognized this by putting a boundary between the conceptual aspects that are modelled in RDF, and the geometries which are expressed in a micro-format in Literals of type http://www.opengis.net/ont/geosparql#wktLiteral or http://www.opengis.net/ont/geosparql#gmlLiteral. See sections 8.5 and 8.6 in GeoSPARQL.
And anyway, GeoSciML shouldn't be trying to solve the geometry/geography problems - stick to your knitting, which is geology and geophysics.
GeoJSON is interesting and important, but is not compatible with JSON-LD.
From: GeoSciML [mailto:geosciml-bounces+simon.cox=csiro.au at lists.opengeospatial.org] On Behalf Of Boisvert2, Eric (NRCan/RNCan) via GeoSciML
Sent: Wednesday, 13 June, 2018 05:32
To: GeoSciml OGC mainling list <geosciml at lists.opengeospatial.org<mailto:geosciml at lists.opengeospatial.org>>
Cc: Boisvert2, Eric (NRCan/RNCan) <eric.boisvert2 at canada.ca<mailto:eric.boisvert2 at canada.ca>>
Subject: [GeoSciML] Vancouver meeting notes
please tell me if I forgot anything
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the GeoSciML