We like the OGC's persistent URI delivery of GML information for Coordinate Reference Systems, e.g. http://www.opengis.net/def/crs/EPSG/0/7844 (GDA2020) however, better than GML-conformant XML data only, human and machine readable, Linked Data accessible Semantic Web data would be maximally useful and not a great deal of effort to implement.
For this, we could use the well-known concept catalogue standard SKOS for simple, human-readable metadata and deliver the GML XML information beside it (same URI, different HTTP Request header) using the new Content Negotiation by Profile spec for profile (SKOS & GML) access. ConnegP use would also address Issue #108.
Optimally, we would also supply a Semantic Web version of the GML information, not just Semantic Web metadata (SKOS), as this would allow Linked Data clients to automatically utilise GML associations such as GMLs's ellipsoidalCS property for association following for extended data gathering and perhaps even automated reasoning.
For this, we would need an OWL version of GML's CRS models.
We like the OGC's persistent URI delivery of GML information for Coordinate Reference Systems, e.g. http://www.opengis.net/def/crs/EPSG/0/7844 (GDA2020) however, better than GML-conformant XML data only, human and machine readable, Linked Data accessible Semantic Web data would be maximally useful and not a great deal of effort to implement.
For this, we could use the well-known concept catalogue standard SKOS for simple, human-readable metadata and deliver the GML XML information beside it (same URI, different HTTP Request header) using the new Content Negotiation by Profile spec for profile (SKOS & GML) access. ConnegP use would also address Issue #108.
Optimally, we would also supply a Semantic Web version of the GML information, not just Semantic Web metadata (SKOS), as this would allow Linked Data clients to automatically utilise GML associations such as GMLs's ellipsoidalCS property for association following for extended data gathering and perhaps even automated reasoning.
For this, we would need an OWL version of GML's CRS models.
Such an ontology would likely be useful for Content Negotiation by CRS.