Definitive 19139/19115 xml schemas? (not -2, -1, -3)

[srstsavage]

Hi, are there any plans to add the original 19139/19115 schemas to this repository/standards.iso.org?

I assumed that the versions here would be authoritative:

http://www.isotc211.org/2005/

but they suffer from a GML namespace problem (the GML 3.1.1 namespace http://www.opengis.net/gml is used, but the namespace should actually be http://www.isotc211.org/2005/gml referencing the embedded GML 3.2.0 version). This causes documents using these schemas to fail validation when namespace based validation is used. The problem is discussed here:

https://inspire-forum.jrc.ec.europa.eu/pg/forum/topic/7339/20070417-version-19139-schemas-for-service-metadata/#Item_1617

A "mopping up" revision of the schemas hosted at http://schemas.opengis.net/iso/19139/20070417/ resolves this issue by using the http://www.opengis.net/gml/3.2 namespace (which points to the since-released GML 3.2.1), and seems to be widely used (e.g. by the ogc-schemas JAXB bindings project).

Many consumers of these documents seem to depend on predictable namespaces, so we want to be able to use an official but valid set of schemas in our documents.

Should http://schemas.opengis.net/iso/19139/20070417/ be regarded as the definitive set of schemas? If not, can a fixed set of 19139/19115 schemas be added to standards.iso.org?

@tedhabermann @dr-shorthair

[srstsavage]

If not, can a fixed set of 19139/19115 schemas be added to standards.iso.org?

BTW I'm happy to set up a pull request if they should be added here.

[srstsavage]

Another variant being used in schemaLocations in the wild:

https://www.ngdc.noaa.gov/metadata/published/xsd/schema/

[srstsavage]

Just found this noted at http://schemas.opengis.net/iso/19139/20070417/ReadMe.txt:

http://standards.iso.org/ittf/PubliclyAvailableStandards/ISO_19139_Schemas/

So I guess that's the offical schemaLocation that should be used? e.g. http://standards.iso.org/ittf/PubliclyAvailableStandards/ISO_19139_Schemas/gmd/gmd.xsd

[smrgeoinfo]

Yes, the xml schema at http://standards.iso.org/ittf/PubliclyAvailableStandards/ISO_19139_Schemas/gmd/gmd.xsd is used as the authoritative implementation of the xml namespace http://www.isotc211.org/2005/gmd. The Namespace URI is an identifier for the namespace, not a location for the xml schema implementing the namespace.

[srstsavage]

@smrgeoinfo Thanks for the verification, that's helpful. You're of course correct, I misspoke above and conflated namespace URIs and schemaLocations, meant to say that some consumers depend on a predictable schemaLocation/set of schemas. I think some validation issues I've been encountering have to do with a mix of the older http://www.isotc211.org/2005/ schemas and newer schemas.opengis.net/iso/19139/20070417/ schemas being used in the pipeline, but now that we can focus on using schemas from http://standards.iso.org/ittf/PubliclyAvailableStandards/ISO_19139_Schemas/ the situation will be much clearer. Thanks again.