I think the schema is mistaken here. If you write, say, a codelist in SDMX-JSON and you want to be explicit about the language names are written in, you would have to duplicate all names, once in the name property and then again in the names.language sub-property.
In comparison, in SDMX-ML it is sufficient to specify the name only once with the appropriate xml:lang attribute (aside: if I read the schema correctly, it defaults to "en" on Name elements, given xml:lang inheritance rules that seems problematic).
According to the prose specification https://github.com/sdmx-twg/sdmx-json/blob/71fe5eaa9fcd29e3c15f2f0216a19b9b650b1dbd/structure-message/docs/1-sdmx-json-field-guide.md#common-sdmx-artefact-properties / https://github.com/sdmx-twg/sdmx-json/blob/71fe5eaa9fcd29e3c15f2f0216a19b9b650b1dbd/structure-message/docs/1-sdmx-json-field-guide.md#item the
nameproperty is optional, but the schema requires it, even when there is anamesentry.I think the schema is mistaken here. If you write, say, a codelist in SDMX-JSON and you want to be explicit about the language names are written in, you would have to duplicate all names, once in the
nameproperty and then again in thenames.languagesub-property.In comparison, in SDMX-ML it is sufficient to specify the name only once with the appropriate
xml:langattribute (aside: if I read the schema correctly, it defaults to "en" on Name elements, given xml:lang inheritance rules that seems problematic).