In the specs, the DV_MULTIMEDIA.media_type is constrained to the openEHR terminology “media types”.
The terminology includes several text/*** types, some which might not be for multimedia data but for parsable data: terminology/openehr_external_terminologies.xml at master · openEHR/terminology · GitHub
Shouldn’t we have a different term sent for parsable and multimedia media types?
I can also see that we don’t have a term set for the DV_PARSABLE.formalism, and if an inline encapsulated data is recorded, it should be contained in a DV_PARSABLE and the formalism doesn’t match analogous type of data when it is recorded as a reference in DV_MULTIPMEDIA Data Types Information Model it uses “json” in DV_PARSABLE and “application/json” in DV_MULTIMEDIA. Shouldn’t the same values be used?
From that I understand when having a reference to some data, even if it’s parsable, it’s modeled as a multimedia because parsable doesn’t have attributes to make references to data stored elsewhere. But, shouldn’t any encapsulated have that feature?
Should we have a term set for DV_PARSABLE.formalism?
If it was me designing this package, I would separate what is parsable (text-based) form strictly multimedia (binary) data, have the formalism/media_type constrained by different term sets (without common codes between them), and have an uri attribute for external references also on both types (that would be inside DV_ENCAPSULATED).