Re: Proposed RDF FHIR syntax feedback

I think it's clear that mime type won't be sufficient to allow a client to
differentiate whether they want the "instance" of a FHIR resource or the
"ontology" for a FHIR resource.  So we're going to need a convention where
the ontology is available from a different endpoint than the instance.
Grahame and I are working at figuring something out that can work for
resource, data types, profiles, valuesets, etc.

And I think we've landed on declaring only one namespace?


*Lloyd McKenzie*Consultant, Information Technology Services
Gevity Consulting Inc.

 E: lmckenzie@gevityinc.com
M: +1 587-334-1110 <1-587-334-1110>
W: gevityinc.com


*GEVITY**Informatics for a healthier world *

CONFIDENTIALITY – This communication is confidential and for the exclusive
use of its intended recipients. If you have received this communication by
error, please notify the sender and delete the message without copying or
disclosing it*.*

NOTE: Unless explicitly stated otherwise, the opinions and positions
expressed in this e-mail do not necessarily reflect those of my employer,
my clients nor the organizations with whom I hold governance positions

On Sat, Mar 7, 2015 at 6:30 PM, David Booth <david@dbooth.org> wrote:

> On 03/06/2015 02:15 PM, Lloyd McKenzie wrote:
>
>> Where we'll have a particular challenge is where the RDF and OWL
>> representations can both be expressed using the same sytnax.  It may be
>> that the solution there is to return both the instance and class
>> information.  Is there a distinct mime-type for JSON-LD from regular JSON?
>>
>
> Yes: application/ld+json
> http://www.w3.org/TR/json-ld/#iana-considerations
>
> However, I think there may be some confusion about the significance of the
> MIME type or the RDF or OWL serialization format.  Any standard RDF
> serialization format (RDF/XML, Turtle, N-Triples, JSON-LD, etc.) can hold
> triples that use any mixture of RDF, RDFS and/or OWL terms.  All are a form
> of RDF data.  And the data might represent an ontology, instance data or a
> combination of both.  The serialization format (or MIME type) is completely
> independent of the role that the RDF content plays (such as representing an
> OWL ontology, or representing patient instance data).
>
> In general, URIs defined in the hl7.org URI space will be for
> *ontologies* -- not instance data -- and those ontologies will use a
> mixture of OWL, RDFS and plain RDF terms.   URIs for instance data will be
> created by the various parties that will be exchanging patient data.
> Various healthcare providers *will* be creating URIs for their own profiles
> (or ontologies) also, but I do not yet see any reason why either HL7 or a
> healthcare provider would need to use the same URI for both a term in an
> ontology *and* an item in some instance data.  If someone has an example of
> where they think that might happen, it would be helpful to post it.
>
> Thanks,
> David Booth
>