Closed jmandel closed 11 years ago
The more things like this you define, the closer you get to redefining hData, you know.
Yup, that's true. If you're interested, there's a group that's been walking through the FHIR spec, section by section, over lunch on Mondays. I'm sure you'd be welcome to call in. A good number of hData folks, including Mark Kramer, will be on the line, so it might be a good place to follow those thoughts.
Is the concern that FHIR is redefining hData, or that BB+ is?
My understanding is that the following sequence:
... is an area of precise overlap between hData and FHIR. (i.e. both hData and FHIR enable this same sequence of operations -- in slightly different ways). Is that correct?
We're using FHIR here on Keith's suggestion (proposed on the BB+ Pull conference call and seemed generally well-liked). But in any case, I don't think BB+ is (or should be) inventing this process.
FHIR isn't redefining hData, it is augmenting it by defining specific resources. hData is content-agnostic, whereas FHIR is a closed world that around its view of resources. There's about 90% overlap in terms of the REST methods used by hData and FHIR. The GET method is certainly defined by both. There is a slight difference in the URL between hData and FHIR, specifically the use of the "@" sign before a resource ID (not done in hData).
OK. So: is there a concern about the currently-documented approach?
Can you be more specific?
Justin had raised a question about "redefining" (see upthread).
On Fri, Apr 12, 2013 at 3:38 PM, markkramerus notifications@github.comwrote:
Can you be more specific?
— Reply to this email directly or view it on GitHubhttps://github.com/blue-button/blue-button-plus-pull/issues/11#issuecomment-16313260 .
Document Retrieval endpoint is now documented (and clearly required in the case of the FHIR DocumentReference feed that we're using.)
As I was documenting the Search endpoint, I realized that FHIR search doesn't return documents; it returns an Atom feed of metadata (document descriptors). We need a way to turn these into documents. I've drafted a quick approach at: http://blue-button.github.io/blue-button-plus-pull/#clinical-document-retrieval
... could use some review to be sure!