Stream: terminology
Topic: 11179 storage issue
Grahame Grieve (Sep 07 2018 at 21:11):
11179 defines that an element can have different value domains - different codes that have the same meaning, but are used in different places. e.g. 1= M, 2=F, 4 = U, 9= O (to use a long defined australian code system for gender). We have 2 extensions defined for this: http://build.fhir.org/extension-11179-permitted-value-valueset.html and http://build.fhir.org/extension-11179-permitted-value-conceptmap.html. I'm not sure that's an optimal set of definitions, and I've just created a task (GF#17828) to review them. Thought I'd alert vocab to their existence, and ask if vocab has any systematic approach here.
Rob Hausam (Sep 07 2018 at 21:26):
Don't know offhand about any systematic approach, but we'll definitely want to look at it.
Lloyd McKenzie (Sep 07 2018 at 21:29):
Actually it says that each data element has both a domain value and a permitted value. One defines the meaning (and apparently the exchange) and the other defines what gets stored.
Grahame Grieve (Sep 07 2018 at 22:50):
that's not what ISO 11179 actually says
Lloyd McKenzie (Sep 07 2018 at 23:14):
That's how it was explained by people who ought to know it pretty well. In any event, I'm fine with whatever solution vocab likes
Grahame Grieve (Sep 07 2018 at 23:35):
the domain value is conceptual, and then there's the syntax you use to represent the vlaue.
Grahame Grieve (Sep 07 2018 at 23:36):
I argued to Ed Barkmeyer (author) that this is fundamentally insane, but the interoperability view point is not part of their world. it's fine if different systems have different non-compatible representations of the same data element, that's their right. as long as they mean the same thing.
Grahame Grieve (Sep 07 2018 at 23:37):
I'm not surprised that in the interop world, this collapses to public/private, since what else could you do? but it's not the underlying notion of 11179
Last updated: Apr 12 2022 at 19:14 UTC
