Stream: netherlands
Topic: Open/Closed modelling in FHIR guidelines
Marc de Graauw (Dec 21 2016 at 09:44):
The section on Open/Closed modelling in the Guidelines is not very clear. What would constitute closed modelling in FHIR? Is this just setting unneeded elements to 0..0 in your profile, or is more involved? Further, this paragraph is a list of pros and cons, but it isn't clear what approach was chosen for the ZIBs and Medmij (and why).
Marten Smits (Dec 21 2016 at 12:59):
We have chosen for "open" modeling for the ZIBs, since they are very general en not meant for one specific sue case. Excluding unspecified elements here is not an option, since we expect derived profiles for more specific use cases that may need those elements.
Marc de Graauw (Dec 21 2016 at 13:11):
Sound choice for ZIB. I'd like to comply with the Profling Guidelines, but I won't use open modelling.
Is it just using 0..0 vs not using this, or is more involved? I.e. valueSets? (And note: open vs. closed modelling is not FHIR terminology, but v3 Template speak. Probably not everyone in the intended audience will understand the section.)
Marten Smits (Dec 21 2016 at 15:02):
I know it isn't FHIR speak, but I actually wasn't aware that this was a v3 thing. I thought it was self explanatory, but some introduction doesn't hurt I guess.
For valueset bindings we have chosen for "extensible" binding instead of "required" in the ZIB, for the same reason: we don't want to exclude any terminology that the ZIB doesn't specify.
Last updated: Apr 12 2022 at 19:14 UTC
