FHIR Chat · consider if CodeSystem display naming is a warning · IG creation

Stream: IG creation

Topic: consider if CodeSystem display naming is a warning


view this post on Zulip Jens Villadsen (Aug 09 2021 at 09:14):

Hi y'all - I'm fiddling with my multi language supported IG here at http://build.fhir.org/ig/hl7-eu/gravitate-health and I've stumled upon something that maybe needs some reconsideration. The IG mainly consists of example document bundles in three different lanugages where the different sections all are marked up with a code from a code system. So far so good. I've started to add the codes to the (newly created) codesystems but was I just found out is that the IG publisher validates the display text from the code as well. Since the document bundles which represents the same text - but in different languages - it does not make sense to validate against the display text from a code in a codesystem as the display text varies from language to language - hence, if validation should have occured, it should have been against a designation and its proper use/value - and not the display value as my display value varies. @Lloyd McKenzie inputs? I guess that I'm basically saying that the display text being the 'default text' is of little use as it is not multifaceted in terms of being able to support multiple languages at the same time.

view this post on Zulip Jose Costa Teixeira (Aug 09 2021 at 10:35):

do I understand correctly that the current code validation defaults to the english language (or whatever is default in the terminology) but should be against the actual resource language?

view this post on Zulip Jens Villadsen (Aug 09 2021 at 11:16):

it defaults to whatever is stated in https://www.hl7.org/fhir/codesystem-definitions.html#CodeSystem.concept.display

view this post on Zulip Jens Villadsen (Aug 09 2021 at 11:17):

not taking the document's language into account. I guess the language could have been stated in https://www.hl7.org/fhir/resource-definitions.html#Resource.language

view this post on Zulip Jose Costa Teixeira (Aug 09 2021 at 12:06):

rigth, i was trying to fing the meaning from the terms.
I would expect the terminology validation to consider the designation in the resource language to see if the code text is valid .

view this post on Zulip Jose Costa Teixeira (Aug 09 2021 at 12:06):

did you state resource language in your instances?

view this post on Zulip Lloyd McKenzie (Aug 09 2021 at 14:49):

Discrepancies with display names should be a warning - and should only trigger if the display name isn't one of the designations in either the code system or value set. I think that's the correct behavior?

view this post on Zulip Jens Villadsen (Aug 10 2021 at 06:38):

Jose Costa Teixeira said:

did you state resource language in your instances?

No - didn't do that

view this post on Zulip Jens Villadsen (Aug 10 2021 at 06:39):

@Lloyd McKenzie what you describe is that the current behaviour or is it what you could wish for?

view this post on Zulip Lloyd McKenzie (Aug 10 2021 at 13:14):

It's what I understand the current behavior is supposed to be - so if it's not doing that, definitely submit a Git issue against the validator.

view this post on Zulip Jens Villadsen (Aug 11 2021 at 11:26):

(I just dropped my jaw) - that works

view this post on Zulip Jens Villadsen (Aug 11 2021 at 11:29):

Is it correctly understood that it 'simply' checks that the display value is among any of the designation values regardless of the designation use?

view this post on Zulip Lloyd McKenzie (Aug 11 2021 at 13:34):

Correct

view this post on Zulip John Moehrke (Aug 11 2021 at 13:54):

the warning the IG builder gives does offer you all the alternative display names you could have used. I have certainly had the warning offer more than one.

view this post on Zulip Grahame Grieve (Aug 17 2021 at 04:55):

multi-language IGs are not supported yet, and this is one likely place for issues to surface.


Last updated: Apr 12 2022 at 19:14 UTC