FHIR Chat · Vocab Block Vote - June 25, 2020 · terminology

Stream: terminology

Topic: Vocab Block Vote - June 25, 2020

Carmela Couderc (Jun 18 2020 at 15:48):

Vocabulary Block Vote Planned for June 25, 2020 Vocab Conference Call (3:30 – 5:30 EDT)
Visit http://www.hl7.org/concalls/CallDetails.aspx?concall=49823 for the full details of this call

Webmeeting Info:
Zoom call: https://zoom.us/j/7183806281?pwd=WHVnUUlkWWhhcnRaYk9sWWQyOEkvUT09

Please let me know if you would like to pull an item out to discuss. You will get another opportunity on the call to remove an item. All items are related to ValueSet and were created as a result of the Value Set Definition Compliant Profile on ValueSet project.
JIRA Summary:

1. J#26551 persuasive, create new extensions for minimum and maximum multiplicity
2. J#26550 persuasive, change extension name (valueset-workflowStatus)
3. J#25395 persuasive with modification, change extension name (valueset-othername)
4. J#25272 – persuasive, add element to resource (scope)
5. J#25179 – persuasive, fix compose.include.valueSet definition to reference intersection, not union
6. J#23786 persuasive, remove extension valueset-system
7. J#23785 – persuasive, update guidance for valueset-trusted-expansion

Michael Lawley (Jun 18 2020 at 18:03):

Re J#23785 , there are two issues:

1. this is currently a URI(dentifier), not URL(ocation) but the definition suggests that this is meant to be directly resolvable -- suggest that the type be changed to URL
2. definition should clarify what this actually refers to (so a tool knows how to use it). Specifically, is this the URL of an actual expansion (ie resolving the link will directly supply a ValueSet with an expansion), or is it intended to be the (base) FHIR endpoint and thus used to construct an $expand operation (or is it something else)?

Jim Steel (Jun 18 2020 at 19:54):

Is the http://www.hl7.org/concalls/CallDetails.aspx?concall=49823 link correct? It appears to have already occurred?

Carmela Couderc (Jun 18 2020 at 20:01):

Sorry - the link is generated over the weekend - I copied the link from last week which was not the right thing to do.

Jim Steel (Jun 18 2020 at 20:06):

I feel like J#25272 might need some discussion. I'm not sure whether I can make the call (I burned a lot of early morning credits this week with DevDays), but I'll try. I'll put some comments in JIRA at least

Carmela Couderc (Jun 18 2020 at 21:19):

@Michael Lawley J#27848 entered to address your issue.

Michael Lawley (Jun 18 2020 at 23:13):

Thanks @Carmela Couderc

Grahame Grieve (Jun 18 2020 at 23:24):

I'm not sure about the change to URL from URI for J#23785 -the language doesn't say that you can't indicate the server by a logical identifier.

Grahame Grieve (Jun 18 2020 at 23:25):

I think it's a server, btw, at least primarily

Michael Lawley (Jun 18 2020 at 23:25):

no, but it's strange and to be useful for tooling needs to say what people are expected to do with it.

Grahame Grieve (Jun 18 2020 at 23:27):

I agree that we should be clear. Just that changing from URI to URL is a good way to introduce substantiative change with the committee realising it. I'm just making it explicit. I don't have a strong opinion either way

Michael Lawley (Jun 18 2020 at 23:27):

I would like it to be a Tx endpoint, but I can imagine that someone might come along and say - no, we have a package registry of all these trusted expansions, we want to point at that

Grahame Grieve (Jun 18 2020 at 23:36):

They can still conform to the URI specification and therefore the FHIR specification, though that might not be convenient

Michael Lawley (Jun 18 2020 at 23:38):

Yes, especially if we're explicit about what is expected at the end of the link.
There are some corner case differences with URIs / URLs around %-encoding etc, but I think we're in agreement that the key thing is to make it explicit.

Last updated: Apr 12 2022 at 19:14 UTC