Ted Klein (chair, minutes)
|1||Roll call and Agenda check||Ted Klein|
No agenda changes. No new business.
|3||FHIR Ballot Item||Rob Hausam/Alexander Henket||ValueSet.compose missing inactive/abstract? (Alexander Henket)|
Rob discussed the fact that this can be accomplished using a Code System Supplement; Grahame noted that we can return hierarchy in an expansion, but we deliberately did NOT put in fancy functions to compose to control hierarchy. There was discussion about extensions. It as agreed that any hierarchy should not be assumed to have any particular meaning, it is really for display only. This is related to item 16814. Grahame suggests that we propose an extension to value set compose to be able to define a hierarchy without specifying any meaning for the hierarchy. Grahame will put together an example, with the suggestion on the terminology zulip channel. We moved to the discussion about the assertion of 'inactive' (or deprecated or other kinds of knowledge about concepts) as part of value set creation, this should probably be included with that. Note that this is distinct from the definition of 'inactive' on codes in the code system. Discussion will be moved to the terminology channel.
Action Item Updates
|6||UTG Items||Ted Klein|
The code systems and value sets that are used in FHIR have had URIs assigned as part of the FHIR development process within the FHIR space, i.e. at http://hl7.org/fhir/v3/... As part of the UTG migration, we want the URIs to be resolvable URLs and to be sensible and easy to manage on the HL7 side. We are in the process of setting up a domain terminology.hl7.org for this, and we believe that the URIs for our material should be here. This would involve a one-time change to the FHIR URIs for all the vocabulary - and this would be a breaking change. We only want to do it once, and now before the final Normative FHIR ballot would be our last opportunity to do so. We need the Vocabulary WG to weigh on this formally. Note this is a key step in unifying the HL7 terminology across the product lines.
We will likely have the URIs be http://terminology.hl7.org/CodeSystem/... and http://terminology.hl7.org/ValueSet/... for all of them. Note we realize we need to deal with some name collisions between V2 and V3 objects having the same name thus the automatic naming software causes the collisions; we will fix this manually one-time.
Discussion: it was noted that this is the big unification step, and this will lead directly to the need to harmonize the content where we have the same semantic space coverage but different codes and granularities. Also noted that we must start the detailed discussion on versioning, and probably needs its own zulip thread. Note that the discussion about this URI change probably also should have a zulip thread. This is a biggie, since it basically requires all implementations to be significantly changed wherever and whenever they use and refer to the existing canonical URIs. This may involve NamingSystem and the redirect things discussed a few WGMs ago. Ted must initiate this zulip discussion today.
After a couple of weeks of zulip discussion, maybe we can be ready for a WG decision on the next call.
(from the FHIR tracker item calls)
|Heather Grain/Susan Barber|
LOINC and TC215: Ted
FTSD & SGB: Russ
Call adjourned at 5:01PM EDT. Next conference call scheduled for two weeks hence on June 28, 2018 at 3:30PM EDT (UTC-4; the clocks in most of North America will have changed to Summer Time March 11).