Skip to end of metadata
Go to start of metadata
HL7 TSC FMG Meeting Minutes 

(voice and screen)

Date: 2020-09-16
Time: 4:00 PM U.S. Eastern
Chair:Note taker(s): Anne W.
Quorum = chair + 4yes/no
Co chairsxDavid HayxLloyd McKenzie
Wayne Kubick, CTO
xHans BuitendijkxBrian Postlethwaite
Paul KnappxAnne W., scribe
xJosh MandelxJohn MoehrkexBrian PechxDaniel Liu
xGrahame Grieve

xLynn Laakso

xLori Wong

xDidi Davis



  • Roll Call
  • Agenda Check
    • MOTION to accept: Hans
  • Minutes from 2020-09-02 FMG Agenda/Minutes
    • MOTION to approve: Hans/John
    • VOTE: All in favor
  • Action items
    • Reviewed
  • Review Items
  • Discussion Topics
    • Formal statement on MustSupport for ONC
      • MustSupport Clarifications
        • Hans give background. With rollout of FHIR US Core IG, ONC's interpretation of MustSupport is not what was expected. One understanding is that MustSupport propogates to each of the attribute's elements defined through the data type and reference or target list. Would someone like HL7/FMG/FHIR-I/Wayne to communicate the original intention, which was that MustSupport is only intended to apply to the FHIR element it is applied to and does not propogate down to subsequent elements, as noted in blue italics in the clarification document.
        • Grahame: The letter of clarification will have to be issued by Wayne in his role as CTO. 
        • The confusion arises from both US Core and FHIR Core. We say we never use MustSupport in Core. But then we use it anyway in context of VitalSigns. FMG will have to clarify what we think it means. Then in US Core, section is a source of confusion. Compels people to support things they're not using. Need to separate the profiles from the MustSupport rules.
        • Josh: There's a clear gap between editorial intent and a hard line interpretation of what the profiles say. There's also a gap between what you can actually do and what people want.
        • Need to say there is confusion around the current IG and the testing regime at present is making one set of interpretations beyond what we intended to say. Can tell them what we did and didn't mean to say.
        • Hans edits the document.
        • Grahame: Labeling an element MustSupport does not mean the whole value domain is MustSupport
          • MOTION that FMG should request that the TSC ask Wayne sending a letter clarifying MustSupport; FMG will continue to work on drafting the technical part of the letter for Wayne's consideration: Grahame/Hans
          • VOTE:  All in favor
    • Policy/process around extensions
    • Extending FHIR R4 - expires in December
      • Lynn advises to submit an extension request on the publication request template
        • MOTION to submit an R4 extension request: Grahame/John
        • VOTE: All in favor
    • Process for retiring IGs
      • Came up when there was a proposed IG that we approved but never went to ballot. Do we need to have anything from an FMG perspective to retire content - whether it went to ballot, or it never went to ballot but still exists in the CI build, IG page, registry, etc.
      • Lloyd; We have a header bar that we could change to flag the fact that the IG has been deprecated/retired, with a blurb about what replaces it (if anything)? Grahame agrees that could be done simply. 
        • What triggers it? Somebody submits a formal request to withdraw a publication, or periodic review that catches content candidates for retirement.
        • Could rely on the TSC process but that doesn't catch things that haven't gone to ballot
        • Is there an FMG role? Or is it a Grahame/Lynn thing? Discussion over current process. FMG process would only apply to things that haven't gone to ballot. Another checkpoint could be when a project gets closed. Grahame could have a process in publication to indicate when it was last updated.
    • Messaging to WGs at WGM - Key objectives for R5 release including any barriers to meeting those objectives in a timely fashion
      • Grahame; Need so solicit WG views on R4B/R5. Do they have content they want to process through the trial use process faster? Do they have an opinion on R5 timelines for the slower content?
        • Josh: Could we provide rough maturity level guidance? Grahame: FMM 3 stuff would be pretty difficult to iterate into R4B context. If you want to argue for something to be in R4B, you need to have implementers who will adopt it. Josh: Also can't pull the rug out from people supporting the old version. Josh: We need to explain our constraints.
        • Lloyd arrives
        • Grahame: There should also be a pressing reason to be included in R4B. 
        • If you want to be in 4B:
          • Pressing need
          • Implementable, with implementers waiting to do it
      • Could mention that we'll be investing significantly in MustSupport infrastructure and policy over the next few months.
    • FMG WGM topics
      • Anything other than wrap-up of the week? FMG is meeting from 2-3:30 on Thursday. No specific other topics noted.
  • Adjourned at 5:30 pm Eastern