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


Date: 2020-09-24
Time: 2:00 PM U.S. Eastern
Chair:Note taker(s): Anne W.
Quorum = chair + 4yes/no
Co chairsxDavid HayxLloyd McKenzie
ex-officioxWayne Kubick, CTO
MembersMembersMembersObservers/Guests

Hans BuitendijkxBrian Postlethwaite
Paul KnappxAnne W., scribe
xJosh Mandel
John MoehrkexBrian Pech


xGrahame Grieve



x

Brittany Brown
Joan Harper
Doug Pratt
Gino Cansessa
Keith Carlson
Oyvind Aasve
Wanda Govan-Jenkins
Reinhard Egelkraut
Joel Francis
Didi Davis
Ciprian-Virgil Gersenberger
Yunwei Wang
Jens Villadsen
Christopher Schaut
Kathy Pickering
Mark Kramer

Agenda

  • Roll Call
  • Agenda Check
  • Discussion Topics
    • Must Support
    • R4B issues to consider
      • can we update maturity levels of resources (tricky discussion about updating stuff that was FMM0, given they were balloted as draft)
      • can we add/remove/correct examples?
      • for resources that were FMM 0 in R4, does that really mean that they can't go normative until R7 - which could be another 8 years from now if we are looking at 3 years between releases.
    • Other issues from the week/wrap-up
  • AOB (Any Other Business)

Minutes

  • Roll Call
    • Numerous guests noted above
  • Agenda Check
    • No additions
  • Discussion Topics
    • Must Support
      • Working on proposed language; reviewed at MustSupport Clarifications#DRAFT
      • Grahame is okay with it; would add the examples to an appendix.
        • MOTION to endorse the wording as a basis for a letter to ONC and/or other relevant parties from the FHIR product director and HL7 CTO: Josh/Brian Pech
        • Discussion over parent vs. child. You only have to support the children if you support the parent. Josh notes that isn't reflected in the language that is currently published. FHIR-I will be tasked with clarifying that issue in the specification. Mark Kramer will log it in JIRA.
        • Confusion over the intent of the last sentence/third paragraph.
          • Recommendation to clarify that sentence before including it in the letter.
        • VOTE: All in favor
    • R4B issues to consider
      • can we update maturity levels of resources (tricky discussion about updating stuff that was FMM0, given they were balloted as draft)
        • Grahame: Can update in current build and draft milestones. The processes that lead to enhanced maturity often happen outside the specification. If committees want to update their maturity in R4B, they can. We would need to indicate that whatever is in R4B is not authoritative unless we sync. Lloyd: There are two levels of authority - the CI build is most current but not official. So a small number of resources can update content, but anyone could potentially update maturity level if warranted. Brian Post is concerned that the changes to maturity levels should be subject to ballot. Lloyd: The assertion of maturity has to be consistent with the content in that release. The objective is to indicate accurately to the community what the maturity level is of the material in 4B - the current reality for the resource. Grahame: Changes to scope would require a ballot; this is a management issue. Need criteria around when and how they do it. Brian Post: They could ask for the change and FMG could review it first.
      • can we add/remove/correct examples?
        • Grahame: They could add a note to any section in the spec that isn't normative alerting people that R4B is a fork and they should refer to what's in the current build. Can't yank material.
      • for resources that were FMM 0 in R4, does that really mean that they can't go normative until R7 - which could be another 8 years from now if we are looking at 3 years between releases.
        • This goes back to the rule that you can't go normative until you've been in two STU releases. The rules were written with the idea of more frequent balloting. Grahame: The resources in question are not part of 4B? Lloyd: Yes, but there are others too, such as FMM0 stuff. Linkage, for example - there is no pressing reason to make it something other than for comment until R5 comes out, but that means they can't possibly go normative until R7. We could have incremental releases. Grahame: What we were seeking with two cycles was that something wouldn't slip through without looking at it. If they can prove quality we could find a different way of meeting the same goal. Lloyd: So the rule stays but we could make exceptions.
      • 4B MedicationDefinition stuff and regulatory projects coming forward have a desire to change the references on some of the other resources to allow subjects and targets to be broader than they currently are in those resources. They are wondering if it's possible to make those sorts of changes, where you're just extending the list of target types for a reference. Wouldn't change how the xml or the JSON looks. Could have a consolidated page in the ballot with resource elements that have had their type lists expanded.
        • Grahame: We have resources that are normative that include references to other resources that aren't at the same level. Therefore, we've already cleared the methodology for this, so this would be reasonable.
      • Brian Post: Is there anything that will be listed as technical corrections? Grahame: There are 4 or 5 things that could be technical corrections to R4 that we could do in R4B.
      • Lloyd: The only other candidate resource that has come forward is AdverseEvent. We are seeking further explanation on the rationale for its inclusion. Who would implement if it went to trial use? What are the ramifications of having to wait?
    • Other issues from the week/wrap-up
      • Grahame: Not sure these are management issues, but CIMI will be bringing a project to MnM that will raise the issue of profile consistency. There is obviously a continuing discussion around conformance and indicating to people on how to match what IGs say with what people expect to be told when they're reading an IG. Some people are looking for simpler material. It's likely that FHIR-I, MnM and Conformance will be spending some time on that going forward.
      • Lloyd: We have talked about going through and actually changing info messages to warnings and warnings to errors in the build process and getting stricter to ensure that WGs go through those things and address them properly. If we're moving towards another normative release and that's starting in May, it would be appropriate to figure out if and when we can tighten those things up. There is tooling work to do to address some of these. Need to decide if and when we're going to do that. Is there a way that we can do this that doesn't take too much of everyone's time? Grahame: We should start communicating it to committees right now that there will be increased pressure on QA for everything over FMM 3. 
        • Anne Wizauer Add standing item to FMG agenda: What are specific QA things that we need to tighten up
      • Hans arrives. He reports that CGP WG reviewed the Must Support language and made some recommendations for wordsmithing. Do other groups need to look at it before sending out the letter? Decision that it's not necessary.
      • Brian Post: Grahame came to PA and they discussed Practitioner and PractitionerRole re: updates and changes for the future.
  • Adjourned at 3:16 pm Eastern
  • AOB (Any Other Business)