FHIR Infrastructure Minutes WGM 201809

FHIR Infrastructure 2018 September Baltimore WGM Minutes

Return to FHIR Infrastructure Minutes

Agenda

• see File:FHIR-I Attendance-Sept 2018.pdf

Monday Q3

Chair: Lloyd McKenzie Scribe: Ron Shapiro

Tracker Items

• general overview of the FHIR Infrastructure WG and WGM Meetings
• [18671] - Persuasive w/ Mod
• [17638] - Persuasive w/ Mod
• [17892] - Not Related

Monday Q6 (Reference Implementations)

Chair: Scribe:

Tuesday Q4

Chair: Lloyd McKenzie Scribe: Ron Shapiro

Tracker Items

• [18673] - Persuasive with Mod
• [18793] - Persuasive
• [18791] - Persuasive
• [18277] - Persuasive with Mod
• [17915] - Not Persuasive (already fixed)
• [17906] - Persuasive with Mod
• [18278] - Not Persuasive
• [18418] - Persuasive
• [18284] - Persuasive
• [18138] - Persuasive
• [17723] - Persuasive
• [17725] - Not Persuasive with Mod

Wednesday Q1

Chair: Josh Mandel Scribe: Michael Donnelly / Josh Mandel

Wednesday Q2

Chair: Scribe:

Wednesday Q4

Chair: Scribe:

Thursday Q2

Chair: Lloyd McKenzie Scribe: Rick Geimer

Tracker 18335: Inheritance from DomainResource isn't clear in renderings - N-Infra #8

• https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=18335
• Discussion about showing elements from DomainResource (id, meta, etc.) in the Structure tab for all resources. Basically a Snapshot tab for resources that mirrors how it is shown for profiles
• Agreed to find it not-related for the ballot, but to create a new tracker item separate from the ballot to move the "Elements defined in ancestors" statement currently under description to below the word "Structure" (directly above the resource) to make it more prominent.
• Motion: Lloyd McKensie/Isaac Vetter: 12-0-0
• New tracker is 19304.
  • https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=19304
  • Motion: Rick Geimer/Yunwei Wang: 12-0-0

Tracker 17791: Invalid UCUM expressions in FHIR UCUM Valueset

• https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=17791
• Ballotter provided a spreadsheet listing some UCUM expressions used in the FHIR spec that she believes to be invalid.
• Found it not related to a ballot, created a separate tracker to review the expressions and fix those that are invalid.
• Motion: Rick Geimer/Richard Ettema: 11-0-1
• New tacker is 19308
  • https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=19308
  • Motion: Rick Geimer/Richard Ettema: 12-0-0

Tracker 17831: Add a new definition of how OPTIONS requests should be handled

• https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=17831
• Balloter would like some description about the use of OPTIONS and the Allow header.
• Agreed to add some clarifying text that FHIR does not provide greater specificity for things such as options beyond what is in the base HTTP spec.
• Motion: Rick Geimer/Michael Donnelly: 11-0-0

Thursday Q3

Chair: Lloyd McKenzie Scribe: Lloyd McKenzie

SMART Publication Request

• Reviewed SMART ballot publication request
• Motion to approve publication request: Grahame/Josh
• Concern expressed by Isaac about the "readiness"/level of experience with the open-id connect portion for single sign-on
• Discussion about nature of concern and how best to address
• Decision to table to call on Monday

Adding "reported" to Request.intent

• Specifically, [GF#17169]
• Lloyd provided a background on the issue
• Jenni clarified that base issue was not "source of truth" but need to indicate that a record was 'reported'
• Concern that tags might be easily changed by man-in-the-middle
• Confirmation from Grahame that tags can contain critical business information, examples of "actionable" tag, security-related tags
• Discussion about whether a "reported" flag would be a "modifier element". Answer: no, because it doesn't change the definition of the element
  • Should include a meaning if missing for number of refills indicating that it's unknown - and thus dispensers can't presume to dispense more than a single fill
• Discussion of whether there's a difference between a "reported" resource - one that comes into being explicitly known as distinct from source of truth vs. "copies" which are clones of the source of truth.
• Lloyd indicated that he'd be comfortable with introducing a "reported" flag that could apply to most resources - orthogonal to Request.intent
• Pharmacy will discuss further.