FHIR Infrastructure Minutes WGM 201805

FHIR Infrastructure 2018 May Cologne WGM Minutes

Return to FHIR Infrastructure Minutes

Agenda

• see FHIR_Agenda_201805_WGM

Attendees

A comprehensive list for the WGM can be seen File:Fhiri - May 2018.pdf

Monday Q2

Chair: Ewout Kramer

Tracker Items

• 16165, 16127, 17001, 17002 - Versioning discussion; prepared motions for voting Wednesday

Prepared MOTION for wed may 16 Q2: "Document the MIME-type parameter as a way to describe the version of the resource in the MIME package. Grahame Grieve/Chris Grenz: 37-0-0"

Straw poll about using "Meta.fhirVersion" versus using "Meta.profile": 8 in favor of "fhirVersion", about 18 for a <profile> based approach.

Below is the example we discussed:

  Content-type: application/fhir+xml;fhirVersion=3.0

   <Patient fhir-version="3.0" xmlns="hl7.org/fhir">
     <meta>
       <profile value="http://hl7.org/fhir/DSTU2/StructureDefinition/Patient" />
       <profile value="http://hl7.org/fhir/StructureDefinition/Patient|1.0" /> 
       <!-- assuming no breaking change to patient in 4.0....  --> 
       <profile value="http://hl7.org/fhir/StructureDefinition/Patient|4.0" />
       <profile value="http://hl7.nz/fhir/StructureDefinition/OurNationalPatient|1.2.1" /> 
       <fhirVersion value="3.0" />
     </meta>
     <!-- more interesting stuff about patient-->
   </Patient>

  {
   "resourceType" : "Patient",
   "fhir-version" : "3.0",
   "status"...
  }

Monday Q3

Chair: Ewout Kramer

Tracker Items

• 16957 - Discussed whether GET and POST support should both be required on search
• 16369 - Discussed semantics of "significant figure"-based equality operations when searching for decimal, integer, and quantity values
• 14961 - Missing naming convention for search parameters assembling two or more words

Mon Q4

Chair: Lloyd Scribe: Lloyd

• Lloyd provided an overview of the workflow project
• Lorraine indicated that OO is starting up a project to map a number of their lab specifications and workflows to FHIR. Wondering if FHIR-I would like to co-sponsor, particularly from a workflow review/alignment perspective. ** Will discuss during Thur Q2 joint. Lloyd will give Grahame a heads-up
• II noticed in workflow alignment that there wasn't necessarily a good match for existing codes for status
  • reviewed current codes for Imaging??? and agreed they seem reasonable and are mapable to workflow states
• Discussed external records - i.e. records that are captured as secondary information that are not considered a source-of-truth
  • Explored the notion of a tag saying "not actionable", but this isn't really about actionability
  • This might be handled as a tag saying "authoritative"
• Jose provided a walkthrough of the ExampleScenario resource and editor

Tue Q2

Chair: GG Scribe: GG

Agenda: Task resolution

• [GF#17087] (Lloyd Mckenzie) - Bundle URL resolution: Persuasive - rework the rules, and add a good example for testing validation
• [GF#17085] (Lloyd Mckenzie) - Bundles and versions: Persuasive with mod - allow duplicates in all packages, but describe risks with non-versioned links
• GF#16862 (Amnon Shabo) - make ElementDefinition.constraint.expression optional: Persuasive with Mod - make expression optional, but make XPath trial use.

Tue Q3

Chair: Lloyd for first item, Grahame for remainder

Scribe: Lloyd

Tracker Items

Lloyd chairs

• [16055] - Drop ElementDefinition.binding.valueSetUri
  • Persuasive

Grahame chairs

• [15879] - Allow canonical to support display
  • Persusasive w/ mod

Grahame leaves Oliver leaves

• [13870] - Allow narrative in contained resources
• Persuasive w/ Mod

Tue Q4

Chair: Ewout

Scribe: Josh

Tracker Items

# Discussion begins at 1342

• Discuss GF#16331 -- process for choosing a canonical URL. Persuasive with mod.
• Discuss GF#16333. Persuasive.
• Discuss GF#17185. Logical models, and the term "logical models", and codes for StructureDefinition.kind. Persuasive.

Wed Q1

Chair: Ewout

Scribe: Josh

# Discussion begins at 0908

• Discuss GF#13989. Persuasive.
• Discuss Searching on references, uris, canonicals... with versions (Discuss GF#16361 and GF#17008)
• GF#16361 — Persuasive
• GF#17008 — Persuasive

Wed Q2

Scribe: Josh

Chair: Ewout

# Discussion begins at 11:13

Start by reviewing notes from Monday's discussion at wiki.hl7.org/index.php?title=FHIR_Infrastructure_Minutes_WGM_201805

We could indicate version information in the mime package, or in resource-specific attributes, or with Resource.meta.profile.

Lenel describes the case for payers to store and share resources across multiple FHIR versions.

Discussion: FHIR provides building blocks to do this kind of thing, but not for mixed versions within a single Bundle.

Detailed discussion about 3 options for meta.profile usage to convey version:

<meta>
  <profile value="http://hl7.org/fhir/STU3/StructureDefinition/Patient" />
  <profile value="http://hl7.org/fhir/3.0/StructureDefinition/Patient" />
  <profile value="http://hl7.org/fhir/StructureDefinition/Patient?version=3.0" />
  <profile value="http://hl7.org/fhir/StructureDefinition/Patient|3.0" />
</meta>

GF#16165 — Persuasive.

Thursday Q1

Scribe: Michael Donnelly

Chair: Ewout Kramer

• GF#16448 - Persuasive with mod. Extensive discussion is captured in the tracker.
• GF#16096 - Persuasive
• HTTP status codes 409 and 412 are swapped in the conditional update
• GF#14444 - Persuasive

Thursday Q2

Chair: Ewout Kramer Scribe: Lloyd McKenzie

• [GF#16868] - Not persuasive w/ mod
• [GF#17214] - Persuasive

Thursday Q3

Chair: @Ewout Kramer

Scribe: @Josh Mandel

Discussion begins.

Discuss Bundle.signature.

GF#16392 —This element allows for signatures within a bundle, rather than outside. There's a long-term recognize of the use case going back to CDA. We haven't seen this field being used anywhere in current FHIR implementations, so we have no practical experience with signing Bundles. This comment from @John Moehrke proposed dropping Bundle.signature.

GF#16519 is a related suggestion to move signatures to .meta.signature.

Or we could try to include a signature in Provenance resource as a Bundle.entry.

GF#16392 — Not persuasive with mod.

Discuss Binary.securityContext. GF#14563 — explanation needs to be more complete. This is a normative candidate. @Grahame Grieve and @John Moehrke will prepare a better description and share back with #fhir/infrastructure-wg .

Discuss Contained tags and security labels. GF#16622 — Persuasive.