MnM Minutes CC 20150312

M&M Conference Call 1:00PM Eastern Time (Date above)

Return to MnM Minutes

Agenda

• Approve Minutes Prior Meeting on 03/05
• Calvin's ballot comments:

5160 Overly-specific models

5156 Inconsistent names

5158 Lack of conceptual & logical models

5157 Document vs. resource

• Identifier Types

5353 Add "type" to Identifier

• Revisit:

5654 Integers for negative numbers

• Items dropped from Block vote for discussion

5664 Identifier use cardinality should be 1..1 (Explain - no change)

5500 Remove reference to range flag (Accept)

5533 Should Given and Family name fields of HumanName be strings, not IEnumerable<string>? (Add better documentation)

• Timing proposals:

3841 Revise Timing Structure

5347 Add range for frequency

5673 With Meals

Attendees

• Lloyd McKenzie (chair)
• Calvin Beebe
• Daniel Lanphear
• William Kinsley
• Grahame Grieve (2nd half)

Minutes

Motion: Approve minutes of 03/05 call: Minutes Daniel/William 3-0-0

Calvin's comments

Tracker item 5156

Discussed the fact that names are allowed to vary - where needed to be clear to domain experts, but that there is also an expectation that names will be consistent where there isn't a pressing domain need for inconsistency and that checking for consistency (across all of FHIR and especially within resource "families") is part of the QA expectations for DSTU 2.

Persuasive with mod Calvin/Daniel 3-0-0

Tracker item 5157

The FHIR group feels that they're still on target to meet their target of 100-150 resources. The larger artifacts are a manifestation of the fact that there are certain "large" constructs that a commonly exchanged and FHIR aligns with these existing patterns (e.g. sections in CDA/CCDA; archetypes in OpenEHR). There are a clear set of criteria used to guide when the creation of a new resource is appropriate and an approval process to help avoid resource proliferation.

Not persuasive Calvin/William 3-0-0

Grahame joins

Tracker Item 5158

Consistency (where appropriate) is a signficant concern with FHIR. The question is what mechanism is used to drive consistency. With v3, the approach was top-down and tooling/model-driven. With FHIR, the approach is more iterative. The risk is being addressed, but via a different mechanism. Certainly mappings will be provided to v3 domain models where there is implementation of them, but the design will not be driven by the v3 modeling.

We will add an initial placerholder for documentation about SAIF to the FHIR Methodology wiki and will include a direct pointer to the methodology page in the spec.

Persuasive with Mod Calvin/Daniel 4-0-0

5160

We're not in a position to address this as part of DSTU 2. However, the strategy and approach will have to be in place before we can go normative.

Not persuasive with mod Calvin/William 4/0/0