This wiki has undergone a migration to Confluence found Here

20120912 WGMQ6 adhoc meeting notes

From HL7Wiki
Jump to navigation Jump to search

HL7 FHIR Gov Board

Ad-hoc dinner meeting Sept 12th Q6

Present: Woody, Ron, Grahame, Ewout, Lloyd

Invited: Bo

Meeting was called by Lloyd when he realized that no one was going to be left in Baltimore for the Friday afternoon Core Team and FGB meetings that were originally discussed.

Grahame: IHE has asked the FHIR team to render XDS as a FHIR resource. PA is now shifting their work plan to be primarily FHIR development


FHIR Gov Session: A lot of FHIR “governancy things” on the FHIR Wiki

  • 2 separate issues
    • Ballot rec on core FHIR
    • Outreach to the domain committees
  • Won’t get through the ballot comments at all
  • Considering offline rec by e-mail as this is just a draft for comment
  • Another topic is the subject of Persistent Resources which is Gary Dickenson is pressing for
  • Can’t do everything at once

FHIR Governance issue: how to manage expectations for pace and scope

Lloyd: there seems to be a move to have FHIR be the representation of Green CDA

Grahame: Not happy with the idea of a wg meeting focused entirely on outreach

Lloyd: I agree but we could overlap outreach simultaneously and then reserve time for core

Tutorials: next meeting ditch the intro, keep the Q4 and ramp up pre-requisites for participants. Problem, Jan meeting tutorial agenda is already set.

Ron: do the Intro in web video and don’t charge, charge for “hands on”. This may occur.

Lloyd: considering moving FHIR tutorials into Education Summits

Grahame: it is time for the Gov Board to take over the mantle on Education (planning) and management to execute. Ron agrees emphatically.

Woody volunteered to teach FHIR content, Patrick will probably be willing

Woody: high level concerns.

Lloyd: FHIR Governance should define who is allowed what to a resource is a management function. It is a critical function. Idea was to leave that function in the hands of the governance board.

Issue is that FGB needs to pull in the core team to build the initial resource list (which would probably take us 1.5 years to 2 years out. FGB will be the point of appeal from external parties on the list, priority, and time.

Ewout: make the resource definitions “good enough”.

Before you can figure out the scope and nature of resource WG’s have to be able to play, so we can’t do top-down planning until this knowledge has been disseminated.

May be a mixture of top-down (core team likely) and bottom-up requests for resources (membership and externals).

There may need to be a pre-requisite list of skill sets that could be embedded in the FHIR PSS template to avoid groups insisting on building and then needing massive hand-holding.

Need both a modeling and vocab facilitator assigned to each project. There may be an assumption that the modeling facilitator may also be the publishing facilitator.

To solve the skills pre-requisite problem, they could take the Q4 tutorial and allow it for only facilitators and co-chairs.

Ron note: I have a concern how we define what constitutes “goodness”? FHIR Methodology.

Resource development: Woody suggests a rule that each team checks in their work weekly.

This may not solve the problem of achieving consistent form, methodology, and quality.

Principle should be that people submit updates as often and as early as they can in order to propagate learning, regardless of the state or quality.

So, how do we achieve quality product? Draft for comment should be allowed to be messy, need it early. DSTU needs to be reasonably right and well formed. Final ballot needs to receive some kind of pass (Ron’s opinion)

We need to define the problem of bad builds and capacity of the core team to deal with them. FGB may want to define a Risk item.

Grahame: balloting… currently expectation is they go draft for comment ballot in Jan 2013, first DSTU in May. Do DSTU again 1, 2 years? Question: once you have done the first DSTU new resources will exist by the time of the second DSTU.

Topic of balloting all resources as a package. With incremental growth of the resource base, they will at least have to flag each artifact as to it’s state. There will need to be a property in the Resource itself that indicates the maturity state of each resource, (Draft for Comment, DSTU, Normative). Will there also be versioning?? The FHIR Profile identifies the version, not the resource instance itself. Conformance statements are mirrored to the profile version.

Schema is normative, schematron is normative, reference implementations are not normative. License is not normative it is a fact. Grahame is tempted to take the license down (now) or at least remove the statement that it is conditional (? Not the right word he used something else, read the current license).

What are we doing in the next WGM? Grahame: core team needs one quarter a day free for their own purposes.

FHIR Methodology has not begun. There needs to be a plan. Grahame will flip the authoring guide to MnM to start the methodology piece. Then figure out what is missing and build a plan to create it. This needs to occur before or at the next WGM. Everything is on the Wiki.

Grahame: if the methodology provides instruction on how to make the spec work, then it is also spec.

The will be FHIR certification for the implementation community. Education traditionally does this after they have normative content.

Discussion of using all day Monday at next WGM. Tuesday Q1 planning for address Q2 FHIR Governance and Management in separate, Q3 Methdology (or maybe on Monday). And then Q4 of each day for the core team. Q1 through 3 on Wed and Thursday will be distributed, with one core team member in a joint meeting with somebody. This may include multiple quarters with one committee. Big committees requiring more than one Q: PA, Patient Care, Pharmacy, OO. Small committees requiring 1 Q: Structured Docs, Devices , SOA, Clinical Genomics, FM, PHER. Woody suggests a FHIR roundtable as part of the facilitators meeting every Thursday night.

Had a meeting with Alean Kirnak in xparadigm. Grahame has put it to her that they can go to the connectathon in January. Alean’s team need to provide the core team with a bit more information. To be there.

FHIR will meet with ArB on Q2 of Sunday in January.