2011-05-15 TSC WGM Minutes

TSC Sunday Evening meeting for 2011 May WGM Lake Buena Vista (Orlando) FL, USA

back to TSC Minutes and Agendas

TSC WGM Agenda/Minutes

HL7 TSC Meeting Minutes Location:		Date: 2011-05-15-DD Time: 7-9PM EDT
Facilitator: Austin Kreisler		Note taker(s): Lynn Laakso
Attendee	Name	Affiliation
x	Calvin Beebe	HL7 SSD SD Co-Chair
x	Woody Beeler	HL7 FTSD Co-Chair
	Bob Dolin	HL7 Board Chair (member ex officio w/vote)
	Chuck Jaffe	HL7 CEO (member ex officio w/o vote)
x	Tony Julian	HL7 FTSD Co-Chair
x	Austin Kreisler (Chair)	HL7 TSC Chair, DESD Co-Chair
x	Lynn Laakso (scribe, non-voting)	HL7 HQ
x	Patrick Loyd	HL7 T3SD Co-Chair
x	Ken McCaslin	HL7 TSS SD Co-Chair
x	Charlie Mead	HL7 ArB Chair
	Don Mon	HL7 Board Chair-elect (member ex officio w/o vote)
regrets	Ravi Natarajan	HL7 Affiliate Representative
	Ron Parker	HL7 ArB Alternate
	John Quinn	HL7 CTO (TSC member ex officio w/vote)
x	Gregg Seppala	HL7 SSD SD Co-Chair
x	Ed Tripp	HL7 DESD Co-Chair
x	Jay Zimmerman	HL7 Affiliate Representative
Quorum Requirements (Co-chair +5 with 2 SD Reps) Met: yes

Agenda

- ArB

• SAIF Book ballot update
• ArB "What's next"

Supporting Documents

See links

Minutes/Conclusions Reached:

Austin called to order at 7:20 PM

1. Ballot reconciliation update
   • Charlie gave an update on the four quarters the ArB met today.
   • Q1 talked about ballot trajectory. John would like the canonical SAIF to go DSTU in September. What conformance statements are necessary; what would it take for an implementation guide to be conformant to that specification. Discussion today on what it meant and was it possible occurred during the first quarter.
   • Q2 discussed reconciliation comments especially Grahame and Lloyd’s. Themes emerged with Lloyd’s concern with the jump from SAIF canonical and the HL7 IG. There was some agreement on a bridge document in between; Ron felt the ArB needed to assist with the HL7 IG. Some suggested an informative appendix to the SAIF book. Canonical SAIF should allow modeling by constraint as well as the Open EHR method of subtypes. Cecil was there, and authored the IF chapter – he admittedly injected HL7 prejudices into the IF and committed to remove them.
   • Q3 had two activities. One was thinking about the rewrite of the SAIF book taking into account all the comments, how would it look? Then they continued on with ballot reconciliation. Some parts of reconciliation process are not about SAIF but about change management. Tony will attempt to negotiate with Implementation and Conformance and Frank Oemig. TSC will have to take the reins of change management. RM -ODP viewpoints as dimensions, with conceptual/logical/implementable as perspectives are the approach. It is not necessarily normalized; artifacts can occupy or overlap boxes. In your implementation guide you define where it lands. One ballot comment was regarding a non-normalized matrix. The TSC needs to support the ArB’s approach. Discussion ensued on whether the ballot reconciliation should be endorsed by the TSC prior to posting as the reconciliation package or should the Arb post the reconciliation and leave the TSC as the appeal body. The TSC can review the package and endorse it.
   • The ArB will probably have a reconciliation package by June 13th.
2. Ballot trajectory to September 2011
   • SAIF canonical should be on normative track according to John because other organizations will take it more seriously. Ballot content would be needed by July 24th. Discussion on the presentation of the ballot information on including it on the V3 ballot site, perhaps as a different name than “V3”, ensued. The proposed approach is to take the current material and rewrite/reorganize, provide conformance statements, and request publication as Informative as Release 1 while we submit for DSTU as release 2 on the Normative track. DSTU can run in parallel with the development of the SAIF IG. The project scope statement will need to be amended to reflect the additional ballot intent and additional documents.
   • Motion: Patrick moves and Woody seconds the approach to the ballot trajectory. Publish release 1 as informative with the current ballot material, reconciliation, and conformance statements; and initiate release 2 as DSTU from that same package, with the modification of the scope statement.
   • Ed asks if there will be additional changes beyond the reconciliation into the DSTU. The conformance statements will be added on top of reconciliation modifications. Conformance statements are going to be asked of all normative track ballots going forward.
   • Discussion ensued on which point the material is frozen for informative release 1. The reconciliation modifications without the conformance statements will be informative release 1. They intend to write a conformance section to each of the four chapters.
   • Vote: unanimously approved
   • Charlie adds the ArB agreed today and John was fine with it: a woman at NCI is technical editor for their SAIF IG, finding technical inconsistencies, 508 compliance, and is willing to do it. 3 of 5 authors with whom it was discussed were willing to work with her even though she’s tough. The other two weren’t familiar with her. Her name is Ann Wiley.
3. Proposed SAIF Canon revision/ToC
   • spent an hour in Q3 based on collective comments discussion. Mapped out a revised structure for the document, quite different from what was there before. Had some negative voters because they had read the wrong document, expecting an IG. Glossary organization is there because there was not enough coordination on the concepts floating around in all the sections in the document. Charlie discussed with the ArB to take on a formal approach to the glossary in the updated document. Concept maps are in the beginning of each chapter. There is an ability to create an OWL from the concept maps; with experience at NCI they uncovered inconsistencies by creating the OWL. Woody’s concern is that glossary management for HL7 in general using a different tool. There are circumstances where a preferred term means something different in a different context. Need a way in publishing as well to include the definition into the body of the document where a singular definition change rolls into all instances where the term lives in the document. Charlie asks if there is a coherent strategy and tool; Woody says there is not. The rudimentary tool they’ve been using produces a rudimentary glossary with the V3 ballot; opportunity to use a more elaborate schema has not materialized. Charlie says the SAIF canonical is posited as building grammars for HL7 IGs, so they wanted to create a more robust tool and more rigorous process. Calvin points out that the OWL components can be discarded when they publish it, using it as a quality check but won’t be part of the specification. Charlie says they’ll maintain it internally in OWL but the OWL is not part of the canonical representation.
   1. Intro
      1. Scope
      2. Goals
      3. Approach
      4. Glossary organization (not part of the document – see discussion above)
   2. Foundational Principles
      1. “Service-Aware”
      2. Describing Complex Systems
         1. Dimension
         2. Perspectives
         3. Interoperability Stack (formally Specification Stack)
      3. SAIF Specifications
         • Grammars and Grounding
           • Charlie described pulling out the various grammars and wanting to describe the intent of the effort, to describe the foundational ‘stuff’
      4. SAIF Implementation Guides
         1. Purpose
         2. Relation to Canonical SAIF
         3. Scope
            1. What is in canonical model?
            2. What is in IG?
            3. What is in “the results of operationalizing an IG?”
            4. Conformity assessment (IG-to-canonical)
      5. ISS (Interoperability Specification)
         • The ECCF is not just a matrix but it collects artifacts for conformance and compliance
         1. Subject and Type (domain of canonical)
            • this discussion was the scope of the domain of the canonical versus the IG; subject can be enterprise, application, service, etc. which you can type and then profile – the profiling doesn’t occur in the canonical but in the IG.
         2. Profile (domain of IG)
         3. Instance (domain of architecture)
   3. Behavioral Framework
      1. Community
      2. Roles
      3. Behavior
      4. Interactions
      5. Policies
      6. etc
   4. Information Framework
      1. Domain
      2. Models
      3. Constraints / Subtypes (Refinements is the term suggested by Woody)
      4. Localization
      5. Classes, attributes
      6. Datatypes
      7. Vocab
      8. etc
   5. Governance Framework
      1. Core Processes
         1. Conformance
         2. Education / Communication
         3. Escalation
         4. Vitality
      2. Precepts
         1. Objectives
         2. Policies
         3. Guidelines
         4. Standards
      3. People / Roles
      4. Process
      5. Metrics
      6. etc
   6. ECCF
      1. IG Conformity Assessment
      2. Compliance
      3. Correspondence
      4. Etc
   7. Glossary

Charlie sent a graphic to Grahame yesterday about the deployment context – interoperability roadmap where the level of the compliance and deployment range from local, intra-enterprise, cross-enterprise, or community. The ISO use of the words conformance and compliance is very specific and are lumped as subtypes under conformity. Charlie notes that we’re going to use the term of conformity; an IG is not really an implementation as a conformance is to a specification, but is instead a ‘compliance’. Woody notes that some say compliance is equated to adherence to regulation in a stronger sense; and cautions with its use. As discussion ensued it was revealed that the clarification of the definitions should be in the glossary to sort the notion of regulatory compliance as compared to the ISO definition of compliance.

Grahame’s quote of the day as “decision-based evidence-making”

Actions (Include Owner, Action Item, and due date) .

Next Meeting/Preliminary Agenda Items .