2010-05-16 TSC WGM Agenda

TSC Minutes

HL7 TSC Meeting Minutes Location: Rio de Janeiro, Brazil Room:		Date: 2010-05-16 Time: Q5 - 6 to 8 pm local time)
Facilitator	Charlie McCay	Note taker(s)	Lynn Laakso
Attendee	Name	Affiliation	Email Address
x	Calvin Beebe	HL7 SSD SD	cbeebe@mayo.edu
x	Woody Beeler	HL7 FTSD	woody@beelers.com
x	Austin Kreisler	HL7 DESD	austin.j.kreisler@saic.com
x	Lynn Laakso (scribe, non-voting)	HL7 HQ	lynn@hl7.org
x	Ken McCaslin	HL7 TSS SD	Kenneth.H.McCaslin@QuestDiagnostics.com
x	Charlie McCay (chair)	HL7 TSC Chair	charlie@ramseysystems.co.uk
x	Charlie Mead	HL7 ArB Chair	meadch@mail.nih.gov
x	Ravi Natarajan	HL7 Affiliate	Ravi.Natarajan@nhs.net
?	Ron Parker	HL7 ArB Alternate	rparker@infoway-inforoute.ca
x	John Quinn	HL7 CTO	jquinn@hl7.org
x	Gregg Seppala	HL7 SSD SD Alternate	gregg.seppala@va.gov
x	Helen Stevens	HL7 TSS SD Alternate	helen.stevens@shaw.ca
x	Ed Tripp	HL7 DESD Alternate	Edward.tripp@estripp.com
x	D. Mead Walker	HL7 FTSD Alternate	dmead@comcast.net
x	Bill Braithwaite			.
x	Linda Fischetti			.
			.
Quorum Requirements (Co-chair +5 with 2 SD Reps) Met: (yes/No)

Agenda Topics

1. Introduction of visitors (including declaration of interests)
2. Agenda review and approval
3. SAIF
4. Evaluate quality plan and U.S. Affiliate efforts, and documenting what we think the work items will be.
5. agenda discussion for the remainder of the week and messages for the cochairs meeting
6. Woody has an IP presentation that could be advanced to the TSC articulating his understanding of the policy
7. Tooling

Supporting Documents

Minutes/Conclusions Reached:
Call to order 6:15pm

• Rearrangement of table for TSC members with guests seated at the perimeter.

1. Intros:
   • Linda Fischetti, Board member, interested in exposure to TSC
   • Gerald Beuschelt, Mitre
   • Bill Braithwaite, ES cochair, board, interested in how ES can help with technical strategy
   • Steve Hufnagel, MHS, SOA WG, interest in SAIF discussion
2. Agenda review
   • Ravi reported to Affiliates and hopes to receive feedback on quality plan by end of WGM. U.S. Affiliate, there were many expectations of artifacts from U.S. coming into international community. Helen asks if ONC cares if these are UV or realm specific. Probably they don’t care.
   1. 45 mins for SAIF
   2. 10 mins on quality plan
   3. 15 agenda for Monday night
   4. 10 mins for IP presentation
3. SAIF
   • Charlie McCay reports we’ve made progress in getting the SAIF materials out, getting content ready for review. Concerns with EA IP not making progress.
   • Charlie Mead reports from the ArB perspectiveCMead reports in Q3 says the step past getting the documents out for review was under discussion. NCI is working towards implementation, and the technology development is occurring, where Ken Buetow announced that SAIF was the stated method. ECCF Implementation Guide part of that, tooling selections and development for traceability.
   • NIH is bigger, not percolated up for acceptance there yet. NCI taking SAIF forward to meet their business requirements.
   • Woody comments there was an AMS meta-model of dynamic requirements a couple years ago but not enough progress. Woody concerned to diverge from NCI’s implementation constraining our specification development. CMead says they are operationalizing the SAIF. A team assigned to develop ECCF IG. The comments on Behavioral Framework are regarding how will they use it. NCI semantic infrastructure will need to be re-done. NCI will funnel their work product back to HL7 for review. Not much is ready for prime-time, they’re not as far at NCI as we may think. RIM, 29010, Vocabulary education is needed in order to proceed. caBIG compatibility guidelines were only operating in two cells in the PISM. ArB agreed that Steve Hufnagel’s draft of populating the ECCF was a good start. CMead says the SAIF alpha project has been less than hoped is the same reason at NCI, that people would have reasons not to do it, and now they are being told they must, top down, non-negotiable.
   • Calvin asks if the NCI product will be considered as suitable for HL7, CMead says it will be non-orthogonal but they are working on their own business requirements. Business architecture has never been fully articulated until now and so the enterprise architecture is now being built so that it can support it at NCI. It’s only an informative level coming from NCI. Austin notes the product quality project plan and he needs the governance framework. CMead notes it is locked in analysis paralysis and they are going to address.
   • Steve states that the VA and DOD in response to the FHA required prototypes of SAIF, so they are doing Immunization Management prototype, adding to an existing infrastructure. Publishing in the Pract Guide to SOA VolII through HSSP, using SampleHealth. They will have a working system in September. Government is reluctant to publish architectures publicly. DOD has a simple governance of decision control, change management, and set of reviews per DOD reg 5002. A straightforward governance model as such would be applicable to HL7. Change management control as with project scope statements, review for quality design review, test review, readiness review etc. Might be a suitable general governance for HL7.
   • Woody notes that we have pretty good DMPs with project reviews, committee reviews of ballots, balloting in multiple cycles. Can we figure it out later when we know what we’re dealing with? Right now it’s only a framework, what’s to govern? Let’s bring the artifacts in and then review the governance structure when we see how they relate to what we already do. #*CMead says NCI has distinguished business or domain governance and the architectural governance with relation of a particular project scope fitting into the enterprise view. Once business view determines a scope then the enterprise governance performs review. ArB sees their role only in the architectural governance not the business governance.
4. Alpha projects to contribute concrete artifacts, and not enough to evaluate vertical slices in relation to other projects.
   • Charlie Mead said he’d provide the NCI governance to Austin to bootstrap the process.
   • Steve adds that in discussion today we may want to have an offsite session to work through the products for the various viewpoints and come to consensus with conformance statements for a certain viewpoint some selection of products/artifacts. This would give a set of formalisms that provide criteria that indicate conformance with ECCF.
   • Ravi notes without the pressure of the top-down approach used at NCI how will HL7 adopt it. Woody says that’s what the TSC is for.
   • Linda notes that the OMB has just stated that all government agencies will use NIEM. Each agency has just created a document that states how they will comply with NIEM, her agency says yes we’ll use NIEM but for the healthcare components may also use the SAIF.
   • It’s extremely complex notes Steve. It cannot be picked up by the casual user.
   • Helen says the concern remains for the work groups as to how this will change their life and we don’t know what that is yet. With new projects and new participants in the next 6 months for ARRA initiatives we need to have our house in order for processes, balloting, what we deliver and what our artifacts look like. Trying to do it in 6 months would be disastrous to implement all at once, but we should identify what we want to have ready in 6 months and what we can tackle later.
   • Calvin asks if we’re choking on something too large; shall we focus on the dynamic model aspects and the relationship of SOA, CDA, messaging and map our existing toolkit into the 12 boxes and then go from there. We can be concrete in HL7 space on this based on HL7 history. We’re not getting the results out of the alpha projects we expected. Austin counters that the OO project was supposed to do just that but it was under-resourced and did not make progress.
   • Austin continues that we’ll tweak the process, not create new process but we are also needing to work on the quality of the product. Helen thinks that there’s a perspective that the SAIF implementation is a major shift. Steve feels this can be an educational outreach; #*Charlie mentioned they only did two boxes initially, and so the WG need to understand a project’s impact on other projects such as EHR-S FM, etc.
   • Charlie Mead states the major change at NCI was top down change, with historic operation in commonality with HL7 for community. Woody notes that we’ve told them what to do all this time but have not the ability to tell them what the impacts in their project are having on some other group. Discovering the overlaps is part of the governance process.
   • Charlie McCay asks what is the plan to deliver efficiently in the next 6 months and the longer term plan to implement the framework, reduce duplicate work, etc. PSS approval has already exposed some duplicate efforts so it can indeed work. Woody notes there is specific address in the roadmap efforts to address duplication and overlap. Charlie McCay adds the SAIF implementation artifacts will further expose such overlap. So, what’s going to happen in the next 6 months for the October and January meetings? What’s going to be developed in parallel for the long term implementation activity?
   • Mead notes with the draft architecture framework in hand, we need an architecture built in the context of that framework or we can’t determine what we need to do in the next 6 months.
   • Ed asks if there is a simplified way to analyze where we’re at and what to do in the short term. Perhaps a high-level process flowchart, showing the artifacts/viewpoints, what exists and where do we need to create them or how many examples of this type already have been used?
   • Ravi suggests we use the NCI framework review as a learning experience to modify for what HL7 needs.
   • Steve says you really need paid staff to do these tasks as volunteer effort results in a hodgepodge of products. NCI and DOD has paid staff/contractors dedicated to the effort. #*Helen asks what is needed to do this? Steve suggests someone work through John Quinn to do a BPMN diagram to document work process and products, identify architectural form that is consistent. Woody notes there are conflicting static models, but they’re internally consistent. Helen feels it’s an imperative that needs to be applied in the next six months, and it’s been said the behavioral model is a big piece that needs to be done. Is there something, like behavioral model, we can positively apply.
   • CMead notes the NCI stumbled into the behavioral model as their initial thinking was about data, but interaction of clinical trials systems they had to evaluate transaction management. The current NCI behavioral model is a nice sketch of something that could work, but it is not ready for deriving the metadata for computational use at run time. Might have a first cut of it in a couple months. NCI goal is to have runtime contract negotiation with computable metadata.
   • As NCI and HL7 are similarly interested in cross system interoperability it might be highly applicable.
   • Woody finds major shortcoming in three paradigms not dissimilar but not aligned. What happens when you receive a document, or a message or receiver responsibilities of a service is the basics of a dynamic model.
   • Mead asks, is it important for the TSC members to read and understand the SAIF documentation? Are we expected to submit comments to the ArB? May be a rhetorical question, but it sends a message. Are we having this debate, whilst really understanding what we’re talking about?
   • CMcCay notes we need to come up with what’s going to happen between now and October, between the 45 mins remaining and Tuesday lunch.
   • CMead asks if the ONC activity comes to fruition as anticipated, what is the expectation going to be? The NCI big turning point is the change management. Much was engaged in why should someone use datatypes, why must someone use an information model.
   • Austin thinks they’re looking for meaningful use guides, V2.51, V2.3.1, CDA, CCD. Will we identify in that process some changes to the underlying standards? Are we prepared to turn around changes to the standards in that time?
   • John states according to what he saw with people showing up at HITSP, you have people from vendors coming to learn about this area of industry as they see there will be profit in that sector. Blumenthal commented that if it wasn’t due between now and September it wasn’t an emergency. Lack of attention from ONC as to what they want until it becomes an emergency is his concern.
   • Woody feels we must build on the HL7 history of development based on the principle of model-based standards development, we have ISO datatypes, RIM, and vocab and binding to data elements; capability of paradigms under the aegis of information framework. Helen asks how do we resist the pressure to take shortcuts and do things a different way. How do we ensure with the quality plan that our products in 18 months show as much pride as the products Woody mentioned?
   • Ed says we think of change management in the organization as a major effort, avoid chicken little syndrome with ‘SAIF is coming’.
   • Alpha project approach still seems to make sense, projects engaged in realizing the framework; separately we have the business-as-usual environment we employ for new work coming in from ONC. Who will take the lead on documenting and presenting to the TSC on teleconference on these items? What volume of changes will there be, what impact on meetings and committees? Austin says we need to at least document those questions.
   • CMead says chances that people will show up to develop V3 messages is Zero. Different implementation than 11179 might cause people to turn and walk away. They will show up looking for V2 and CDA, and a few looking for services. Helen asks do we separate this work onto a separate track or shelve the existing work or some other approach? Calvin notes to enable the shift we need to pull out the V3 content and bind it to services; and separate the semantic content from the vehicle of exchange. Need to take the step to tell folks how to take their valuable messages and extract the content from the messaging.
   • CMcCay says who will take ownership? Tracking of a risk analysis document, what might happen, what do we need to do?
   • Ravi asks will the efforts be towards enabling a U.S. affiliate or will it be specifically to support a U.S. initiative. Helen feels it needs to be encouraged to come in at the international table.
   • Helen recounts the difference in the UK and Canadian approach the development needs to be done in HL7 Intl and not take it away into an affiliate and bring back only small pieces as with UK.
   • Production of status document presented to TSC each Monday on what those work items and scale, reading the tea leaves as it were and identifying our requirements. Helen likes the risk assessment and mitigation strategy idea but we need to focus on the quality project. What tools do we need or communication to the work group to promote the separation of content from messaging. Calvin states we need to stay on top of the education to assimilate new members to working with the new direction.
   • Gregg notes his TC with the Information Models they’ve developed, the registry they’re working on is a service candidate, and they need a SOA facilitator as well as messaging and documents facilitator.
   • CMcCay is more comfortable with the grasp we’re taking of the situation but does not have a message for the Co-chairs. Helen says CMead should take ½ hour and tell the committees concretely what they need to do at the work group level, starting THIS WEEK, differently than what they were doing in January. CMcCay says that needs to be part of a communication plan but may not be ready for tomorrow night.
   • CMead asks what would the TSC like to get from the NCI on this? Perhaps something from the ArB? Commonality between the paradigm stacks, suggests Woody. Structures usable for messages, documents or services? What do they need to consider to create services from existing message standards?
   • CMead says the best person is probably Ron Parker, with Canada’s V3 messaging experience. Andy Bond in Aus, Cecil Lynch in NCI, etc. ArB could bring those organizational contributions forward. They all are 3 months or more to the first usable implementation guides. NCI has 11179 perspective. SAIF just takes the content out and then worries about the transport mechanism separately.
   • CMcCay says we need time Monday night to describe.
   • Woody says Helen or Charlie McCay would be better to remove the technical bias. John notes it’s a tsunami warning and we’re watching for the impacts but it’s not an air raid siren just preparation.
   • CMcCay will present the context setting and impact analysis tomorrow night, with Charlie Mead and Ron there to answer questions about SAIF contexts. Gathering requirements as they become available and we’ll let them know right away.
5. Forward IP presentation and tooling to Tuesday lunch. Working out verification issues based on the state of the SMD, and may be resourcing projects based on our needs.
6. Monday night, Project Services has slides that have not been reviewed with Publishing, Vocab, and MnM to change processes. He’d like them removed from the slide deck. It’s the SAIF and Sound presentation and it was sent to the TSC, and Austin has commented. It should not be presented until after socialization.
   • Rene asked for time on the agenda for best practices of the Wiki.
   • Helen volunteers to give up her slot on the PIC report and allow Rene to speak. Ken keeps the Electronic Services Online Presence, Dave Hamill loses five minutes on SAIF and Sound. Cut Woody by 5 minutes. Electronic Services has not reviewed Rene’s presentation but Helen endorses it.
7. Recommendation: Search for HL7 Video, MIF for dummies. Rene has recorded some good videos.
8. Adjourned 20:14 BZT.

-->

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

Next Meeting/Preliminary Agenda Items .