20090115 arb orlando minutes

Architecture Board

Thursday, January 15, 2009

Q3, Q4 Gardenia Room

Error creating thumbnail: File missing

Attendees

Name	Present	With	Affiliation	E-main address
Curry, Jane	Yes	ArB	Health Information Strategies	janecurry@healthinfostrategies.com
Grieve, Grahame	?	ArB	Kestral Computing	grahame@kestral.com.au
Julian, Tony	Yes	ArB	Mayo Clinic	ajulian@mayo.edu
Koisch, John	Yes	ArB	NCI	koisch_john@bah.com
Lynch, Cecil	Yes	ArB	ontoreason LLC	clynch@ontoreason.com
Mead, Charlie	Yes	ArB	Booz Allen Hamilton	charlie.mead@booz.com
Orvis, Nancy	Yes	ArB	DOD	nancy.orvis@tma.osd.mil
Parker, Ron	Yes	ArB	CA Infoway	rparker@eastlink.ca
Quinn, John	?	ArB	Health Level Seven, Inc.	jquinn@HL7.org
Shakir, Abdul-Malik	?	ArB	Shakir Consulting	ShakirConsulting@cs.com
Walker, Mead	?	ArB	Health Data and Interoperability Inc	dmead@comcast.net
Yongjian, Bao	?	ArB	GE Healthcare	yongjian.bao@med.ge.com
Boyes, Paul	Yes	Guest	NCI/BAH	boyes.paul@bah.com
Maclsee, Peter	?	Guest	?	peter.macisaa@eds.com
Mukooyo, Eddie	?	Guest	?	eddie.mukooyo@healthgo.ug
Neel, Lyssa	Yes	Guest	HL7 Canada>	pneel@infoway.ca
Jorgenson, Don	?	Guest	Inpriva	djorgenson@inpriva.com
van der Zel, Michael	Ues	Guest	University Hospital Groningen, The Netherlands	m.van.der.zel@ict.umcg.nl
Bear, Yogi	?	Guest	US Dept. Interior, Park Service	yogi@jellystonepark.gov

Agenda

1. Call to order
2. Approval of Agenda for quarter(s)
3. Wrap Up, Open
4. Next Steps
5. Telcon Schedule
6. Preliminary Agenda for Kyoto WGM
7. Preliminary Schedule for Kyoto WGM
8. Adjournment

Call to order

The meeting was called to order at 13:51 U.S. EST by John Koisch with Tony as scribe.

Approval of Agenda for quarter(s)

The agenda was approved by affirmation.

Wrap Up, Open

• Nancy Orvis: Joint SAEAF implementation pilot project between EHR/Government projects Work Group. Will discuss with SOA/HSSP. Done initial discussion with Don Mon. DOD has an in-kind resource. Believe that Galen Mulrooney will add services perspective. Will forward next steps to TSC. Will need to validate 2.5.1/3.0 to EHR functional model.

• Mead Walker: Will be asking HL7 to fund the project?

• Nancy Orvis: Would like to ask HL7 to fund gap analysis. End of year fuller analysis, and relationships to service. Don Mon can get donations to do the work if

• John Koisch: What are the phases?

• Nancy Orvis: Public health/Emergency response - to go through more detailed. Phase 2 will be a candidate pilot project.

• Ron Parker: Will be doing alpha project in services metaphor: just doing business requirements. Need subsequent project to do messages et. al. Elapsed timeline very long - one year.

• John Koisch: We need an end-to-end project. Nancy is about to show a project which will define a non-SAEAF project. PHARMA wants to put their messages through SAEAF.

• Mead Walker: Who?

• John Koisch: Hugh Glover was interested in that. Then the OO composition work that Patrick Loyd wants to bring forth, CTS2, NCI proposed services to support national patient registry. 4 projects, that occupy different places. PHARMA will use existing artifacts as will EHR and OO.

• John Koisch: Question? Do we want to recommend to the TSC that they need to adopt a more general notion of Alpha projects, the re-factoring project, or turn folks away, or third option.

• Nancy Orvis: Project to start with EHR-FM to propose services of interest to the healthcare domain, and propose a portion as a health domain model, that would show the health lines of business across government.

• Nancy Orvis: The relationship includes a definition of services to healthcare, mapping the HL7 2.5.1, missing the V3 messaging map, other artifacts, and a map. Are there missing references. There would be a reference service taxonomy. Individual entities could have their own services, relate to the outside world through the HL7 taxonomy.

• Nancy Orvis: There are multiple service layers. The mapping will expose the cross reference. There are pieces in the SAEAF.

• Nancy Orvis: PHER would like to be a case study for a second domain of a system model.

• Mead Walker: What this would do would be to find the gaps in the EHR-FM?

• Nancy Orvis: To create a set of services in the healthcare domain that relate to EHR-FM functions. Multiple cross links - . Used Rubics cube of life as a reference.

• Cecil Lynch: there is a need across the healthcare realm to map to this.

• Ron Parker: You are right.

• John Koisch: There are parts of your IT platform will do parts of this.

• Cecil Lynch: The other thing is the tremendous activity to specify requirements in HITSP.

• Nancy Orvis: waiting for outcome of discussion. Starting with list of descriptions, tie to 2.5.1 data dictionary. The Domain experts/EHR work group will be starting. Phase 1 probably could be funded with current funds. How do we relate the EHR-FM to other artifacts in HL7? This would become an IT portfolio tool. What mechanism do we use to test a new interface to the specs? It would be good for transition planning.

• Nancy Orvis: If NHS/VA/Canada could contribute it would be good.

• Ron Parker: We have done some low level stuff - we have a SOA model, but the ESB was designed to take whatever, we have done the mapping: soap headers, wrappers. We have not determined what to expose as services?

• Nancy Orvis: Do I do it coupled, or as services?

• Ron Parker: we have not evolved prescriptive expression of the contract: we now have a framework. I have a guy who is competent in SOA to craft the contracts. Is the expressive, can we surface it? We will be doing services for management.

• John Koisch: Arging with Grahame that the dynamic model does not explain human workflow. Had day-job meeting to discuss cross/institution cancer research.

• John Koisch: We have put in front of everyone a different way to think about these things. What you are doing is going to be creating chunks an pieces of this. If you go through your thing, and all you get is models, it seems to me that someone is going to go 'I can use that'.

• Ron Parker: I have to produce a gap analysis of what the standards industry can do. The key thing is that this will be one of the first times the requirement will be fully expressed.

• John Koisch: So what would Mark Koehn say?

• Ron Parker: What would be important is that they articulate the timeline/envelope of the project. He will want to constrain it. The consumers don't want to wade through a pile of great stuff, so 3-4 are reasonable.

• John Koisch: I acknowledge that, but I am seeing something else - great, are we done yet?

• Ron Parker: People are ready to go. Give me the template artifacts. They love the idea:

• John Koisch: In InM Paul said this is an organizational framework that deals with standardizing. He did not hear it. You have to repeat this 10 times.

• Ron Parker: Once Mark gets going

• Cecil Lynch: People are trying to find out where they fit. Regardless of a reference implementation, take a use case, and map it to the boxes in the SAEAF matrix.

• Nancy Orvis: How does this relate.

• John Koisch: What is Mark's timeline?

• Ron Parker: 4-6 elapsed weeks to get it framed and circulated.

• John Koisch: Deliverables are in Kyoto? Kyoto is process review.

• Ron Parker: There will be a large contingent not in Kyoto.

• Nancy Orvis: What is the formal decision?

• John Koisch: Presented the TSC minutes of the Wednesday meeting. [[1]].

• John Koisch: The timelines et. al. is up to Mark. We have DOD, NCI, Info way, OO. There will be an issue in Kyoto - we are tired of waiting.

• Ron Parker: Lets let him mull it over - he will deliver the implications of the choices to the TSC. He pointed out that his timeline/budget needs a preliminary communication plan. We need to communicate again & again & again.

• John Koisch: Do we need F-F before Kyoto?

• Mead Walker: We need to discuss the ties to the existing standard, and how it makes a difference to what people do.

• John Koisch: One thing - the ArB is not going to be doing the business architecture modeling - it will be informed by the alpha projects.

• Nancy Orvis: TSC will do the process modeling

• Mead Walker: It will be in parallel. If things go

• John Koisch: The alpha projects are a recognition that the merry-go-round is spinning going downhill. They have the business architectural model.

• John Koisch: We need concrete representations of the containers.

• Mead Walker: Do we have an architectural framework?

• John Koisch: To be delivered Monday.

• Mead Walker: It will explain how the framework is important to HL7?

• John Koisch: Yes

• Nancy Orvis: Will not be fleshed out, but should have all of the pieces.

• John Koisch: If there are artifacts that the alpha projects need, the ArB is responsible, but may not do the work.

• Mead Walker: Ioana wants to see the documentation of the disposition.

• Tony Julian:: They are published on the wiki in the minutes.

• John Koisch: The behavioral framework(BF) was supposed to be published this week but it will be rolled in the SAEAF.

• Mead Walker: why not a separate document?

• John Koisch: We have discussed that - we will roll them together, and then split them apart.

• Mead Walker: It should be multiple documents - BF should be a priority task, the roll back in.

• Mead Walker: Should have a review of the BF , distribute, and discuss in a face-to-face.

• John Koisch: Dedicated time to discuss that. What about other parts of the SAEAF?

• Mead Walker: Don't need a F-F to discuss the pilot projects.

• Ron Parker: BF needs to be surfaced to the membership. To get them to own the problem with the BF. I agree that it is a high priority. Between now and Kyoto I will have some problems.

• John Koisch: Meeting adjourned at 3:00pm

Q4 Next Steps

Milestone Dates	Date
Projects to ballot in May cycle need to be submitted	January 29, 2009.
Notification of Intent to ballot due	February 22, 2009.
OHT face-to-face	March 9-11.
Harmonization	April 14-17.

• John Koisch: Two interesting discussions: Suggested meeting dates: Harmonization is April 14-17 2009. There will be less brainstorming and harder deliverables. There will be candidate artifacts emerging from Infoway and NCI that have a place on the SAEAF matrix.

• Jane Curry: We could say that those who are candidate early adopters identify your place on the matrix.

• John Koisch: We should do 4/15 - 4-17 - wednesday through friday. Taxes due day! We can request funding.

• Charlie Mead: They are ok with funding a F-F every calendar quarter. We must have concrete goals.

• John Koisch: I will get confirmation.

• John Koisch: We have 6 alpha projects. The notion was we have submitted criteria for what an alpha project should be.

• Jane Curry: We have not communicated.

• John Koisch: Saturday TSC meeting.

• Jane Curry: There was no motion to that effect.

• Charlie Mead: One of the notions of the Phase zero project is to develop criteria for alpha project:
• Crosses 2 or more work groups
• End-to-end business daces
  • Crisply defined scope
  • Dedicated resources
  • Not already under way
• Not about foundational or structural frameworks - ISRS and/or behavioral specification.
• Hit as much of the SAEAF as possible

• Ron Parker: HITSP is struggling the relationship with use cases and SOA. Ed Larson is pushing the SAEAF to HITSP. Expressing EHR-FM in SAEAF would be a HUGE help.

• Mead Walker: EHR-FM is starting to realize that they must understand messaging.

• John Koisch: Candidates are:
• DOD VA EHR-FM
  • Message portfolio
  • Creates analysis specs for prioritized EHR profiles
• Pa Registies (More than one)
  • Most of the analysis done, service interfaces
• COPPA - National Person Registry
  • Captures RIM semantics, uses instances of the Behavioral Framework(BF).
  • Set of artifacts already done
• OO
  • Complex Behavioral models(instances of the Behavioral Framework(BF)) and access patterms
• Infoway - Patient access to quality care (PAQC)
  • Messaging expressed in a services metaphor
• CTS2
  • Terminology Services
• PHARMA Medicine Identity Reconcillation
  • Pushes on OMG harmonization
  • Uses EIS for medicine identification
• PASS
  • Directing OHT to pull together the work from MCI, Infoway, NeHTA, NHS
• PHER
  • Immunization registry.
• John Koisch: TSC wants list.

• Mead Walker: If ArB is not involved can they still go forward.

• John Koisch: ArB is responsible for the specifications necessary, even if produced by someone else.

• Charlie Mead: Requirements for each of the projects depends on the project. They need paid fulltime management to oversee the alpha projects, Education.

• Charlie Mead: The paid person will supervise the alpha projects - like a project manager. We are not on the hook, but we will need to help the phase0 team to determine the requirements and success of an alpha project.

• John Koisch: I see some overlap - PA registries should be able to advise COPPA. PAQC will be using PA registries, possibly COPPA. OO will use PA Registries.

• Charlie Mead: What is OO doing?

• John Koisch: OO is trying to capture the lab work into the SAEAF.

(Add to project list) DOD is nancy, PA is gregg seppala, COPPA charlie Mead, OO Patrick LLoyd, PAQC is ron parker, CTS2 is russ hamm, MIR is Hugh Glover

• Mead Walker: I dont think we should block just because there are so many. We need to know more about them, and they should know about the SAEAF. We should meet with them in telcon.

• John Koisch: Should be part of the rollout.

• Mead Walker: Nancy was thinking about asking for funding for first project?

• John Koisch: I am not sure. ArB should not have a say in funding or prioritzation. We think PA Registries should be done before CTS2.

• Ron Parker: We should have a good idea of the nature and process we put to a alpha project. Mark Koehn will have a say on how many, and ArB involvement.

• Jane Curry: We still have a stream of work.

• Ron Parker: John Quinn says we need to be an archecture review board.

• Tony Julian: Backward compatibility?

• John Koisch: Yes, in parallel with the BF. Lots to do, lots to talk about. Maybe we need to say something to TSC about that - maybe we need dedicated resources.

• Jane Curry: We need to go through the SAEAF, and identify the mature artifacts, with methodology and quality,: Identify candidates for improvement; Which ones don't exist.

• Mead Walker: We should say what in this project will demonstrate the usefulness of what we did? What are they testing?

• Ron Parker: If there is a problem with the project, was it a function of the SAEAF.

• Jane Curry: If you have a framework, you have the project define which artefacts they are going to:
  • Leverage
  • Test
  • Contribute

• John Koisch: This tests the artefact as well as the SAEAF.

• Jane Curry: We need to identify artefacts that are mature, and which are missing.

• John Koisch: Security would be talking about policy language.

• Jane Curry: It would be like CDA R1 - cover the concept, not the rigor.

• John Koisch: Who brought up looking at this?

• Cecil Lynch: We need to define these for componentization - but not in this session. I am thinking ontologically about how these need to be done.

• John Koisch: We can evaluate a CIM/LIM with great rigor.

• Mead Walker: With great rigor?

• John Koisch: With some rigor.

• Cecil Lynch: Cleaning up the mess - we need to document the failures if we don't succeed, or learn from them.

• Ron Parker: Problem with alpha is hand holding. They will expect a lot of it.

• Jane Curry: We will have enough problems supporting the rollout projects. Are we going to commit an hour a week?

• John Koisch: How many man hours? How much hand holding? It was HARD. We started on week 1 'we need to be on the same model. On week 4 'we need to be on the same model' by week 8, the development team finally decided 'we need to be on the same model'.

• Ron Parker: How many alphas, under what circumstances, with a mentor for each.

• Jane Curry: Since this is a semi-volunteer group, we have the project designate a participant as a ArB liason.

• John Koisch: Which was what Paul did - advocate.

• Jane Curry: The advocate reviews with us.

• John Koisch: Should we offer a day in april for them to sit with us.

• Mead Walker: No, we should offer them conference calls. We would get a 2hour overview of a project, which would be waste of our time. On a telcon, we could be more focused. What do you want to know.

• Cecil Lynch: Before the concall they should have a template.

• Jane Curry: Explain how they fit in the following
  • Leverage
  • Test
  • Contribute

• John Koisch: We have a matrix, but we have a bit to tackle. NCI and infoway should bring forth artifacts in a metered way "we have a candidate for this square". Each member should do a square on the wiki.

• Mead Walker: What?

• John Koisch: If I put forth a candidate recommendation for Collaboration analysis, someone must review it. IF we have one person do it, one review it, then we will be good in April.

• Jane Curry: We can use the wiki for things that are collected elsewhere.

• John Koisch: SAEAF is M2, projects are doing M1, and we need to do Mzero.

• Mead Walker: How to's?

• Jane Curry: We don't have the how-to's?

• John Koisch: I will volunteer a few pieces

• Jane Curry: On the main page document the owner

• Tony Julian: I will have page ready tomorrow morning.

• John Koisch: Just get something out there. by the time we meet in april.

• John Koisch: That is our artifacts.

Q4 Continued

• John Koisch: IN the remaining 30 minutes should we prioritize?

• Ron Parker: Not up to us.

• Charlie Mead: We should give the TSC a notion of the focus of the project (why we care). We should tell them whether or not from our understanding of the project what the maturity of the project is.

• Cecil Lynch: Each of us should, as we review our boxes, determine whether it applies to our box.

• John Koisch: By next week?

• Charlie Mead: The TSC is hoping to have the funding question figured out by next week.

• John Koisch: When you were gone, we talked about the projects appoint a ArB Liason. There is concern about enough bandwidth to manage, as well as the hand-holding. COPPA has been through this and has artifacts, as does CTS2, the MIC is imature, as is OO, and will struggle.

• Jane Curry: They are looking about guidance.

• John Koisch: Patrick Loyd sees the possibility that the behaviorable patterns as well as service patterns.

• Jane Curry: What happened to Grahames Idea?

• John Koisch: It was endorsed, but not enthusiastically.

• Jane Curry: The idea that we simplify payload across domains. If you work from the clinical statement, you could use CDAR3 as the payload. Then you could work out which is documents, messages, or services.

• John Koisch: The human process goes with which paradigm, the granularity, state machines.

• Jane Curry: Those patterns

• John Koisch: presented the work From INM/SOA/ITS/IC/MnM Q1 Thursday. Tony will post.

Telcon Schedule

Stays as is

Kyoto Schedule

TBD -

John Koisch will send an e-mail to the list to determine who will attend in Kyoto so we can do room requirements NLT 26Jan2009.

Adjournment

• Meeting was adjourned at 5:00pm U.S. EST. Tony 22:01, 15 January 2009 (UTC)