20090212 arb telcon minutes

Architecture Board

February 12, 2009

Logistics

• DTM: Thu Feb 12, 2009 03:00 PM (US Eastern Time, GMT -5)
• Phone Number: 770-657-9270
• Participant Passcode: 854126

Agenda

• Call to order
• Roll call
• Approval of agenda
• Update: SAEAF
• SOA workgroups practical guide
• Adjournment

Attendance - Roll Call

Name	Present	With	Affiliation	E-mail address
Curry, Jane	Yes	ArB	Health Information Strategies	janecurry@healthinfostrategies.com
Grieve, Grahame	No	ArB	Kestral Computing	grahame@kestral.com.au
Julian, Tony	Yes	ArB	Mayo Clinic	ajulian@mayo.edu
Koisch, John	No	ArB	NCI	koisch_john@bah.com
Loyd, Patrick	No	ARB	Gordon point Informatics LTD.	patrick.loyd@gpinformatics.com
Lynch, Cecil	Yes	ArB	ontoreason LLC	clynch@ontoreason.com
Mead, Charlie	Yes	ArB	Booz Allen Hamilton	charlie.mead@booz.com
Nelson, Dale	Yes	Arb	II4SM	dale@zed-logic.com
Ocasio, Wendell	Yes	ArB	Agilex Technologies	wendell.ocasio@agilex.com
Parker, Ron	No	ArB	CA Infoway	rparker@eastlink.ca
Quinn, John	Yes	ArB	Health Level Seven, Inc.	jquinn@HL7.org
Shakir, Abdul-Malik	No	ArB	Shakir Consulting	ShakirConsulting@cs.com

Call to order

The meeting was called to order at 3:00pm U.S. Eastern by Charlie Mead with Tony Julian as scribe.

Roll call

The chair called the roll as reflected in the attendance chart.

Approval of agenda

The agenda was approved by affirmation.

Update: SAEAF

CM: I spent a fair amount of time to come to terms with the conformance/compliance framework (deck 3). I have to say I have hit a brick wall. Jane and John helped me, as a result of a number of things, to come to a naive deck. I hope to send out the new&Improved deck in the next 2-3 days. What is the difference between conformance and compliance. Wendell pointed out the ISO usage. The current definition of Conformance when you certify conformance, the various viewpoints have collected conformance assertions, which can be bound to the specification. We have layered conformance with the viewpoints. The passionate issue is that most implementers only care about conformance.

CL: We have a definition in RM-ODP.

Charlie Mead: Thanks to Andy Bond and Wendell, what compliance is is a measure given a standard that is declared, someone else decides to modify in a compliant way extending it or constraining it: you can only talk about compliance between specifications. Wendell Ocasio: According to ISO, there is no such thing as compliance.

Charlie Mead: Yes

Wendell Ocasio: The rest of the colloquial use of the compliance is outside the definition.

Charlie Mead: We had five definitions. You are worried about compliance on a single layer. I am hoping that the slides I am sending out will be clear and concise, and that we have it nailed. Alan Honey sent me his latest tweaks. I am hoping the behavioral deck is done.

Jane Curry: Working on the governance deck: More complex than I thought. There are a number of governance concepts that must align. It will be 15 slides instead of the 6 I originally envisioned. It will help describe the different governance/scope.

Charlie Mead: In the next week we will have the remaining 3 decks. Once they are stable, the documents will have the same structures and pictures, and the documents will be straightforward if tedious. Does that sound ok?

Jane Curry: We should include going back over the peer-review to make sure we answered everything.

Charlie Mead: We will put the decks and documents back out for peer review.

Cecil Lynch: Do you have examples of compliance/conformance?

Charlie Mead: Yes

Cecil Lynch: The problem I have had is the HSSP service models - can you claim compliance or conformance? IT is build as a functional model. You can never claim conformance to CTS2, because the specification is a functional model, not an implementation.

Charlie Mead: The HSSP process stops at the blueprint specification layer. There is no traceability through the other layers. You can claim compliance by constraining it, but not conformance - since the assertions are not traceable down through the layers.

Cecil Lynch: Does that not go against the idea of claiming conformance?

Jane Curry: Some are observed by humans, so there is a subjective aspect.

Cecil Lynch: There is a missing piece - we speak of conformance/compliance with different definitions. There is a relationship there, but I am not sure what it is. I am trying to align them to II4.

Dale Nelson: What is at issue is that there needs to be a claim to a use-case as to the parts of the service models. They have to be per use-case, in sequence, with certain behavior.

Cecil Lynch: Testing is between use-cases, with claims of conformance, but there is a gap - maybe in the crossing of the use cases.

Jane Curry: Can you be patient, and take a look at the deck, to see if we have addressed this? We have tried to perceive this through implementer, standards body, and developer, and have traceability up and down. We will double-check to see if these are taken care of.

Charlie Mead: I agree that there is another piece - one of the things that stopped me cold. I will go through it tonight with Jane. There is more than we have talked about.

Discussion: SOA workgroups practical guide

Charlie Mead: Why do we have this matrix?

There is a document called the soa practical guide, which has multiple authors, John Koisch being one, as well as Ken Rubin produced by the SOA WG. It wants to be a sort of "Here's how you do SOA in health care". The problem is that according to the TSC it is sending a mixed message now that the TSC has committed to SAEAF. A couple of weeks ago at TSC there was an involved discussion about what to do about it. The TSC has asked the SOA WG to align it with SAEAF. SOA WG thought the guide was OK. The ArB will provide to the TSC the discrepancies, and misalignment. It is neutral, almost a marketing document. Because John Koisch: was one of the first authors, he produced a started matrix. I did my priorities, John Koisch and Charlie Mead had a difference in the priorities. The TSC will be the bad guy delivering the message, we are to give them the ammunition. In Kyoto there will be an educational message - the Guide for part one, and the SAEAF for part 2. We need to provide the TSC some substance for their communication with the SOA WG.

Charlie Mead: Go through the matrix, and read the document, providing the gaps. We will present to the TSC on Monday. You should read the comment, and determine the gaps.

Wendell Ocasio: What is the purpose of the document?

Charlie Mead: The document was trying to establish the notion of the importance of services, a practical look at services, and advertise HSSP as the purveyor.

John Quinn: The document was created before SAEAF, so we need to correct.

Wendell Ocasio: What are the goals of the document?

Charlie Mead: Does there need to be a document? HSSP thinks so. From HL7 perspective the document must align with SAEAF.

Wendell Ocasio: We are filtering for consistency, not owning the document?

Charlie Mead: Yes. SOA WG does not see a conflict. We as the ArB are to provide the TSC with the discrepancies.

Wendell Ocasio: I understand it now.

Charlie Mead: This is the first time of ArB with teeth.

Charlie Mead: Sections are the sections in the document. Line 2, SAEAF is not mentioned in the Guide. the real detail starts after Item 5. 1 is low, 5 is high. Nothing under three is required.

Jane Curry: I agree with everything up to line 8

Wendell Ocasio: The stuff that is on the screen is the document?

Charlie Mead: Yes, the spreadsheet i sent along with the original document.

Wendell Ocasio: I am looking at that also.

Wendell Ocasio: I do not disagree, but many of these are about strategy, not necessarily conformance to the SAEAF.

Charlie Mead: Johns perspective about what is reflected in the matrix, are about the governance of HL7 in the SAEAF. The TSC want the alignment issues, and will draft the final statements.

Cecil Lynch: I would change #6 to a 5 - I am involved in a spec where the data-types of the spec do not align with the vocabulary.

Dale Nelson: I am wondering whether #9 currently a 3, it needs to be higher -

Jane Curry: I am sure the definition of compliance is not aligned with the SAEAF.

Charlie Mead: We are moving to the point where everything needs to be a 5, otherwise it will send a mixed message.

Jane Curry: The language HL7 has used for years is static/dynamic.

Charlie Mead: We have a more robust framework that needs to be referenced.

Jane Curry: Instead of saying a dynamic view.

Jane Curry: Is this how standards till hit the industry?

Charlie Mead: Yes

Jane Curry: It(16) is a higher priority than a 4.

John Quinn: Think of HSSP being half HL7.

Charlie Mead: The original HSSP framework was the HL7 would provide the main expertise, and OMG the know how. HSSP could continue that and limit their input to the blueprint or the platform independent level, with no input to the computational viewpoint. They were interested as to how they would align with the SAEAF. The part that comes out of HL7 needs to do so.

Dale Nelson: Line 14, not knowing the real context, the phrase "art rather than science" - whether it is a 4 or five, it is a disturbing phrase.

Wendell Ocasio: I do not know what our role is on the methodology, their methodology may not be the best one for service methodology.

Charlie Mead: We should not care about HSSP methodology, except that their artifacts committed to produce before they throw them over the wall to OMG. We need to control any artifact that has relevance on their side of the spec should be in our stack, and compliant with the SAEAF stack.

Charlie Mead: John Quinn: do you agree?

John Quinn: yes

Wendell Ocasio: Do we need a disclaimer?

Charlie Mead: We need a disclaimer that we separate ourselves from the methodology.

Wendell Ocasio: The methodology - i would not feel comfortable endorsing it as a way that HL7 practices SOA.

Charlie Mead: There should be verbiage in the header of how it aligns with SAEAF.

Wendell Ocasio: We should add HL7:

Charlie Mead: Tony will add disclaimer to the first line that Charlie Mead: sends him.

Tony Julian: Disclaimer reads:

Health Level Seven (HL7) does not endorse the processes stated or implied in this document.  
In particular, the HL7 Services Aware Enterprise Architecture Framework (SAEAF) and the processes that it both implies and  
defines explicitly emphasizes the importance of taking a holistic view of an organization and its environment preceding service
identification and specification, providing traceability from a core understanding of business objectives and processes to a  
final service specification.   In particular, the term “enterprise architecture” – as it is utilized by the SAEAF – encompasses
much more than the technology architecture involved in deploying specific service implementations.  

Wendell Ocasio: This is a different universe than the HDF: the analysis of the business comes first - people are making the mistake of focusing on candidate services before understanding the business.

(projecting Lines 18-20)

Jane Curry: One of the things that bothered me is that we do not have the same idea of an Enterprise Architecture framework. what is here is a technical framework, not enterprise.

Charlie Mead: I will add to disclaimer.

Wendell Ocasio: Where

Charlie Mead: On Line 1.

Jane Curry: ON page 16-18 - no way to measure the benefits.

Wendell Ocasio: Top of page 16 they say architecture framework, when the are designing a bus. Different use of framework. Bad use of the term framework.

Charlie Mead: 26 is the last one?

Tony Julian: Correct- and we have 2 minutes.

Jane Curry: on 23 the unity of purpose should be in a governance section, to create a focus of unity of purpose.

Charlie Mead: Send additional comments to me or John Quinn. If you cannot do it in the next day, we will assume that everything is OK

John Quinn: do not send at the last minute. I will rely on John Koisch: or Charlie Mead:. I will be chair.

Charlie Mead: John, by sending this the ArB has met the goal?

John Quinn: I am also trying to figure out the next step of the Arb - which I suspect will be to provide an alternate draft of the document.

Charlie Mead: Can we ask them to do so with this input?

John Quinn: I do not think they will do so. I do not know that there is a prejudice within the TSC. Once the SAEAF is in place

what do we need to do to reshape the TSC/Work groups to meet the goals? What is the organizational/process structure?

HSSP is one of several elements of the WG community that will be pushed in another direction as far as moving forward.

Charlie Mead: Thank you - send additional comments to me and John. 5decks and 5 documents. I will send you the latest of the introduction.

• Adjournment

Meeting adjourned at 4:05 pm u.s. eastern Tony Julian