2012-05-13 TSC WGM Minutes

From HL7 TSC
Jump to navigation Jump to search

TSC Sunday Q5 meeting for 2012May WGM in Vancouver BC CAN

back to TSC Minutes and Agendas

TSC WGM Agenda/Minutes

HL7 TSC Meeting Minutes

Location: Port Alberni room

Date: 2012-05-13
Time: Q5 - 5:15 PM to 6:30 PM local time
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
x Giorgio Cangioli HL7 International Representative
x Lorraine Constable ArB member
x Freida Hall HL7 TSS SD Co-Chair
Chuck Jaffe HL7 CEO (member ex officio w/o vote)
x Tony Julian HL7 FTSD Co-Chair
x Austin Kreisler (Chair) HL7 TSC Chair, Ad-hoc member
x Lynn Laakso (scribe, non-voting) HL7 HQ
x Patrick Loyd HL7 T3SD Co-Chair
x Lloyd McKenzie MnM/FHIR
x Grahame Grieve MnM/FHIR
x Charlie Mead HL7 ArB Chair
Don Mon HL7 Board Chair (member ex officio w/ vote)
regrets Ravi Natarajan HL7 Affiliate Representative
x Ron Parker HL7 ArB Alternate
regrets John Quinn HL7 CTO (TSC member ex officio w/vote)
x Helen Stevens Ad-hoc member
x Ed Tripp HL7 DESD Co-Chair
x Pat Van Dyke HL7 SSD SD Co-Chair
x Mead Walker HL7 DESD Co-Chair
Quorum Requirements (Co-chair +5 with 2 SD Reps) Met: (yes/No)


Sunday

  • ArB report
  • SAIF update
  • FHIR update and Governance Council discussion


Supporting Documents

Minutes/Conclusions Reached:

  • FHIR update and Governance Council discussion
    • Austin in this morning's tutorial heard some more about it. Let's hear what Grahame and Lloyd's ideas are around governance of FHIR.
    • Helen wishes to hear the parameters of governance.
    • Grahame notes necessary governance in two parts: technical governance ensuring processes leading to definitions published that are stable usable and reliable. Working on that in MnM. The other is who gets to do what, that the larger interests are represented and Work Groups respond in a timely manner. First set is internal process but more interested in the second set.
    • The board and Grahame have agreed that Grahame would gift FHIR to HL7 on condition that its first full normative version be publicly available and free to implement. Existing implementers are invested in current standards. You have to see that a new proposition is working, first. Second much integration is happening now that we're not involved in. After the first full normative release HL7 may evaluate how subsequent releases are controlled in IP. Derivative works are likewise. Wes Rishel quote is that FHIR is Fixing HL7's International Reputation. Grahame is trying to help implementations along so that FHIR is led by implementations.
    • Question is where is the revenue model? Not taking away existing implementations. He proposes OMG model on specifications are free but implementations by governments have revenue. He proposes TSC has a governance committee with oversight of FHIR. TSC ensures it runs well and transparently. Committee exists in default right now with Lloyd, Ewout and himself. Committee needs to reflect organizational interests but also commitment by the members of the committee.
    • Helen asks what examples we currently have. The SAIF AP is similar but the ArB is the closest we have but this committee would have more power than ArB to delegate work to MnM or other groups.
    • Woody notes that we have internal processes such as harmonization with overarching authority on key artifacts which functions through MnM but is really an organizational function.
    • Mead finds this split into two groups: what resources are created and who gets to create them, as well as the responsibility for definition of what is a good resource. Determination of whether a proposed resource is acceptable is another facet. Grahame agrees that is the key operational function of the committee.
    • Grahame adds that Bo Dagnall is working on a way to approach that question of what resources are created by whom. Patrick notes that the HL7 organization generally attributes "person" ownership to the Patient Administration Work Group and if that is somewhere different then Patrick is concerned. Grahame notes that if they don't produce something in a reasonable time that perhaps it's going to happen somewhere else anyway. Helen notes one purpose of the TSC to ensure the work is happening in the right work group. PA works on V3 parts of person as well as V2 parts, that they should also have responsibility with FHIR. Patrick's fear is what happened with CDA that someone else was defining Lab and didn't bring it back to the OO group.
    • Grahame clarified that he created a lab report because he needed something to show. He's not attempting to take that out of OO. He's proposing a governance committee which is intended to alleviate the problems caused by Structured Documents as a de facto 'governance committee' for CDA. Calvin notes that SDWG asked for use cases they did not get sufficient input or didn't get use cases that others wanted to build.
    • Woody recounts that Grahame said we couldn't have begun this 4 years ago without the experience we have between work groups. CDA IGs and implementation that tell you what's needed, as well as V2 background. What we need to bring this together at the TSC level to oversee it. He notes that we need a governance committee for FHIR just as we need a governance committee for HL7. We need to be able to bring this together as soon as possible.
    • Austin notes that if it's just another work group they don't have authority. We don't want it to be a Board Committee. Mead thinks the TSC itself ought to be that committee. If we can't undertake the work and put in the time then we delegate a group to make recommendations to the TSC and the TSC adopt them. Is this now to be the main thrust of activity, we must decide. Helen recommends we talk to the members. Austin thinks the TSC should be the appeals body for decisions by this governance group.
    • Woody harkens back to 1996 to the group building V3 and its relationship to the TSC. Mead notes that the way the RIM bifurcated is a way the group determines who does what. Mead wants them in separate bucket, for what resources to be developed and who builds them.
    • Lloyd find the base constructs for quality criteria versus the fine grained quality criteria, where the former should be subject to ballot. You need a body to state what the criteria are that will be enforced. Mead notes that we need to set which criteria but when two bodies want to develop a resource this proposed group has to deal with that.
    • Helen sees the idea with the TSC as arbiter but she'd like to see the TSC involved in establishing the terms of reference for this group and it's M.O. and composition, communication with membership, rollout to Work Groups and so on. We should not just name them and let them go off. Elected TSC representatives need to be involved with setting up the process.
    • Calvin asks how much of the V2 organizational structure was still applicable when V3 came in? Woody saw it very little change. Attrition happened on the V2 road and many participants especially vendors have left meetings and not come back notes Helen. Ed notes that is it an either/or or does FHIR become another tool in the toolbox.
    • Woody notes a key question is that what is the real value proposition for V3 messages today? Canada and UK are a fraction of our membership. Lloyd notes that large organizations like Infoway using the tooling and the methodology and foundational artifacts. He finds little direct use of the committee artifacts other than as informing what those organizations do for themselves. There are few implementations that are conformant to the RMIMs and interactions published by HL7. Helen disagrees, that universal RMIMs are intended to be a basis for realm implementations with constraints. Woody notes the NHS is using a RIM that is three years old, so it's not that tightly tied. CDA is the V3 success story. Canada fed into the V3 universal specs.
    • Patrick notes that for the implementers to be noted on project scope statements in order for them to pass. He thinks this effort needs to be disruptive for this organization. Woody feels that many projects are coming in for 'edge' space. PA has hit a lull. OO and Pharmacy work in a generic space for HL7. Even clinical research may be considered an edge space. Need to be prepared to invest in this so that it makes the inroads we expect. Need to encourage those key work groups to produce what we want from them - Helen reminds that they work at the pace of their members. Woody notes that we need a set of specifications with the same energy as CDA has displayed the last 5 years.
    • Ron notes that Infoway bet on V3 at their decision point. Concern now is this many years (10) later, what they hoped would happen in the market space has not happened. Worse yet, ONC did not select standards not compatible with the vendors in their market. He wants to see something to support the promise like V3 to enable interoperability to make a product to fit the market space. We need to harvest the thinking that went into RIM based modeling. TSC needs to have a sense, a set of criteria, of when to make a commitment to a new product space. If we have a limited resource space, and this holds interest to vendors, we may increase our resource space. What are the metrics or measures where we hit the boundary in transition from one state to another on the product offerings. Grahame is aware of a number of attendees that are here specifically for FHIR. Ron notes that the metrics now provide the information that we might use for such determinations. We need to recognize the trend where participants vote with their feet and recognize when to advance a brand.
    • Helen viewing the project database does not have many projects in V3 messaging currently under development. Woody notes that as publishing committee the decline in V3 messaging projects that come up. Helen notes that some Work Groups are in maintenance mode on their V3 materials that a healthy injection of interest on setting their attention to creating the FHIR resources. Ron now answers from the ArB perspective, the essence is to create a stable definition of an information resource that can use multiple transports to go between systems to do 'things". PA needs to weigh in on behavioral dynamics on resources.
    • Lloyd notes that in the tutorial for what FHIR means for HL7 WGs. Need to determine the essential numbers of resources. They're not huge or super complicated. Helen advocates we revitalize PA who have only three active projects. Lloyd notes that committees have varying levels of interest but don't know if it's official, are we allowed to do this and should we do this. We'd like to determine tonight what that message is. Are we in the early stages not ready to endorse; we are ready for certain committees to start work; or FHIR is open for business and anyone interested can engage and here's how. Lloyd advocates option 2 or 3. We may not be ready for #3. Some committees need special attention, and they're meeting with PA for example to talk with them about FHIR and have had a conference call already. Need endorsement from the TSC that this activity is okay and what are the boundaries. Mead asks if there is already a list of the resources we know we'll need. Lloyd notes that MnM will be reviewing that list this week and noting their logical ownership. Austin agrees he wishes to make an announcement to the membership this week, probably on his Tuesday general session address.
    • Patrick says this smells like the SAIF AP with critical tasks and critical path with critical involvement. Austin notes that he wished to organize this under the SAIF AP umbrella to use the existing coordination process, where we follow strategic projects. It may become a program of its own at some point. Dave Hamill has time dedicated for it and PSC has a project for coordination.
    • Calvin disagrees and that empowering the group that is working on it to do what they're doing can sometime be counterproductive. We want them to consider SAIF each step of the way but we may not want to bundle this in the SAIF space.
    • Ron asks Grahame if he'd be comfortable if governance for FHIR was not exclusive for FHIR but a division between governance, management and methodology? Governance presumes the existence of the management such as the SAIF AP. It's really an Enterprise Architecture Program. Charlie offers the first page of the governance framework of the SAIF CD as the distinction and definition. What is the boundary line between governance and management, Woody asks. Charlie says that management is the process of daily operational decisions, deciding value propositions, Governance is authority to make decisions and the purpose of governance is to ensure management succeeds. FHIR mgmt would be what resources to build and who builds them. FHIR methodology is are the artifacts complete.
    • Mgmt should be in an open WG for participation, notes Helen. Grahame notes that the core group with vested authority to make that decision must have continuity and consistency but their discussions are open and consultative. Their authority comes from respect within their community. He cites OpenEHR examples. Helen cites the MnM harmonization process; Grahame feels that outside HL7 the MnM Harmonization process doesn't hold respect.
    • Freida arrives.
    • Calvin notes strategy for ArB when building SAIF. TSC had close tabs on who participated. We have both models. It takes a dedicated and focused group to pull these things off.
    • Patrick notes the current structure of ArB is the type of group that would work for what we're looking for this Governance group to have. Ron agrees, and supports Grahame and enters into the difficult conversation on governance for the organization not centered in a work group, with TSC as appeal capacity. He has no problem injecting this into something like the MnM group and their experience and skill set and interest with view points to inform that process. Someone must be imbued with the responsibility to create the methodological direction for FHIR. The governance thingy we're talking about now does not exist now, though the ArB is as close as we come as an organizational group.
    • Woody reiterates Grahame's notion we need technical governance, MnM we have that it's there. We need content governance for question of who defines what resources consistency and completeness and that they are reliable. Need to get the core definitions done of resources needed and engage work groups to get them done. That group needs to respond to the TSC. Can have open meetings but not open membership.
    • Concern with possibility of buying a seat on the committee with commitment based on time and or money. Lloyd notes that the Board would be involved in the development of such a committee based on the acceptance in the market. FHIR is a free product and we can't put that genie back in the box but considerations on revenue generation with a separate body if we determine it's appropriate to move to it - we must be able to have that flexibility.
    • Grahame notes it's not appropriate now but may be in the future and want it to remain possible. That group for now should be those with skin in the game but not money; must be effort. Need to avoid controlling interests as well on the committee. Helen notes TSC assigned chairs can help with that without locking the membership. Woody feels you have to have appointed voting members.
    • Patrick asks how do we get to a decision? Woody has a motion, a draft position for the TSC:
      • a Message that FHIR appears to be a significant opportunity that HL7 must explore,
        • RIM Based,
        • Extends V3,
        • Aligns with V2 and extant CDA guides.
      • FHIR needs a core governance group,
        • TSC will empower a governance group
          • Expects it to demonstrate best-practice management and governance of itself (with good DMPs and excellent consultation).
          • To undertake content governance issues of:
            • Collaboratively what are the needed resources
            • Who can define what resource (right to allocate and withdraw responsibility)
            • How to assure it will be consistent and complete,
            • How to assure it will have maintenance and reliability.
      • Woody further identifies Membership as Chair, Lloyd McKenzie, Grahame Grieve, Ewout Kramer, Bo Dagnall, CTO, ArB, SD reps?, TSC rep
      • Skill sets required are are methodology, terminology, SOA, and International influence.
    • They will not create things themselves but determine who is supposed to define it.
    • Ed asks if we have some Work Groups that are supposed to define resources that might run like hell, and others with their hands up to ask to play. Ron notes that projects will have a matrix of resources towards production of FHIR resource.
    • Resources created will be balloted as HL7 products? FHIR will be balloted as a monolithic standard similar to the way we ballot V2.
    • If we had a SAIF IG, then our SAIF AP would be an Enterprise AP. Ron notes that this creates compelling reason to complete the IG.
    • Lloyd is more concerned that we have appropriate skill sets from various work groups rather than naming representatives from certain Work Groups to the membership such as Patient Care, Methodology, Terminology, SOA, International representation. SD groups are okay if they are there to provide the committees with a direct stake in the committee. Woody suggests DESD FTSD and SSD SD. Patrick and Freida are okay with not including T3SD. Woody thinks the chair should be a TSC member. Woody and Austin would both be willing to serve as chair. Grahame is concerned with Woody's reputation with V3 in leading the committee. Lynn requests they include a vice chair; Lloyd would like some administrative support. Lynn will check.
    • Lloyd would like to see a TSC rep as a named role but allow the group to name their own chair. Patrick asks if we then need the SD reps also? Helen suggests that one person should not hold more than one role since Patrick could represent ArB, TSC and SD, for example. Grahame would like to see 8-10 members. The continuity is important to the committee who knows why we did what was done.
    • Motion: Helen moves that the TSC request the named individuals Austin, LM, GG, EK, CTO, Dagnall, create a DMP and composition and terms of reference for the group. When they do, we'll then empower them. Austin notes that Woody might serve instead of John for the short term. Patrick seconds. The message is the number 1 part, and the number 2 part, with the membership and terms of reference (Mission and Charter, scope of practice) TBD by that committee. Then the TSC will empower the group. Charlie Mead opposes. Vote: Approved 6/1/0. Charlie feels Grahame has done something that HL7 isn't good at, not inventing everything from here and incorporating ideas and inspiration from elsewhere. 99% has been defined and operationalized outside of HL7. The notion of setting up a governance office, about resources as business-centric, context-neutral are a service taxonomy as noted by Thomas Erl. Patrick encourages the named committee to harvest from Thomas Erl's work.
  • Adjourned 7:00 PM


Actions (Include Owner, Action Item, and due date)
  • .
Next Meeting/Preliminary Agenda Items
  • .