This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

Difference between revisions of "MnM Minutes CC 20070615"

From HL7Wiki
Jump to navigation Jump to search
 
(24 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
[[Category:2007 MnM Minutes]]
 
==Attendees==
 
==Attendees==
 
*Kathleen Connor
 
*Kathleen Connor
Line 16: Line 17:
 
Mottion to accept agenda: (No objections)
 
Mottion to accept agenda: (No objections)
  
===Administrative Topic: [[Dynamic_Model|Dynamic Model]]===
+
==Administrative Topic: [[Dynamic_Model|Dynamic Model]] Discussion ==
  
Desinging system behaviors is pervasive in HL7.  
+
Concerns regarding desinging system behaviors are pervasive in HL7 (from application roles to committee-specific '''complex interactions''' to functional service specifications). We have a variety of constructs (Applications Roles, SOA Services, interactions) that require dynamic views and behavioral model constructs such as interfaces, ports, and subsystems.
We have a variety of constructs (Applications Roles, SOA Services, interactions)
+
Ioana summarized the concerns expressed on the list because we don't have a generic way to deal with behavioral modeling. The following is a set of objectives voiced on the list:
There was a general consensus that changing the identifier of an artifact should not be regarded as substantive. Some felt that a move to non-meaningful identifiers would be useful.
 
  
We need to separate the maintenance of semantic domains from the maintenance of publishing topics.
+
*"All implementations I've experienced to date include both messaging and services to accomplish the application solution.  We need to begin to understand and include those requirements for dynamic model methodology for services as well as messaging."
 +
*"We need to tackle the generic modeling considerations before we can delve into any specific model. Order management is one instance of a behavioral model and I expect it to be developed by the appropriate TC not by M&M during harmonization. I think M&M has the responsibility to develop the generic methodology that can be applied by any committee to their needs to communicate how the interactions are intended to be applied to the fulfillment on business processes."
 +
*"The agenda for these two days was intended to address changes to the dynamic model methodology to meet the requirements initially identified by the LabSIG (who is seeking to address negative ballots re: the current dynamic model).  We're obviously interested in everyone's needs re: documenting the dynamic model and appropriate artifacts to support.
 +
We are intending to meet and discuss the methodology itself including diagramming, etc.  This particular group does not have the same makeup as the group working through OO on order management.  I expect that exercise will continue within OO."
  
Gregg asked when the tooling for publishing could support levels of models as the MIF does?  Woody suggested that this would probably be about 18 months out.
+
*[[OO_Dynamic_Model|A link provided by Hans to on-going work in Orders and Observation/Lab SIG]]
 +
* MITA Project Service Specifications demonstrate the use of UML for describing behavior and service [http://www.bptrends.com/publicationfiles/03%2D05%20WP%20WS%2DCDL%20Barros%20et%20al%2Epdf choreography] (rather than [http://weblogs.java.net/blog/johnreynolds/archive/2006/01/service_orchest.html orchestration] of processes. This project produces WSDL and XSDs for runtime execution.
  
For now, Clinical Decision Support will publish as one domain with two topics.
+
The agenda for next week must address the needs of a variety of stakeholders:
  
Agreed to publish an item on the Wiki relative to "[[Domain_Message_Information_Model#What is a domain|What is a Domain]]".  Among the points of agreement were:
+
*Support for the processes specified in the [http://hdf.wikispaces.com HDF] regarding behavioral models
*All artifacts should derive from some coherent set of knowledge as represented in a single model.  This is a "semantic domain" and is the focus of an HDF DIM.
+
*SOA Service support and relating interfaces to Application roles
*Publishing should be able to aggregate content from semantic domains into collections that are useful for committees that address multiple semantic domains
+
*Ability to describe complex interactions for specific domains
*Future (18 moths or more out) tooling will allow such publishing, as the MIF allows for such distinctions
+
*Leverage the notation and design to describe a variety of in, in/out, output parameters and return types
*Today, several committees are using a single "publishing domain" (really just a package) to publish multiple "semantic domains" each as a topic, including Decision Support, Patient Administration, Common Messages, and Patient Care.
+
*We need to ensure that our design constructs are supported by implementation technologies (e.g. WSDL, WS-CDL, WS-BPEL)
*Other committees are maintaining correct semantic domains, such as Pharmacy with Medication and RX. and Lab with OO, LB, SP.
 
*Publishing should determine which pattern is being followed in each of today's "Domains" and should require:
 
**That the style be declared at the front of the domain introduction
 
**That a committee '''not do both'''. That is, in each publishing domain either every topic is a semantic domain or all of the topics are in a single semantic domain. '''Note:''' This may represent an issue for Patient Care.
 
  
'''From the Secretary''' If you take issue with this list, please use the new Wiki entry - [[Domain_Message_Information_Model#What is a domain]] to comment.
+
===Dynamic Model Agenda (Harmonization Meeting June 21-22, 2007) ===
 +
The agenda was created in a '''[[Dynamic_Model_Agenda_June_21_2007| separate page]].'''
  
===Dynamic Model Agenda===
+
----
Edited the form on-screen. Agreed to submit as marked.  Beeler will finish and submit the form. Nelson will submit new Project Scope Statement for CMETs Release 6 (those that were in Committee) - The new Project was moved by Nelson, seconded by Beeler and Approved unanimously.
 
  
The CBC form, as submitted is available [[Media:CBC_INFO_MNM_2007MAY-Complete 20070119.pdf|on this Wiki]] and also at [http://www.hl7.org/library/committees/mnm/minutes/CBC_INFO_MNM_2007MAY-Complete%2020070119.pdf http://www.hl7.org/library/committees/mnm/minutes/CBC_INFO_MNM_2007MAY-Complete 20070119.pdf].
+
Minutes by --[[User:Ioana13|Ioana Singureanu]] 14:50, 15 June 2007 (CDT)
 
 
 
 
 
 
Minutes by Ioana Singureanu [[User:ioana13|ioana13]]
 

Latest revision as of 01:59, 21 May 2010

Attendees

  • Kathleen Connor
  • Lee Coller
  • Ioana Singureanu
  • Lloyd McKenzie
  • Rick Chesnik
  • John Cooper
  • Dale Nelson


Agenda

  • (Lloyd): Administrative topic: Dynamic Model - agenda
  • (Lloyd): Hot Topics: Act Reference - postponed for next WGM because this topic requires input from INM.

Mottion to accept agenda: (No objections)

Administrative Topic: Dynamic Model Discussion

Concerns regarding desinging system behaviors are pervasive in HL7 (from application roles to committee-specific complex interactions to functional service specifications). We have a variety of constructs (Applications Roles, SOA Services, interactions) that require dynamic views and behavioral model constructs such as interfaces, ports, and subsystems. Ioana summarized the concerns expressed on the list because we don't have a generic way to deal with behavioral modeling. The following is a set of objectives voiced on the list:

  • "All implementations I've experienced to date include both messaging and services to accomplish the application solution. We need to begin to understand and include those requirements for dynamic model methodology for services as well as messaging."
  • "We need to tackle the generic modeling considerations before we can delve into any specific model. Order management is one instance of a behavioral model and I expect it to be developed by the appropriate TC not by M&M during harmonization. I think M&M has the responsibility to develop the generic methodology that can be applied by any committee to their needs to communicate how the interactions are intended to be applied to the fulfillment on business processes."
  • "The agenda for these two days was intended to address changes to the dynamic model methodology to meet the requirements initially identified by the LabSIG (who is seeking to address negative ballots re: the current dynamic model). We're obviously interested in everyone's needs re: documenting the dynamic model and appropriate artifacts to support.

We are intending to meet and discuss the methodology itself including diagramming, etc. This particular group does not have the same makeup as the group working through OO on order management. I expect that exercise will continue within OO."

The agenda for next week must address the needs of a variety of stakeholders:

  • Support for the processes specified in the HDF regarding behavioral models
  • SOA Service support and relating interfaces to Application roles
  • Ability to describe complex interactions for specific domains
  • Leverage the notation and design to describe a variety of in, in/out, output parameters and return types
  • We need to ensure that our design constructs are supported by implementation technologies (e.g. WSDL, WS-CDL, WS-BPEL)

Dynamic Model Agenda (Harmonization Meeting June 21-22, 2007)

The agenda was created in a separate page.


Minutes by --Ioana Singureanu 14:50, 15 June 2007 (CDT)