2009-08-10

Meeting Information

Attending: Susan Albright, Rachel Ellaway, Valerie Smothers Luke  Woodham,

Note: Jörn Heid and Jeroen Donkers attempted to dial in but were unable to stay connected due to teleconferencing problems. We are working to resolve these issues.

Agenda Items

1 Results of ballot on moving spec to the MedBiq Standards Committee

Rachel reviewed that the ballot was an action item from the last call. It provides the working group with an opportunity to come to consensus, saying we are done with the specification and passing it to the MedBiquitous standards committee. Valerie commented that so far the vote was unanimous in favor of moving the specification to the standards committee. 10 votes have been received of the 22 members of the committee.

2 Results of ballot on undertaking profiling

Rachel commented that the second ballot measure was related to the working group undertaking the activity of profiling the standard for different types of VPs and VP systems. This reflects the fact that systems implement the specification quite differently based on their pedagogical design. Valerie commented that so far the vote on this measure was unanimously in favor as well, with 10 affirmative votes out of 22 members. The ballot closes Friday August 14th.

Rachel commented that we would work on the basis that these would be approved. Rachel asked Valerie there would be opportunities for comments going forward. Valerie replied that there would be. Following the standards committee review, there will be public review of the specification prior to ballot. Any substantive changes to the specification would require a new public review.

3 Planning for profiling activity

Rachel summarized that we've identified the following potential profiles for the virtual patient specification:

-          Schema design/string of pearls. This model is used by web SP and Campus. The activity model is simple, virtual patient data and data availability model are more advanced.

-          Branching. This model is used by open labyrinth and VP Sim. The activity model is more advanced, and branching is the key activity.

Susan commented that Tufts would be a schema string of pearls design as well. She questioned how the web SP systems ability to jump around would be reflected in the profile. Rachel commented the profile design would be an iterative process. Profiles will be further restrictions of the existing specification. These restrictions should improve interoperability between like systems.

Rachel asked Luke about the experiences in eViP. He commented that they work with open labyrinth and use only the VPDText element within virtual patient data. They use lots of branching in the activity model. It is very difficult to map elements from a branching case to other systems. Bringing open labyrinth cases into campus and Casus is very difficult. The reverse is true as well. There cases require manual reworking before importing. They don't have much of an automatic import.

Rachel commented that eViP has posted 4 MVP packages, one from each system, on the project website. Campus uses that backslash character in the file name, which is illegal for Windows systems. She asked Valerie if it was legitimate in terms of the spec. Valerie offered to research the question and push the response to the working group and the eViP TRG.

Rachel added that the schema group may be less homogenous than originally seemed. Susan questioned whether global navigation would be a third profile. If Tufts and web SP are distinct from campus and casus, we can create third profile. She offered to have an off-line conversation with Nabil.

The group revised the lists of profile types to:

-          Schema design/string of pearls (campus, casus)

-          Branching (open labyrinth, vpsim)

-          Global navigation (Tufts, WebSP)

-          Native player (pivote, VPSIm)

Rachel commented that for our call on September 17, the primary agenda item would be defining the first steps for profiling.

Susan commented that Tufts may be both a schema design and global navigation profile system. Valerie questioned whether global navigation would dramatically change how the data is represented. Luke offered to talk to Susan further about it.

Rachel commented that we would need packages representative of the current version of the specification. Valerie offered to put together a list of the changes from version 0.48 to the current version, 0.55. Luke offered to see how their most recent export would validate against the current version of the specification.

4 Compliance and certification update

Rachel asked Valerie if there was any update on compliance and certification. Valerie replied there was not. Rachel requested that we keep this item as a standing item on the agenda.

5 Extensibility - Jörn's paperetc

Rachel commented that we should wait until Jörn can participate in the call to have this discussion.

6 Open discussion

Rachel asked the group for implementation updates. She commented that she has been working on OpenLabyrinth import. It now works for CASUS and WebSP as well as for OpenLabyrinth. It will import the content, but it does not present in the same way and requires adaptation order to be useful. Susan asked if it required a lot of work to be usable. Rachel commented that it did. Base content is available, but it doesn't look like it did in the originating system.

Decisions

Action Items

-          Valerie will research the legitimacy of using backslash in a file name and push the response to the working group and the eViP TRG.

-          Susan will talk with Nabil about possibility of a profile for global navigation systems.

-          Valerie will put together a list of the changes from version 0.48 to the current version, 0.55.

-          Luke will provide an up-to-date package for open labyrinth.

-          Valerie will inform the group that we've identified 4 profile groups

-          Participants will present their profiling discussions at Sep 17 meeting (how tightly, loosely they see themselves conforming, etc)