Learning Objects Working Group, April 18, 2007
Attending: Morgan Bantly, Chair, Charlie Atkinson, Gabrielle Campbell, James Dalziel, Susan Engelmann, Ralph Halpern, Jim Martino, Terry Poulton, Cindy Sheffield, Carl Singer, Daniel Trottier, Bob Waddington
The group began with introductions and review of health care along in its use cases. Healthcare LOM is a customization of the IEEE's learning object metadata standard. Healthcare lom adds healthcare specific elements to the IEEE LOM. Metadata is data that describes data, so healthcare LOM is a specification for describing learning resources and activities. Healthcare LOM can facilitate search and discovery of learning resources and activities. Healthcare LOM also serves as a building block and is used by other MedBiquitous standards.
Morgan asked Valerie to provide an update on the status of healthcare LOM. Valerie explained that Healthcare LOM is currently in public review. The public review ends April 30. After April 30, the Standards Committee will review any comments received, work with the working group to make appropriate changes, and then ballot the specification.
Morgan then reviewed the problems related to consent and licensing of clinical data within learning objects. Often clinical data including images are shared as teaching resources or are included in learning content. This data often has privacy restrictions and sharing constraints associated with it. The group had compiled a list of problems that illustrate this dilemma. Many of them relate to managing data over time and changing requirements.
Morgan then asked Gabrielle to provide an update on the AAMC's work to address these problems. Gabrielle explained that she works with MedEdPORTAL, a resource that facilitates sharing of educational resources. Many of the resources registered on MedEdPORTAL have medical images or recordings. Often it's not clear whether the authors have the consent required to distribute. The AAMC has decided to put together a colloquium to examine these issues. Earlier today, she spoke with the Association of University Technology Managers. They are talking about the same thing with cell lines. She's speaking with creative commons tomorrow to see if they will be involved.
Tim asked how they currently address author rights. Gabriele answered that they use a creative comment license. James added that the creative Commons share alike license is commonly used by academic initiatives like MedEdPORTAL. No derivatives is another option, which may be important in healthcare. For example, one can imagine a patient being delighted to have their image used when referencing a relative, but unhappy when the same image is used to describe impotence. The context is important.
Tim commented that with the learning object design. One could compile a course at of several learning objects coming from different sources. How do you negotiate all the licensing and consent along the way? Who owns the final course? Gabrielle replied that it depends on the licensing. Morgan added that the parent course or SCO needs to have the most limiting rights of all the assets.
Tim asked if licenses could be machine-readable, and James replied that many projects have tried but it is too difficult. Creative Commons allows for a base level, it's easy to understand.
Morgan explained that many organizations have a consent form, with certain fields required. On the VA consent form, these include how material will be used, for what length of time, for what purpose, and on what media. They also have statements for broadest possible reuse that are recommended to be used on the consent forms. Developing guidance that provides a standardized release with broadest possible re-use should be something we strive for.
One area of concern is when patients retract consent. The asset has been broadly reused, there's no way to prevent the assets use in those derivative learning objects. Terry asked if this case was so infrequent that it's not worth pursuing. Bob agreed that as a policy, you can say that there's no further distribution, but there's no way that you can impact past distribution.
Morgan Bantly, Chair, email@example.com
Charlie Atkinson, firstname.lastname@example.org
Gabrielle Campbell, email@example.com
James Dalziel, firstname.lastname@example.org
Susan Engelmann, email@example.com
Ralph Halpern, Ralph_Halpern@tufts-health.com
Jim Martino, firstname.lastname@example.org
Terry Poulton, email@example.com
Cindy Sheffield, firstname.lastname@example.org
Carl Singer, email@example.com
Daniel Trottier, firstname.lastname@example.org
Bob Waddington, email@example.com
Tim Willet, firstname.lastname@example.org