E-mail
Attending: Joel Farrell, Chair; Peter Greene, Dan Rahak, Carl Singer, Valerie Smothers.
1. Leap 2A discussion
Valerie provided a summary of the educational trajectory work and the issues for the technical steering committee. The educational trajectory working group is seeking to create a specification for facilitating the transition from medical school to residency. The group wants to transfer data on extracurricular activities, things that don't show up on the transcript, that are often valuable enrichment activities of interest to residency program directors reviewing applications. The group has evaluated the leap 2A specification created by JISC and found it to be a fairly good fit for its requirements.
Leap 2A was designed for portfolio entries. Portfolios are often used in medical school for evaluating competence. Atom was leveraged for the leap 2A specification in all likelihood because there is a great deal of similarity between a blog and a portfolio. Both allow the individual to reflect on activities or the environment on a regular basis and to include other content, including images, documents, and multimedia files. Portfolios add to this the concept of assessment and the connection to competencies, among other things. Valerie had forwarded a sample XML file developed by Simon Grant, the developer of the leap 2A specification who is also an invited expert in the working group. The questions Valerie had for the committee related to the overall soundness of the approach as well as a few particulars related to competencies and categorization. She added that she had difficulty in validating the sample XML file against an unofficial atom XSD file because XML spy tripped up on the RDF reference.
Carl commented that atomenabled.org references a relax NG schema for atom. Joel clarified that they did not want to use XSD in the atom specification. There are atom XSDs to be had, they are just not official.
Joel added that the RDF reference is valid. He added that he personally would not have architected the extension and that manner. Typically any extensions are put under the content element. Content allows you to establish a type with a new namespace. That's what elements in other name spaces would be expected. For example, one could have defined activity as a type and put that within the content element of atom. Carl agreed that was consistent with the approach to atom that he had used.
Joel asked what the current status of the leap 2A specification was. Valerie clarified that the specification was to be updated in February of next year. That said, there is an existing e-portfolio tool, Mahara, that implements leap 2A. University of California San Francisco has implemented Mahara, so they are very keen to use leap 2A as it would facilitate their data transfer.
Carl asked about the leap start date and end dates and how they related to publication date. Valerie commented that publication date would be the date that the entry was published. Leap start date and leap end date provide metadata about the activity. For example, the XML file contains a description of the Spanish class. The start date and end date indicate the dates the class began and ended. These dates are necessary in order to accomplish the goal of creating a visual schematic that shows when the learner is engaged in full-time study and when the learner had overlapping enrichment activities or enrichment activities that necessitated a leave of absence. If they learner is attending medical school full time and takes a night class in Spanish for professional enrichment, that is an impressive accomplishment that a residency director would like to know.
Joel commented that given the status of the specification it is likely too late to change where extensions occur. In addition, if the specification is used for import and export, it's less important.
Carl commented that the atom enclosure tag could be used for attachments.
Valerie questioned whether is evidence of was sufficient to identify a link to a competency. Her concern was that is evidence of could link to anything; how would you know that the link referenced was actually a competency reference? Joel recommended having a relationship is evidence of competency to clarify that the link referenced is in fact a competency reference.
Joel asked how we would address gaps between currently leap 2A specification and MedBiquitous requirements given the ambitious time line of the accelerated development project and corresponding pilot. Does MedBiquitous extend the standard or otherwise enhance it, or do you live within the confines of the current specification? Valerie commented that that was the primary question she had for the technical steering committee.
Carl commented that their approach was to leverage existing standards and use RDF to express what is missing. He added that qualifying relationships like is evidence of competency may not be the right way to go.
Joel commented that the simplest way to specify what we needed without using RDF was to add an attribute to the link element. He added that the value of the rel attribute is up to the atom implementer, so that could be used as well. The link element is extendable, so an additional attribute would be possible. You could add a type attribute to indicate that the link is a competency. MedBiquitous could define the extension or request that LEAP make the extension for us.
Joel recommended that in the short term MedBiquitous draft a set of extensions to leap 2A and put together a document that defines our requirements.
Peter asked Dan where leap 2A is in terms of adoption and strategy. Dan commented that he didn't have a sense of where it was overall. He recommended going back to JISC and seeing how much they would be willing to change, even going back to fundamental issues around atom. If the rest of the world is using atom differently than they are, that is something they may want to consider. He added that Cetis is heavily involved in web 2. There may be a Cetis-based rationale for why they did it this way as opposed to another way. Either way it should reflect an overall Cetis strategy rather than something specific to leap 2A.
Valerie commented that she still needs to obtain appropriate license to use leap 2A in this context. Dan recommended speaking to Adam Cooper. He offered to provide an introduction. Another possibility is Tish Roberts, the e-learning program manager. He commented that we should explain the MedBiquitous timeline and find an approach to licensing that is realistic.
Peter questioned whether we should be use relax NG. Joel replied no; we want to be able to validate extensions. If you have a foreign namespace extension where it is allowed, it will validate. There are XSD files for atom; the question is their status. Joel offered to do research on atom xsd files. He knows the leader of the atom working group and can get some information.
2. Web services testing forum
Joel commented that there had been discussion on having a Web services testing forum presentation to the technical steering committee and having ADL technical staff join us. This would be related to Web services harmonization. Valerie asked what the Web services testing forum was. Joel commented it is where people get together and perform interoperability testing. It was originally put together for people implementing Web services specifications and allows vendors to test in a more general way around WSI profiles. MedBiquitous develops vertical industry specifications. We could say here are some web services, implementers can come together to see if they can interoperate on web services we define. Doug Davis of the IBM is in charge of WSTF.
The group agreed to proceed with a joint call on WSTF January 19 at 11 AM Eastern.
Joel commented that this is the last meeting of 2009. He wished everyone a happy holiday. The next call for the group will be January 13, 2010.