search

AICC / CMI-5_Spec_Current

cmi5
http://adlnet.gov
Apache License 2.0
181 stars 91 forks source link

The AU id is defined by the publisher id #711

Closed thomasturrell closed 3 years ago

thomasturrell commented 3 years ago

This is potentially a spec change.

The roles of AU publisher and course author seem to be distinct in some place and not in others. However I am fairly sure that the AU publisher ID should not be set by the course designer (unless of course one person has both roles).

@brianjmiller and @cawerkenthin I would appreciate your thoughts on it, since I am not aware of the original intention of the quartz spec. If it is simpler and quicker we can discuss on the working group call.

cawerkenthin commented 3 years ago

I don't particularly have a problem with the change. Unfortunately, the course designer, developer and publisher are often the same person as many departments are not adequately staffed. Consistency of how we reference it, however, would be good.

brianjmiller commented 3 years ago

I think this better aligns with how we've used those terms in the past. The person working with the course doesn't really have the control to adjust the identifier in so far as that AU is intended to be the same globally. If they change that identifier they are in effect becoming an AU publisher and publishing a new, globally unique AU that is now 100% distinct from the original source. It is the intention that the person working with the course is assembling AUs and adjusting their expected behavior via the other controls that they are expected to edit, such as moveOn, masteryScore, etc. because those values aren't globally unique and are captured via statements so that any comparison of data with the same publisher id can be cross compared.