CMI 5 Working Group Meeting Minutes – July 24th, 2015 - AICC/CMI-5_Spec_Current GitHub Wiki

cmi5 Working Group Meeting Minutes – July 24th, 2015

Attendees

  • Bill McDonald - cmi5 Working Group Leader
  • Andy Johnson - ADL
  • Ben Clark – Rustici Software
  • Art Werkenthin - RISC
  • Henry Ryng – inXsol
  • David Pesce – Exputo
  • Charles Touron - ASTC
  • Ben Clark – Rustici Software
  • Patrick Selby – LexisNexis
  • Andrew McGarrity – Thrivist

Notes

cmi5 promotion Bill McDonald presented on cmi5 to the xAPI Camp in Seattle (Hosted by Amazon) July 21, 2015. There were approximately 60 people in attendance (with a broad mix of skills) and the presentation was well-received.

See presentation on slide share: http://t.co/AprmGtpJkN

cmi5 compatibility with other xAPI profiles

It is the intention of the cmi5 specification to allow AU developers to make additional statements outside the scope of the cmi5 specification. This is a key concept for extensibility/flexibility. However in order to have plug-and-play interoperability, such “non-cmi5” statements still have to conform to “some rules” to fit with a cmi5 AU session.

The group reviewed last week’s discussion on “non-cmi5” statements and agreed that:

  • ALL statements made within a cmi5 AU session MUST conform by including the context template provided by the LMS (using the State API).
  • The specification language regarding the LMS provided context template will remove the reference to the “cmi5 category ID”.
  • The AU will be required to add the “cmi5 category ID” to the Context Activities in statements that are “cmi5 specific”.
  • Statements without the “cmi5 category ID” are allowed (but they do not affect moveon, or session management or other cmi5 functions). Such statements are not “cmi-defined” and could use other verbs (or could “re-use” cmi5 verbs).
  • No statements can be made before the cmi5 defined Intialized statement or after the cmi defined Terminated statement.

The “normative” language in the specficiation will enforce these rules but, since they are confusing, some “informative” language will be need to be added.

A new section called “7.2.4 Types of cmi5 statements” this section will describe the following:

The types of statements:

  • cmi5 defined - Statements using cmi5 defined verbs and cmi5 category Id's (and cmi5 content templates)
  • cmi5 allowed - Statements using any verbs and cmi5 content templates (and NOT including cmi5 category id's)
  • cmi5 not-allowed - Any statements non conforming with the cmi specification.

Andy Johnson will draft appropriate language for a pull request.

cmi5 Runtime

Issue #217 - Why can't the actor be a group?

The group decided to close this issue. It is a core requirement of cmi5 the that actor is a individual learner. The cmi5 use case is specifically for a individual learner using a single system.

**Issue #224 - Fix /cmi/adl purl URLs **

This issue was already resolved in the Quartz draft. However there was a strangler URI with “adl” which will be removed.

Issue #216 - There's no such thing as an actor object

There are some instances where “actor object” should listed as “actor” in section 9.2 and 10.0.

In section 8.1.1 – this Value space: A JSON Actor object (as defined in Section 9.2) should read Value space: A JSON object of type “Agent” (as defined in Section 9.2)

cmi5 Course Structure

**Issue #223 - Can langstring values be used that are not listed within languages? ** After discussion this issue, the group decided to remove the “Languages” element from “7.1.1 Course Level Metadata” as it is redundant and could be derived from the AU metadata in other parts of the course structure.

Next Meeting:

Next Meeting: July 31st, 2015 – 10:30 am US Eastern Time

⚠️ **GitHub.com Fallback** ⚠️