Thank you for this very well thought out and presented paper.  I enjoyed going over it.   There is a lot here to digest and I'll have more comments or join in later discussions, I'm sure, as the days go on.
"We also need to be quite clear about the role of SCORM in learning (ie what it should enable), and remain clear about those things that should be outside of the scope of SCORM."
I think this is very important.  SCORM should have boundaries with regards to what we are trying to accomplish.  I think many times SCORM is looked towards as both the cause of and solution to things that that are outside it's intended scope.  I'm not sure where that scope begins or ends, but I know it needs to be more clearly defined than today.

With regards to "True Interoperability", I completely agree that interoperability should be a very high priority_. _ However, it should be noted that interoperabilty can be quite successfuly achieved today in 2004 3rd Ed.  The shortcomings you mention related to tools seem to be more related to the tools and not SCORM itself.  I am someone who does not use tools at all.  If LMSes are truly compliant, I have had great success delivering very complex, sequenced content that runs on those LMSes. 

Now, I think that SCORM 2.0 should be created with tool vendors in mind.  We should help ensure things are defined well enough that quality tools can be built on top of the specifications.

Allyn, I think this was an excellent and thought provoking paper. You make some good arguments and in fact have swayed me in a direction I didn't expect to find myself going. This quote is what got me:

"This paper is neutral to the reasons for LETSI seeking to implement a new content aggregation model for SCORM 2.0, rather it takes the approach that if this is necessary, then replacing the aggregation model alone does not provide sufficient value to warrant the effort or resulting impact across the LET communities, including vendors."

Your other arguments around the problem statements make a good case for including DITA, but the fact that we likely need to replace the current content packaging specification anyway pushes it over the top. I'm not very familiar with DITA yet, but from the sounds of things, I believe this definitely merits exploration.

Some specific comments from the paper:

-"It is also important to consider a broader definition of reuse where learning content is liberated from purpose-built LET technologies alone in favour of being mainstreamed within organizations. Rather than being treated as a separate body of content locked away in applications unable to participate in 'single source publishing' models the content is liberated for reuse according to need."

I agree that this is happening, but I don't know why. Why has learning been silo-ed?

-a) True Interoperability - "'true interoperability' it should be possible for content to be both interoperable at the level where courses are consumed 'as is' by LMS-type applications and yet are also adaptable by tools other than those used in the original creation."

I like the thinking here, but if the discussion progress I would argue that we should separate out the two types of interoperability and use different terms. Interoperability that leads to tools being able to rework content is different than interoperability of courses being consumed 'as is'. I hate to be the guy parsing words, but they I think they are two different discussions.

-b) Reusability in an organization-wide context - "online learning is that it has largely become divorced from the rest of the content in organizations. It requires different tools that create and deliver content that is often limited to the LET domain."

Absolutely, this is a problem. What should we do about it? What are the "inhibitors resulting from the current versions of SCORM (that) should be removed"

-c) The need for structured content

Your argument later in the paper convinced me that we should considering moving in this route. My initial reaction to this section however was that we need to be very careful to not tell people how they HAVE TO author a SCO, but rather provide them with a new and more powerful way that the CAN author a SCO. This goes beyond the need for simply a smooth transition, I don't think that forcing people into a structured content paradigm will ever be appropriate.

-d) Service orientated approaches

Sounds good in theory. Do you have any thoughts on how these manifest themselves in practice?

-f) Smooth transition

Absolutely agree. The new model needs to be consist enough with the old model to permit migration. Drastic change however can sometimes be better than small change. For example, changing an API call from "LMSInitialize" to "Initialze" looks nonsensical to an observer, however changing from an ECMAScript API to a web services API makes sense. I think this is the same point you are making in saying that transitioning from current content packaging to a similar but slightly different content packaging doesn't make much sense.

-g) Accomodation of new approaches

Agreed, this is the right attitude to have.

h) Metadata, CORDRA and syndication

I'd like to hear more about the unmet needs in the metadata area.

i) Cross domain scripting

I think this is the 4th time in the last two days that I've posted that I absolutely agree that this must be fixed in SCORM 2.0.

Allyn --  This is a very thorough treatment of many of the issues awaiting consideration by LETSI.  I think the content data model is a very important issue, and I agree that LETSI should consider DITA as an option.  It is LETSI's objective to use existing standards and specs (as long as they're open) as much as possible.    I agree that the value of SCORM 2.0 will be much greater if its implementation brings that 'locked away' content into the forefront or an organization's assets, enabling its use for a multitude of purposes. 

 I'm glad you embrace the focus of the TLWG and invite you to join us. 

 Jackie Haynes

It's good to see a paper addressing SCORM applied across the fields of military, academic, and corporate training. It sometimes seems to me there's a big gap between k-12 and training (in my work environment it's training development for the military) such that each thinks the other is a foreign country. But our foundations are the same: instructional design principles are the same, most of the technology used is the same, and the need to 'entice' the learner toward engagement with the content is the same.  In my master of education (instructional technology) studies, when I mention SCORM in discussion it tends to draw a blank with my classmates who are mostly in the k-12 world. I think it's a shame because educational technologists and creators of digital k-12 learning content would certainly benefit from using SCORM standards and specs.  My other reaction to this article is that it correlates with Len Murphy's Low Hanging Fruit white paper, even though Allyn's article speaks specifically of reaching beyond the LMS UI to data within an EPSS and Len's speaks toward a learner reaching through the UI to outside web sources - both papers address similar facets of the same concept. It reminded me of the relatively-new learning theory of connectivity, where access to learning comes through multiple pathways facilitated by people networking digitally. A web of learning rather than a pathway. Very interesting. Thanks for this thorough, well-structured paper.

Naima Haviland