Silvers - Content Authoring - SCORM 2.0

	SCORM 2.0	Log In \| Sign Up
	Silvers - Content Authoring

Browse Space

Added by Avron Barr, last edited by Daniel Ingvarson on Sep 10, 2008 (view change)

Home	About LETSI	Events	SCORM 2.0	Contact LETSI
Standing Committees:	Policy & Procedures	Strategic Communications	Technical Roadmap	Membership
Working Groups:	Architecture	Business Requirements	Sequencing	Teaching and Learning Strategies

Content Authoring Standards and Services

Aaron Silvers, Grainger

Summary

SCORM 2.0 should expose content authoring and packaging services to promote interoperability, relying on existing and developing technologies, standards and specifications.

Requirements/Needs Outlined

Content authoring using complex presentation frameworks (such as the RTE and sequencing and navigation) provides for less interoperability than is desirable. Improve these higher order functions and simplify their use and adoption

There are two elements proposed

That a content packaging service could be abstracted (ie provided by a expert third party eg LETSI) such that the final steps of packaging could be done by this service thereby improving the interoperability of the content as each LMS would receive content that is consistent.
That a template system to simplify a number of items that are implemented differently across LMS's.

This is because many developers have found the difficulties of using the more advanced features to great and have bypassed the use of the SCORM standard and done those functions from within a flash object. This reduces the branding, LMS navigation and use tracking from the LMS. By adding agreed templates that can be overridden and simplifying the creation of packages more authors would use LMS accessible (an modifiable) features rather than hiding these complexities within the objects.

Recommendations

Create templates or Services which can transform raw content into correctly formed objects. Examine the Content Authoring Standard from OASIS for structural ideas

The bottom line is, make it simpler by introducing templates and an online service that would simplify the creation of consistent packages ie whats the point in ultimate flexibility if the evidence shows that it is not used and can even lead to a reduction in the elements of the spec that are used.

Comments (Hide)

Aaron, interesting idea and well done! Any chance of pulling together a diagram, happy to help. I want to make sure That I am interpreting this correctly. The idea is good, make stuff easier with the help of templates. I will drop you an e-mail

Posted by Daniel Ingvarson at Sep 01, 2008 03:39

Aaron I pretty much agree with the idea presented here. It is just a matter of choosing how something like this would be done. In actual fact we created a version of this just using html back in 2001. There are still some versions of the original paper kicking around on the web. I found one of them here - SCORM_SSS. The LMS we used (Avilar) provided the skinning we needed as templates. (The other piece of work that emerged from what we did in that implementation was one of the solutions for cross domain scripting (the SCO fetcher model, but I think we all agree there needs to be a better solution than a workaround.)

DITA would fit pretty nicely into this model as far as I can tell.

Posted by Allyn J Radford at Sep 01, 2008 16:56

A couple points come to mind after reading this paper:

1) Some kind of defined content structure seems to make sense in many cases. Whether or not this should be part of Core SCORM or an profile of SCORM is still uncertain in my mind, but you've made the case well that there are definite benefits.

2) You start the paper off by pointing out the complexities and interoperability problems caused by different interpretations of runtime behavior. Then, you go on to suggest that we encode our conent in XML and use N templates to render that content. Wouldn't that exacerbate the interoperability problem by adding even more interfaces and more potential interpretations into the equation (1 content to N LMS's becomes 1 content to M templates to N LMS's). And, these interpretations would be in a highly visible area (content rendering) that would likely lead to more interoperability complatins than the subtle interactions that happen behind the scenes in the SCORM runtime. I'm not saying that it's a bad idea, but I don't necessarily see it as simplifying interopability (correct me if I'm missing something here).

Posted by Mike Rustici at Sep 03, 2008 14:49

Mike, I don't like to point out problems without highlighting "a" solution. I advocate the use of, as you call them "n templates" to solve interoperability issues at the platform level -- which *could* allow for greater re-use. This isn't the only way to make reusable content, even if we agree that a shared content format schema is a generally good idea.

I posit that if we have one defined content format to interpret, that will improve interoperability. It *may* improve reusability and portability if a packaging service was structured to take content and its required assets from one system and transport them somewhere else.

My goal here is to push the conversation forward. I would not pretend to have a definitive answer. Allyn, for example, is far more familiar with the ins and outs of DITA than me.

Posted by Aaron Silvers at Sep 09, 2008 15:03

Aaron,

What you have written sounds very useful. I will read it more thoroughly and give more comments, but my first gut reaction is that the complexity of content structure and template system will not be widely adapted without tools that help developers create these packages. This is certainly true in higher education where faculty are the people creating the content and do not usually do complex content development. This is one (of many) of the reasons SCORM has not taken off in higher education.

generally speaking the development of good tools to help the practitioners is needed for any standards or specifications to be widely adopted (for those that do not have access to programmers).

Posted by David Wirth at Sep 09, 2008 11:45

David,

Point well taken -- what I'm hoping to see accomplished should this idea come to fruition is that a suite of services is provided to everyone -- the alpha nerds who will roll their own solution to small development houses to large-scale big box retailers of tools like Adobe -- all using the same set of services to ready content for SCORM 2.0.

I'm not advocating for any particular authoring solution -- I'm advocating for a profile that is shared across the board -- LETSI provides the hooks for content readiness regardless of your tool -- as long as you can fill in the blanks, the services handle the lifting. This is the same logic we've been talking about at a high level for Sequencing. I'm just applying it a step further down the chain to the common problems of content interoperability.

If everyone needs to format their content templates to meet the same specs, it's just going to be way easier to innovate and differentiate in other ways.

Posted by Aaron Silvers at Sep 09, 2008 15:07

Aaron,

Interoperability is one of the "ilities" which SCORM 2.0 should strive to improve. This can be one of the approach.

Aaron, can you clarify few things:-

1. How content authoring tools would interact with the packaging and template services?

2. What is the expectation from the LMSs if we adopt this approach?

Posted by Jayashree Pandeya at Sep 11, 2008 07:01

> 1. How content authoring tools would interact with the packaging and template services?

In my mind, where all my ideas just compile perfectly? Say you take an off-the-shelf authoring tool like Articulate, but it could be any E-Learning authoring tool. It may not need the content authoring "service" but it could use the schema(e) to validate and format the core of the content and any required content-level media.

Further, it *could* use the/an authoring service to validate the content structure and make sure that any required hooks into system-provided templating or styling would be available to it.

Now if the authoring tool was built into the LMS, or was an extention of the system (whatever that is), that's all you would need. If the authoring tool was standalone, or at least separate from the run-time, it could then use the packaging service to appropriately package content for the scorm 2.0 profile, or to whatever the "current" profile was.

My point being that in keeping it in the cloud and by keeping it an open technology, it's available for everyone to use. This saves everybody time and money -- and at the same time promotes interoperability.

> 2. What is the expectation from the LMSs if we adopt this approach?

I guess that depends on a number of factors. If the cloud is a good idea for authoring and packaging, maybe the cloud is a good idea for run-time and sequencing, too. Then LMS vendors can spend more time on the experience of their products as differentiators instead of whose run-time "works" (quotes intentional).

Posted by Aaron Silvers at Sep 11, 2008 14:16

Aaron,

How does this logic work with the notion in Ellaway (No.22) and Wiley (cited all over the place) that we should get past the idea of instruction as content?

Maybe it has nothing to do with that idea, but I thought I would ask anyway.

Thanks

Posted by Eric Roberts at Sep 12, 2008 14:02

I am late to the show. The paper's point might be fulfilled by an improved Reload that applies S&N to a package of content.

Posted by David Searcey at Sep 29, 2008 16:15

Thanks for the paper! A standardized format would greatly increase resuability of content. At heart I am all for this idea, but I am afraid it may be limiting. For example, how would I use DITA to make a SCORM conformant flight simulator? Or a SCORM conformant classroom lecture?

While I am initially in favor of the ideas you suggest, it may be better to consider this as a community extension to the CoreSCORM product.

Posted by Mark Ewer at Oct 13, 2008 07:54