1.2 Scorm 2.0: problem scope

Scorm 2.0: problem scope

In the following sections a number of key issues are highlighted and their requirements elaborated. At the outset, and for context, the following perspective is aguiding principle in what is presented. SCORM has always been an educational technology standard that has sought to remain agnostic to pedagogy (after all, its focus is uponcontent). While additional support may be possible for good practice, SCORM should remain agnostic to pedagogy and andragogy and should not enforce any particular approachto the access of content for the purposes of supporting learning. Individuals and organizations should remain free to choose the optimal learning, education or trainingapproach(es).

The following are the main components upon which this paper will focus prior to proposing a solution: (a) True Interoperability, (b) Reusability in an organization-widecontext, (c) The need for a structured content model, (d) Implications of service-orientated approaches, (e) Granularity, aggregation, independence and rendering,(f) Smooth transition between current and future SCORM, (g) Accommodation of new approaches to LET, (h) Metadata, CORDRA and syndication, and (i) Cross-domainscripting.

1. True Interoperability

   The level of interoperability that is most frequently available in SCORM 1.x (especially 1.3.x or SCORM 2004) may best be described aspartial interoperability. Many vendors have sought to make the creation of SCORM 2004 content simpler for course developers. As a result, achieving adaptabilityand interoperability at the same time is not normally possible without being limited to the tools of a single vendor for assembly, maintenance andadaptation. This is due to the use of wizard-like authoring/assembly tools that add proprietary scripts for aspects of sequencing and navigation. While it hasbeen important for vendors to simplify the course development process, there has been a negative impact on interoperability.

   For the most part, it is possible to upload packaged SCORM-conformant courses into a SCORM conformant LMS and run the course. The course will launch andexchange data with the LMS as required. If, however, there is a desire to adapt the course, then this must generally be done using the same tools with whichthat course was originally created. In the case of ‘true interoperability’ it should be possible for content to be both interoperable at the level wherecourses are consumed ‘as is’ by LMS-type applications and yet are also adaptable by tools other than those used in the original creation. It should be possibleto change the most granular parts of a course, reorganize the content, alter the organizations within a course or change their sequence and sequencing rules inapplications other than those in which the course was originally created.

   True interoperability should be an objective of SCORM 2.0.

2. Reusability in an organization-wide context

   Originally SCORM (and the specifications from which it was constituted) was focused on the ability toexchange courses between LMSs. In that original time and context this was an entirely appropriate goal, and one that required significant effort to achieve.In the current context, however, there has been an increasing focus on the reuse of content beyond the domain of LET alone. Initiatives such as S1000D and SCORMalignment in the aerospace and military sectors have sought to bring content that was not originally training content into the training domain. Similaractivities are evident from corporate sector under the banner of “Single Sourcing” or “Single Source Publishing”. (A useful paper by Jim Nugent isavailable here .)

   One of the obstacles currently faced by online learning is that it has largely become divorced from the rest of the content in organizations. It requiresdifferent tools that create and deliver content that is often limited to the LET domain. Content that supports LET objectives should be useful more broadlyacross the organization in areas such as decision support, online help,marketing, print-based communications, mobile devices etc. It should also be possible to source LET content from other parts of an organization or itsbusiness partners.

   The notion of creating course aggregations from content other than content specifically created for or by LET applications (eg LMSs and/or authoring andassembly tools) has not been possible without significant effort, changes and duplication of content across organizations and/or their business partners.While there are both infrastructure and content issues at play, the inhibitors resulting from the current versions of SCORM should be removed or LET contentand infrastructure will always be seen as an external ‘add-on’ to the core business.

   Organization-wide reusability of content should be an objective of SCORM 2.0.

3. The need for structured content

   SCORM has long been in need of a structured content approach that would support delivery of content to a variety of devicesand applications without the need to provide such services outside the context of SCORM itself or in a proprietary manner. The inclusion of a structuredcontent model within SCORM would also be a key factor in satisfying the objectives of (a) and (b) described above. In addition, a standard that involvesa structured content model will better provide for both the current and future needs of LET communities and would be a key component of bringing the LETfunctions of an organization back into the mainstream.

   It is not likely to be worthwhile or timely to consider developing a new structured content model for SCORM. More importantly this would tend to impedeprogress towards the achievement of organization-wide reuse as described in (b). There are already a number of potentially suitable structured content standardsand/or architectures that are likely to satisfy the requirements LET communities may have for learning content. Where existing standards are not able to fullysatisfy the requirements, they may be adapted over time.

   Incorporating a structured content model within SCORM should be an objective of SCORM 2.0.

4. Implications of service-orientated approaches

   Service-orientated approaches fundamentally “change the game” in relation to content and metadata. Usingmultiple applications to provide a service-orientated approach for LET functions is characteristically different to that of using a single monolithic application(eg LMS or Course Management System). In a service-orientated approach a number of different, loosely-coupled applications provide functionality in the form ofservices that are exposed to the user through an appropriate interface. (The result of this approach would be a Service-orientated Architecture (SOA)). The e-Framework project continues to provide a wide range of resources to support this approach. During the design stage the range of services and supportingapplications would be defined.

   Given the range of possibilities in this type of approach, it is increasingly likely that a larger number of applications would be involved in various aspectsof the content lifecycle. This in itself forces change in the way we consider reusability and interoperability of content where the primary mode of access mayor may not be a Learning Management System, and the purpose for access of that content may vary between performance support, decision support, online help,learning, education, training etc. In addition, if the issues raised in items (a), (b) and (c) are included, it would stand to reason that multipleapplications (or services) may access the same content for different purposes and possibly, simultaneously. This model contains implications for: thegranularity of content and how it is stored and managed; the purposes and levels of aggregation of content; and, the structure of the content. It should also benoted that the content may also be accessed from different devices that provide features related to mobility and accessibility (in the sense of W3C WebAccessibility).

   Within this context there should also be an awareness of the implications of distributing functions across services and applications, and ultimately theirreliance on different types of content, data and metadata repositories. The metadata that describes a learner are more likely to be stored in dedicated butfar more independent and possibly distributed repositories that can be accessed by multiple applications for a variety of inter-related purposes. Such uses mayinclude, for example, metadata concerning competence, work history, course progress, career management (internal and perhaps external),accessibility/disability, etc. SOAs will tend to rely more heavily on distributed metadata that would assist the variety of applications within asystem to control the ‘what and how’ of content access and rendering as well as improvements in the discovery and reuse. These requirements would imply adifferent type of focus on the exchange of data and metadata associated with content and users within a service-orientated infrastructure. Awareness andsupport for these approaches will be critical to the utility and successful implementation of the next generation of SCORM.

   Accommodating the broad range of requirements resulting from service-oriented approaches should be an objective of SCORM2.0.

5. Granularity, aggregation, independence and rendering

   Several of the previous sections have already suggested requirements for structured content, support forservice-oriented approaches and the same content being accessed by a variety of different application types for different purposes, and that of ‘trueinteroperability’. This section will argue that the result of the previous sections requires us to think differently about granularity, aggregation,content independence and rendering (amongst other factors). If one accepts the utility of being able to reuse content across organizational boundaries and requirements (training, marketing, tech docs, EPSS etc) then it suggests thatnot only is it necessary to move towards a structured content model, but it is also necessary to be flexible in the way that granular content assets arecombined to create a meaningful experience within a particular context. For example, one may wish to use some of the technical specifications of a productin marketing brochures or combine them different sequences or aggregations for training or within different courses, or provide access to the unique componentsfor EPSS.

   These types of scenarios are by no means new (cf S1000D-SCORM alignment; single source publishing etc). Where these requirements exist it is common touse a structured content model so that authoring and aggregation are separated from processing and rendering thereby allowing for content transformation. Astructured content architecture provides the flexibility to engage in a single sourcing approach.

   In approaches such as this, the range of uses and combinations of granular content is not necessarily known at the time of content authoring, therefore, itis also important to be able to work with different levels of aggregation as well as being able to create or edit aggregations for different purposes. Anadditional level of utility is achieved by enabling the linking of smaller aggregations of content to form larger aggregations of content. (An examplewould be reusing ‘Virtual Learning Objects’ In 2005 the author coined the term Virtual Learning Object to describe the concept of creating dynamiclinks of granular content to a manifest file and then linking that manifest to a ‘parent’ manifest. In this approach the ‘child’ manifest may be used todescribe the content of a learning object but because the content is linked and not packaged it becomes ‘virtualized’, hence it becomes a VirtualLearning Object. across different courses or outside the context of any form of sequenced, course-based activity) Of course, logic andsuitability will naturally apply some constraints on the extent to which content and aggregations are reused.

   On the one hand, the implication for content is in some ways no different than it always has been where reusability has been an objective. Content must bedeveloped for reuse. On the other hand, there is an added dimension because the reuse context could be substantially different. This places a higher premium onthe independence of content.

   The European Learning Industry Group ( ELIG ,), is recognized as the major EU association “representing the ecosystem for 21st century learning solutions”. Itacts as the main consultation body for the European Commission concerning future learning industry trends. The Chair of ELIG, Fabrizio Cardinali, recentlyannounced the establishment of new, dedicated work streams within ELIG to better focus and support the requirements of the European Publishing Industry in itsmove towards innovative technical, architectural and business solutions. According to Cardinali, “these ELIG work streams are intended to better supportthe increasing pressure from the users for open and flexible content, emulating what happened in the digital music publishing Industry in its turn towardssingle-tunes-based business models at the beginning of the millennium A dedicated workshop on “Digital Educational Content Marketplaces” (DECOM2008) has been announced by ELIG and will take place in Italy in October 2008 (For details visit the DECOM 2008 site.) .” Again, the suggestion of educational publishing to accommodate smalleraggregations of content rather than entire textbooks and the ability to reuse such content within the context of different educational activities alsosuggests a structured content model with new levels of support for granularity, aggregation, independence and rendering.

   The use of an appropriate structured content architecture can allow for all of the requirements described in this section as well as the ability to render thecontent in different ways for different uses, and indeed, different devices, applications etc.

   The ability to aggregate structured content at different levels and transform or render that content according to thecontext of its use should be an objective of SCORM 2.0.

6. Smooth transition between current and future SCORM

   Any change to SCORM will cause some measure of disruption throughout the cycles of both content andtechnology. It is important to soften the impact of such a change as much as reasonably possible. The extent to which disruption will occur and the abilityto soften such impact is very much dependent upon the composition of SCORM 2.0.

   Taking reasonable steps to minimize the impact of changes to SCORM should be an objective of SCORM 2.0, however, futurebenefit should take priority.

7. Accommodation of new approaches to LET (including instructional design objectives)

   At the beginning of this section it was stated that SCORM should remain pedagogy/andragogy neutral. The perspective in this sectionattempts to define a little more clearly that boundary between technology and support of good practice.

   In the past it may have been misinterpreted that being pedagogy neutral also meant that there was little or no work to be done in supporting ‘good practice’(good? better? best? different? appropriate? etc – pick all that apply...) in teaching and learning. An opposite view is proposed here. While SCORM should notdictate any particular pedagogical approach, it should, nonetheless, provide adequate support for a variety of pedagogical/andragogical approaches. At riskof this seeming to be an insurmountable task, some further comments on the role of SCORM in this approach will be provided. The key messages may be statedas:

   1. SCORM should provide useful structures to support different approaches to teaching and learning but not enforce any particular method.
   2. SCORM should not be seen as ‘elitist’. If the simplest and possibly mechanistic approaches to teaching and learning are all that an individualor organization can achieve, then they should not be precluded from capturing the broader benefits of standards-based content. Hopefully, overtime and with improved tools it will be easier for more learning content to be developed in line with an appropriate definition of good practice.

   The LETSI Teaching and Learning Strategies Workgroup has provided a thoughtful outline of the sort of practices and requirements that are likely to need support from SCORM 2.0. Itseems likely that the efforts of that group will result in refinement of the existing requirements and the addition of new requirements as work progresses.Due to the scale of their task, it is unlikely that the work of that group will be ‘complete’ before the end of 2008 when a decision on the composition of SCORMas the base for the future is scheduled to be made. Even if their work was complete, that should be regarded as a temporary status. There will continue tobe new approaches to andragogy and pedagogy that are both supported or challenged by technology. This should re-stimulate the question concerning thescope of SCORM.

   In the past SCORM has been about the aggregation and sequencing of [primarily] packaged courses and the ability to deploy them to multiple Learning ManagementSystems or Course Management Systems. The premise was that an LMS or CMS would provide all the functionality required for web-based course delivery. Bycontrast, many of those involved in Web 2.0 learning would question the value of an LMS or CMS for any of their requirements. In reality, these exist as the twoopposite ends of a spectrum of requirements for the role of learning technology and standards supporting learning (including training).

   Possibly the biggest difference between the requirements of the past and those from now into the future from the technology side is that there is a far greaterlikelihood of a larger number of applications with very different functionality accessing [the same]content and metadata from a larger number of sources than in the older models. From the human stakeholder perspective we could draw uponNigel Paine’s “5 Key Shifts” and say that while in some situations courses will continue to be delivered to an individual under tight control and primarily forthe acquisition of skills, the shift is towards the creation of environments for the free-flow sharing of knowledge among communities to support changes invalues and attitudes as well as the acquisition of actionable knowledge.

   SCORM should not be about method (ie how to facilitate learning), it should be about liberating content and making it available (existing or created as part oflearning) to a broad set of applications so that any learning method may be supported.

   Currently we are sorting through the issues and challenges posed by requirements to support collaborative learning, social learning, Open EducationResources, discovery learning, adaptive content, immersive experiences, simulations and the impact of these challenges on learning content. In thefuture, a new set of challenges will emerge along with a new set of applications to support them. SCORM should both support and survive these challenges. Ofcourse, the levels of success are dependent not only upon SCORM itself, but how it is implemented at both the application level and the system level.

   Supporting the different ways for accessing, adapting or creating content in the context of new models for learning and trainingshould be an objective for SCORM 2.0.

8. Metadata, CORDRA and syndication

   While the IEEE LOM (LOM) metadata has become the default schema for describing learning content, it is also widely acceptedthat it is no longer adequate. Reliance on LOM alone would prove to be of very limited value in the context of enterprise-wide content reuse. Thinking aboutmetadata has moved on considerably since the creation of LOM and it is now important to revisit metadata in the context of the issues presented here andthose raised through the existing work on complimentary architectures such as the Content Object Repository Discovery and Registration/Resolution Architecture(CORDRA). In addition, it is important to consider that SCORM and CORDRA may also need to be aware of other types of related metadata (competencies,profiles, etc) and that these may need to be factored in to the overall design.

   While work on CORDRA has been undertaken at ADL with related agencies, content syndication using standards such as RSS has proliferated widely across the web.In some cases this is seen as an alternative to some CORDRA-like functions, however, it may be more useful to consider RSS as complimentary to CORDRA’sobjectives. Enabling a comprehensive set of syndication functions related to learning content and repositories may also require some consideration of theseneeds within the composition of SCORM.

   Revision of the approach and schemes used for metadata and support for content syndication should be an objective of SCORM2.0.

9. Cross-domain scripting

   The cross-domain scripting issue has been a pain point in SCORM for anyone dynamically aggregating content from distributedrepositories for delivery via SCORM conformant applications such as LMSs. A range of alternative approaches is possible and the need to find suchalternatives increases as service-orientated approaches become more widely deployed within infrastructures to support learning activities. The problemneeds to be resolved. (The content of this paper will not make any contribution to solving this issue.)

   Solving the cross-domain scripting issue should be an objective of SCORM 2.0.