background image
DigiCULT 23
content design which can ultimately lead to
difficulties in making content, content repositories
and Learning Management Systems (LMSs)
comply to e-learning specifications;
| discusses how some restrictions rather than making
available a wider range of reusable high quality
learning materials may achieve exactly the opposite
| proposes a more open approach to the issue and
outlines the implications to content suppliers, as
well as content repository and LMS vendors.
his paper is based on one originally written by
us in the summer of 2002, in the specific
context of the UK education sector.There have been
developments since then in the IMS Simple
Sequencing specification.These may result in a
greater ability in future to address the problems of
appropriate navigation within a delivery system
rather than within content.This revised paper does
not attempt to review the impact of such changes in
detail; however, we believe that the basic contentions
of our original paper still stand.
Learning Object Interpretations
The main objective of creating learning objects is to
achieve the goal of maximum reusability, leveraging
the high cost of production of quality materials
without sacrificing the learning meaning. For the
benefits to be available outside the publishing
community, learning objects need to be able to be
individually broken out and delivered in a wide
variety of systems, rather than being delivered as part
of fully constructed courses via proprietary systems.
The meaning and precise definition of learning
objects has been the subject of much debate which
it is not the intention of this document to repeat.
Stereotyping viewpoints somewhat:
| at one extreme is the `publisher' view in which an
object is little more than an asset to be incorpor-
ated in a larger piece of material;
| at the other extreme are practitioners who insist
that a single learning object should completely
address a specific learning objective, by delivering
the learning AND assessing whether it has been
| in-between are a range of views that often take a
pragmatic stance based on the size of the learning
object and its practical ability to be reused in other
ome argue that `all of the above' are valid learning
objects which is a philosophically generous
approach, but does not provide content producers or
LMS implementers with any help as to what they are
supposed to be building and supporting.
Views on the ideal size and length of learning
objects vary with the proponents' position in the
debate above, but the consensus appears to be around
a range of 1-15 minutes. More flexible delivery
systems allow existing assets such as documents and
presentations to be included `as-is' and there will
inevitably be simple content based on typed text,
but content developers are encouraged to avoid
`page-turner' content where the material is basically
pages of a book replicated as Web pages, with
interaction being limited to `turning the page'.
Learning Objects and e-Learning
Given that there is a range of interpretations of
`learning object', specifications groups have tended
to avoid the term, in favour of other terms with
definitions that can be agreed by producers and
The Advanced Distributed Learning Network's
SCORM specification (ADL SCORM) defines a
`Sharable Content Object' as follows:
A Sharable Content Object (SCO) represents a
collection of one or more Assets that include a specific
launchable asset that utilizes the SCORM Run-Time
Environment to communicate with Learning Management
Systems (LMS). A SCO represents the lowest level of
granularity of learning resources that can be tracked by an
LMS using the SCORM Run-Time Environment.
(SCORM V1.2 Content Aggregation Model,
October 2001).
SCORM defines a Run-Time Environment which
allows content to exchange data with the delivery
environment.This includes launch parameters,
interim data such as bookmarks (saved when the
SCO is suspended and retrieved when the SCO is
re-launched), feedback to the LMS of objectives
achieved, and an assessment score and maximum
score which can be compared with a pre-defined
mastery score.The API and its data model only
operate within a single SCO, i.e. one cannot begin a
SCORM interaction within one SCO and complete
it within another. A SCO's use of the run-time API
can theoretically be limited to a simple `initialise call'
followed by a `complete' call. Except in this trivial
case, the need to complete an interaction within the
SCO can limit its minimum size.
Digicult_THI4_backup_13_10_03 24.10.2003 11:54 Uhr Seite 23