Social, Personalized Lifelong Learning - E-Infrastructures and Technologies for Lifelong Learning - page 99

metaphor: DM represents the topic(s) on which the
presentation (such as a PowerPoint presentation
represented by the GM) is built. From one topic
(or DM) one can construct several presentations
(here, GMs), depending on the goal. This goal, in
a learning environment, can be a set of learning
objectives, which are either implicit, or would need
to be expressed separately. A presentation does
not contain a whole topic, just some (constrained)
part of it. Furthermore, a presentation can contain
information from several topics. The separation
therefore gives a high degree of flexibility, based
on the DM-GM multi-multi dependency.
Another important difference is given by the
notion of 'concept' used in the domain model. In
LAOS, concepts have different representations
defined via attributes, and are restricted to repre-
senting a semantic unity (unlike in AHAM). This
is further enforced by allowing only self-contained
attributes (without direct or indirect dependen-
cies). This setting allows attributes to be flexibly
re-ordered, and links are therefore external and
can be dynamic.

Unlike some of the other models, such as
XAHM or WebML, LAOS does not prescribe
a unique representation for each layer, but just
specifies its contents. Thus, each layer could be
represented by databases, XML, state machines,
etc. Moreover, the adaptation model, LAG, only
specifies the different entry levels for reuse (whole
strategy, high level adaptation language patterns,
or low level adaptation 'assembly' language pat-
terns such as if-then rules) but does not enforce a
specific language. An example authoring system
built on LAOS is MOT (Cristea & De Mooij, 2003).
To summarize the main features examined
in the previous models and how they compare
with LAOS in short, we provide Table 1 with a
comparison between these models.
For the reasons above, and due to the fact that
it provided most of the desired features, as shown
in Table 1, we have selected the LAOS framework
for further development in our research.

Table 1. Comparison between models of adaptive (educational) hypermedia

AHAM Munich WebML XAHM LAOS

Separation of concerns Yes Yes Yes Yes Yes

Reusability No No To some extent To some extent Yes

Different user roles No No Yes No Yes

Flexibility (different
formats, etc.)

Yes No No No Yes

Pedagogic layer No No No No Yes (via Goal

and Constraints
model)

Group representation No No Yes No No

Social interaction No No No No No

Approaches Database /XML Object-oriented UML XML Database /XML

Target A(E)HS A(E)HS Web App AHS A(E)HS

Notes:
A(E)HS: Adaptive (Educational) Hypermedia Systems
AHS: Adaptive Hypermedia Systems
Web App: Web Applications