Information Technology Reference
In-Depth Information
analyses the MUPPLE pages automatically and
extracts typical action-outcome-tool triples into
an own database.
This information is used for inexperienced
users if they want to start a new action using the
corresponding facilities in the navigation area
(below the widgets for creating new activities).
As shown in Figure 13, MUPPLE suggests a list
of tools that were used by other learners within
their activities. So far, the list of recommended
tools for the statement 'publish call-for-paper' is
unordered, but additional information (usage in
activities derived from the same pattern, closeness
to the action and/or the outcome, etc.) could be
used to rank and filter these recommendations.
Overall, these aspects of personalization in
MUPPLE outline the differences and similarities
to traditional streams like adaptive educational
hypermedia or intelligent tutoring systems. On
the one hand, MUPPLE is based upon a simple
pedagogical model of learning activities that
consist of a set of action-outcome-tool triples.
This semantic model is easy to understand and a
solid base for planning and realizing further ad-
aptation strategies. On the other hand, adaptation
mechanisms are not realized according to design-
er-driven, top-down, ex-ante design. They are
either based on services, like the action-object-tool
recommendations, or come from propagation
effects within the learning network, e.g., when
pages are derived from patterns or during col-
laboration on shared artefacts. In the end, the
personalization effects on the learning environ-
ment are reflected through the visible surface of
the mash-up PLE.
FUTURE RESEARCH DIRECTIONS
Despite the possibilities of our MUPPLE approach,
a few disadvantages have to be outlined here.
Primarily, these problems concern technological
issues.
First of all, it could be even nicer to have a
high degree of interoperability between web ap-
plications, which now is not always the case. This
specifically relates to single-sign-on procedures
and communication channels to transfer both data
and events from one application to the other. For
example, a the wiki and the webmail client require
authentication, so learners right now have to log-
in separately in each application (not that they
were not used to it). The approach would benefit
from authentication mechanism such as OpenID
(http://openid.net), to avoid these repetitive logins
in each participating application.
Regarding communication channels, we have
proposed a specification how to realize distributed
feed networks with buffered-push capabilities (cf.
Wild & Sigurdarson, 2008). We intend to further
investigate these means (beyond the 'connect'
statements available today). We can think of other
Figure 13. Recommendation of tools within an activity
Search WWH ::




Custom Search