OPERATIONALIZATION OF LEARNING SCENARIOS ON
EXISTENT LEARNING MANAGEMENT SYSTEMS
The Moodle Case-study
Aymen Abedmouleh, Pierre Laforcade, Lahcen Oubahssi and Christophe Choquet
LIUM, IUT de Laval, 52 Rue des drs Calmette et Guérin, F-53020 Laval Cedex 9, France
Keywords: Learning Management System, Moodle, Learning Scenario, Visual Instructional Design Language,
Operationalization, Domain Specific Modeling, Model Driven Engineering.
Abstract: Despite the increasing number of Technology Enhanced Learning platforms (eg. MOODLE) and their wide
spreading, the operationalization of learning scenarios is still a problem for teachers. We aim to facilitate
their implementation on existent platforms. We propose an approach based on the formalization of the
implicit instructional design domain language embedded by these Learning Management Systems. It
consists to identify and formalize this specific language in order to use it as a mean of communication with
external design tools without losing the semantic of the designed scenarios. The originality of our approach
relies in performing the scenarios operationalization by the development of a communication API based on
the formalized language of the platform. Our proposal is also based on the application of theories and
practices from the Domain Specific Modeling domain in order to formalize the domain language, to specify
some graphical languages on top of it, and to help in the development of dedicated graphical editors. This
paper details the implementation of our proposal (API and first editor) on the MOODLE platform.
1 INTRODUCTION
Most of academic organizations provide teachers
with some Learning Management Systems (LMS)
for improving or completing their face-to-face
courses by some additional activities from simple
resources access to scheduled communication or
online assessments. However, the management and
the appropriation of theses platforms by practitioners
are complex tasks (Al-Ajlan and Zedan, 2007).
Some LMSs, such as MOODLE, do not hide this
complexity and struggle to provide assistance and
appropriate means despite their numerous and active
communities. Also, each platform embeds both a
specific instructional design paradigm and a specific
pedagogy. However, practitioners are not familiar
with this implicit instructional design domain
(Martinez-Ortiz et al., 2009). They are not able to
implement scripts required by these platforms
(Mekpiroona et al., 2008) or to adjust the many
parameters of the form-based interfaces. The
teachers-designers could be disappointed by the
semantic distance between Educational Modeling
Languages (EML) and the LMS because they cannot
implement their scenarios specified by these EMLs.
Several technical approaches provide design tools
which do not instrument the operationalization. As a
result, the practitioners are still looking for design
and implementation approaches more appropriated
to their practices and expertise.
We discuss in this paper the issue of the learning
scenarios design for educational platforms. We
present a new approach that facilitates the
implementation of learning scenarios on these
platforms. This approach consists (1) in identifying
and expliciting the instructional design languages of
platforms and (2) exploiting them by providing
practitioners with some new tools based on these
external languages in order to facilitate the design of
learning scenarios.
Next section presents the operationalization
activity and an overview of the current approaches
for automating this activity.
2 EXISTING APPROACHES
The operationalization of learning scenarios consists
in implementing teachers-intended scenarios on one
TEL (Technology Enhanced Learning) environment.
It resumes some manipulations as creation of
activities, selection of participants, allocation of
143
Abedmouleh A., Laforcade P., Oubahssi L. and Choquet C..
OPERATIONALIZATION OF LEARNING SCENARIOS ON EXISTENT LEARNING MANAGEMENT SYSTEMS - The Moodle Case-study.
DOI: 10.5220/0003486001430148
In Proceedings of the 6th International Conference on Software and Database Technologies (ICSOFT-2011), pages 143-148
ISBN: 978-989-8425-77-5
Copyright
c
2011 SCITEPRESS (Science and Technology Publications, Lda.)
roles and selection of required services and content
for scenarios (e.g. pedagogical resources). However,
the operationalization of learning scenarios is not
only an engineer activity. It aims to translate
teacher’s intention and relative pedagogical
semantics on a TEL system. Two approaches can be
followed: manual or automatic ones.
Most of learning scenarios are manually
implemented. It consists firstly in choosing the
participants, then attributing roles foreseen by the
scenario to the proper participants, and finally
selecting the services and contents required by the
scenario. This approach requires the intervention of
a pedagogical engineer or an expert to make the
necessary manipulations in the target platform.
In contrast, the other approach aims to
implement automatically learning scenarios on TEL
environments. The intervention of LMS experts or
engineers is not necessary. Nevertheless, this
approach requires an infrastructure for interacting
with the LMS and taking in charge the creation and
configuration of the working spaces, as well as the
activity performance, starting from a formalized
description of the task. To this end, such approaches
require a 'language' for specifying the learning
scenario, and a binding technique (or a formal
language mixing both) for allowing the machine-
readability of scenarios.
We are focused on this last approach. Relevant
works fall into four categories: standard oriented
approach (as IMS-LD (De Vries et al., 2006) or
CopperCore (Berggren et al., 2005), practitioners
oriented approaches (as COLLAGE (Hernández et
al., 2006) or LDL (Martel et al. 2007)), approaches
proposed for specific platforms (as LAMS (Delziel
et al., 2006)) and the hybrid approaches based on
Model Driven Engineering techniques (as Bricoles
(Caron et al., 2005)). The analysis of these
approaches leads us to observe that:
The different proposed solutions do not fit with
the teachers-designers needs we mentioned,
excepting the LAMS one which partially satisfies
them with its user-friendly interface.
Nevertheless, LAMS editor integration into
existent LMSs does not take advantage of the
potential internal semantics embedded in the
platform: it requires to add to LMSs a new
runtime engine with its dedicated course format.
The COLLAGE proposition is interesting
because the collaborative design patterns
proposed to practitioners have been specified and
developed on top of the IMS-LD standard:
semantics about concepts/relations transfor-
mations have been taken into account when
building the patterns; these patterns are this fully-
compatible with IMS-LD. In the other hand,
operationalizing COLLAGE models relies on
operationalizing IMS-LD ones. Unfortunately,
most of existing platforms are still not compatible
with this standard (Berggren et al., 2005)(Burgos
et al., 2007).
Research works dealing with exportation or
transcription of learning scenarios have highlighted
the semantic learning design gap that appears when
considering learning scenarios concepts and
platforms features (Caron et al., 2005). Such
scenarios transcriptions lead to lose some
informations when binding the source scenario on an
LMS. This conceptual gap is inherent to the
transformation process when both languages have
been elaborated with no reciprocal relations.
Teachers naturally decline any tools or
approaches that are not able to facilitate the course
design on their platform. It seems that current
research propositions have not yet reached a level of
maturity for allowing teachers-designers to
implement theirs scenarios.
3 OUR APPROACH
Current propositions rely on a same underlying idea
about evolving existent Learning Management
Systems by large add-ons (editors or runtime
engines) and new semantics in order to integrate
learning design standards or to improve the design.
We do not aim to add new semantics to LMSs.
In contrast, we assume that each LMS is not
pedagogically neutral and that it embeds an implicit
language for describing the process of designing a
learning activity. Thus, our proposition is based on
the idea that this language can be identified and
explicitly formalized in a computer-readable format.
We propose in addition to use this formalization as a
binding format for various external tools which will
focus on different designing facets as well as
interoperability purposes between various LMSs
expliciting their instructional design semantics.
Nevertheless, LMSs have to be able to
import/export learning scenarios in conformance
with their own language: current platforms have
notwithstanding to evolve in order to offer this new
functionality. From an LMS viewpoint, our
proposition is to add a similar 'import/export'
functionality like the SCORM (Gonzalez and Anido,
2010) one but with the LMS language itself (as it is
already proposed for specific sub-domains like the
quiz one for MOODLE). We propose so a kind of
self-compliance format specific to each LMS. This
will warrant e-learning tools developers that they
could exploit this explicit language (that will have to
ICSOFT 2011 - 6th International Conference on Software and Data Technologies
144
be accessible through an XML schema for example)
for communicating with the LMS.
Operationalizing a learning scenario from this
LMS-centered viewpoint will consist then in the
importation of a learning scenario formalized in
conformance to this explicit LMS language. It will
not require the addition of a new runtime-engine like
the SCORM/LAMS or other approaches.
Figure 1: Our external approach for improving learning
designs on existent Learning Management Systems.
We also propose an original TEL-centered
Model-Driven Engineering and Domain-Specific-
Modeling (DSM) (Kelly et al., 2008) approach both
to identify/formalize the LMS language and to use it
as a basis for the elaboration of LMS-centered
Visual Instructional Design Languages and their
dedicated graphical authoring-tools. From a
metamodeling viewpoint, every LMS language can
be considered as composed of an abstract syntax
(formalized as a metamodel and additional well-
formed rules), a concrete syntax (the machine-
readable textual notation that will be used for the
binding of learning scenarios), and some semantics
for both syntaxes.
The explicitation of LMSs languages allows the
specification of VIDLs/EMLs on top of them. This
approach will propose also a new opportunity to
design and operationalize learning scenarios. A first
step for this approach is to provide practitioners with
some external learning design editors based on the
LMSs languages. It is also important to provide
practitioners with some learning design editors
dedicated to the VIDLs built on top of the LMS
language. Many VIDLs can be proposed for a same
LMS language according to the LMS semantics they
will include (whole or part). These LMS- centered
VIDLs have to be composed of the same abstract
syntax than the LMS language (same domain meta-
model), but have to propose a visual notation (e.g. a
concrete syntax) in order to facilitate thinking and
communication for practitioners (human-interpre-
table formalism). In contrast, the dedicated editors of
these VIDLs have to manage the persistence of
produced learning scenarios in the machine-
readable format of the considered LMS (binding).
Our proposition is focusing on a DSM approach
that aims to offer a practical solution to produce
scenarios according to the semantics of the LMS
language. DSM tools will manage the binding to the
LMS machine-readable format. We propose to use
these DSM tooling in order to elaborate some LMS-
centered VIDLs and dedicated user-friendly editors
based on the meta-model of the identified LMS
language. We experimented such DSM tools, the
ones from (Eclipse Project, 2011). They are able to
specify all these artifacts (domain meta-models,
graphical and textual notations, generation of
dedicated editors, etc.). These tools have been
experimented within several projects of different
scopes and following practitioners centered
viewpoints as well as TEL-centered ones (Laforcade
et al., 2008; Laforcade, 2010).
Our LMS-centered solution guarantees that
future produced scenarios will be implemented on
the concerned LMS taking account the probability
that this solution may restrict the pedagogical
expressiveness of learning scenarios. But we assume
that the explicitation of the internal LMS language
will create the opportunity to build more
practitioners-directed but LMS-centered on top of
the initial LMS language. This external approach
can also exceed the technological constraints and
limits of existing TEL systems and offer more user-
friendly computer artefacts to work with. Figure 1
illustrates our proposal detailed within this section.
4 THE MOODLE CASE- STUDY
We have chosen to apply our proposal on the
MOODLE LMS. It is a distance learning platform
based on a socio-constructivist pedagogy (Cocea,
2006). Our choice is motivated by: its open source
code and its modular and extensible architecture, its
large community of users and developers, and its use
in our university.
4.1 Internal Language Identification
First of all, we had to identify concepts, attributes
and relationships composing the abstract syntax of
the MOODLE language. The relevant instructional
design facet relies on the teacher’s course space
functionalities. We have performed an analysis work
by combining three viewpoints: user interfaces
analysis (what the designer sees), MOODLE
database analysis (persistent objects and data
associated to courses), and functional and technical
OPERATIONALIZATION OF LEARNING SCENARIOS ON EXISTENT LEARNING MANAGEMENT SYSTEMS -
The Moodle Case-study
145
analysis of the platform specific mechanisms linking
human machine interfaces (HMI) and database (e.g.
backup techniques, etc.).
By analysing the user interfaces, we have
identified a set of challenges faced by teachers when
using platforms. For example, changing informa-
tions in a Moodle course requires to fill a long list of
operationalization-oriented parameters: full name,
short name, identification number, course format,
number of topics, etc. Some parameters are required
and mandatory (full name and short name) for the
creation of courses but others are optional or too
technical (course registration parameters, file size,
etc.). We have set up concepts used in space courses
and also links or relations between them according
to screens sequences. Briefly, a course is a set of
editable sections that serve as structural components
for defining and positioning activities and resources.
By analysing the MOODLE database, we have
identified the system data organization. We have
analysed about 266 tables of MOODLE 2.0. This
has highlighted relationships between different data,
in particular those used by many types of activities
and resources available in a course. Crossed with
HMI analysis, data analysis let us to identify
properties, attributes, concepts (linked to some
portions of screen-forms), and to verify and validate
relationships between concepts. We have also
checked off required data and optional ones (i.e. can
be omitted without making database inconsistent or
incoherent).
By analysing the platform functionalities, which
has consisted to successively test all possible
handling of the identified objects, attributes and
relationships, we have characterized the MOODLE
provided backup functionality for saving existing
courses into an external format (here an archive
containing many XML files and course resources,
organized in folders). MOODLE also provides a
restore functionality to import courses from external
packages previously saved.
4.2 Language Explicitation
The explicitation of the specific MOODLE
instructional design domain requires the choice of a
concrete syntax for the representation of the
language vocabulary and grammar.
Its widely use in interoperability standards and
its use by the backup/restore functionality of
MOODLE has oriented us to choose XML. Firstly,
we have chosen to develop an instance of an XML
document representing a complete course in order to
help us to set representation choices (tags for
concepts, sub-tags for attributes, etc.). Then, we
have specified an XML schema that have been
completed and refined in order to clarify all possible
semantics (limited to the XML Schema expressive-
ness) that was not explicit in the previous XML
instance: multiplicities of relationships, enumerated
types, etc. These two steps have been primarily
directed by the analysis of the package contents
generated by the pedagogical course backup. We
have proceeded by successive refinements:
modifying/adding/deleting data in order to observe
the generated course package files and to isolate the
XML fragments to deduce the concrete syntax of our
XML document. We have also taken into account
the results of our previous analysis to specify
relationships between different concepts. Finally,
our XML instance which specify a full course, has a
different structure and formalization in comparison
to the numerous XML files generated by the
backup/restore functionality.
The associated XML schema has been specified
by using the more relevant Russian dolls design
pattern. This schema formally represents the
MOODLE domain language we propose as a
communication format between external tools and
this LMS. It may also be used to validate XML
documents that will be imported or exported.
4.3 Import/Export Communication
API
In order to import (and to export) learning scenarios
into (from) the MOODLE platform, it is necessary
both to develop and to integrate a new
communication module (IN/OUT Course) in its
design space. This new module provides an API
between the external design tools and the LMS. It
ensures the operationalization of learning scenarios.
We have taken into account the possibility to
successively perform import/export operations in
order to adapt a course with external design tools.
Therefore, we also took into account that external
tools could only be compliant with all or part of the
formalized language (to focus on specific design
facets for example).
The communication module integrates two
processes: an export process to generate an XML
instance which specifies all course details and an
import process able to operationalize scenarios
specified in conformance with the platform
language. In order to ensure the cooperation of these
two processes in terms of first creation (first import
into an empty course) and adaptations (by addition
and deletion of concepts), we have initially tested it
on the basis of two concepts (forum and label).
Then, in a second step, we have extended the
development throughout the whole language.
ICSOFT 2011 - 6th International Conference on Software and Data Technologies
146
The first process has to deal with the course
backup and its formalization in the specified
platform language. Concretely, it relies on the
backup functionality to generate the MOODLE
course in an external package. Considering the XML
format of the data package, we have developed an
XSLT transformation using the XML schema we
defined, in order to generate and to apply the
necessary transformations rules on the different
XML files. This process leads to generate an XML
document instance conformed to our XML schema
and handleable by the specific external design tools.
The second process leads to import scenarios in
the platform. We have chosen to use the existing
restore function of MOODLE to operationalize the
scenarios. The general idea of the transformation
algorithm we specified is to: (1) perform a course
backup, (2) complete and modify this package with
information specified within the XML scenario
instance (with XSLT transformations), then (3)
achieve the restore of this modified package. This
process takes into account some constraints when
the API is used to adapt an existing course.
Concretely, the XSLT transformations take into
account four possible cases: (a) modification of
some informations from an existing course concept;
(b) creation of a new scenario concept not already
present in the course (c) deletion of a course concept
(i.e. concept appearing in the backup but not in the
scenario to import with the additional information
that the external tool was able to handle this
element); (d) omission of a concept (i.e. concept
only appearing in the backup files while knowing
that the external tool was not able to deal with it).
External tools that do not support all the LMS
language have then to precise their "domain
language perimeter of understanding", an imposed
constraint for future external tools, by providing a
subset of the XML schema we defined.
4.4 A First DSM-based Editor
From the XML schema dedicated to the MOODLE
learning management system, we automatically
generated an equivalent metamodel (figure 5) thanks
to the EMF tooling. This metamodel can then be
used as a basis for the development of Visual
Instructional Design Languages, and their dedicated
graphical editors, according to the DSM approach.
We have developed a first visual editor. It aims
to graphically ease the specification of a course
structure for MOODLE. We used the EMF DSM
tool already used to formalize the MOODLE
metamodel. EMF is a Java framework and code
generation facility for building tools and other
applications based on a structured model. Once you
specify an EMF model (or metamodel), the EMF
generator can create a corresponding set of Java
implementation classes. In addition to simply
increasing your productivity, building your
application using EMF provides several other
benefits like model change notification, persistence
support including default XMI and schema-based
XML serialization (the feature that interests us in
order to offer a binding towards the platform XML
schema), a framework for model validation, and a
very efficient reflective API for manipulating EMF
objects generically. Most important of all, given an
EMF model definition, the EMF code generator can
produce a fully functional editor tool with a tree-
based interface that will allow to view instances of
the model using several common viewers and to add,
remove, cut, copy, and paste model objects, or
modify the objects into property sheets.
Although we also plan to use GMF (Graphical
Modeling Framework) as a complementary
framework to EMF in order to generate user-friendly
and graphical editors, we decided at first to focus on
the EMF preliminary editor. By so, a full- generated
prototype, as a plug-in for Eclipse, has then been
generated by the EMF tooling in accordance with
the MOODLE domain model. This editor does not
require some computer skills but a graphical version
with GMF will be more adapted for
teachers/designers. From this EMF-based editor,
visual scenarios are automatically serialized in an
XML machine-readable format in compliance with
the XML schema we propose (ie. the tree-based
view is synchronized with the XML-formatted
scenario; no other binding functions or services have
to be performed).
We have experimented and tested this editor in
relation to the previous communication API. We
verified the efficiency of the four manipulations of
MOODLE course scenarios (creation, deletion,
omission, modification) and succeeded in
operationalizing the successive models. Its is quite
important to notice that appearing named resources
have to be concretely added manually: only their
labeled name is set. We planed others experiments
with best-practices MOODLE courses as well as
direct use with teachers-designers.
5 CONCLUSIONS
Increasing the spreading and the using of computer-
based learning is a crucial issue of this decade.
Nowadays, the supply of TEL environments as
LMSs is effective and the technology is mature for
their use. But most of teachers do not effectively use
OPERATIONALIZATION OF LEARNING SCENARIOS ON EXISTENT LEARNING MANAGEMENT SYSTEMS -
The Moodle Case-study
147
these new artifacts and claim for more suitable and
easy-to-use tools. We do not think the solution to
this problem only relies on teachers’ training or on
institutional initiatives and supports. Our proposal is
(1) to simplify the authoring tasks and (2) to take
more into account the teachers’ requirements by the
automation of the operationalization process of a
course on a given platform. The idea that underpins
our work is the explicitation of the domain specific
language of a platform in order to emphasize the
underlying pedagogical approach embedded in the
platform while allowing the courses designers to
focus, at the knowledge level, on pedagogical
concepts rather than technical ones.
To reach this goal, we have decided to explore
the potential of Domain Specific Modeling techni-
ques. Our proposal is to externalize the design of a
learning scenario by the mean of editors, reifying a
Visual Instructional Design Language based on the
instructional design language of the targeted
platform, and to develop a bridge between this editor
and the platform for ensuring the operationalization
of the scenario. This paper details the architecture
and the functionalities of this kind of bridge, an API
for a platform based on its implicit language. We
have tested this approach with MOODLE. This API,
is able to both import and export learning scenarios.
For going farther this first result, we actually
work on two directions. One is to define a visual
instructional design for MOODLE and to develop its
dedicated graphical editor. To this aim we are
working on exploiting the Graphical Modeling
Framework tool in order to specify a graphical
notation and a mapping model with the existent
domain model. We also plan to evaluate the usability
of the editors by teachers-designers and to
experiment them on concrete learning situations case
studies. The second direction is to study at least one
another LMS, to repeat the same approach (ie.
identifying and formalizing the internal instructional
design language, develoment of communication
API, defining of VIDLs and externals editors) in
order to evaluate the possibilities of interoperability
between two different technical frameworks, helped
by Model Driven Engineering techniques of models
transformations.
REFERENCES
Al-Ajlan, A., Zedan, H., 2007. E-learning (Moodle) based
on service oriented architecture. In EADTU'07.
Lisbon, Portugal, 8-9 November, Lisbon-Portugal, vol.
1, 2007, pp. 62-70.
Berggren, A., Burgos, D., Fontana, J.M., Hinkelman, D.,
Hung, V., Hursh, A., and Tielemans, G. 2005. In
Practical and Pedagogical Issues for Teacher Adoption
of IMS Learning Design Standards in Moodle LMS. In
Teacher Adoption of IMS Learning Design Standards
in Moodle LMS. Journal of Interactive Media in
Education, 2005.
Burgos , D., Tattersall , C., Dougiamas , M., Vogten , H.,
Koper, R., 2007. A First Step Mapping IMS Learning
Design and Moodle . Journal of Universal Computer
Science, vol. 13, no. 7 (2007), 924-931 .
Caron, P.A., Derycke, A., and X. Le Pallec, 2005.
Bricolage and Model Driven Approach to design
distant course. In E-learn 2005, pp. 2856-2863.
Cocea M., 2006. Extendibility of Educational Systems to
Include a Learner-Adaptive Motivational Module. In
EATA NETTIES'06, Timisoara, Romania, pp. 195-198.
Delziel, J.L., 2006. Lessons from LAMS for IMS
Learning Lessons from LAMS for IMS Learning
Design. In ICALT'06, Sydney,5-7 July, pp. 1101-1102.
Eclipse, 2011. Eclipse Modeling Projects, http://
www.eclipse.org/modeling/, retrieved from May 2011.
De Vries, F., Tattersall, C., and Koper, R., 2006. Future
developments of IMS Learning Design tooling. In
Educational Technology & Society, 9 (1), pp. 9-12.
Gonzalez, V. B., and Anido, L. R., 2010. From SCORM to
Common Cartridge: A step forward. In Computers &
Education 54 Issue 1, January, pp. 88–102.
Hernández, L. D., Villasclaras-Fernández, E.D., Jorrín-
Abellán, I.M., Asensio-Pérez, J.I., Dimitriadis, Y.,
Ruiz-Requies, I., and Rubia-Avi, B. 2006. Collage, a
Collaborative Learning Design Editor Based on
Patterns Special Issue on Learning Design.
Educational Technology & Society. 9(1), pp. 58-71.
Kelly, S., Tolvanen, J.P., 2008. Domain-Specific
Modeling. ISBN: 978-0-470-03666-2. Paperback. 427
pages. Wiley-IEEE Computer Society Press.
Laforcade, P., Zendagui, B., Barré, V. 2008. A Domain-
Specific-Modeling Approach to Support Scenarios-
Based Instructional Design. In: ECTEL'08, Sept. 16--
19, Maastricht Netherlands.
Laforcade, P., 2010. A Domain-Specific Modeling
approach for supporting the specification of Visual
Instructional Design Languages and the building of
dedicated editors. In: Journal of Visual Languages &
Computing, vol. 21, issue 6, pp. 347-358.
Martel, C., Vignollet, L., Ferraris, C., David, J.-P., and
Lejeune, A. 2007. Modeling collaborative learning
activities on e-learning platforms. In ICALT'07, pp.
707-709.
Martinez-Ortiz, I., Sierra, J.L., and Fernández-Manjón, B.,
2009. Enhancing IMS LD Units of Learning
Comprehension. In the 4th International Conference
on Internet and Web Applications and Services. May
24-28, Venice, Italy.
Mekpiroona, O., Tammarattananonta, P., Buasrounga, N.,
Apitiwongmanita, N., Pravalpruka, B., and Supnithia,
T. 2008. SCORM in Open Source LMS: A case study
of Learnsquare. In ICCE'08. Taipei, Taiwan, pp. 166-
170.
ICSOFT 2011 - 6th International Conference on Software and Data Technologies
148