Specification of Learning Management System-centered Graphical
Instructional Design Languages
A DSM Experimentation about the Moodle Platform
Esteban Loiseau and Pierre Laforcade
Laboratoire d'Informatique de l'Université du Maine, Avenue Olivier Messiaen, Le Mans, France
Keywords: Domain-Specific Modeling, Meta-modelling, Instructional Design, Learning Management System.
Abstract: This paper presents a 6-month explorative research work about the specification of specific instructional
design languages. These languages have to tackle a double objective: capturing teachers-designers’ needs
and practices and guarantee a model in conformance with an existent Learning Management System.
Domain Specific Modeling techniques are used to both specify these languages and to provide practitioners
with some graphical authoring-tools. This explorative research work has been conducted in relation with a
pedagogical engineering team, specialized in Moodle, from Le Mans University. Three different DSM
approaches have been experimented and analyzed. This research is part of the French ANR GraphiT Project.
1 INTRODUCTION
Many institutions provide teachers and students with
some Learning Management Systems (LMS). These
educational platforms are not restricted to online
uses. They are also useful when combined with face-
to-face teaching for blended learning. Teachers can
use them for providing students with some materials
or to support the some complex collaborative
learning situations involving a strong use of the
platform communication features. In order to set up
such complex activities, teachers must develop some
designer skills about how and when managing and
sequencing the available features and tools. Such
skills can be acquired through specific education
programs generally focusing on the technical aspects
of the platform. They are rarely about designing
learning situations on the considered LMS. Because
of the multiplicity of educational theories and
approaches, as well as the lack of tools and
processes dedicated to existent LMSs, teachers
develop some ad-hoc and individual learning design
techniques. In such contexts, it seems relevant to
help teachers in focusing on the design for the
specific LMS they have at their disposal. A focus on
the instructional design possibilities and how they
can rely on the platform features should encourage
individual reflection about the design of learning
situations.
The GraphiT project (funded by the French
Research Agency) is based on an LMS-centered
designing approach. Within this starting project, we
have conducted a 6-months exploratory research. Its
main objective was to investigate some Domain
Specific Modeling (DSM) techniques for helping the
specification of LMS-centered graphical instruc-
tional design languages as well as the development
of dedicated editors. This paper focuses on its
presentation and on the analysis of its results.
2 RESEARCH CONTEXT
2.1 The GraphiT Project
This project (Graphit, 2013) has started in February
2012. Its main goal is to study the possibilities and
limits about the pedagogical expressiveness of
operationalizable languages to specify: future
leaning scenarios could be fully deployed and
automatically operated on an existent LMS. Such
instructional design languages aim at promoting and
improving the uses of current LMSs by providing
practitioners with some LMS-specific designing
language and authoring-tool. To achieve these goals,
the Graphit project is based on three assumptions: 1/
LMSs implicitly embed their own instructional
design paradigm (vocabulary, rules, constraints etc.);
504
Loiseau E. and Laforcade P..
Specification of Learning Management System-centered Graphical Instructional Design Languages - A DSM Experimentation about the Moodle
Platform.
DOI: 10.5220/0004491505040511
In Proceedings of the 8th International Joint Conference on Software Technologies (ICSOFT-PT-2013), pages 504-511
ISBN: 978-989-8565-68-6
Copyright
c
2013 SCITEPRESS (Science and Technology Publications, Lda.)
Figure 1: Targeted contributions of the Graphit project (tools in blue, processes/methods in green, models/techniques in
orange, specific results in grey) and perimeter of the exploratory work.
2/ it is possible to make these instructional design
domains explicit; 3/ their identification and
formalization allow to build and to develop external
LMS-centered tools. We also assume that
instructional design domains are stable enough
(through versions, extensions, integrated external
tools) to guarantee the durability of the instructional
design artifacts that will be specified and developed.,
The project scientific approach is illustrated in
Figure 1. For every considered LMS, we propose: a/
to make explicit the platform instructional design
domain; b/ to extend the learning platform with an
adaptative import/export API; c/ to make explicit the
teachers' designing needs & practices for this
platform; d/ to specify and support some LMS-
centered and practitioners-directed instructional
design languages and editors.
The main challenge of this project is to abstract
enough the LMS instructional design to propose
teachers some higher design blocks. The LMS
expressiveness and limits have to be overcome in
order to offer teachers some instructional design
mechanisms closer to their practices and needs for
specifying and sequencing the learning activities to
perform. Our idea is to conduct the platform
abstraction in accordance with the formalization of
future learning scenarios. This LMS-centered
approach guarantees that learning scenarios could be
operationalized into the LMS. The underlying
scientific issue relies on the identification of the
inter-relations between pedagogical expressiveness
and operationalization support.
The identification of practitioners’ needs (Figure
1 - left) and the formalization of the LMS
instructional design domain (Figure 1 - right) are the
prerequisites and first blocks for the GraphiT project.
They have already provided some results (Clayer et
al., 2012; Abedmouleh and Laforcade, 2012). The
project main issue and objective (Figure 1 - center)
has been investigated through the initial exploratory
works, explained in this paper.
2.2 Targeted Instructional Design
Languages
The project aims at proposing tools and instructional
design languages in conformance with these
properties (1) graphical formalization (specific
visual formalism for representing the learning
scenarios) and (2) operationalization ability (output
scenarios are machine-readable through the API to
integrate to existent LMSs - cf. Figure 1).
The second property relates to the Educational
Modeling Languages (EML) and their binding
concept (Koper and Manderveld, 2004)
. Such EMLs
often aim at being generic: their educational
expressiveness is independent from Technology-
Enhanced Learning systems like LMSs, and neutral
about the instructional design practices they cover.
EMLs focus on the scenarios formalization and
executability towards LMSs. The experiments about
the extension of the MOODLE LMS for importing
learning scenarios conformed to IMS-LD (the EML
de facto standard), proved that adapting existent
LMSs requires some complex and heavy re-
engineering (specific runtime-engine to integrate) in
order to overcome the limits of the platform features
(Burgos et al., 2007)
. EMLs fail to provide a support
for operationalizing EML-conformed learning
scenarios into existent LMSs.
The graphical property fits more to VIDLs
(Visual Instructional Design Language); (Botturi and
Stubbs, 2008). These languages offer some visual
notations from simple drawing with a few symbols
to complex diagrams. VIDLs focus on higher-level
languages, i.e. with syntaxes and semantics closer to
some instructional theories or to some specific
communities of instructional designers' practices.
VIDLs aim at supporting imagination, creative
thinking, communication, etc. Because VIDLs are
rather visual domain-specific languages focusing on
human-interpretations, they do not systematically
provide some binding techniques. Learning models
SpecificationofLearningManagementSystem-centeredGraphicalInstructionalDesignLanguages-ADSM
ExperimentationabouttheMoodlePlatform
505
are generally saved as proprietary files mixing
learning scenarios data with graphical information.
Some VIDLs or authoring-tools provide
instructional designers with a standardised EML
binding by the means of a dedicated export service
(Dodero et al., 2010). Nevertheless, resulting models
are generally less expressive than the original
scenarios because of the semantic gap between the
source/target languages.
One can consider the instructional design
languages targeted by the GraphiT project as a mix
between VIDL (because of the graphical
formalization) and EML (because of the binding to
the LMS explicit instructional design domain).
Within the project the operationalization will be
realized thanks to a dedicated adaptative import/
export API (presented in (Abedmouleh and
Laforcade, 2012)). According to the elaboration
criteria for VIDLs classification (Botturi, et al. 2006),
the GraphiT project languages will be designed at
implemen-tation (LMS-centered design) and
specification (design directed towards teachers’
needs and practices) levels. The main issue relies on
studying the limits of the visual expressiveness, in
accordance to teachers’ practices, while constraining
the operationalization on a specific LMS.
2.3 Domain Specific Modeling
Domain Specific Modeling (DSM) (Kelly and
Tolvanen, 2008) is a software engineering
methodology involving the systematic use of
domain-specific languages to represent the various
facets of a system. They are specific to a domain and
can be defined as the set of concepts and their
relations within a specialized problem field. They
offer primitives whose semantics are familiar to all
practitioners in that domain (in opposition to
generic-purpose languages like UML). This
methodology aims at reducing the software
engineering costs by automating the generation of
the application source code and by allowing the
handling/execution of the produced models.
Potential challenges and issues regarding the
application of DSM techniques and tools within the
instructional design domain have been discussed in
(Laforcade, 2010).
The Graphit project uses the DSM principles as a
methodological framework to specify languages and
as a practical framework to guide the development
of the related graphical editors. It provides a very
challenging trend for supporting the specification of
human-interpretable visual models with machine-
readable persistence.
We conducted a very first experiment about the
Moodle LMS (Abedmouleh and Laforcade, 2012).
We apply the DSM techniques and tools to specify a
graphical language and editor on top of the Moodle
metamodel (produced through the identification
process): semantics are voluntary limited to the LMS
expressivity in terms of instructional design abilities.
This experimentation succeeded in tackling the
binding and operatio-nalizing objectives.
Nevertheless the added value of such external
authoring-tools is limited because of the semantics
restriction: the learning situations designed are too
close to the LMS semantics.
2.4 Objectives of the Exploratory
Work
This paper focuses on a further exploratory research
work. Its main objective is to explore several DSM
approaches to specify instructional design languages
at a first abstraction level from the LMS-specific
expressivity (Figure 1 center). In order to achieve
this goal, we chose a specific LMS and defined a
first need in terms of pedagogical practices. The
LMS metamodel, capturing its instructional design
abilities, as well as the operationalization API, are
considered as known and functional (obtained from
(Abedmouleh et al., 2012)). In order to encompass
the metamodeling subjectivity, our results will focus
on concepts, relations and properties of the LMS but
not on their syntactic representation formalized by
the metamodeling technique.
3 REQUIREMENTS ANALYSIS
3.1 The Moodle Platform
Moodle is currently deployed and available for
blended learning at Le Mans University. It is a
widespread educative platform (Moodle, 2013),
largely used by public institutions. Moodle is
supported by a large, still growing, users community.
It is an open-source, modular, easy to extend PHP
Web application. Moodle was designed with a social
constructivist approach in mind (Dougiamas and
Taylor, 2003). It provides several features (resources
and activities) like collaborative tools and services
(forum, chat, wiki and others).
3.2 Pedagogical Activities
In order to identify some very first needs to capture,
we worked with a pedagogical engineering team,
ICSOFT2013-8thInternationalJointConferenceonSoftwareTechnologies
506
named PRN, from Le Mans University. Their
missions include managing the deployed Moodle
instances and training teachers. Recently, the PRN
has trained teachers about instructional design
aspects. They also take in charge the manual
operationalization of learning scenarios designed by
some learning teachers-designers involved in distant
learning programs. The PRN engineers are skilled
and experimented Moodle users. We considered
them as appropriate interlocutors considering our
exploratory research work. With their approval, they
gave us an access to their teachers training materials
and to some relevant online courses. The analysis of
these materials let us identify a first abstract need
from the platform's features. From a single tool, for
example a forum, one can design several
pedagogical uses, depending on its configuration.
We then compiled a non-exhaustive list of
pedagogical activities, from these concrete sources
and from other uses found in the domain literature
(Conole et al., 2004). In order to complete these
future elements for the language to specify, we also
added some VIDLs recurring structural elements
(selection, sequence, conditional activities, etc.).
3.3 Manual Binding
We first manually match the pedagogical needs and
the Moodle features in order to later formalize
within the instructional design language. Assessment
activities such as self-assessment, summative
assessment or formative assessment rely on the same
Moodle quiz feature but on different settings of its
parameters, including answering modalities
(examples : number of attempts, “adaptative mode”).
Some pedagogical activities can be set up on the
platform in a variety of ways. According to the users
choices while designing a scenario, the equivalent
properties/values will drive the elicitation of the
most appropriated LMS feature to use. For example,
a debate activity can be conducted synchronously
with a chat, or asynchronously with a forum. The
Writing a report activity is a more complex example,
with three properties to value (collaborative versus
individual modality, online versus offline, and
iterative or finalist writing). Four different Moodle
tools can be chosen (with proper settings in the 8
cases) : online text assignment, wiki, blog or file
submission. To implement the structural elements of
a pedagogical scenario, we have to exceed the
Moodle limitations making uncommon uses of
Moodle features: there are no activity-structures and
conditional branching in the 1.9 version used at Le
Mans University. To implement sequence and
choice activity-structures we used labels to provide
students with instructions. Moodle does not provide
conditional activities before the 2.3 version: it is
possible to use groups and groupings to limit
availability of an activity to a particular group of
students and to assign each groups to an alternative
activity. Nevertheless, branching conditions cannot
be automatically checked: a label, not visible by
students, can be used as a reminder for the teacher
about assigning students to groups.
4 IDENTIFYING THREE
APPROACHES
According to the DSM definitions, the graphical
instructional design languages we propose to specify
are specific modeling languages. Every modeling
languages can be defined as a tuple <AS, CS*, M*ac,
SD, Mas> (Chen et al., 2005) where AS is the
abstract syntax, CS* are the concrete syntaxes, M*ac
is the set of mappings between abstract syntax and
concrete syntaxes, SD is the semantic domain, and
Mas is the mapping between abstract syntax and
semantic domain. The abstract syntax (AS) defines,
in a structural way, the concepts and relations of a
modeling language. It is concretely formalised with
a metamodel. This metamodel also specifies the
future conformed models (binding).
Our instructional design languages have to meet
two requirements: providing a human- interpretable
graphical notation (a specific graphical concrete
syntax), and also providing a machine-readable
notation (serialisation format) for the scenarios
conformed to the Moodle metamodel, in order to be
handled by our Moodle import/export API. A first
approach consists in defining the abstract syntax as
the exact Moodle metamodel (considered pre-
existent for this exploratory research work) in order
to achieve, first and foremost, the machine-readable
requirement. The graphical concrete syntax will be
derived from the Moodle metamodel. Nevertheless,
this approach has to tackle the expressivity limits of
models when only based on the graphical notation
expressiveness. The second approach is about
extending the abstract syntax, initially the Moodle
metamodel, with new syntactical elements adding
the required semantic (pedagogical activities and
structural elements) non-covered by the initial
abstract syntax. By adding new elements to the
abstract syntax, we break the conformity to the
initial metamodel for the future models. This
approach must address such issue within the
SpecificationofLearningManagementSystem-centeredGraphicalInstructionalDesignLanguages-ADSM
ExperimentationabouttheMoodlePlatform
507
Figure 2: Partial representations of the abstract syntax according to the three approaches.
design-time (i.e. A post-design processing is not
relevant in this approach), in order to study the dsm
tooling capabilities. The third approach is opposed
to the first one, focusing on a language firstly
capturing the teachers-designers' needs and
practices: the abstract syntax is defined with no
relations with the moodle metamodel. This approach
will produce non-conformed models that cannot be
directly operationalized in moodle. Such approach
will have to provide some model transformations in
order to restore the platform-conformance.
5 TOOLING EXPERIMENTS
5.1 DSM Tooling
We chose to use additional frameworks from the
open-source project (Eclipse, 2013). EMF allows to
define the abstract syntax (metamodel) and to drive
the business code generation, GMF to specify the
graphical modeling languages and to generate the
editor code, ATL to specify and to execute some
transformation rules. The GMF framework provides
the functionalities to define modeling languages
according to the DSM methodology. Every syntax
and mappings are models, modifiable through an
Eclipse integrated editor: the abstract syntax
(domain model) is a metamodel conforming to the
Ecore meta-metamodel; concrete syntax can be
defined through two models: gmfgraph, defining the
graphical notation, and gmftool, specifying the
concepts and relations available in the diagram
editor toolbar. Abstract/concrete syntax mapping
(Mac) is described through the gmfmap model. The
semantic domain and its mapping towards the
abstract syntax are not explicitly defined in their
own specific models, but can be specified through
OCL rules in addition to the domain model. Two
additional models are involved in the editor code
generation: genmodel model, precising the code
generation parameters (for example the models
persistence format) and gmfgen model, parameters
about the editor code generation. The three
approaches we propose are based on an initial
metamodel formalizing the Moodle instructional
design domain (Abedmouleh et al., 2012).
5.2 Tooling the First Approach
This approach is about specifying concrete syntax of
the language through the gmfgraph and gmftool
models, using the Moodle metamodel as the abstract
syntax (figure 2 left). Figures, icons, labels,
properties, etc. representing the toolbar and the
graphical elements can be specified to hide the
underlying platform concepts, in order to provide a
more significant representation (pedagogical in our
particular case). On one hand this approach is
appropriate for situations such as the assessment
activities, when a single platform tool is used (1-1
mapping), with different pre-configured properties.
Indeed, GMF provides an initialisation feature,
allowing instantiation of several domain concepts
triggered when creating an element. On the other
hand it is an issue when some specific information
are required to identify the right platform feature.
GMF cannot automatically triggers some specific
initialisation actions when the properties of an
existent instance are dynamically modified. In order
to overcome this difficulty, we have to define one
tooling element for each combination of settings of a
pedagogical activity in order to maintain the same
level of expressiveness. For example Writing a
report activity should exist in 8 different variants,
according to the 3 criteria values, available in the
editor toolbar.
5.3 Tooling the Second Approach
Some techniques of the first approach can be reuse
to process the simple binding cases. For the others,
the domain model have to be extended with new
concepts (for example the debate pedagogical
activity with the synchronous property - cf. figure 2
ICSOFT2013-8thInternationalJointConferenceonSoftwareTechnologies
508
Figure 3: Partial editors screenshots of editors obtained through (approach 1 on the left, approach 2 in the middle and
approach 3 on the right).
center) and mapped with graphical representations.
Unfortunately, the GMF framework relies on this
metamodel to drive the persistence of the future
models created with the graphical editor. Extending
the initial metamodel breaks the conformance of
future models with the dedicated operationalization
API. However, we can address this issue by mixing
meta-modeling techniques (for example we use the
transient property in order to disable the persistence
mechanism) and code modification (for example if
the synchronous property of a debate activity is set
to true, a chat element is created and the
corresponding forum element, if previously created,
is removed).
5.4 Tooling the Third Approach
This approach relies on the formalization of a
metamodel specific to teachers' needs and practices
presented in 3.2. This new metamodel specifies
some relevant pedagogical activities (non-
exhaustive), the structural/sequencing elements and
allows the definition of pedagogical objectives
(figure 2 right). These objectives were on purpose
added because this approach does not focus, in the
first place, on the binding towards any LMS
conformance. The specification is only limited by
the teachers' needs and our ability to formalize them
by using metamodeling techniques. The GMF
processes we conducted are the same as for the other
two approaches: definition of the graphical notation,
of the toolbar, etc.
The models produced with such DSM generated
editor do not comply with the initial Moodle
metamodel, thus the learning scenarios cannot be
operationalized on Moodle. Because the Moodle
metamodel is not included in this specific
metamodel, we cannot reuse the techniques
(metamodeling plus code modification) from the
second approach to restore a Moodle conformance.
The use of model transformations, through ATL
rules for example, is a potential solution but it
requires an additional step after the learning scenario
design and before the operationalization. Models
produced with GMF-based editors are saved in an
XML/XMI format, enabling also some XML-
oriented transformation (with XSL for example).
This transformations must be defined and specified
in accordance with an educational engineer and can
be very complex to produce. Some instructional
design concepts could not have a matching set of
elements in the LMS language (incomplete
mappings lead to inconsistent scenarios), or, on the
contrary, source elements may not be enough
detailed to make a transformation decision. For
example in our case study, the pedagogical
objectives have no matching concepts in the Moodle
language; thus, they have to be skipped in the
operationalization phase (semantic losses). Such
ignored elements, if important for the scenario, can
make other elements inconsistence.
6 DISCUSSION
We analysed and compared the 3 DSM approaches
with four criteria: 1/ visual expressiveness (the
semantic expressiveness from models as perceived
by the teachers when manipulating visual elements);
2/ abstract expressiveness (the real model semantics,
according to underlying abstract syntax, not directly
perceived by teachers); 3/ the potential of
operationalization for the produced models; 4/ the
operated scenario semantics (i.e. the semantics of the
SpecificationofLearningManagementSystem-centeredGraphicalInstructionalDesignLanguages-ADSM
ExperimentationabouttheMoodlePlatform
509
resulting course setting up on the LMS during the
operationalization step). The first criteria does not
require some specific computer-science skills and is
considered as an end-user point-of-view. We
gathered feedbacks, collected during an informal
meeting, from the involved PRN members. Table 1
summarises the results, marked +, / or – according to
the two languages requirements of this study
(pedagogical expressiveness and LMS binding).
From a graphical expressiveness perspective,
approaches 2 and 3 are more relevant because they
do not require end-users to anticipate the most
appropriate detailed concepts at the design-time
when picking up elements from the toolbar. For
example in the approach 1, the teacher-designer has
to choose between Synchronous debate and
Asynchronous debate activities before adding the
activity to the drawing space; a later change of mind
will require to delete the previous chosen activity
from the diagram and to pick up / instantiate the
other one. In opposition, approaches 2 and 3 only
require for designers to choose the most relevant
activity from the toolbar, and to decide its properties
later. The third approach does not deal with the LMS
binding or conformance at first (when design
scenarios). Instructional design languages from this
approach provide some elements, properties and
relations directly addressing the teachers' needs.
There are no constraints or limits like in the second
approach focusing on extending the LMS
metamodel.
The abstract expressiveness of the produced
models is directly related to the initial choices
defined in accordance with the three approaches.
The compliance of the produced models is
straightforward for the first approach, while it still
requires some adaptations in the other two ones. In
the second approach we used metamodeling
techniques and code modifications to maintain the
models persistency in conformance with the LMS
metamodel. For the third approach, a more complex
binding is required after the design-time (with
models transformation for example). Our experiment
showed that such transformations can easily become
more complex and time-consuming. Also, this
approach cannot guarantee the LMS conformance
for the produced scenarios. Previous researches on
model transformations between practitioners-
oriented learning scenarios and LMSs-specific
instructional design metamodels already revealed
such issue (Nodenot et al., 2008; Abdallah, Toffolon
and Warin, 2008). In opposition, the second
approach maintains the initial semantics by using
both metamodeling techniques and compliance
restoring techniques ; it is strongly dependent on the
weaving used to extend the initial platform
metamodel. The first approach guarantee the
conformance of the semantics by directly using the
initial metamodel.
The third approach corresponds to the usual way
to specify a VIDL with its main advantage
(expressiveness) but also inconvenience (difficulty
to operationalize). The first approach reveals the
limits of the concrete syntax expressiveness when
only defined by derivation of the abstract syntax:
this approach can map several representations with a
single concept or relation from the LMS metamodel
but this relation of derivation is immutable and
cannot be dynamically changed (DSM limit) to
improve the user-friendliness. The second approach
is intermediate on all criteria: best expressiveness /
LMS compliance ratio. However, it requires a strong
metamodeling expertise to reduce the developing
cost while restoring the LMS compliance. This
approach highlights the importance to drive the
expressiveness (and semantics) extension of the
initial metamodel with the binding capacity.
Matching the teachers' needs and practices to the
LMS features cannot reduce to a programming task.
It has to be made explicit by involving teachers
(validation of matching rules and constraints) and
DSM experts. Such explicit informations can be
used as a base for the formalization of the
metamodel extensions.
Table 1: Comparison of the three approaches.
ICSOFT2013-8thInternationalJointConferenceonSoftwareTechnologies
510
7 CONCLUSIONS
This paper presents a six-month exploratory research.
Its main objective was to study several Domain
Specific Modeling approaches to specify/ develop an
instructional design language/editor specific to an
existent Learning Management System. The targeted
language had to meet two requirements:
expressiveness directed towards teachers-designers'
needs and practices, and binding/persistence
centered on the platform metamodel. For the study
we choose Moodle as the LMS to comply with, and
we restricted the teachers' needs to some
pedagogical activities and usual activity-structures.
From the DSM theoretical framework we identified
three approaches to put into practice with the DSM
tooling from the Eclipse Modeling Project. The three
results have been analyzed and compared. The
approach offering the best compromise for both
requirements consisted in extending the initial
platform metamodel in order to include the first
abstraction level from the platform features. This
research was part of the starting Graphit project.
Further researches are currently conducted, still
meeting the same two requirements than from this
first study but with a scope expanded to several
LMSs and to more complex teachers-designers'
needs and practices. The Domain Specific Modeling
methodology, as well as Model Driven Engineering
techniques, will be deeply studied, in particular the
models weaving techniques. Indeed, such techniques
allow to explicit the mappings relations between
teachers' practices and platform features. Such
explicit and formal relations could help us in
identifying a specific way to extend the LMS
semantics while maintaining the models binding to
the LMS metamodel.
ACKNOWLEDGEMENTS
This work is funded by the French GraphiT project
[ANR-2011-SI02-011] (http://www-lium.univ-
lemans.fr/~laforcad/graphit/).
REFERENCES
Abdallah, F., Toffolon, C., Warin, B., 2008. Models
transformation to implement a PBCL Scenario :
Moodle case study. IEEE ICALT 2008.
Abedmouleh A., Laforcade P., 2012. Improving the design
of courses thanks to graphical and external dedicated
languages: a Moodle experimentation, 1
st
Moodle
Research Conference, Heraklion.
Abedmouleh A., Oubahssi L., Laforcade P., Choquet C.,
2012. An analysis process for identifying and
formalizing LMS instructional language, ICSOFT
2012, Rome.
Botturi L., Stubbs T., 2008. Handbook of Visual
Languages in Instructional Design: Theories and
Pratices. Hershey, Information Science Reference.
Botturi L., Derntl M., Boot E., Figl K.A, 2006.
Classification framework for educational modeling
languages in instructional design, IEEE ICALT 2006,
Kerkrade.
Burgos D., Tattersall C., Dougiamas M., Vogten H.,
Koper R., 2007. Mapping IMS Learning Design and
Moodle. A first understanding, Journal of universal
computer science, 13(7).
Chen K., Sztipanovits J., Abdelwalhed S., Jackson E.,
Semantic Anchoring with Model Transformations,
ECMDA-FA 2005, Nürnberg.
Clayer J-P., Toffolon C., Choquet C., 2012. A pattern
based and teacher-centered approach for learning
design, CATE 2012, Napoli.
Conole G., Dyke M., Oliver M., Seale J., 2004. Mapping
pedagogy and tools for effective learning design,
Computers & Educations, 43(1-2), pp.17-33.
Dodero J., Martinez del Val A., Torres Jorge, 2010. An
extensible approach to visually editing adaptive
learning activities and designs based on services,
Journal of Visual Languages & Computing, 21(6).
Dougiamas M., Taylor P. C, 2003. Moodle: Using
Learning Communities to Create an Open Source
Course Management System, EDMEDIA, Honolulu.
Kelly S., Tolvanen J-P., 2008. Domain-Specific
Modeling:Enabling Full Code Generation, Wiley-
IEEE Society Press.
Koper R., Manderveld J., 2004. Educational modelling
language: modelling reusable, interoperable, rich and
personalised units of learning, British journal of
educational technology, 35(5), pp.537-551
Laforcade P., 2010. A DSM approach for supporting the
specification of Visual Instructional Design Languages
and the building of dedicated editors, Journal of
Visual Languages & Computing, 21(6), pp.347–358
Nodenot T., Caron P-A., Lepallec X., Laforcade P., 2008.
Applying Model Driven Engineering Techniques and
Tools to the Planets Game Learning Scenario, Journal
of Interactive Media in Education, 23.
SpecificationofLearningManagementSystem-centeredGraphicalInstructionalDesignLanguages-ADSM
ExperimentationabouttheMoodlePlatform
511