A Method for Identifying and Formalizing the Underlying
Instructional Design Language of Existing LMSs
Nour El Mawas, Lahcen Oubahssi and Pierre Laforcade
Laboratoire D’informatique De L’université Du Maine, Lunam Université, Université Du Maine, Le Mans, France
Keywords: Learning Management System, E-learning, Instructional Design, Operationalization, Technology Enhanced
Learning, Process, Moodle.
Abstract: The use of existing LMSs presents many difficulties related to the design and operationalization of learning
scenarios. Teachers have to encompass the LMS technical features and services in order to understand the
underlying way of designing. Generic instructional design editors fail in bridging the gap between how they
design a learning scenario and how the learning session can be set up within the target LMS. If LMSs could
be able to make explicit their intrinsic and implicit learning design model, it can be exploited as a
proprietary format to build tools and facilities dedicated to this LMS. The research presented in this paper
aims to present our method in terms of necessary analysis and steps for the identification and the
formalization of such LMSs’ instructional design languages. The method takes into account three different
viewpoints: a viewpoint centred on the LMS macro-HMIs (Human-Machine Interfaces), a functional
viewpoint and a micro viewpoint. We concretely illustrate the proposed method about the Moodle platform.
1 INTRODUCTION
Our research work focuses on the field of
Technology Enhanced Learning (TEL) engineering
and re-engineering. TEL is a scientific domain
where different disciplines such Computer Science,,
education, psychology, philosophy, communication
or sociology intersect (Tchounikine et al., 2009). We
are particularly interested in applying and adapting
Computer Science solutions for providing
practitioners with some customized instructional
design solutions.
Instructional Design (ID) is the systematic
development of instructional specifications using
learning and instructional theories to ensure the
quality of instruction. It is the entire process of
analysis about learning needs and goals as well as
the development of a delivery system to support
those needs. It includes development of instructional
materials and activities and delivering and
evaluation of all instruction and learner activities
(Berger and Kam 1996). It has been a well-
established discipline for several decades (Gimenes
et al., 2014).
TEL is a large domain for research and practice,
including e-learning, mobile learning, and Learning
Management System (LMS). An LMS is the
framework that handles all aspects of the learning
process. An LMS is also the concrete infrastructure
that delivers and manages instructional content,
identifies and assesses individual and organizational
learning or training goals, tracks the progress
towards meeting those goals, and collects and
presents data for supervising the learning process of
organization as a whole (Szabo and Flesher, 2002).
LMSs support the use of standards for describing the
learning objects, packaging them into larger content
and learning units (such as lessons and courses), and
applying various instructional design strategies and
techniques (Jovanovic et al., 2007). Nowadays,
LMSs are not restricted to distant learning only.
Teachers use them for blended learning which
combines traditional face-to-face learning with
computer supported learning (Graham, 2005). LMSs
have created remarkable opportunities for higher
education to expand the educational process beyond
the traditional classroom to include geographically
dispersed students (Brito et al., 2014).
The research work presented in this paper is a
continuity of other former works in our lab
(Oubahssi et al., 2010) (Abdallah et al., 2008) by
proposing a new implementation approach of
learning situations and pedagogical scenarios. It
takes place into the context of the GraphiT project
5
El Mawas N., Oubahssi L. and Laforcade P..
A Method for Identifying and Formalizing the Underlying Instructional Design Language of Existing LMSs.
DOI: 10.5220/0005405800050014
In Proceedings of the 7th International Conference on Computer Supported Education (CSEDU-2015), pages 5-14
ISBN: 978-989-758-107-6
Copyright
c
2015 SCITEPRESS (Science and Technology Publications, Lda.)
(Graphical Visual Instructional Design Languages
for Teachers). Its main goal is to study the
possibilities and limits about the pedagogical
expressiveness of operationalizable languages to
specify future leaning scenarios that could be fully
deployed and automatically set up upon an existing
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. Despite
many existing standards (Martinez-Ortiz et al.,
2009) (Mekpiroona et al., 2008), approaches (De
Vries et al., 2006), languages (Baggetun et al.,
2004), architectures (Alario-Hoyos et al., 2013), and
tools (Baggetun et al., 2004) (Al-Ajan and Zedan,
2007) to facilitate the instructional design, they are
often not compatible with existing LMSs, or require
a costly reengineering of the LMS (new web service
API, new runtime engines, etc.). Moreover, they do
not simplify the operationalization of the produced
models. Some translations, leading to information or
semantics losses, are still required to operationalize
them into a targeted LMS.
In this paper, we are focusing on the
identification and formalization of LMSs implicit
instructional design language. Indeed, the expected
result will be the base for the development of
binding solutions and will simplify the instructional
design on platforms. These solutions must insure
that future scenarios formalized in accordance with
the language to identify will be operationalized
without semantics losses into the LMS internal
structures. This process is dedicated to LMSs active
communities and more specifically to designers with
a competence in IT and the service of information
technology and communication for education
(pedagogical engineers) who meets difficulties in
appropriating the instructional design language of
LMSs.
The paper is organized as follows. Section 2
presents related works about identifying and
formalizing LMS languages. Section 3 highlights
our motivation to extract the pedagogical LMS
language. Section 4 details our approach. Section 5
is dedicated to the application of our method on
Moodle 2.4. Section 6 concludes our paper and
presents our perspectives.
2 RELATED WORK
In recent years, researchers have begun to formalize
LMSs instructional languages in order to specify
models in conformance with the infrastructure
design languages of LMSs.
In an E-learning context, (Caron et al., 2005)
defines three features that a meta-model must have:
(1) Limitation of the functionalities consisting in
restricting the modeling domain to the web services
without global settings like security, (2)
Identification of the element factories consisting in
identifying element factories and their capacity to set
elements which can be used by the web services, and
(3) Definition of the factorization mechanism based
on the fact that a model is a simplified view of a
system. This meta-model enables a team of
designers to describe what should be learnt from a
scenario, the characteristics of students that will use
the scenario (learner models), how the learners will
face this knowledge (teaching and available learning
strategies), etc.
(Graf, 2007) has proposed a meta-model for
adaptive courses that can be easily integrated into e-
learning platforms. The meta-model is based on the
Felder-Silverman Learning Styles Model (FSLSM)
describing a single student in accordance to four
dimensions: active & reflexive learning style,
sensitive & intuitive learning style, visual & verbal
learning style, and sequential & global learning
style. Other learners’ characteristics like the state of
knowledge and the learning goals are not taken into
account. For presenting the content of the course,
content objects are considered to include the relevant
learning materials. Furthermore, (Graf, 2007)
incorporates examples as course elements. Examples
are used for better illustration and provide students
with more concrete material. Moreover, students can
check their acquired knowledge by the use of self-
assessment tests. Another element includes exercises
that serve as practice area where students can try
things out or answer questions about interpreting
predefined solutions or developing new solutions.
(Abdallah et al., 2008) were interested in
specifying and designing learning situations
supported by PBCL (Project-Based Collaborative
Learning). To allow teachers to elaborate a PBCL
scenario, they propose a meta-model dedicated to
the PBCL. In this approach, teachers can design a
learning scenario based on the PBCL meta-model.
Then, this scenario is adapted to a chosen platform:
a models transformation approach is proposed
allowing the integration of PBCL scenarios in a
platform. (Abdallah et al., 2008) applies his proposal
on the Moodle platform.
All these presented works and many others
(Drira et al., 2012) propose a meta-model to
formalize LMS instructional language but to our
knowledge, there is no proposition that focuses on
CSEDU2015-7thInternationalConferenceonComputerSupportedEducation
6
identifying an explicit process or method to
formalize it.
The next section emphasizes on the importance
and the utility of defining an explicit method to
formalize LMS instructional languages.
3 MOTIVATION
Many universities have adopted web-based LMSs as
the TEL system. They use them to offer teachers a
range of pedagogical and administrative tools for
supporting teaching and learning activities (Coates
et al., 2005). However, many teachers have
difficulty using LMSs to create learning designs that
are truly engaging to their students (Steel, 2009).
They are not familiar with the implicit learning
design domains of LMSs (Martinez-Ortiz et al.,
2009). Most of open source LMSs are very difficult
to apply in real schools, because teachers are not
familiar to using an LMS which needs to take an
effort to appropriate it (Mekpiroona et al., 2008).
Due to the complexity of LMS functionalities,
users are expected to have some pre-existing
knowledge of these functionalities. Despite online
forums, it is still difficult for a teacher to design his
courses on platforms. LMSs are in continuous
evolutions and discussions regarding different
versions of a platform are interwoven. In addition,
many forums, if not all, have input from developers,
programmers, and software architects. That is why
forums are difficult environments for non-expert
LMSs users to make sense of.
In addition, there is no support (neither human
nor software products) able to help teachers in
clarifying, defining and then specifying their
learning situations before setting them up within the
LMS. They have to appropriate the various screens
and form-based interfaces to abstract some low-level
details to think about their global design courses.
Teachers need solutions to narrow the gap
between their educational intention and the
pedagogical features proposed by the LMS at their
disposal. They ask for appropriate tools helping
them in understand the underlying “way of thinking
and designing” of this LMS.
In our work, we aim at supporting practitioners
to overcome these LMSs’ obstacles in order to help
them in focusing on the design of learning situations.
Our contribution consists in extracting,
identifying, and formalizing the LMS implicit
instructional design language. We also on purpose
propose a meta-model formalism to capture it. The
meta-model is obtained by the abstraction of
pedagogical features and services provided by the
considered LMS. This meta-model acts, according to
the language theory, as an abstract syntax. It will
then be used as a basis for the development of
external editors (Loiseau and Laforcade, 2013)
(Laforcade and Abedmouleh 2012).
4 OUR APPROACH
We propose a method to identify and formalize the
instructional language of LMSs. Our approach takes
into account a macro-HMI analysis, a functional
analysis and a micro-analysis. In this section, we
sketch an overview of our approach then we explain
in details each step of the method.
4.1 Overview of Our Approach
In our work, we focus on pedagogical tasks and
functionalities of a specific LMS. Our hypothesis is
that LMSs are not pedagogically neutral and they
embed an implicit language based on the LMS
specific paradigm to specify the design of a learning
activity (Abedmouleh et al., 2012). Our work aims
to define the necessary analysis and steps for the
identification and formalization of an LMS
instructional design language.
The first attempt to define the method was
presented in (Abedmouleh et al., 2012). However,
the proposed process did not take into account the
presence of common elements between pedagogical
activities/resources on LMS. The final meta-model
excludes elements that are relevant for instructional
design such as activity completion conditions, as
well as outcomes and grade conditions.
Our method is specified according to three
different viewpoints: a viewpoint centred on macro-
HMI, a functional viewpoint and a micro viewpoint.
The first viewpoint consists of HMIs analysis
according to two strategies: (1) the analysis of
existing situations on the platform and (2) the
analysis of interfaces related to the specification of
new situations. After the macro-HMI analysis, we
factorized the macro-HMI model in order to obtain
the simplified macro model. The second viewpoint
focuses on the identification of LMS existing
functions. The third viewpoint concerns the micro
analysis of the LMS instructional design language.
Figure 1 shows the proposed process. It is
composed of the macro-HMI analysis, the
factorization of HMI-macro model, the functional
analysis and the micro analysis. The micro analysis
is based on the micro-HMI analysis and technical
AMethodforIdentifyingandFormalizingtheUnderlyingInstructionalDesignLanguageofExistingLMSs
7
Figure 1: Analysis process of the instructional design language.
analysis. The final model results from a
confrontation of micro-HMI and technical models.
In the next sections, we present in details
different steps of the process.
4.2 Macro-HMI Analysis
The macro-HMI analysis consists in identifying
platform interfaces related to the Instructional
Design (ID).
LMSs are usually composed of many interfaces,
developed for different purposes and users’
categories. In our work, we have ignored interfaces
related to administration and management purposes;
we are only interested in interfaces related to
instructional design usages. The instructional design
language is identified using two methods: the
analysis of interfaces titles and the analysis of the
navigation paths.
The first analysis step is to choose the main
interface. Then, the analysis must determine whether
or not the interface provides a pedagogical aspect.
Interfaces related to ID are taken into account. The
main interface concept is identified and presented on
the macro-HMI model. Relations between model
concepts are finally identified and defined.
Interfaces identification is an iterative process.
When a new interface is identified, the analyst
studies existing links inside this interface in order to
access to new interfaces. Only Interfaces related to
ID are analyzed and added to the macro-HMI model.
The macro-HMI model is presented by the meta-
model format. We have chosen the meta-model
format because it allows presenting clearly platform
elements, their attributes, relations between them
and their cardinalities.
4.3 Factorization
Factorization is the process of finding common
attributes shared between two or more pedagogical
elements (classes) in the macro-HMI model and
moving them into an existing or a new abstract
parent element. The non-common attributes will not
change place. The difference between an abstract
class and a concrete class is that a concrete class can
be instantiated. The role of an abstract class is that of
possessing concrete subclasses. This is important for
the factorization of the attributes and common
methods realized by the sub-classes. Visually, an
abstract class is represented implicitly with a cursive
formatting (in italics) of the name of the class (cf.
figure 2, Activity/Resource class).
Many works shows the relevance of classes and
associations factorization in modelling languages
(Dao et al., 2004). The factorization we propose is
based on the fact that a model is a simplified view of
a system. Therefore a model element can factorize
the system collection of elements (Al-Ajlan and
Zedan, 2007).
This step aims to find common elements in
pedagogical activities/resources and common
relations between them. Factorization is applied on
the Macro-HMI model. The macro model, resulting
for the factorization, is clearer and more simplified
than the Macro-HMI model.
4.4 Functional Analysis
In the software engineering field, a software life-
CSEDU2015-7thInternationalConferenceonComputerSupportedEducation
8
cycle model includes a functional analysis in the
requirements and specification phases. Functional
requirements are associated with specific functions,
tasks or behaviours the system must support.
Functional specifications describe what the system
must do as well as requested properties of inputs and
outputs.
In our context, the functional analysis aims to
identify the functionalities dedicated to the course
instructional design. The HMIs of the Macro-HMI
model are analyzed from both functional and
pedagogical perspectives. Administrative
perspectives (like display functions, etc.) are
rejected from the functional model. The
functionalities are implicitly embedded in interfaces
via HMI widgets (buttons, links, etc.) facilitating the
interactions between users and system. Each widget
has to be tested in order to determine its pedagogical
features. Then, the analyst has to give a function
name for each pedagogical widget. The functional
analysis is an interactive process, every time we
identify a new function, we must verify its
pedagogical use. Only functions with pedagogical
use are presented on the model. Sub-functions are
also added to the functional model.
4.5 Micro Analysis
The micro analysis is based on the macro and the
functional models. It takes into account two different
viewpoints: micro-HMI and technical viewpoints.
We propose a confrontation of micro-HMI and
technical models to formalize the final model.
4.5.1 Micro-IHM Analysis
The micro-HMI analysis consists in analyzing the
concerned interfaces at a finer scale. It aims to
identify all elements relevant to the instructional
design, including their features (attributes, types,
etc.). To conduct this analysis, we propose many
steps. After choosing an element of the macro
model, the analysis concerns the interfaces for
realizing/defining a dedicated use case of the
functional model. The concerned interface is break
down into many areas. Each component of each area
(titles of blocks, menus, forms, etc.) has to be
analyzed in order to determine its pedagogical
features. The analysis concerns also many
pedagogical elements which are described by the use
of various forms, widgets and software components
(buttons, links, etc.). Two main categories of the
forms elements/attributes can be identified: required
elements and optional elements. The required ones
have to be identified because they form the main
elements of the LMS instructional design language.
The non-setting of these elements prevents the
ordinary working of system. These characteristics
have to be identified: it presents an important feature
about the instructional design language of learning
platforms.
4.5.2 Technical Analysis
The second step of the process concerns the
technical analysis (Abedmouleh et al., 2012).
Several technical analyses are possible: databases,
source code, courses backup/restore, etc. During this
step, the main source of information for identifying
the instructional design language is the LMS
database. The other technical analyses will be used
during the confrontation step.
This analysis consists in specifying a reduced
Conceptual Data Model from the one available by
LMS providers if it exists. In our approach, the
database analysis has to be restricted to the
tables/columns in relation to instructional design
data. The main obstacle is to identify these data.
Information from the micro-HMI analysis could be
useful to achieve this goal.
This technical analysis consists in (1) looking
over all database tables in order to sketch a first draft
of the model, (2) focusing on tables embedding
elements in relation to instructional design concepts.
These tables can be identified through the semantic
analysis of their titles or their record fields. Some
tables could be identified through their dependencies
with others or through the foreign keys. The analysis
consists then in specifying the database schema on
the basis of the databases reverse engineering rules.
The Conceptual Data Model can be finally specified
from this schema. This model is relevant to represent
the technical-model viewpoint because it hides ill-
structured databases, misconceptions or
redundancies.
4.5.3 Confrontation and Formalization
The last process step concerns the confrontation of
both micro-HMI and technical models, and the
formalization of the final model. The micro-HMI
and technical models are compared in order to (1)
refine the micro-HMI model, (2) detect and correct
the difference between models, (3) ensure that the
final model can be easily bind to a computer-
readable format for the existing LMS.
The confrontation conducts verifications on the
definition of the instructional design elements on
both models. Some differences or ambiguities (like
AMethodforIdentifyingandFormalizingtheUnderlyingInstructionalDesignLanguageofExistingLMSs
9
the definition of similar elements, the non-existence
of some attributes, divergences about the types of
attributes, etc.) are so identified. They require a
deeper and finer analysis of both HMI and technical
analysis. At this step, other technical-centred
analysis (source code, backup packages, etc.) can be
useful. For example the source code analysis
consists in directly reviewing the LMS code.
It primarily concerns the code of the HMI
definition and the queries for inserting / selecting
data. This analysis can reveal many details that
developers have chosen to encode for effectiveness
or portability reasons. The aim of this process step is
to formalize the instructional design language.
5 MOODLE CASE STUDY
In this section we present the application of our
process on an LMS. We have chosen Moodle 2.4 as
a use case for many reasons: (1) Moodle is
increasingly used in schools, universities and
companies, (2) Moodle is also used in our
university, and (3) Moodle has an active community
who continuously develops APIs that provide tools
for its scripts (so once the editor is finalized, we will
share it with the community). Note that the version
2.4 is the installed version in our university.
5.1 Application of the Macro-HMI
Analysis on Moodle
The application of macro-HMI analysis on Moodle
consists in identifying interfaces related to course
design. We analyzed interfaces titles and navigation
paths / URLs. We studiously browse all the links in
a specific interface. These links often point to new
interfaces. Moodle is designed based on a top-down
approach: the main interface is about specification
and presentation of the course content, other
interfaces (like add a forum, a label...) are accessible
from the main interface.
The figure 2 shows the result of applying the
macro-HMI analysis on Moodle. A course is
composed of categorie(s), outcome(s), scale(s),
section(s), group(s), grouping(s) and one question
bank.
Course sections are organized into resources and
activities for students. Moodle 2.4 offers 7 resources
(Book, Page, Label, IMS content package, File,
Folder, and URL) and 13 activities (Forum,
Database, Glossary, Assignment, Lesson, Quiz,
Workshop, SCORM package, External tool, Choice,
Survey, Wiki, and Feedback). In figure 2, we present
only one resource (Label), and 5 activities (Survey,
Chat, Workshop, Quiz, and Forum) for clarity
reasons.
In the page specification of each concept,
attributes are divided into different parts. For
example, for the Chat activity, its fields are divided
into 4 parts named: general, common module
settings, restrict access, and activity completion.
These parts names are presented in the macro-HMI
model.
Note that there are only two types of
relationships within this model: composition
relationship and inheritance relationship.
5.2 Application of the Factorization on
Moodle
After the macro-HMI analysis, we applied the
factorization process. We noticed that all
activities/resources had the common attributes:
“commonModuleSettings”, “restrictAccess”, and
“activityCompletion”. So we moved these attributes
to the Activity/Resource class. All activities had the
common attribute “general” according to the macro-
HMI model, that’s why we created a class called
“Activity” and we moved the attribute “general” into
it. Some Moodle activities could have outcomes like
Chat activity, Workshop, and Quiz. We added in the
macro model a class named
“ActivityWithOutcomes”. This class had
“outcomes” as an attribute. We noticed that some
activities with outcomes could be graded. Therefore,
we added the class “GradedActi-
vityWithOutcomes”. Among
“GradedActivityWithOutcomes” class, some
activities had the common attributes “grade”. The
“ActivityWithGradedSection” class is created and
contained the “grade” attribute. Some activities from
the “ActivityWithGradeSection” had the common
attributes “ratings”. The class
“ActivityWithRatingsSection” is added to the macro
model with the attribute “ratings”. All coming steps
are carried out on the basis of this analysis.
5.3 Application of the Functional
Analysis on Moodle
Based on the macro-HMI model, we proceeded to
the functional analysis on Moodle. We divided each
interface to several areas. Then, for each area, we
studied the graphical interface components to
identify functionalities related to instructional
design. For example, from the main interface of a
Moodle course, a teacher can show/hide/move a
CSEDU2015-7thInternationalConferenceonComputerSupportedEducation
10
Figure 2: An extract of Moodle macro-HMI model.
section. He can modify the course description, and
manage different groups. He can also add an
activity/resource in a specific section. If the teacher
adds a forum, he will be pointed to a new page about
forum specification. He can add files,
add/modify/delete/separate a discussion and also
reply to a discussion.
We have grounded the formalism of the
functional model on the SADT (Structured Analysis
and Design Technique) Model (Dao et al., 2004).
SADT is a multi language supporting the
communication between users and designers. It is
based on simple concepts in an easy graphical and
textual formalism. This language is conformed to
our functional analysis approach: top-down,
hierarchical, modular and structured.
This analysis is very important in our process; it
can verify existence and relation between macro-
HMI elements.
5.4 Application of the Micro Analysis
on Moodle
As explained in section 4.5, the micro analysis
consists the micro-HMI analysis, the technical
analysis, and the confrontation and formalization
process.
5.4.1 Micro-IHM Analysis
The application of IHM-micro analysis is about
characteristics identification of instructional design
elements. It is based on the macro and functional
models.
For example, the “Course” class has “general” as
attribute. In this phase, we study in details fields
with pedagogical use related to this attribute.
“Fullname” and “shortname” are these fields, so we
replace “general” attribute in the macro-HMI model
by “fullname” and “shortname” attributes in the
micro-HMI model.
The figure 3 shows an extract of Moodle micro-
HMI model (without taking into account corrections
in red).
Reference relationships appear in this model. For
example the abstract class “ActivityWithOutcomes”
refers to “Outcome” class: a teacher can define
outcomes to a course then he can associate a specific
outcome to Moodle activities except for Choice,
Survey, Wiki and Feedback activities.
5.4.2 Technical Analysis
The technical analysis consists in analyzing the
Moodle database. Our goal is to identify the Moodle
instruction design language from a technical
viewpoint to approve the relevant of specific data for
this language.
This analysis consists in specifying the reduced
Conceptual Data Model for Moodle in relation with
the instruction design. We have reviewed all Moodle
database tables. Titles semantic analysis of tables
and fields allows to (1) gather the tables related to
the ID, and (2) ignores those related to technical
specifications (users’ management, learners’
tracking…). Then we studied dependences and
relations between database tables. The generated
Conceptual Data Model is based on reverse
engineering rules. Foreign keys enable the
specification of required multiplicities.
In the next section, we present the confrontation
and the formalization of the Moodle instructional
AMethodforIdentifyingandFormalizingtheUnderlyingInstructionalDesignLanguageofExistingLMSs
11
Figure 3: An extract of Moodle micro-HMI model (without corrections in red), an extract of Moodle final model (with
corrections in red).
design language.
5.4.3 Confrontation and Formalization
The micro-HMI analysis and the technical analysis
have specified two Moodle instructional design
models according to two different viewpoints. In this
step, we are interested in the confrontation of these
models to formalize Moodle instructional design
language.
This step is very important in our process. We
think that the use of only one analysis method
presents many negative points. For example, the
micro-HMI model depends directly on the Moodle
analyst competence. This means the possibility lack
of pedagogical attributes. Similarly, the technical
analysis is not an easy task. Many data structures are
not explicitly reported when creating the database.
From the 2 models comparison, we notice that
every element/ attribute existing in the micro-HMI
model is certainly presented in the technical model.
But some elements exist in the technical model
without being present in the micro-HMI model. That
is why we refer to the PHP source code analysis of
Moodle to verify the presence of these elements.
Figure 3 (including corrections in red) shows an
extract of Moodle final model. Corrections in red
present the confrontation result of the two models.
For example thanks to the technical analysis, we
found that every section has an order. This attribute
has not been detected by the micro-HMI model. The
code source analysis confirms the presence of this
attribute. The attribute “SectionOrder” is presented
in the final HMI model.
The confrontation phase allows also rectifying
information on the micro-HMI model. Figure 4
shows an example about relationship verification
between the “GradeCondition” class and the
“Activity/Resource” class.
Based on the micro-HMI analysis, the
“GradeCondition” class refers to the abstract class
“Activity/Resource” while the same class refers to a
graded activity in the technical model. The code
source analysis of Moodle conditionlib.php file
confirms that the grade condition refers to a graded
activity. That is why the reference relationship is
between the two classes “GradeCondition” and
“GradedActivityWithOutcomes” in the final model.
The final model resulting from the confrontation
phase formalizes the Moodle instructional design
language.
6 CONCLUSIONS
In this paper, we present a meta-model-based
approach and method for identifying and
formalizing LMS languages. We apply our proposed
method on the Moodle platform. The meta-model
will be used as a basis for the development of the
external editor by using a Model Driven Engineering
CSEDU2015-7thInternationalConferenceonComputerSupportedEducation
12
Figure 4: An example about relationship verification between the “GradeCondition” class and the “Activity/Resource”
class.
tooling like EMF-GMF. It will guide and generate
most of the final code for the editor. We have then
been able to propose a graphical and external editor
communicating with the system thanks to a
dedicated API developed and integrated to the LMS.
It will offer more user-friendly and soundly
computer artefacts when development is freed from
the technological choices related to the initial design
of the TEL system considered. This will facilitate
the use of LMS and allow to teachers and
pedagogical engineers (service information
technology and communication for education) of
becoming more familiar with the specific design
upon this LMS. Through the final model, we can
confirm that every LMS is not pedagogically neutral
but embeds an implicit instructional design language
relying on specific paradigms and educative theories
followed by the LMS providers. Note that we are
applying this method on three other LMSs: Ganesha,
Dokeos and Sakai.
Our research work aims to reduce the gap
between TEL and teachers-designers community and
allows to these practitioners designing their entire
courses, outside platforms, basing on their
pedagogical needs without technical difficulties. Our
approach promotes the use of all LMS activities and
resources and expands LMS pedagogical concepts
not by adding new concepts to users but by
facilitating and clarifying the existing tools thanks to
the external editor.
ACKNOWLEDGEMENTS
This work and submission are 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.: Models
transformation to implement a Project- Based
Collaborative Learning (PBCL) scenario: Moodle case
study. IEEE-ICALT, Santander, Cantabria, Spain
(2008) 639-643.
Abedmouleh A., Oubahssi L., Laforcade P. et Choquet C.:
Expressing the implicit instructional design language
embedded in an LMS: motivations and process. In:
Computers and Advanced Technology in Education,
Naples, Italie (Juin 2012).
Alario-Hoyos, C., Bote-Lorenzo, M. L., Gómez-Sánchez,
E., Asensio-Pérez, J. I., Vega-Gorgojo, G., Ruiz-
Calleja, A., GLUE!: An Architecture for the
Integration of External Tools in Virtual Learning
Environments, Computers & Education. 60(1), (2013)
122-137.
Al-Ajlan, A., and Zedan, H., “E-learning (Moodle) based
on service oriented architecture”, In Proc. of the
EADTU's 20th Anniversary Conference, Lisbon,
Portugal, 8-9 November, Lisbon-Portugal, vol. 1,
(2007) pp. 62-70.
Baggetun, R., Rusman, E., and Poggi, C., “Design Patterns
for collaborative learning: from practice to theory and
back”, In Proceedings of World Conference on
Educational Multimedia, Hypermedia and
Telecommunications, pp. 2493-2498, Chesapeake,
VA: AACE 2004.
Berger C. et Kam R. (1996). Definitions of instructional
design. [On-line]. Available: http://www.umich.edu/
~ed626/define.html.
Brito, I. S., Tavares, M., Rodrigues, E. (2014). Using ICT
to support e-Learning in higher education. CSEDU
2014 - Proceedings of the 6th International Conference
on Computer Supported Education, Barcelona, Spain.
AMethodforIdentifyingandFormalizingtheUnderlyingInstructionalDesignLanguageofExistingLMSs
13
Caron P.-A., Derycke A., Le Pallec X., (2005), The
Bricoles project: support socially informed design of
learning environment , 12th International Conference
on Artificial Intelligence in Education (AIED 2005),
Amsterdam, IOS Press, (2005) p 759 - 761.
Coates, H., James, R. & Baldwin, G.: A critical
examination of the effects of learning management
systems on university teaching and learning. Tertiary
Education and Management, 11, (2005) 19-36.
Dao, M., Huchard, M., Rouane Hacene, M., Roume, C.,
Valtchev, P.: Improving Generalization Level in UML
Models Iterative Cross Generalization in Practice,
ICCS'04: 12h International Conference on Conceptual
Structures, Huntsville, USA (2004) 346-360.
De Vries, F., Tattersall, C., and Koper, R., “Future
developments of IMS Learning Design tooling”,
Educational Technology & Society, 9 (1), (2006) pp.
9-12.
Drira, R., Laroussi, M., Le Pallec, X., Warin, B.:
Contextualizing Learning Scenarios According to
Different Learning Management Systems. TLT 5(3),
2012, 213-225.
Gimenes, Itana; Barroca, Leonor; Barbosa, Ellen and
Júnior, Edson A Oliveira (2014). Learning design in
software engineering courses. In: International
Conference on Computer Supported Education,
CSEDU 2014, 1-3 April 2014, Barcelona.
Graf. S. (2007), Adaptivity in learning management
systems focusing on learning styles. PhD thesis,
Vienna University of Technology.
Graham: Blended learning systems: definitions, current
trends and future directions. In Bonk, C. J. & Graham,
C. R. (Eds.) Handbook of blended learning: Global
Perspectives, local designs. San Francisco, CA:
Pfeiffer Publishing, (2005) 3-21.
Jovanovic,J., Gasevic, D., Brooks, C., Devedzic, V.,
Hatala, M.: LOCO-Analyst: a Tool for Raising
Teachers' Awareness in Online Learning
Environments. The 2nd European Conference on
Technology Enhanced Learning, Crete, Greece (2007)
112-126.
Laforcade P. et Abedmouleh A. (2012). Improving the
design of courses thanks to graphical and external
dedicated languages: a Moodle experimentation. In:
Moodle Research Conference 2012, Heraklion,
Greece, 14-15 septembre 2012, p. 94-101.
Loiseau E, Laforcade P.: Specification of learning
management system-centred graphical instructional
design languages - A DSM experimentation about the
Moodle platform. In ICSOFT’13, Reykjavik, Iceland
(2013) 29-31.
Martinez-Ortiz, I., Sierra, J. L., and Fernández-Manjón,
B.: Enhancing IMS LD Units of Learning
Comprehension. In the 4th International Conference
on Internet and Web Applications and Services,
Venice, Italy (May 2009) 561-566.
Mekpiroona O., Tammarattananonta P., Buasrounga N.,
Apitiwongmanita N., Pravalpruka B. et Supnithia T.:
SCORM in Open Source LMS : A case study of
LEARNSQUARE. In ICCE2008, Taipei, Taiwan,
(2008) 166-170.
Oubahssi, L., Laforcade, P., and Cottier, P.: Re-
engineering of the Apprenticeship Elec-tronic Booklet
: Adaptation to new users requirements.The 10th IEEE
International Conference on Advanced Learning
Technologies, Sousse, Tunisia (July 2010) 511-515.
Steel, C.: Reconciling university teacher beliefs to create
learning designs for LMS envi-ronments. The
University of Queensland. In: Australasian Journal of
Educational Technology (AJET2009), vol. 25(3),
(2009) 399-420.
Szabo, M., & Flesher, K.: CMI Theory and Practice:
Historical Roots of Learning Managment Systems.
Paper presented at the E-Learn 2002 World
Conference on E-Learning in Corporate, Government,
Healthcare, & Higher Education, Montreal, Canada
(2002) 929-936.
Tchounikine, P., Morch, A. and Bannon, L.: A computer
science perspective on technology-enhanced learning.
The Netherlands: Springer. (2009) 275-288.
CSEDU2015-7thInternationalConferenceonComputerSupportedEducation
14
