A Meta-Modeling Approach for Extending the Instructional Design

Semantics of Learning Management Systems

Esteban Loiseau, Pierre Laforcade and Sébastien Iksal

Laboratoire d'Informatique de l'Université du Maine, Avenue Olivier Messiaen, Le Mans, France

Keywords: Meta-Modeling, Domain Specific Language, Composition.

Abstract: Nowadays Learning Management Systems (LMS) are not restricted to distant learning. Nevertheless, the

pedagogical expressiveness of courses designed by teachers is strongly dependent on their knowledge and

level of expertise on the LMS they use. The GraphiT project aims to help teachers design pedagogically

sound and technically executable learning designs. To this end, we propose to support teachers by providing

them with an LMS-specific Visual Instructional Design Language, according to a Domain Specific

Modeling approach and tooling. This paper focuses on the abstract syntax of such language. We propose a

specific LMS-centered approach for raising the pedagogical expressiveness of their implicit learning design

semantics. We discussed how the LMS low-level parameterisations could be abstracted in order to build

higher-level building blocks. Based on the Moodle LMS, we present and verify our meta-modeling

approach by formalising the abstract syntax of a Moodle-dedicated instructional design language.

1 INTRODUCTION

Nowadays, Learning Management Systems (LMSs)

are widely spread in academic institutions and are

not restricted to online and distant courses. Indeed,

they can also be useful during, or in completion of,

face-to-face learning sessions (Garrisson and

Kanuka, 2004). Nevertheless, the results of a study

we conducted with 214 teachers, put forward their

heavy form-oriented human-interfaces and

tools/content-oriented instructional design lead to

reduce their uses. In order to set up complex

learning activities, teachers must develop high-level

skills for managing and sequencing the LMS’s

available features and tools. Such skills can be

acquired through specific teacher education

programs that generally focus on the technical

aspects of the platform and not the way they can be

used to support pedagogical practices. Because of

the multiplicity of educational theories (Ormrod,

2011) and approaches, as well as the lack of tools

and processes dedicated to existing LMSs, teachers

develop ad hoc and individual learning design

techniques.

In such contexts, it seems relevant to help

teachers understand the instructional design

possibilities offered by the LMS at their disposal.

This should encourage individual and collective

understanding about the pedagogical uses of the

targeted LMS. the GraphiT project we present

(funded by the French Research Agency) is based on

an LMS-centered designing approach. Its main

objective is to investigate several Model Driven

Engineering (MDE) and Domain Specific Modeling

(DSM) techniques to help specify LMS-centered

graphical instructional design languages and develop

dedicated editors. This paper focuses on the main

challenge: raising the pedagogical expressiveness of

the LMS learning design semantics by using meta-

modeling techniques. Indeed, our past research led

us to identify and formalize, according to a specific

process, the LMS instructional design semantics as a

dedicated metamodel. However, this metamodel

needed to be extended, in order to provide the

semantics of future learning scenarios. This article

presents our current results related to identifying and

formalizing the pedagogical semantics for this

metamodel extension. Because it is widely spread,

we have chosen the Moodle LMS to verify, as a first

validation step, the feasibility of our proposal.

We discuss, in Section 2, the current approaches

for instructional design and operationalization on

LMSs. In comparison with them, we then detail the

original position of the GraphiT project regarding

MDE and DSM. We also detail the teachers’ design

Loiseau E., Laforcade P. and Iksal S..

A Meta-Modeling Approach for Extending the Instructional Design Semantics of Learning Management Systems.

DOI: 10.5220/0005002800720080

In Proceedings of the 9th International Conference on Software Paradigm Trends (ICSOFT-PT-2014), pages 72-80

ISBN: 978-989-758-037-6

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

requirements and needs that we collected during

interviews. Section 3 is dedicated to the presentation

of our abstraction approach. We also formalize as a

metamodel extension, the application of our

approach for the Moodle LMS. We then explain and

illustrate the 4-levels architecture we propose and

illustrate it with a representative example in Section

4. This first validation step is necessary in order to

verify that pedagogically sound learning scenarios,

that meet the designers' requirements, can be

formally described with our model. Finally, we

discuss how we plan to use specific DSM tools in

order to elaborate the graphical instructional design

language from this abstract syntax.

2 RESEARCH CONTEXT

The main issue we aim to tackle is studying the

pedagogical expressiveness, in terms of possibilities

and limits, of operationalizable instructional design

languages to specify, i.e. languages allowing to

formalize executable learning scenarios that can be

automatically set-up into existing LMSs. After

discussing the existing approaches and techniques,

we present the original approach we chose for the

GraphiT project.

2.1 Research on LMS Instructional

Design

The Technology Enhanced Learning (TEL) research

domain has provided many solutions to support

instructional design: Educational Modeling

Languages facilitate the specification of learning

situations as formal learning scenarios for delivering

and exchanges purposes (Koper and Manderveld,

2004), Visual Instructional Design Languages

(VIDL) (Botturi and Stubbs 2007) support

practitioners communities to communicate and

imagine new learning situations and finally,

Learning Management Systems (LMS) provide

operational environments for delivering online

learning situations (Muñoz-Merino et al. 2009).

These solutions consider LMSs as a final generic

environment, providing LMS-independent

approaches that focus on the instructional design

aspects rather than on how they can actually be

operationalized with existing LMSs.

Unfortunately, most design languages do not

propose direct binding and operationalization with

existing LMSs. Standards such as IMS-LD (Koper,

2006) have therefore not succeeded in being

integrated into widely spread LMSs (Burgos et al.

2007). Some researchers have proposed partial

transformations from practitioners-centered

scenarios towards LMS-centered models (e.g. from

PPC to Moodle (Abdallah et al. 2008), from IMS-

LD to Moodle (Burgos et al. 2007)). However, these

models are based on a subjective and incomplete

Moodle metamodel specified by researchers. Such

transfor-mations attempts show a semantic gap

leading to information loss or incomplete target

models. Nevertheless, they have also highlighted the

relevance of applying techniques and tools from the

Model-Driven Engineering domain.

Recent attempts to operationalize LMS-

compliant models have been tried by following a

similar binding/translation approach. For example

the Glue! architecture, including the Glue!PS editor

(Alario-Hoyos et al., 2012), and the CADMOS

editor (Katsamani et al, 2012) are LMS-independent

solutions offering LMS deployment features towards

the widely spread Moodle LMS (Moodle, 2014).

They both realize the deployment by generating a

Moodle course backup with all the information,

mapping their own data model concepts to Moodle

data model concepts. This backup is then imported

and deployed within a Moodle course, using the

Moodle restoration process. Such approaches result

in semantic adaptations and losses during their

internal mapping, because of the gap between the

instructional design language and targeted LMS’

learning design capabilities and features.

For now, the LMS-independent approach

therefore reduces the operationalization issues but

raises challenges such as specifying a transformation

model, capturing the LMS metamodel, reducing the

semantics losses during translation and providing a

tool that can embed the scenarios into various

existing LMSs.

2.2 Overview of the GraphiT Project

from an MDE and DSM

Perspective

Our approach in the GraphiT project, is different to

current ones. Indeed, we propose an LMS-dependent

architecture. It only focuses on one existing LMS in

order to provide an instructional design language

that will be specified and tooled according to the

future mappings to realize (interoperability of

generic learning scenarios is out of our scope). In

other words, the main idea is to drive the design by

taking into account, at first, the LMS semantics (and

then the future mappings).

We do not aim at extending the LMS semantics

with new add-ons/plugins, enriching it with more

AMeta-ModelingApproachforExtendingtheInstructionalDesignSemanticsofLearningManagementSystems

Figure 1: Global overview of the GraphiT architecture.

pedagogical-oriented features. Our objective is to

support the design of learning scenarios in

conformance with the LMS’s semantics (its abilities

as well as its limits). We also do not aim at only

providing a notation layer, on top of the LMS

metamodel. By extending the LMS metamodel we

also extend the abstract syntax of the instructional

design language, resulting in losing the LMS-

compliance format. We plan to restore this format by

DSM techniques (weaving and transformation

models) we are currently experimenting. We aim at

guarantying that learning scenarios could be fully

operationalized, into the LMS, without semantics

losses. Obviously, our approach has the advantage

being LMS-dependence (operationalization support)

but it also has the inconvenience of being restricted

to one LMS and one of its versions (reengineering

cost). We will particularly study how MDE/DSM

tools can be useful to reduce that cost.

A global architecture of our solution is illustrated

in Figure 1. The LMS instructional design semantics

first has to be identified and formalized as a domain

metamodel. Then, this metamodel drives the

elaboration of an XSD schema that will be used to

develop the API. This API will be used through an

import facility, accessible by teachers-designers, in

their LMS courses. It will take in charge the XML-

based scenario parsing and the LMS's databases

filling-up. The LMS metamodel will also act as a

basis for the elaboration of the visual instructional

design language. According to DSM techniques and

tools (like the EMF/GMF ones for example), this

language will be composed of an abstract syntax,

from which the graphical, tooling and mapping

models will be driven. The editor will be also

developed using the code-generation facilities of

DSM tools.

Past works have focused on the LMS meta-

model formalization (Abedmouleh et al., 2012). We

are currently focusing on the abstract syntax of the

instructional design language.

2.3 Restoring the LMS Semantics

The produced scenarios need to be compliant with

the initial LMS meta-model, in order to be deployed

by the API. In order to reach this compliance, we

propose to modify the model with two models

transformations. The first transformation consist of

various, fine-grained transformations during the

design process: it will provide several LMS

mappings to teachers in order to guide them through

their design. They are endogenous transformations

because source and target models will both be

conformed to the instructional design metamodel.

The second transformation will be realized as an

export feature that can be used after the design

process. This exogenous transformation will produce

a scenario/model conformed to the LMS-metamodel.

Unlike other LMS-independent approaches,

using transformation models techniques, we are

particularly interested in making the underlying

mapping models explicit. Indeed, these mappings

models are at the center of our approach: their

validation, a priori of their machine-translation, by

experts of the considered LMS, will mainly

participate in the learning scenario expressiveness.

These explicit LMS bindings will control the

translations at runtime. They will guarantee the

semantics preservation.

2.4 Focus on the Instructional Design

Abstract Syntax from a

Metamodeling Perspective

The main challenge is to abstract the LMS

instructional design semantics enough to provide

teachers with higher, pedagogically-sound, design

blocks. The LMS expressiveness and limitation

therefore have to be overcome, in order to offer

teachers instructional design mechanisms that are

closer to their practices and needs for specifying and

sequencing learning activities. Concretely speaking,

the issue is to find a way to specifying the relations

between the instructional design language

metamodel (that we will refer to as MM-ID) and the

LMS metamodel (that we will refer to as MM-

LMS).

To this end, we already led several experiments

(Loiseau and Laforcade, 2013) on three different

approaches: 1/ MM-ID and MM-LMS are two

different metamodels without any structural

relations, 2/ MM-ID and MM-LMS are the same, the

ICSOFT-PT2014-9thInternationalConferenceonSoftwareParadigmTrends

ID-language being only built as a notation layer on

top of the metamodel, 3/ MM-ID is an extension of

MM-LMS. The first approach corresponds to the

usual way of specifying instructional design

languages with its main advantage (expressiveness)

but also inconvenience (difficulty to operationalize).

The second approach reveals the limits of the

concrete syntax expressiveness, because it is only

defined by derivation of the abstract syntax. Finally,

the third 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 also highlights the

importance to drive the expressiveness (and

semantics) extension of the initial metamodel with

the binding capacity. This paper focuses on our

further results and propositions about this issue.

3 EXTENSION BY

METAMODELING

According to the practitioners' needs, presented in

the next sub-section, we propose to direct the

abstraction of LMSs semantics to the LMS uses

supporting learners and tutors activities. The

following sections present these abstractions in

relation with their formalizations for the Moodle

LMS (Figure 2). We used the Ecore metamodel

format because it will be handled by the EMF and

GMF metamodeling tools (EMP, 2014) in order to

drive the specification of the instructional design

language and the development of its dedicated

graphical tool. The metamodel from Figure 2 can be

considered as the general abstract syntax of the

instructional design language to be developed.

3.1 A Practical Overview of Teachers'

Requirements

Before tackling the LMS metamodel extension we

first have to collect the LMS-specific pedagogical

needs and practices. We therefore conducted several

theoretical studies from literature sources (Conole et

al., 2014), and practical studies with surveys and

iterative interviews of 203 teachers and pedagogical

engineers. These interviews covered a large variety

of Moodle use contexts: online learning, local

learning within universities, full distant courses as

well as blended learning. Although the GraphiT

project deals with different LMSs for guarantying

the reproducibility of its results, we propose to focus

on the Moodle platform which is the most popular

open-source license free LMS. The analysis of these

different sources aimed at collecting pedagogical

practices or needs, and requirements about the

languages and editors to specify and develop.

This study highlighted the fact that practitioners

do not really have complex practices to capture,

because of the heterogeneity of their Moodle

expertise and pedagogical background. Nevertheless

they all need to design their course by adapting

Moodle’s tools to their basic pedagogical uses.

Indeed, 88% of of respondents point out the heavy

parameterization of tools and resources; 46%

requiring an abstract view of the pedagogical uses,

in order to help them in select and configure the

right implementation activities.

The advanced studies we conducted with

pedagogical engineers, allowed us to identify several

specific requirements concerning the language and

the authoring-tool we will develop. First, they

mention the need for the graphical authoring-tool to

allow designers to select pedagogical blocks on top

of the LMS semantics as well as with Moodle

building blocks to compose with. In their mind, the

editor will not have to strictly follow a top-down

process from abstracted specification elements to

implementation one expressed in terms of Moodle;

abstractions from Moodle and its own concepts

should be mixed up together according to

practitioners' expertise about instructional design

(mix of specification and implementation

concepts). Secondly, they are interesting in the idea

that mappings from pedagogical design blocks to

Moodle concepts can be showed to practitioners

(default mapping) and adapted if required

(mapping adaptation). This design approach could

help practitioners in the appropriation of the

pedagogical constructs and guide them in designing

more abstract learning scenario, while mastering the

translations into LMS elements.

Another highlighted need is to declare

information, within the learning scenario, that does

not require to be mapped into LMS concepts point

(declarative non-visible information). This would

allow the designers to write information that is only

visible by them and not by students or tutors such as

information about face-to-face sessions mixed up

with the LMS-centered ones, indications about

pedagogical strategies or pedagogical objectives or

information about activities to realize on the LMS at

a specific runtime moment, according to concrete

data (enrolled students, dates, etc.). Finally, another

identified need is to facilitate the course sequencing

with advanced structures (choices, sequences with

AMeta-ModelingApproachforExtendingtheInstructionalDesignSemanticsofLearningManagementSystems

Figure 2: The 4-levels abstract syntax of an instructional design language on top of the Moodle metamodel.

elements showed one-by-one according to their

progress (advanced activity structures). Indeed,

these structures can be designed manually but it

requires to parameterize many low-levels and

technical-oriented properties (achievements, restrict

access conditions...) that teachers would appreciate

not to set up by themselves.

3.2 Fine-Grained Pedagogical Activities

as First Abstraction

The first LMS-abstract building block we propose is

the pedagogical activity. This activity is defined as

an abstraction of parameterizations one can realize

when using a LMS tool or resource for a specific

pedagogical usage. For example from a single tool,

for example a forum, one can design several

pedagogical uses, depending on its configuration:

provide information to students, set up group work,

propose a peer reviewing activity etc.

To be used appropriately, this first abstract block

requires a name, a description, and some specific

properties that are set by the practitioner, during the

design process. For example an exchange activity,

involving student communication, could either rely

on a forum or a chat, depending on a synchronous

property. The mappings will not be limited to the

parameterization of a tool. For example with Moodle

it will also impact other elements in relation with the

tool/resource: grades, objectives, groupings,

restriction access and achievements rules, etc.

3.3 Large-Grained Pedagogical

Activities as Second Abstraction

The second LMS-abstract building blocks are of two

kinds. We propose to adapt and integrate some

pedagogical patterns and templates from literature

(Bergin et al., 2012) (Heathcote, 2006) (for

examples as high-level blocks to use and combine

for building learning sessions involving instructional

strategies: inquiry, problem solving, role-playing,

exploration, etc. Although practitioners from our

studies do not use to compose with them, we aim at

integrating them to encourage some pedagogical

reflection and guide designers towards new ways of

supporting their didactic and pedagogical objectives.

This kind of pedagogical pattern will also have a

description of their context, problem and solution

uses. They will rely on a mix of structural activities,

low-levels blocks (pedagogical activities) and LMS

elements to be realized.

In order to ease and assist the practitioners, when

assembling and setting-up combinations of activities

or resources, we propose a set of usual structural

elements (selection, sequence, conditional activities,

etc.). These blocks will be composed of other

blocks, from high or low levels, including

themselves. In the case of Moodle they will be

concretely translated into complex combinations of

labels (stating the structure kind and use for users)

and shifted content (move left/right Moodle feature)

according to the activity structure components in the

learning scenario. After various translations and

mappings until reaching the LMS low-level

elements, all its content parts will be parameterized

ICSOFT-PT2014-9thInternationalConferenceonSoftwareParadigmTrends

(restrict access, visibility, achievement...) with

appropriate properties in order to set up the desired

behavior.

3.4 A 4-Levels Abstract Syntax

The global architecture we propose for the abstract

syntax of the Moodle-centered instructional design

language is composed of four levels. Figure 2

illustrates our proposition.

Level 1 fits to the Moodle metamodel. Readers

have to consider the Figure 2 illustration (right part)

as an incomplete representation of the whole

metamodel. Indeed, in order ease the

comprehension, only the important structural

relations and concepts are depicted

. Level 1

elements (restricted to the Moodle activities – the

Moodle name given to the tools - and resources) can

be directly used by teachers and parameterized for

building a learning session. From the Moodle

metamodel point of view, these elements require a

global Course and a Section container to be attached

in. In the extended metamodel, they will be specified

at first as children of level 4 elements. The large

model transformation, after the design process, will

deal with producing a model in full-compliance with

the Moodle metamodel: creation of the global

Course instance, of its Section instances, attachment

of all the corresponding Moodle elements according

to the orders and positions deductible from the

source scenario.

Level 2 (part 2) corresponds to our first high-

level blocks about pedagogical activities. They are

composed of Level 1 elements, i.e. Moodle activities

and resources. Level 3 (part 3) captures the second

abstract blocks about pedagogical patterns and

activity structures. The first one will be composed,

after automatic design-time models transformations,

by Level 3 elements, i.e. elements from levels 1-to-3,

including structural activities and other Pedagogical

Activities. The activity structures are also composed

of Level 3 elements but their content will be

specified by teachers-designers during the design

process. Indeed, there is no default content to obtain

by models transformation. Finally, the fourth level

(part 4) is the contextual level, focusing on the

global structure of the learning session, in relation to

the different face-to-face, complementary or distant

sessions.

1 A global overview of the Moodle 2.4 metamodel we

captured can be retrieved at http://www-lium.univ-

lemans.fr/~laforcad/graphit/wp-content/uploads/2014/05/

Moodle-2.4_GeneralMM.png

Such Level 4 elements rely on the Moodle

section concept. Indeed, Moodle only proposes some

sections into the space of the course for aggregating

the tools and resources. However, designers have at

their disposal an indentation feature (position

property in the Moodle metamodel) to shift activities

and resources in order to visually indicate their

collective relationships. This position property will

be used by the dynamical mappings, in order to

position the corresponding elements in accordance

to the source element position in the global learning

scenario.

The composition-relations, annotated with a (1),

indicate that the content will not be showed in the

future concrete syntax (notation) as nested elements

but will be shown in another sub-diagram where the

parent container will play the role of the root canvas.

On the contrary, the composition annotated with a

(2) symbol, indicates that content will be showed as

nested elements of the parent container in the same

diagram. Finally, the nextE reflexive relation allows,

by inheritance, to provide a previous/next

information to sequence the various elements within

their dynamic pedagogical context (the ordering

concerns the child elements sharing a same Level

Element parent).

The future authoring-tool will directly propose to

practitioners the level-4 elements in the tool palette.

Indeed, these elements are necessary to map to

Moodle sections in order to sequentially structure

the course skeleton. Sessions that do not rely on

Moodle features can also be described if designers

need an overall view of a global module/course

larger than the ones involving the use of an LMS.

Other level-4 elements will then open an empty sub-

diagram when double-clicked. It can then be used to

arrange levels 3-to-1 elements from the new palette.

Indeed, practitioners can then choose the method

(top-bottom, bottom-up), the description level

(specification versus implementation) and the

elements to select, combine and adapt. Except

activity structures, other levels 3-to-2 elements can

be opened up as another sub-diagram containing the

default mapping to levels 2-1 elements. Every

mapping can be adapted and modified by

deleting/adding new elements (according to those

accepted under the parent element) or modifying the

elements properties.

The leaf meta-classes from figure 1 (dark

elements) sketches some examples of future

elements. For ease of reading, we choose not to

show these attributes. However, each of them owns

specific properties in accordance with the different

in-progress formal specifications we are studying

AMeta-ModelingApproachforExtendingtheInstructionalDesignSemanticsofLearningManagementSystems

about the Moodle instructional design semantics,

pedagogical activities and patterns, and activity

structures.

The current abstract syntax proposition still has

to be improved in order to allow the declaration of

didactical objectives to the various Level 4-to-1

elements. Such objective will be mapped into

Moodle Objectives, attached to the root Course and

referenced by the direct or indirect corresponding

Level 1 elements. Similarly, roles or groups have to

be included in order to allow the division of labor in

the learning scenario. Mappings to the Moodle

concepts of Group and Grouping will be studied.

Our 4-Levels architecture meets the practitioners'

requirements depicted in section 3.1. from a static

perspective. The dynamical aspects will be tackled

by the transformations models and are out of the

scope of this article.

4 FIRST VALIDATIONS

The validation of our proposal requires several steps.

First, we need to verify that our 4-levels metamodels

can be used, in a declarative way, to formalize a set

of learning scenarios, identified as relevant use-

cases. Because we have not yet integrated the

automatic execution of models transformation into

the EMF-based editor, we specified the different

default mappings by defining them directly using the

“add child” service of EMF-based editor (manual

definition). Then, we will still have to verify that the

semantic meaning is maintained when the automatic

weaving/ transformations models will be added.

Finally, the graphical aspects of the editor

(usefulness, user-friendliness, etc.) will have to be

validated when the final concrete syntax will be

specified and developed from our abstract syntax

architecture proposal. Even though we have

extended our research, this article is mainly focused

on the first step of this project.

To this end, we propose to illustrate our proposal

by formalizing one of the simple and representative

use-cases for the Moodle LMS. First, we propose a

brief textual description of this use-case and then,

the equivalent specification with the dedicated

metamodel we proposed in section 3 (Figure 3 is a

caption-screen of the EMF-tree-based model editor,

annotated to highlight the elements' levels).

The learning scenario is composed of two

learning sessions. The first one is a lecture session

for which the teachers simply want to provide

Resource consultation activity that contains their

lecture presentation material. This pedagogical

activity has the quantity property set to “one” and

the location one set to “local”. These properties will

lead the dynamic mapping process to propose the

File Moodle element. The learning scenario then

continues with a face-to-face practical work sessions

in a room with computers. The teachers would like

to use the Moodle platform for supporting the

pedagogical pattern “Write a synthesis” with the

collaborative property set to “true”. This pattern is

automatically mapped to be composed of a sequence

activity structure embedding 4 sub-components. The

first one is another Resource consultation. This time,

the properties set to “several” (quantity) and “local”

(location) by the teacher will lead the transformation

process to add a Folder tool. The second sub-

element is a Brainstorming pedagogical activity. Its

orientation property, set to “discussion”, leads to

propose a Forum tool. Similarly the third sub-

element is Report writing activity leads to a Wiki

tool because of the collaborative property set to

“true”. Finally, the fourth sub-element is a Guidance

activity that aims at reminding the teachers to

evaluate the synthesis in the wiki. The public

property set to “tutor”, leads the mapping process to

make the corresponding Label invisible

(visible=”no”) to students (it will be only visible to

the teacher).

The teacher can change the activities properties

at any time, leading to other mapping adaptations.

For example, by changing the

collaborative property

of “Write a synthesis” to “false”, the default values

for the sub-components 2 and 3 properties in relation

to individual work will be changed to new mappings

for individual-oriented Moodle tools. The teacher

can also manually delete the mapping elements, re-

arrange their order, or add extra elements. Figure 3

shows a global overview of the learning scenario

elements, including all the automatic mappings,

according to the various properties and values (not

depicted within the figure).

Figure 3: Example of learning scenario composed of

elements from the 4 levels.

L4 element

L2 element

L3 element

L1 element

L3 element

ICSOFT-PT2014-9thInternationalConferenceonSoftwareParadigmTrends

5 CONCLUSIONS

This paper proposes a metamodeling approach for

raising the pedagogical expressiveness of learning

design semantics of existing LMSs. To do so, we

propose to extend the LMS metamodel with specific

concepts, properties and relations, in order to meet

practitioners' requirements. We discussed how the

LMS low-level parameterizations could be

abstracted in order to build higher-level building

blocks. Based on the Moodle LMS, we present and

illustrate our approach, by formalizing the abstract

syntax of a Moodle-dedicated instructional design

language, following a specific 4-levels architecture.

Based on one illustrated use-case, we discuss how

we validate, as a first step, our metamodel extension

to formally describe Moodle-specific learning

scenarios. Such abstraction of LMSs semantics may

be a promising approach to develop a new

generation of LMS-centered learning design

languages, enabling teachers to develop

pedagogically sound and technically executable

learning scenarios.

The complete version of our metamodel

proposition is currently exploited to specify a

concrete syntax (graphical notation), a palette and

mappings models, in order to develop the final

authoring-tool. Because of our former experiences

with EMF/GMF frameworks, we will also have to

pay attention to the abstract syntax adjustments,

required in order to realize specific visual

representations.

We are also currently experimenting different

frameworks for weaving and transforming models.

Indeed, the different default mappings to realize

during the design, require a contextualized

transformation model to perform. We are studying

weaving tools that will allow us to specify the

mappings and automatically generate these

transformation rules (during the design process).

Also, in our approach, the 4-levels extended

metamodel will not allow to serialize the future

learning scenarios in conformance with the LMS

format (source metamodel): a global transformation

is required to restore this conformance. This

transformation will be realized as an export feature

from our authoring-tool.

ACKNOWLEDGEMENTS

This article is part of the GraphiT project, a 42-

months funded project of the French research

agency. We thank the various individuals that

participated to the interviews and surveys. We also

thank the project members involved in the

identification and formalization task that helped us

for this publication.

REFERENCES

Abedmouleh A., Oubahssi L., Laforcade P., Choquet C.,

2012. An analysis process for identifying and

formalizing LMS instructional language. In

Proceedings of the 7th International Joint Conference

on Software Technologies (ICSOFT 12), Rome.

Abdallah, F., Toffolon, C., Warin, B. Models

transformation to implement a Project-Based

Collaborative Learning (PBCL) Scenario: Moodle

case study. In Proc. 8th IEEE International

Conference on Advanced Learning Technologies

(ICALT 08). Jul. 2008. pp. 639–643.

Alario-Hoyos, C., Muñoz-Cristóbal, J.A., Prieto-Santos,

L.P., Bote-Lorenzo, M.L., Asensio-Pérez, J.I.,

Gómez-Sánchez, E., Vega-Gorgojo, G., Dimitriadis,

Y. GLUE! - GLUE!-PS: An approach to deploy non-

trivial collaborative learning situations that require the

integration of external tools in VLEs. Proceedings of

the 1st Moodle Research Conference. 77-85.

Heraklion, Crete-Greece, September 2012.

Bergin J., et al., Pedagogical Patterns: Advice For

Educators, Joseph Bergin Software Tools, 2012.

Botturi, L., and Todd, S., (Eds.). 2007. Handbook of Visual

Languages for Instructional Design: Theories and

Practices. Information Science Reference.

Burgos, D., Tattersall, C., Dougiamas, M., Vogten, H., and

Koper., R., 2007. A First Step, Mapping IMS Learning

Design and Moodle. In Journal of Universal

Computer Science 13 (7). pp. 924-931.

Conole G. et al., “Mapping pedagogy and tools for

effective learning design”, Computers & Education.

Vol 4. 2004. pp. 17–33.

Eclipse Modeling Project. Official Website

http://www.eclipse.org/modeling/ Retrieved from april

2014.

Garrisson, D.R., Kanuka, H. Blended learning:

Uncovering its transformative potential in higher

education. The Internet and Higher Education. Vo l . 7 .

2004. pp. 95–105.

Heathcote, Elizabeth A. Learning design templates – a

pedagogical just-in-time support tool, in Minshull,

Geoff and Mole, Judith, Eds. Designing for Learning.

Chapter 4, pages pp. 19-26. JISC Development Group.

2006.

Katsamani M., Retalis S. and Boloudakis M. Designing a

Moodle course with the CADMOS learning design

tool, Educational Media International. 49:4. 2012. pp.

317-331.

Koper, R., and Manderveld, J., 2004. Educational

modelling language: modelling reusable,

interoperable, rich and personalised units of learning.

AMeta-ModelingApproachforExtendingtheInstructionalDesignSemanticsofLearningManagementSystems

In British Journal of Educational Technology Vol 35

N°5. Pp 537-551.

Koper, R., 2006. Current Research in Learning Design. In

Educational Technology & Society, 9 (1), pp.13-22.

Loiseau E., Laforcade P. Specification of learning

management system-centered graphical instructional

design languages - A DSM experimentation about the

Moodle platform. Proceedings of the 8th International

Joint Conference on Software Technologies (ICSOFT

13), Reykjavik (Iceland), 29-31 juillet 2013.

Moodle. Official Website, https://moodle.org/ Retrieved

from april 2014.

Muñoz-Merino, P.J., Kloos, C.D., and Naranjo, J.F., 2009.

Enabling interoperability for LMS educational

services. In Computer Standards & Interfaces 31

(2009). pp. 484-498

Ormrod, J.E. Human Learning. Pearson College Division.

2011.

ICSOFT-PT2014-9thInternationalConferenceonSoftwareParadigmTrends