Towards
normalization in e-learning for collaborative work environments
Theme 1. Information sciences in the
digital age: learning, education and research.
Everardo REYES 1-2, Abderrazak MKADMI 1-3
1 Laboratoire Paragraphe, Université Paris
VIII.
2, rue de la Liberté 93526 - SAINT-DENIS cedex 02, France.
Tel. +33.1.49.40.67.58
2 Departamento de Comunicación, Tecnológico de
Monterrey (ITESM) Campus Toluca.
Eduardo Monroy 2000. Col. San Antonio Buenavista.
CP. 50110. Toluca, Mexico.
Tel. +52.72.22.79.99.90
3 Unité de recherche : Bibliothèque numérique : pour
la valorisation du patrimoine, ISD, Campus Universitaire de la Manouba, Tunisie
Tél.: +216.71.600.
everardo.reyes@itesm.mx ; amkadmi@univ-paris8.fr
Keywords: digital
documents, collaborative work, e-learning recommendations, learning objects
Abstract:
Contemporary approaches on information and communication technologies and their implication on knowledge domains reveal needs to improve our manners for treating information. The possibilities of XML technologies and hypermedia allow the development of new systems to distribute knowledge in local or global fashion. This perspective enhances the role of authors who have now access to open channels for sharing and working with other authors.
This
paper presents an exploration of digital documents in the educational field and
its implications within common problematic such as
normalization, modeling, sharing and reuse, with the intention to present an integration
of two software applications developed at the Paragraphe Laboratory of the University of Paris VIII: ICRS and HTT
Online.
Taking into account the works lead by W3C, IEEE, and IMS Global, our study aims to link the robustness of XML technologies with the support of collaborative work strategies in distance learning environments where Learning Objects and Content Packages seem to become the common currency between universities and educational institutions.
1. Introduction: an approach to digital
documents
Creating and exchanging information has
established common practices in most of today's knowledge domains. Either we
lead a project on digital libraries or on distance education we deal with the
so-called digital documents.
New ways of producing and making
available information demand to reconsider the notion of document itself.
Contrary to traditional documents - or analog
documents - digital documents are not static objects regarded in terms of
its physical form or its editing processes [Chartron, 2002]. In terms of our
research, we consider digital document an
object created by an author that can also be reconstructed by a potential user
with the intention to take advantage of it. Current discussions recommend the
term resource to refer to digital
documents and document to refer to
traditional documents, which should be seen as finished entities [Cruzel,
2004], however we
regard digital documents right from their content and we consider that the use
of XML technologies for its modeling should be employed.
The field of digital documents is broad.
[Le Cosnier, 1997] has proposed a typology consisting of seven categories,
sorted by priority: 1) electronic journals; 2) pedagogical documents; 3) public
administration documents; 4) heritage documents of great historical value; 5)
personal documents; 6) electronic press; and, 7) media and Internet.
Our research mainly focuses on the second
category: pedagogical documents for computer-based training (CBT). For [Le
Cosnier, 1997], these documents develop its potential through distance and
asynchronous environments; they are commonly modular and could be articulated
in different ways depending on the learner's path. From this perspective, the
employment of learning platforms - or Learning Management Systems (LMS) - is
required, yet the development of tools for authoring content is needed for
granting best practices and results to acquire, to share and to produce
knowledge.
Taking into account the works lead by
W3C, IEEE, IMS Global, and SCORM, our study aims to link the robustness of XML
technologies with the support of collaborative work strategies in distance
learning environments where Learning Objects and Content Packages seem to
become the common currency between universities and educational institutions.
This paper presents an exploration of
digital documents in the educational field and its implications
within common problematic such as normalization, modeling, sharing and reuse,
with the intention to present an integration of two software
applications developed at the Paragraphe
Laboratory of the University of Paris
VIII: ICRS and HTT Online.
2. Learning and teaching in the digital
age
E-learning, as we know it today, has been
closely attached to the evolution of information and communication technologies
(ICT). The inclusion of educational technology at school allowed teachers to
enhance learning from different supports besides the spoken word. In that
manner, radio, television, movies, photographs, and, more recently, CD-ROMs,
DVD-ROM's and Internet have gained a great success during last two decades.
However, current recommendations
demonstrate that digitizing learning materials and delivering them to students
does not suffice. It is important to conceive those materials from a more
public and collaborative perspective, which means creating pedagogical
documents that can be used and reused in different contexts. One of the
solutions proposed by international organizations are Learning Objects (LO).
2.1 Learning Objects
Defining the concept of Learning Object
has not been simple; actually, there exist as many definitions as literature on
the theme. Furthermore, it is necessary to explicit that Learning Objects are
also called differently depending on the context or institution, most notably:
"pedagogical objects" [Kenley, 2000], "online learning
materials” [MERLOT, 2000], "pedagogical documents" [ARIADNE, 2000], "sharable
content objects" [SCORM, 2004].
It has been noted by [Wiley, 2001] that
LOs are grounded in the object-oriented paradigm of computer science. This
paradigm values the creation of components (called “objects”) that can be
reused in multiple contexts. Adapting those ideas, the main characteristics of
Learning Objects are:
-
they are small, structured and standalone pieces that compose a
course ;
-
they can be used and reused in different contexts;
-
they contain high quality information in a manner that facilitates learning
and pedagogy; - they have a stated
objective and a designated audience;
-
they have ownership and associated intellectual property rights;
-
they are transportable (digital entities deliverable over the Internet);
-
they are addressable and available for others through metadata.
We agree with the definition of Learning
Object proposed by the E-learning Consortium [ELC, 2003]: "A re-usable,
media-independent chunk of information used as a modular building block for
e-Learning content. Learning objects are most effective when organized by a
metadata classification and stored in a content repository such as a Learning
Content Management System". And, in the same line, with the five abilities
that should be assured by standard bodies and specifications in e-learning
[ELC, 2003]: 1) Interoperability; 2) Re-usability; 3) Manageability; 4)
Accessibility; and, 5) Durability.
3. Towards normalization in e-learning
E-learning is still in its developing
phase. As we have seen, even the term has no official definition and there are
plenty points of view. Besides, concepts such as Learning Object, Learning
Management System, or Content Package, still cause confusion from one context
to another.
The works lead by international
organizations that regularize the use and reuse of learning technologies have
pointed out specifications that aim to assist best practices for their
implementation. Among those organizations we take into account in our research
the W3C, the IEEE and the IMS.
3.1. IEEE LOM:
defining metadata
Nowadays, the only learning technology
standard is the IEEE Learning Object Metadata (IEEE Standard 1484.12), which
was approved in 2002 and is based on the IMS LOM (2001).
IEEE LOM defines a base schema
distributed in nine general categories to specify a data framework to define
the metadata structure of the instructional components. Each category contains
different sub-elements (figure 1).
Fig.
1. IEEE LOM elements and sub-elements [IEEE, 2000]
The objective is to guide learners,
instructors and software developers to search, evaluate, obtain and use
Learning Objects and facilitate their exchange. Nevertheless, in the real
practice it has been perceived that IEEE LOM lacks of a well-defined learning
perspective.
3.2. IMS CP: distributing learning
objects
IMS Global was initiated in 1997 by the
non-profit consortium EDUCAUSE, its goal is to detail a set of specifications
that build a framework to interchange educational elements. IMS covers six
specifications for learning environments: 1) metadata; 2) content packaging 3)
assessments design; 4) learner profiles; 5) content sequence, and 6) learning
design.
Our research focuses on the Content
Packaging Specification [IMS CP, 2003]. The concept of content package implies
the distribution of a single and standalone compressed binary file (zip, jar,
cab) containing a course, a lesson, a module, or simply a collection of related
content objects. A content package bundles content objects with a content
organization that is described in a manifest.
The manifest, on its side, is an
independent XML file that must be named imsmanifest.xml.
It describes four parts of the learning object: metadata, organizations,
resources and sub-manifests. As recommended by IMS CP, the manifest should be
stored at the directory 's top level, from where other sub-folders are created
and organized along with the physical files of the learning unit (e.g. PDF,
HTML, JPG, MPEG, etc.) (Figure 2)
Fig.
2. Conceptual Content Package
Currently, the distribution and
integration of packages in different systems is not yet well supported, even if
they offer fully compliance to the specification.
3.2. XML: modeling documents
The World Wide Web Consortium (W3C)
developed the eXtensible Markup Language in late 90s as a protocol for containing and managing information, however, it is also regarded as a family of technologies that can do
everything form formatting digital documents to filtering data.
XML is considered as the future language
for electronic exchanges because of its capacity to associate a semantic
structure to data (by means of elements and attributes) and the rigid
separation of format from content. Furthermore, XML offers the possibility to
present data in different formats, to transform its document structure into another
(using the eXtensible Stylesheet Language) and to add multimedia assets (e.g.
SMIL).
Since the beginning, it has been
necessary to accurate practices in order to benefit from semantic possibilities
of digital documents. Since XML allows us to describe our own elements, the
panorama conceived by Tim Berners-Lee requires an integration of document
models (e.g. XML Schemas) with ontologies. In this sense, it becomes important
not only to create those models but also to share them.
Current approaches for sharing document
models take us to the notion of repository.
A repository of XML models is a database that facilitates the access to
existing document models in a determined area, responding to a certain need.
The main idea behind repertories is to foster the free exchange between
professional users and to get informed on the work that is being done by others
[Mkadmi, 2003].
4.
HTT Online and ICRS, an approach to collaborative work environments
ICRS and HTT Online are two software
applications created from an educational context at the Laboratoire Paragraphe of the University of Paris VIII. This
section sketches an approach to collaborative work environments based on
reflections on normalization of technology for modeling and assisting the
creation of learning materials. By integrating both tools, we propose a
hypermedia perspective to simplify the presentation and interaction of actors
with digital documents.
4.1. Introducing ICRS: collaborative
research of information
The Information Collaborative Retrieval
System (ICRS) was developed as a result of field studies and a state-of-the-art
in the subject [Mkadmi, 2004]. This prototype is a proposition to users' needs
in digital resources. Among the services proposed are:
- Access to a XML models database, with
the intention to provide a series of models on determined fields with
particular needs. This repository fosters the free exchange of models between
professional users, who stay informed on practices and methods lead by other
users in specific domains. It also represents a collaborative workspace for
assuring the exchange of information.
- Collaborative research and access to a
XML structured database that contains documents described according to the
structure type defined in the XML repository. The graphical interface supports
collaborative work between users and researchers.
The objective of ICRS is to promote
collaborative work between users to integrate a virtual community of
researchers. By means of repositories, users are able to know what others are
doing and to participate in two modes: reader
(by searching and accessing existing models) and author (by creating own models and making them available to
others). Documents are on XML format, which facilitates parsing data from other
systems.
There
are two ways of collaboration in ICRS:
- Collaborator:
consists of making a collaborative research in its strict sense. This mode
considers similar users' profiles to show to them queries and results
reciprocally. It is possible to annotate documents and evaluate its pertinence.
These annotations may be available for other users, as well.
- Observer:
this mode deactivates the onscreen presence of other queries, however user's
researches, results, annotations and comments remain visible to other users.
Observers are meant to be domain experts, persons who would extend their
competencies to students or practitioners.
From the main interface of the system,
users can choose the mode of collaboration, their profile (educational level
and disciplinary field), and, search parameters (keywords, author and text)
(Figure 3). Synchronic collaboration is also supported from this interface;
users select contacts from a list to send instant messages and to display
others' researches sorted by search parameters and the number of results
gathered. This functionality allows users interested in the same subject to
click right into other's query to get the same results, avoiding to redo
operations.
Fig.
3. ICRS: Collaborative research
interface.
Results of queries are listed as
hyperlinks and ordered by pertinence. They show the title of document and
another collaborative interface is launched if clicked.
Annotations and evaluations are two
elements that can be easily added to digital documents, this is of great
importance when dealing with collaborative environments. Annotations are
normally regarded as a technique for appending content (so to share ideas) to a
document; on the other hand, evaluations are important indications on the way
the author qualifies the document. ICRS can display annotations found in
documents and allows to add more, as well as evaluate the content (figure 4).
The function of annotations is to emphasize certain parts of the document by
underlining them with different colors, depending on the theme. ICRS propose
two kinds:
- Public
annotations: a user highlights some parts that will be visible for all
users. These annotations may be in form of comments, of quality judgments or
any other that reflects the content of the document.
- Private annotations: only the
author-reader can see them. They are in form of remark points and/or
indications that researchers would work later.
For
evaluations, the interface gives a note scale to evaluate the current document,
regarding its pertinence and consistency with the query passed. Notes go from 1
(not pertinent) to 5 (very pertinent).
4.2. Introducing HTT Online: creation of
learning materials
HTT Online stands for HyperTectol Online, which is the offline
system presented in 2004 as an assistant tool for creating distance electronic
courses [Reyes, 2004]. However, one of the major constraints was its dependency
to Macintosh platform and operating system Mac OS X 10.2. HTT Online represents
a sort of evolution of HyperTectol and is meant to assist the creation of
Learning Objects in form of pedagogical digital documents and to distribute
them as content packages. HTT consists of an interface that allows authoring
content direct from a web browser (Gecko browsers suggested), which makes it
platform-independent. In addition, one of the main improvements from the
offline version is that HTT Online outputs documents defined in XML format,
which allows a semantic treatment of information and facilitates its transformation.
HTT has been conceived for distance
education environments where faculty members have experience with learning
platforms and need assistant tools for creating their own learning material.
From this point of view, our system considers two main kinds of users: teachers
and learners, in other words, authors
and readers.
The creation of digital documents is
based on a window management system, where each window is presented inside the
browser (simple HTML layers controlled with JavaScript). In this sense, authors may load several
windows at the same time for reading previously released content, authoring new
documents, or linking between documents. The text-based content of windows is
added and edited through a word processor-like tool, which defines the element
markups in semi-automatic way (e.g. the author selects headers to select if
they head a document, a section or a paragraph) (figure 5)
During the creation it is possible to
include media assets: images, audio, video and animations. Asset files are
stocked inside the final content package unless they are described as URL
addresses. Moreover, any window is capable of loading URLs, making them to
appear as virtual windows.
Every document created with HTT is output
in XML format and follows the structure proposed by A. Mkadmi [Mkadmi, 2004]
for defining scientific documents and technical reports. The model regards four
parts in a document: metadata, content of the document, footnotes, and
bibliography references. Anyhow, our current version focuses on the content
definition (figure 6).
HTT Online offers a metadata editor in
compliance with the IEEE LOM standard. The editor is represented through HTML
forms in two flavors: shortcut (most used elements) or extended (almost eighty
elements to be filled in are presented). This interface (figure 6) generates
two files: the xmlmanifest.xml and
the start.html, both files located at
the folder's top level.
Readers
(students) access the system in read-only mode. They are able to search for
learning objects and verify metadata upon the following parameters: name of the
author (instructor), title of document, subject, date, id entry, and type of
media asset included.
Finally,
the start.html file allows readers
and authors to have a quick view of the content of the learning object, based
on the metadata manifest. Furthermore, access to learning material can be
directly done from here. For all documents found, users are able to choose from
a series of stylesheets in order to display the document according to his/her
characteristics.
4.3. Integration of ICRS with HTT Online
The employ of the systems presented above
implies a first step towards semantic pedagogical learning materials, where XML
technologies combined with standards and norms are the cornerstone for their
development. In this part we analyze the integration of ICRS and HTT Online
from a collaborative work in distance education environments perspective.
So far, one of the major projects on
creating educational elements and attributes in XML format for education is the
Educational Modeling Language (EML), developed at the Open University of
Netherlands in 1998. The EML is a notational method for e-learning environments
based on a pedagogical meta-model and since 2000 it has been used to describe
courses [Koper, 2003]. On February 2003, when the final specification of IMS
Learning Design (LD) was approved, EML has not been maintained any longer.
Consequently, IMS LD is based on ELM. Its
objective is to allow authors to create various types
of educational designs, following a consistent notation, that can be
implemented homogeneously in several courses or learning contexts. The
implementation is made through three levels:
- Level A: contains the vocabulary to support pedagogical diversity,
- Level B: adds attributes and conditions to level A (this allows
personalization and more elaborated interaction sequence based on students'
portfolios), and
- Level C: adds notifications to Level B.
In spite of this, disadvantages have been
related to its recent implementation, i.e. it does not consider
student characteristics or student evaluation, which requires design with other
IMS specifications (e.g. IMS Learner Information Package Specification, IMS
Question and Test Interoperability), and does not give guidance to define
educational contents.
As it was noted before, the system ICRS
offers to instructors and domain experts an interface to work together to
construct XML models for documents. In addition, every author can closely
observe the work of others with the intention to make improvements on his own
research. The importance of such system in e-learning environments is to
promote a channel for collaborative work to exchange document models that could
be used in specific situations. For digital documents the model is the most
important part; syntax can be transformed anytime
with XSLT to suit specifications, but models need human interaction with
serious consideration for future transportability.
Once a document is found or
annotated in ICRS, its treatment in HTT Online enhances the content with media
assets, learning standard metadata and delivered mode in form of content
package. In the same way, importing digital documents created in HTT into ICRS
is achieved transparently, according to the Mkadmi model.
In order to benefit from semantic
documents in e-learning, it is fundamental to promote the use of document
models in faculty members as an approach to normalization. Collaborative work
as a technique for granting interchange may promote greater advancements not
only in local situations (e.g. laboratories, universities), but also in global
circumstances (e.g. countries, consortiums).
5. Conclusions and perspectives
Contemporary approaches on information
and communication technologies (ICT) and their implication on knowledge domains
reveal needs to improve our manners for treating information. The possibilities
of XML technologies and hypermedia allow the development of new systems to
distribute knowledge in local or global fashion. This perspective enhances the
role of authors who have now access to open channels for sharing and working
with other authors.
From this context we have presented in
this paper the integration of two software applications, developed at the Laboratoire Paragraphe of the University
of Paris VIII, to promote collaborative work based on reflections on
normalization of technology for modeling and assisting the creation of learning
materials: ICRS and HTT Online. By integrating both tools, we propose a
hypermedia perspective to simplify the presentation and interaction of actors
with digital documents.
In one hand, ICRS proposes an interface
for collaborative research of information. It consists of two main services:
access to a XML models database (repertory for free exchange of models between
professional users) and access to a XML structured database that contains
documents described according to the structure type defined in the XML
repertory. ICRS also supports annotations and evaluation to the documents
found, which aggregate value to documents.
In the other hand, HTT Online is an
online tool for creating learning objects in form of pedagogical digital
documents and to distribute them as content packages. HTT consists of an
interface that allows authoring content direct from a web browser. In addition,
HTT Online outputs documents defined in XML format, which allows a semantic
treatment of information and facilitates its transformation, it also takes into
account IEEE LOM for metadata definition and IMS CP for content packaging.
The integration of both systems has the
intention to be a first step towards normalization in e-learning in collaborative
work environments, where pedagogical digital documents are meant to have their
own educational structure described in XML format. The main advantage of such
paradigm is to render documents understandable both by humans and computers.
Currently, we cannot discuss field
experiences because the implementation is in its prototype stage and will fully
start the next academic year at the Département
Hypermédias of the University of Paris VIII; however, further work in our
research is concerned with a transparent compatibility between systems (e.g.
importing document models from ICRS into HTT Online) and with document models
in the e-learning domain.
6. References
[ARIADNE, 2000] ARIADNE (2000). Alliance of remote instructional authoring and distribution networks
for Europe [On-line] Available: http://ariadne.unil.ch/
[Chartron, 2002] Chartron, G., Guyot, B., Lafouge, T., et
al. (2002). Le document numérique : un
objet fédérateur de recherche en sciences de l'information. Documentaliste Sciences de l'information, 39 (6), December 2002.
[ELC, 2003]
E-learning Consortium (2003). Making
sense of learning specifications & standards: A decision maker’s guide to
their adoption [Electronic Version] Saratoga Springs, NY: The Masie Center.
[IEEE, 2000] IEEE LOM (2000). LOM working
draft v4.1
[Electronic Version]. Available: http://ltsc.ieee.org/doc/wg12/LOMv4.1.htm
[IMS CP,
2003] IMS CP (2003). IMS Content
Packaging specification v1.1.3. [Electronic Version] Available: http://www.imsglobal.org/content/packaging
[Kenley,
2000] Kenley, R. (2000). Pedagogical
objects. Case studies towards a different perspective on collaborative
education. [Electronic Version] Paris: École d'Architecture.
Available: http://www.elia-artschools.org/ejhae/Issue0/Kenley.pdf
[Koper, 2001]
Koper, R. (2001). Modelling units of
study from a pedagogical perspective.
The pedagogical meta-model behind EML. Available: http://eml.ou.nl/introduction/docs/pedmetamodel.pdf
[Cruzel,
2004] Cruzel, S.L. (2004). Documents, ressources, données : les avatars des
documents numériques. Information-Interaction-Intelligence, n°
1, 2004.
[Le
Crosnier, 1997] Le Crosnier, H. (1997). Les
bibliothèques numériques. October 1997.
[MERLOT, 2000] MERLOT (2000). Multimedia educational resource for learning and on-line teaching
[Electronic Version] Available: http://www.merlot.org/
[Mkadmi,
2004] Mkadmi, A. (2004). Recherche
collaborative d’informations : repenser
l’architecture des SRIs à l’ère numérique. (Doctoral thesis). Paris:
Université Paris VIII.
[Mkadmi,
2003] Mkadmi, A., Bouhaï, N., Langlois, M. (2003). Partager des modèles
XML:Quel intérêt ? BBF, n°5, September 2003.
[Reyes, 2004]
Reyes, E., Saleh, I. (2004). HyperTectol, an assistant and authoring tool for
using multimedia in Learning Objects creation. Proceedings of the 5th. International Conference on Information
Technology Based Higher Education and Training. Istanbul: IEEE.
[SCORM, 2004]
SCORM (2004). SCORM 2004 Overview. Alexandria,
VA: ADL. [Electronic
Version] Available: http://www.adlnet.org
[Wiley, 2001]
Wiley, D. (2001). Connecting learning objects to instructional design theory: A
definition, a metaphor, and a taxonomy. In D. Wiley (ed.), The Instructional Use of Learning Objects. [Electronic Version]
Available: http://reusability.org/read/chapters/wiley.doc