DESIGN AND IMPLEMENTATION OF A DEVICE-INDEPENDENT

PLATFORM FOR MOBILE LEARNING

A Thesis

Presented to

The Faculty of Graduate Studies

of

The University of Guelph

by

RAZIEHNIAZI

In partial fulfillment of requirements

for the degree of

Master of Science

December, 2007

©RaziehNiazi, 2007
1*1 Library and
Archives Canada
Bibliotheque et
Archives Canada

Published Heritage Direction du

Branch Patrimoine de I'edition

395 Wellington Street 395, rue Wellington

Ottawa ON K1A0N4 Ottawa ON K1A0N4
Canada Canada

Your file Votre reference

ISBN: 978-0-494-36552-6
Our file Notre reference
ISBN: 978-0-494-36552-6

NOTICE:
The author has granted a non-
exclusive license allowing Library
and Archives Canada to reproduce,
publish, archive, preserve, conserve,
communicate to the public by
telecommunication or on the Internet,
loan, distribute and sell theses
worldwide, for commercial or non-
commercial purposes, in microform,
paper, electronic and/or any other
formats.
AVIS:
L'auteur a accorde une licence non exclusive
permettant a la Bibliotheque et Archives
Canada de reproduire, publier, archiver,
sauvegarder, conserver, transmettre au public
par telecommunication ou par Nnternet, preter,
distribuer et vendre des theses partout dans
le monde, a des fins commerciales ou autres,
sur support microforme, papier, electronique
et/ou autres formats.

The author retains copyright L'auteur conserve la propriete du droit d'auteur

ownership and moral rights in
this thesis. Neither the thesis
nor substantial extracts from it
may be printed or otherwise
reproduced without the author's
permission.
et des droits moraux qui protege cette these.
Ni la these ni des extraits substantiels de
celle-ci ne doivent etre imprimes ou autrement
reproduits sans son autorisation.

In compliance with the Canadian Conformement a la loi canadienne

Privacy Act some supporting sur la protection de la vie privee,
forms may have been removed quelques formulaires secondaires
from this thesis. ont ete enleves de cette these.

While these forms may be included Bien que ces formulaires

in the document page count, aient inclus dans la pagination,
their removal does not represent il n'y aura aucun contenu manquant.
any loss of content from the
thesis.

Canada
 ABSTRACT

DESIGN AND IMPLEMENTATION OF A DEVICE-INDEPENDENT

PLATFORM FOR MOBILE LEARNING

Razieh Niazi Advisor:

University of Guelph, 2007 Professor Qusay H. Mahmoud

Mobile learning is an emerging trend in educational and training systems. Current

mobile learning systems, however, are impractical due to many challenges. Consider
for example the wide variety of mobile devices, that exist today with different
capabilities and constraints, which has made device-independent delivery of learning
content difficult. In this thesis we present the design and implementation of a device-
independent platform for mobile learning. Some of the novel features of the platform
include: (1) enabling educators to create the educational materials once and generate
content for a variety of devices, including cellular phones, personal digital assistants
(PDAs), smart phones, laptop computers, and PCs; (2) allowing educators to share
materials with others through a knowledge-based repository; and (3) offering learners
on-line as well as off-line access to the materials from any device.
 Acknowledgements

I would like to extend my most sincere gratitude to my advisor Dr. Qusay Mahmoud
for his ideas and excellent guidance throughout the writing process.

I would also like to thank my husband, my beautiful daughter and my parents for their
continued encouragement and support, without which this thesis would most likely
not have been written.

1
Table of Contents

Acknowledgements i
Table of Contents ii
List of Figures v
List of Tables vii
List of Abbreviations viii
Chapter 1 1
Introduction 1
1.1 Forms of Education ; 1
1.2 Mobile Technology and Mobile Learning 3
1.3 Opportunities in Mobile Learning 5
1.3.1 Potential of Mobile Technology for Learning 5
1.3.2 Opportunities in Mobile Learning 7
1.4 Problems Facing Mobile Learning 9
1.4.1 Issues in Wireless Networks ...9
1.4.2 Issues with Mobile Phones 10
1.4.3 Issues in Content Creation 11
1.4.4 No Standards for Mobile Learning Systems 12
1.4.5 Security 12
1.5 Organization of the Thesis 12
Chapter 2 '. 14
Background 14
2.1 Related Work 14
2.1.1 WELCOME 15
2.1.2 MLE 16
2.1.3 Mobile ELDIT 17
2.1.4 MOBILE 18
2.1.5 An Adaptive Mobile Learning System 19

ii
 2.1.6 MobiLP 20
2.2 What these Systems Fail to Address 21
Chapter 3 23
Proposed Mobile Learning Model 23
3.1 Motivation 23
3.2 Proposed Definition of Mobile Learning 23
3.3 Proposed Mobile Learning Engine Model 26
Chapter 4 30
Proposed Platform 30
4.1 Goals and Platform Specifications 30
4.2 Proposed Platform Architecture 32
4.3 XML Metadata 35
4.4 Platform Classification 38
Chapter 5 40
Implementation and Evaluation 40
5.1 Overview 40
5.2 Technical Details 40
5.3 Operating Scenario 42
5.4 Mobile Technologies in our Implementation 52
5.5 Considered Criteria 53
5.6 The Processing Flow 54
5.7 Evaluation 55
5.8 Lessons Learned 57
5.8.1 Customized Environment 57
5.8.2 Text Inputs 59
5.8.3 Content Material Design 60
5.8.4 Wide Spread Use of M-Learning Applications 60
Chapter 6 61
Conclusions and Future Work 61
6.1 Conclusions 61
6.2 Future Work 62
iii
Bibliography 63

IV
List of Figures

Figure 1.1 Potential of mobile devices in mobile learning 7

Figure 1.2 A mobile learning system broken up into 4 objects 9

Figure 3.1 Face-to-face education model 25

Figure 3.2 Proposed m-learning model 25

Figure 3.3 Proposed mobile learning engine model based 26

on the definition of Wayne Hodgins

Figure 3.4 The extended mobile learning engine model 27

Figure 4.1 The architecture of proposed platform 32

Figure 4.2 The proposed platform specification model 39

Figure 5.1 3-tier system architecture 41

Figure 5.2 Operating scenario from the instructor's viewpoint 43

Figure 5.3 Quiz configuration 45

Figure 5.4 Sample quiz question 46

Figure 5.5 The quiz generated for various devices 46

Figure 5.6 An offline quiz on the Blackberry 8800 device simulator 49

Figure 5.7 An offline quiz on a Java-enabled device simulator 50

v
Figure 5.8 An offline quiz on Motorola L7I emulator

Figure 5.9 An offline quiz on Motorola vl 100 emulator

Figure 5.10 A browser-based quiz (WML format) on

Nokia Mobile Browser 4.0 simulator

Figure 5.11 A browser-based quiz (XHTML format) on

Nokia Mobile Browser 4.0 simulator

Figure 5.12 The work flow diagram of learners

Figure 5.13 Evaluation roadmap to examine current

mobile learning systems

Figure 5.14 Online access to learning materials using

native technology

vi
List of Tables

Table 3.1 The relationship between Hodgins's definition 27

and the proposed mobile learning engine model

Table 5.1 Offline quiz formats 48

Table 5.2 Supported markup languages 51

vn
List of Abbreviations

GSM Global System for Mobile communications

GPRS General Packet Radio Service

MMS Multimedia Messaging Service

PDA Personal Digital Assistant

LAN Local Area Network

3G The third generation of mobile phone standards and technologies

WML Wireless Markup Language

WAP Wireless Application Protocol

XML Extensible Markup Language

XSLT Extensible Stylesheet Language Transformation

CLDC Connected Limited Device Configuration

MIDP Mobile Information Device Profile

JSR Java Specification Request

JDBC Java DataBase Connectivity

viii
KXML An XML parser with a small footprint for all Java platforms

XHTML Extensible Hyper Text Markup Language

IrDA InfraRed Data Association

IX
Chapter 1

Introduction

1.1 Forms of Education

Today, there are two forms of education and training. One is conventional education and
the other is "distance education". In conventional education, the learners are assembled in
a class, in a particular place, at a particular time and with a particular curriculum and
taught by a teacher. Unlike conventional education, in distance education there is no
particular class, and no particular time. In this form, teachers and learners can
communicate at their own choice of time by exchanging either printed or electronic
media, or through technology in a real-time communication [33]. In this system, the
teacher is separated from the learner and the learner from the learning group [24]. In fact,
in distance education the focus has always been on the individualization of teaching,
learning and even communication between students. Interestingly, it has been a very
successful education system and proved that other forms of education systems can work
out. Today there are many strategies in education and training which can be called in
various labels such as "distributed learning", "online learning", "Web-based learning",
"e-education" or "e-learning" [24].
The next development of distance education was e-learning. E-learning which
means "the provision of education and training electronically, on the Internet and the
Web" [24], removes boundaries of the classes, saves costs of education and makes
education available at every time and just for the right person. E-learning helps both
instructors and learners to customize and personalize learning content targeted toward
enhanced performance. It enables us to take various electronic courses as much as we
want and we need. With e-learning, we take advantage of the evolving technologies in
Internet including hyper text, video, audio, email, chat, discussion panel and

1
conference system and combine them together, to make a virtual learning
environment which has been very successful in the last decade. The following
statistics presented in [24] demonstrate the impact of e-learning as a major sector in
education and training:
• "There were about one million courses on the Internet, 30,000 of them complying
with a scientific definition of online, 70,000 of these were listed on the
Telecampus portal, with many of them making didactic use of the World Wide
Web"
• "WebCT kernel alone was used by 5,100,000 students in 123,000 courses,
developed by 33,000 university and college faculty at 1,100 institutions in 48
countries"
• "CISCO systems stated that more than half of all technical training will be done
by e-learning by the year 2003."
• "The Irish e-learning company Riverdeep was launched on the New York Nasdaq
exchange in March 2000 for the market capitalization of $1,000,000,000."
These statistics demonstrate that e-learning has been a very successful strategy
throughout the world. However, significant changes have happened in the economic
sector which may demand a new form of education and training system.
Today, we live in service-based economy. Many product companies are
turning into service companies. This has caused a fundamental change in the balance
between skilled and unskilled jobs [6]. In other words, the economy has required
skilled workforce rather than hard physical labours. Thus, in order to cope with the
economic changes, and continue to compete, we need to offer new learning
opportunities for individuals, for people who can benefit from education and training
and for those who do not.
Recently, we have seen significant advancements in mobile technologies and
mobile devices. In fact, we have entered in an era of global mobile communication.
We are in the Mobile age. This provides us a great opportunity for a new
transformation from "electronic" to "mobile" services. As we can see today e-
business is moving to m-business and e-learning to m-learning.

2
 M-learning is becoming the third form of education and training. However,
there are several issues in mobile learning. For instance, there are various definitions
about m-learning. In addition, given the wide varieties of mobile devices available
with different capabilities and constraints, there are many technical challenges in
developing mobile learning systems. These challenges include device-independent
delivery of content, support for a wide variety of mobile devices, and reusability of
educational materials. Hence, in this thesis, we consider these issues. We propose a
mobile learning model and the design and implementation of a system to demonstrate
that it is possible to create a device-independent mobile learning system for educators
and learners that facilitates teaching, learning, and supports a wide variety of mobile
devices.

1.2 Mobile Technology and Mobile Learning

By the end of 2007, it is estimated there will be 3 billion mobile subscribers world
wide [8] whereas the number of PC users worldwide is expected to reach 1 billion by
2010 [9]. This means more people have mobile phones than PCs. People are very
comfortable with their cell phones because of mobility, size and weight. They support
different capabilities and services such as voice, SMS for text messaging, email,
Internet, MMS for sending, receiving photos, videos, and camera for taking pictures.
They can also act as an mp3 player and a mass storage of music, photos, images, and
documents. In other words, mobile phones have progressed as a device for both
communication and entertainment. However, they still have lots of potential for new
additional revolutionary applications and usages. For instance, in Japan [10, 11],
mobile phone companies provide instant emergency earth quake alerts to their
customers. In Sri Lanka, GSM cell phone signals were used to track tsunami survivors
and to identify victims [12]. In Ireland, Catholics now sign up for daily inspirational
text messages from the Pope on their cell phones and Muslims around the world can
use their phones to find the direction of Mecca using a built-in compass [12]. In
pervasive healthcare, it is expected a person wears biological sensors which gather
vital sign data and transmit them to his/her cell phone for further processing, and in

3
the case of life threatening situations, the cell phone generates an alarm and/or calls
an emergency center. All of these changes and evolutions show that we are in the
Mobile age.
Mobile learning utilizes mobile devices into learning. The term mobile device
includes not only cell phones, smart phones and PDAs, but also every portable device
which has mobility and portability like laptops and tablet PCs although there are some
debates on the inclusion of tablet PCs and laptops in mobile learning [28]. Generally,
there are various definitions of mobile learning given in [3], including:

"Mobile Learning is e-leaming through mobile computational devices: Palms,

Windows CE machines, even your digital cell phone."

"The term mobile learning (m-learning) refers to the use of mobile and
handheld IT devices, such as PDAs, mobile phones, laptops and tablet PCs,
in teaching and learning."

"According to software vendors, mobile learning is 'the point at which mobile

computing and e-leaming intersect to produce an anytime, anywhere learning
experience.' Translation: It's the ability to enjoy an educational moment from a
cell phone or personal digital assistant (PDA)."

We can consider these definitions from two aspects: one aspect is the use of mobile
handheld devices in mobile learning, and the other is that mobile learning (m-
learning) is a natural extension of e-learning that goes beyond the e-leaming
experience and makes learning even more available and accessible [25]. Futurist
Wayne Hodgins said, mobile learning provides "just the right content for the right
person in the right place on just the right device at just the right time" [14]. In our
opinion, what Wayne Hodgins proposed as a definition of mobile learning is an
intelligent definition of this term which we follow as a roadmap in this thesis.

4
1.3 Opportunities in Mobile Learning

In order to realize the impact of mobile learning to establish pervasive learning

environments, facilitate and enhance learning, we would take two questions into
account:
1. What is the potential of mobile technology to establish pervasive learning
environments?
2. What opportunities will mobile learning offer?

1.3.1 Potential of Mobile Technology for Learning

Mark Weiser [15, 29], the father of pervasive computing, talked about a third way of
computing, in which technology will be hidden in the background of our daily lives.
This means technology will support us at every time, in every where and we always
have access to it. This definition can be extended for pervasive learning as well. In
pervasive learning, education will be available for all people in all scopes from
toddlers to seniors, even people who are hard-to-reach, people in rural areas with
limited learning resources or people who have been geographically isolated. At this
point, it is important to note the distinction between mobile learning and pervasive
learning. In our opinion, mobile learning can be employed to establish pervasive
learning environments to make learning experience pervasive. However, in order to
establish pervasive learning environments, we take the following items into account:
• We need pervasive communication networks. Recently, we have seen
advancements in wireless LANs like WiFi. But they are only available in
hotspots. However, according to some statistics [8], the number of mobile
users is growing exponentially and this requires the exponential growth of
wireless networks as well. Hence, in the near future the wireless infrastructure
will cover many places. Another phenomenon about to happen is the arrival of
3G technologies. These technologies eliminate many of the encountered
obstacles in bandwidth, speed, connectivity, data transfer rate and create a
great opportunity to establish pervasive environments.

5
 • We need connected devices. Ideally, in pervasive computing and consequently
in pervasive learning we should have access to technology at any time, from
any where. Hence, mobility and portability are two important requirements of
connected devices in the context of pervasive computing. Here we discuss how
mobile devices have lots of potential in establishing pervasive learning
environments:
o Today, mobile devices are light-weight, cheap, small in size, and
portable. They support lots of capabilities and services and can be
always carried by people. In other words, mobile devices have
remarkably improved as a multi purpose device for communication and
entertainment,
o Mobiles are ubiquitous today. So far, there is a high rate of mobile
phone ownership and still increasing,
o Another important aspect to consider is the personal nature of mobile
phones which are used as a device for entertainment, communication
and building relationships with friends. In fact, the young define their
cell phones as a personal space that can give them an opportunity to
communicate with friends, explore new relationships and provide a
sense of freedom. According to the Wireless Kids project [13] in the
UK, most teenagers considered the mobile phone as being an important
part in their lives. In this survey, the word 'need' occurred in over 75
percent of the responses.
Due to the qualifications mentioned above, many educators are optimistic about the
potential of mobile devices to establish pervasive mobile learning environments.
Figure 1.1 depicts this potential:

6
 A Device for Fun &
Ertertainment ADevicefor
Oast (Cheap)
, (lotsof capabilities) Communication

! Fersonal Space
ttn.il I Sa:& Hcih Fbnclralion
| (private)
LMrt-Weiiilit ule(Fferv;iavp)

Connection
Adbvicx. for Alwnys Connect A.-ailabiht/ina
Exploration (Every Time) broad range
(Everywhere)

Multi-Fur pose Fteady Asa Mass

Always Carry Sorage
(All-in-One) Infrastructure

Figure 1.1: Potential of mobile devices in mobile learning

1.3.2 Opportunities in Mobile Learning

In recent years, there has been an explosion in the growth of mobile learning in all
education sectors. Many communities have defined it based on their particular
assessments, backgrounds and experiences. This has increased the variety of views on
mobile learning which has made the identification of the nature characteristics of
mobile learning difficult [16]. Using mobile learning, however, can make many
positive contributions, including the following:
a) Mobile learning can provide the facility to access and review course materials,
labs, assignments and lectures from every where at every time while learners
are waiting for a bus, "on the go", at a coffee time or in the recess time. This
method of learning can be an extension of e-learning.
b) Mobile learning can help learners to get the "right content", at the right time in
the right place. In other words, learners can have intelligent context-aware
learning. The example of this kind of learning can be seen when you go to visit
museums, exhibition or science centers. For instance, currently, there are text

7
 descriptions for the arts in a museum. It would be useful if you can listen to a
voice or audio description about each particular exhibit through your cell
phone just when you arrive at a particular place. According to [30], we can
only remember 10% of what we read whereas we can remember 50% of what
we see and hear. Hence, this kind of learning experience can affect the
learning process. MOBIlearn is an example of this scenario [31, 32].
c) Mobile devices can be employed for collaborative learning. Technically,
mobile learning can provide tools for group discussions, group interactions,
and sharing experiences and knowledge using the evolved technologies in
mobile devices like SMS messaging, phone calls, discussion panels, forums
and other features.
d) Mobile learning can provide tools for interactive learning. The examples of
these kinds of tools are quizzes and games.
e) Mobile learning also promotes lifelong learning. Lifelong learning is defined
as "all learning activity undertaken throughout life, with the aim of improving
knowledge, skills and competences within a personal, civic, social and/or
employment-related perspective" [4]. Job training is an example of this kind of
learning.
f) Because mobile devices are defined as a private space and provide a sense of
freedom, owners focus on the content materials and available resources in the
devices for a longer period of time in their spare time.
g) Mobile learning can have a tremendous impact on coursework activities and
make learning more comprehensive. Perhaps, one potential scenario can be of
a student in Botany. Every week, s/he has to go to Nature, identify and gather
the sample of some particular plants. For this purpose, s/he uses her/his cell
phone to surf the Web, identify samples, and take photos of them. When s/he
comes back to the lab, s/he has lots of information and the vastly relevant
pictures stored on her/his phone.
h) Mobile learning can provide tools for informal learning. Informal learning is
an important aspect of our learning experience that occurs in a variety of ways
such as completion of work-related tasks [2]. Informal learning is defined as

8
 "learning that takes place outside a dedicated learning environment and which
arises from the activities and interests of individuals or groups, but which may
not be recognized as learning" [5].

1.4 Problems Facing Mobile Learning

Although mobile learning can offer many opportunities to facilitate and enhance
learning experiences, there are several challenges and obstacles. In order to identify
these problems, we have broken up a mobile learning system into 4 objects including
wireless networks, mobile devices, users and educational material (See Figure 1.2).

MOBILE LEARNING SYSTEM

Figure 1.2: A mobile learning system broken up into 4 objects

In this way, we are able to identify the major problem areas across each object
clearly. The following sections define problem areas:

1.4.1 Issues in Wireless Networks

In the last decade, we have seen significant advancements in wireless networks and
wireless technologies. However, there are still many challenges in wireless network
environments. One of these major problems is limited bandwidth. For instance, GPRS

9
(General Packet Radio Service) enables users to use high speed transmission from 56
up to 114 kbps and continuous connection to the Internet via their cell phones.
However, in current network implementations, GPRS data transfer speeds are
between 30 and 40 kbps. This transmission rate of data will not be enough for mobile
applications while a number of people using mobile phones as a device for learning
will be increasing. Hence, there is a need for 3G technologies which enable service
providers to offer users a wide range of high quality services in a greater network
capacity.
Another issue in wireless networks is the cost, including the cost of online
connections, MMS and SMS messages. This issue should be considered carefully
from this aspect: people who are using mobile learning technology are mostly
students with limited budget. They can not afford using expensive systems. Hence, a
mobile learning system should be cost effective.
Disconnectivity of wireless networks is another problem. In mobile learning,
learners are defined as objects which are on the move constantly, as suggested in [16]
mobile learning is about "a society on the move". Technically, it is possible to lose
the mobile wireless signals due to congestion, node mobility or signal disruption.
Thus, we need some strategies to make reliable systems that make learning experience
available any time any where.
Lack of location-aware services is another issue in developing an intelligent
mobile learning system. As we will discuss later in Chapter 3, in order to make the
"right content" available for learners, we need contextual information including
location-aware data.

1.4.2 Issues with Mobile Phones

Problems in mobile phones can be considered from two aspects. One is the diversity
of mobile phones. The other is mobile phone constraints. Today, there are thousands
of different brands of mobile phones with different capabilities and constraints not
only in hardware such as memory, screen size, processing power, battery
consumption, resolution, display color and bandwidth limitation; also in software

10
such as operating system and Web browser. Even though, some new generations of
mobile phones like smart phones have been improved in processing power, memory
and bandwidth, they are still restricted in some other aspects such as screen size,
display color and user interface. Hence, from a developer perspective, if we want to
develop mobile applications and specifically mobile learning applications for mobile
devices, they should support all kinds of mobile devices with various capabilities and
constrains. Of course, writing applications for each brand of mobile device is tedious
and requires a lot of effort. In addition, maintenance of these applications itself, is a
big challenge. We need a strategy to overcome these issues, a strategy to create
content once and then reuse it for various target devices.
Battery consumption in mobile phones is another issue. The mobile applications
require more processing power which demands more of the battery on mobile devices.
In order to save battery life on mobile devices we have to make a balance among the
processing power, usability, and performance of the mobile application.
Small screen size is another obstacle. Perhaps, the key question in mobile
learning is .how can we fit learning material onto these small screens without any
impact on learning performance?

1.4.3 Issues in Content Creation

As mentioned earlier, there are various brands of mobile devices with different
capabilities and constraints. The term "mobile device" here refers to any portable
device including smart phones, PDAs, mobile phones, tablet PCs and laptops.
Technically, creating educational materials which can be rendered in all of these
devices is a big challenge. It needs methods and strategies to fit content into an
appropriate format for each device. In addition, it is hard and time consuming for
educators to create materials several times customized for each device. We again
emphasize on this idea that creating educational materials should be done once and
then generated for various target devices automatically.

11
1.4.4 No Standards for Mobile Learning Systems

An addition to the issue in developing mobile learning systems is the lack of

specifications and standards [18]. There is even no framework which can help
developers to develop mobile learning systems and educators to create their
educational materials appropriate for mobile as well as PC platforms. Thus, there is a
need to provide a roadmap for educators, a framework for -developers and identify
systems requirements and specifications.

1.4.5 Security

Generally, one of the major problems in mobile phones has always been the security
issue. Today, mobile phones are used for a mass storage of pictures, documents,
messages and digital records. In fact, mobile phones are defined as a private and
personal space needed to be kept secret. Due to the limited resources in mobile
devices and limitation in processing power and memory, security in mobile devices is
a big challenge. Processing, encrypting and decrypting security data on mobile
devices demand more processing power which can impact the mobile devices
performance and battery life. Technically, more processing power requires more
battery on mobile devices.

1.5 Organization of the Thesis

This thesis is organized as follows:

Chapter 2: Examines the related work in the area of mobile learning.

Chapter 3: Presents our definition, model and framework for mobile learning.

12
Chapter 4: Discusses the architectural design and how our system will address the
problems raised.

Chapter 5: Demonstrates the implementation of our proposed solution to the

problems listed earlier and the evaluation process.

Chapter 6: Concludes the thesis with a conclusion and an outlook at future works.

13
Chapter 2

Background

2.1 Related Work

As discussed in Chapter 1, there are no specifications or standards for mobile learning

systems yet. The reason is that many communities have had their own experiences of
mobile learning systems based on their assessments and backgrounds. This has
increased the variety of views on mobile learning which has made the identification
of the characteristics of mobile learning difficult.
Due to the variety of current mobile learning systems, we consider them from
two perspectives: first, what capabilities and services they support, and second, how
they address the issues discussed in Section 1.4. According to the capabilities and
services, we evaluate them using some of the indicators specified in [18], among
others, as follows:
1. The supported kind of mobile devices: Quinn states that m-learaing solutions
must work for a broad range of devices, not only for a set of devices [23]. We
adopt this specification. Thus, in this critic we specify what kind of mobile
devices these systems support. For instance, some of the current m-learning
systems work only in PDAs or Tablet PCs.
2. Support of e-learning: we consider them whether they support e-learning or
not.
3. Availability of content: whether the educational materials are always available
or not. In other words, whether they support online as well as offline access to
content materials or they support only one method.
4. The type of data they support: For instance, some current systems only support
flash type.

14
In the remaining sections, we discuss the various mobile learning systems that exist
today.

2.1.1 WELCOME

WELCOME (Wireless E-Learning and Communication Environment) [17] is a

project at the University of Regensburg which aims to support both students and
teachers by providing services that facilitate teaching, learning and education- related
administrative tasks. This system has been built based on existing experiences in the
field of e-learning from VUR (Virtual University Regensburg). VUR is the Internet-
based e-learning, widely accepted and used in the university. It is also the
fundamental software architecture for all VUR wireless learning and administrative
services. Selected features of this platform such as news, distribution of learning
materials and personalization are available for mobile devices.
In applying our criteria discussed above (Section 2.1), we evaluate this system in
the following areas:
1. The supported kind of mobile devices: Basically, PDAs are selected as a
device of choice. In attempt to remain platform independent, PDFs, RealVideo
streams and Flash animations are used. These formats are often not fully
supported by all mobile devices or supported with limitations.
2. WELCOME does not support e-learning specifications and standards [18].
3. It is possible to have both online and offline access to Audios and Videos just
on PDAs. In fact, it uses AvantGo, a Web page / file synchronization service
for PDAs and smart phones. AvantGo [19] consists of browser client software
running on the PDA, a synchronization server, and (optionally) an Internet-
connected PC to transfer Web pages into the PDA. Using AvantGo, users are
able to synchronize. In other words, Web pages, information and data can be
stored on their device during synchronization and, later stored pages can be
read using the built-in Web browser without reconnecting to the Internet
(offline).

15
 4. The supported information types are PDFs, Audio, Video, Flash animations
and WML. The WML content format is supported by the WAP protocol.
The mobile learning content materials are mostly generated by adapting third-party
Web sites such as the Cafeteria page and so on. The content are as HTML which can
be rendered on PC without any problem. However, on the mobile devices they must
be displayed in an adapted form. For this purpose, WELCOME converts and compiles
the content. The basic procedure of translating Web content for mobile devices begins
by parsing several HTML pages at regular intervals using PERL programs and
extracting the information contained there in. Then the extracted data is stored in a
relational database. On the other hand, mobile users request WML pages which are
dynamically filled with the data stored in the database.
The downside is that content adaptation requires a long process to parse HTML page,
store in the database, restore data from database and then make WML page. In
addition, in this way the quality of data should be evaluated. The other problem is
maintaining content in the database. Data are extracted from HTML Web pages: a
complicated strategy is needed to keep up to date these data.

2.1.2 MLE

MLE (Mobile Learning Engine) [20] enables students to use cell phones as a medium
for learning. It is used in combination with a learning-platform (MLP) which is an
Internet-based server. MLP is a learning management system (LMS) in the e-learning
field which hosts the available learning objects and offers several functions. MLE
communicates with the learning platform over HTTP and XML. In other words, MLE
and MLP communicate together over HTTP and XML. However, in applying our
criteria, it has the following advantages and drawbacks:
1. It uses a smart phone (a mixture between a mobile phone and a personal digital
assistance) as a device of choice. MLE does not support a broad range of
mobile devices.
2. It is possible to reuse existing learning management systems to support e-
learning.

16
 3. Users are connected to the server (online) and download the learning objects
from the server. If the learning objects have been downloaded before, they can
be used without a new connection to the server (offline).
4. The supported data format is XML.
Learning objects are written in XML format making it possible to present many kinds
of learning content. It needs, however, the author of learning objects to spend time
planning and designing these objects, otherwise the result will be a vast frustration to
students. In fact, the authors of the learning objects are responsible for quality of their
works. They determine the success and the acceptance of the MLE [20].

2.1.3 Mobile ELDIT

Mobile Eldit (m-Eldit) [21] is a system under development at the University of

Trento, Italy. This is a mobile version of an existing online language learning
platform that can be accessed from mobile devices like PDAs and Pocket PCs. In the
Web-based version of this application, there are two types of data: a searchable words
dictionary in German and Italian, plus a set of text in both languages. The texts have
specially been designed to prepare learners for the examination in bilingual
capability. The user can search for a particular word in the dictionary or browse the
texts to study. In the mobile version, however, searching is not supported, because the
raw data of m-Eldit is beyond the capacity of available memory on mobile devices.
Technically, data are in XML file format. The Web server transforms data
from XML into HTML and Java-Script. The learner then can browse the HTML files
by clicking on hyperlinks. However, in applying our criteria it has the following
advantages and disadvantages:
1. It uses some particular mobile devices like PDAs and Pocket PCs. It does not
support a broad range of mobile devices such as cell phones or smart phones.
2. It supports e-learning. In fact, if mobile devices used are laptops, the hardware
and software characteristics are the same as e-learning.
3. M-Eldit supports online access to educational materials. It can connect to the
Internet through GPRS, WAP, WiFi and browse HTML files. It also supports

17
 offline accessibility. For this purpose, m-Eldit uses caching technique and in
the case, when the data is bigger than the device available memory, it uses
Hoarding technique. Hoarding technique is a technique for automating
selection and caching of the data needed by the user in offline periods.
4. This system only supports HTML files plus Java Script. This format of file is
not supported by all mobile devices.
In Mobile Eldit, in order to support content adaptation, context discovery should be
done. For this purpose, m-Eldit adds contextual information including hardware and
software limitations (such as screen size in pixel, color resolution, CPU, supported
language, operating system and etc) to the HTTP request message. In the proposed
solution, however, it is not clear, given the wide varieties of mobile devices, how the
contextual information will be provided.

2.1.4 MOBILE

MOBILE [34] is a MObile-Based Interactive Learning Environment which aids

elementary school English Learning. This system consists of a server and a set of
mobile learning tools which supports in or outdoor learning activities. It is composed
of the following components:
• MMS (Mobile Management System): it involves content management, course
management, dictionaries, and learning records.
• MAS (Mobile Assessment System): it supports online assessments.
• MIS (Mobile Instruction System): it provides services for browsing material,
bulletin board, learning community, chat, quiz, notebook, and FAQ.
• MLDB (Mobile Learning Database): students' learning statuses and records
are stored in this database.
• LCMS (Learning Content Management System): it handles storing and
managing the digitized teaching and assessment materials.

In this system, each student uses a PDA to send and receive learning materials from
the mobile learning system. In addition, teachers are able to download the teaching

18
and assessments materials from LCMS through Internet to the MLDB while
conducting indoor or outdoor activities.
In applying the specified criteria, it has the following advantages and disadvantages:
1. It supports mobile devices including PDAs and notebooks. It does not support
a broad range of mobile devices.
2. No information has been provided to support e-learning
3. It supports online as well as offline access to educational materials. In fact,
educational materials can be downloaded to be used for offline learning.
4. It supports HTML-based educational materials for diverse learning activities.
In this system, the teacher has access to a notebook and each student uses a PDA that
is provided by the school. In fact, MOBILE does not support a wide variety of mobile
devices, and thus does not face many of the obstacles described in Section 1.4.

2.1.5 An Adaptive Mobile Learning System

In [35] an adaptive mobile learning system is proposed. It suggests an architecture

and prototype quiz system based on XML/XSLT technologies. The prototype consists
of five dimensions as follows:
• The content dimension: it represents the actual content containing course
modules, content and multimedia representation of the content (audio, video,
animation and so on) sub-dimension.
• The user model dimension: it contains two sub-dimensions including learning
model and user preference. Learning model dimension includes attributes such
as module completed, weight, score and time taken. User preference contains
attributes like learning style and preferred difficulty level.
• The device dimension: it includes attributes such as the device types, audio,
and video capabilities as well as memory, bandwidth, and operation platform.
• The connectivity dimension: it includes four operating sub-dimensions as
follows:
o A real-time online mode: some attributes like throughput determine
representation of text or multimedia.

19
 o Pre-fetching capability: it specifies how much the pre-fetching depth
should be? It completely depends on the device capability, network
reliability and connection type.
o Offline synchronization mode
o Channel mode: it represents the actual mode of the connection between
the user and the server. For instance, a user can experience a longer
delay in a satellite connection than a cable connection.
• The coordination dimension: it represents the logic and algorithm sub-
dimension of the application.
In the development of the quiz system, the multi-dimension framework has been
applied. This system is capable of adapting to the device and the user profiles. It
works on both PC and mobile Platforms. However, in applying our specified criteria,
it has the following advantages and disadvantages:
1. It supports a variety of mobile devices including hand phones, PDAs and
Pocket PCs.
2. Probably supports e-learning.
3. It is a dynamic quiz system which does not support offline learning.
The current prototype has been tested only for PC users, PDA (iPaq) users, and
mobile phone users with Openwave browser.

2.1.6 MobiLP

MobiLP (Mobile Learning Platform) [22] is a Web-based learning system which

supports access from both mobile and non-mobile devices. It aims to provide
educational materials and communication services to teachers and students any time,
any where. Totally, MobiLP functions are classified into three groups:
• User functions available to students are designed for mobile devices access.
They include the display of Web materials specified by the teachers, online
chat rooms and online quiz systems.

20
 • Teacher functions include input of teacher-specified Web materials, online
quiz management, online chat and user profile management. These functions
can be accessed either by mobile devices or non-mobile devices or both.
• Administrator functions include general system management functions such as
access control and user account management. All of these functions are to be
accessed by non-mobile devices.
This system does not support offline access to educational materials and quizzes. In
addition, it supports browser-based technology and the supported data format is XML.
No information has been provided on customization process of XML data on the
supported browsers.

2.2 What these Systems Fail to Address

As discussed earlier, we examined the current mobile learning systems from different
points of view: The first viewpoint was their capabilities and facilities which they
provided for learners and educators. We evaluated their capabilities based on some of
the indicators specified in [18]. These indicators were as follows: supported mobile
devices, support of e-leaming, availability of content (offline and online access) and
type of data. The second viewpoint was that how these systems challenge with the
identified obstacles in mobile learning (discussed in Chapter 1). As a result of our
investigation on the current mobile learning systems (those mentioned above), we
reached the following conclusions:
In spite of the growing number of mobile learning systems implemented so far,
none of those are fully functional. They have only partially solved the obstacles in
mobile learning. For instance, some of these solutions like WELCOME have only
been implemented for a specific device like PDAs or Pocket PCs. Many of these
systems like MLE and MobiLP use XML to avoid device-dependency of data format
while customization of XML data on different mobile devices with different
resolution is still a problem. In mobile learning, the cost of online connections and
online access to learning materials are still an issue. In addition, mobile learning from
the perspective of an educator has not quite been defined. In fact, mobile learning is

21
not only about learning. It includes teaching as well. Thus, mobile learning systems
should facilitate teaching. In other words, in m-leaming we need innovative tools for
teachers. We need an efficient strategy to create the educational materials for both
PCs and mobile platforms.
In contrast, our platform addresses most of these issues. It supports a wide
variety of mobile devices, including PDAs, smart phones, cell phones, tablet PCs,
laptops and PCs. It supports device-independent delivery of content to various target
devices as well. Also, it customizes content materials by generating learning materials
in midlets or Blackberry API format. In developing the platform, the cost of online
connections is considered as well. For this purpose, the platform generates both
offline as well as online materials. In addition, it supports the reusability of
educational materials. This means learning materials are created once and then the
same content are generated for various device platforms.
Our platform provides tools for educators to facilitate the creation of
educational materials. These tools include a quiz generator and a knowledge-based
database. In addition, it supports e-learning as well as m-learning.

22
Chapter 3

Proposed Mobile Learning Model

3.1 Motivation

Based on our analysis of the current mobile learning systems in Chapter 2, we

identified mobile learning obstacles in the following areas:
• Various definitions of mobile learning.
• No standard and no specification for mobile learning.
• No framework for mobile learning systems.
• Many obstacles (discussed in Chapter 1) in developing mobile learning
systems that have not yet been addressed.
• Many debates on mobile learning from the viewpoint of a learner, but those of
an educator have not been clearly defined.
• Effects of the wireless network limitations such as limited bandwidth and costs
on the usability of the mobile learning systems.
In the following sections, we identify our roadmap in mobile learning and propose our
mobile learning model and framework. In addition, we identify the general
specifications which a mobile learning system should support.

3.2 Proposed Definition of Mobile Learning

We consider the term "mobile learning" in two areas: mobile and learning. The term
"mobile" comes from mobility and portability but it is not quite obvious that it
involves either the mobility of the data (knowledge, information), or the mobility of
the technology or the mobility of the learner. Early definitions [3] of mobile learning
focused on the mobility of technology. However, today there are new perspectives on

23
the term "mobile" in mobile learning. Recently, the focus has been on the mobility of
learners. This means that even if a learner is away from his/her usual learning
environment and uses fixed technology, then learning is considered to be as mobile
because of the mobility of the learner [3]. More recently, Taylor and Vavoula in "A
Theory of Learning for the Mobile Age" [1] did not focus on either the learner or the
technology. Instead, they focused on "the communicative interaction between these to
advance knowing" [1]. We adopt this definition. However, we go one step further. We
propose that in the domain of mobile learning, the term "mobile" refers to the
mobility of knowledge during the communication and interaction between learners
and their technologies, or learners and other learners that this mobility can be cause of
knowing. In addition, we argue that among the technologies, mobile technology has
the potential to make these communications and interactions more effective. But how
does this happen? In order to answer this question, we emphasize on the Wayne
Hodgins's definition [14] again and argue that in the context of mobile learning,
providing a model of learning based on this definition, along with supporting other
learning methods including collaborative learning and interactive learning, can make
learning experiences more effective. In the next section, we propose our model and
identify the general specifications for mobile learning.
As described earlier, in conventional education, also known as face-to-face
education [24], learners are assembled in a particular place, at a particular time and
with a particular curriculum taught by the teacher. We take the conventional system
and divide it up into the following components illustrated in Figure 3.1. This model
demonstrates that in conventional education, learners take "right content"
provided/communicated by teachers with respect to the course's curriculum. The term
"right content" here means the content which are supposed to be taught at a particular
time by the teacher. In conventional education, students are able to use other methods
of learning, including group projects, group studies, labs, assignments, quizzes and
exams.

24
 A knowledge-based teacher according to the
curriculum teaches and presents the "right contents"

Knowledge-
Curriculum based
teacher

Learners take rjfjfn

contents^

Interactive Collaborative
Learning Learning

Figure 3.1: Face-to-face education model

Similarly, we can have the same model for mobile learning. Figure 3.2 shows a
corresponding m-learning model:

knowledge-based data according to the current

context can make the "right contents" available

Knowledge-
based Data

Rght Content:

Interactive Collaborative
Learning Learning

Figure 3.2: Proposed m-learning model

In this model, knowledge-based data plays the role of teacher and contexts are instead
of curriculum. Hence, the right content can be drawn from knowledge-based data with
respect to the current context in a semantic way and learned by learners. In addition,

25
there are many evolved technologies such as chat, conference rooms, forums,
discussion panels and many others which can support interactive learning as well as
collaborative learning.

3.3 Proposed Mobile Learning Engine Model

According to the definition of Wayne Hodgins and our proposed m-learning model,
we propose a mobile learning engine model as follows (See Figure 3.3):

Device Adaptive
Model
V J
- A - " - -

Mobile Context-Aware
U9er Model u~ ^-*|
Learning Model .
...y . -_ - ' • •

Engine

Content Adaptive
Model

Figure 3.3: Proposed mobile learning engine model based

on the definition of Wayne Hodgins

This engine model can provide the right content, for the right person, at the right time,
in the right place and on the right device.
The following table shows the relationship between Wayne Hodgins's definition and
the proposed learning engine model:

26
 Table 3.1: The relationship between Hodgins's definition
and the proposed mobile learning engine model
Definition
Right Content
.?^£gH9RH
Content Adaptive Model
Right Person User Model
Right Device Device Adaptive Model
Right Time Context-Aware Model
Right Place Context-Aware Model

In the proposed learning engine model, we identified 4 major components, including a

device-adaptive model, a user model, a content-adaptive model and a context-
awareness model. We extended the model and extracted the other elements
encapsulated in these components. Figure 3.4 shows the extended model of the mobile
learning engine:

Figure 3.4: The extended mobile learning engine model

27
In what follows, we descript each component and the related elements:
• Context-aware model: Schilit [36] refers to contexts as "any information that
can be used to characterize the situation of an entity where an entity can be a
person, place and a physical or computational object". In our model, we use
this definition but narrow it as follows:
Context in the domain of mobile learning is referred to as any
information that can be used to characterize the situation which can
affect extracting "right content" from the knowledge-based learning
resources.
This definition explains the importance of contextual information in the
effectiveness of m-learning because productivity of mobile learning heavily
relies on providing "right content" for learners as discussed before. In the
extended model, we specify some of this contextual information, including
time, location, user profile, user preference, user leaning style and curriculum.
(Curriculum is applicable for mobile learning applications that support
learning under an organized curriculum). It is important to note that contextual
entities are not only limited to those mentioned above.
• Content adaptive model: it extracts the "right content" from knowledge-
based learning resources and repositories available for this model. For this
purpose, it uses a semantic search to match data with the specified criteria and
extract the knowledge-based content. The criteria are determined by either the
contextual information obtained in the context adaptive model or learner's
keywords or both.
• Device adaptive model: while the context-aware model determines what
materials should be extracted from knowledge-based repositories, this model
specifies how these content materials should be rendered on the mobile
devices. For instance, it determines whether they should be displayed in text or
multimedia format? What resolution should they have? What layout will they
use to display? The answer to these questions fully depends on the device's

28
 capabilities, networks specification and their constraints as determined in this
model as well.
• User model: peers, study groups, mentors and tutors are defined in this model.
The model specifies their roles, duties and the interactions between them as
well. In addition, it is responsible for user authentication and authorization.
Technically, there are various tools and services which can be used in this
model as follows:
- Collaborative tools
- Interactive tools
Synchronous services
- Streaming tools
- Administrative services

29
Chapter 4

Proposed Platform

4.1 Goals and Platform Specifications

Our approach is the integration of formal education into our daily lives. In other
words, we want to design and develop a mobile learning platform which supports both
learners and educators. Using this platform, learners are able to manage their courses,
take quizzes, access course materials, labs, assignments and view their scores,
messages and notifications from any target device. In addition, our platform provides
tools for educators to create their educational materials for both PC platform and
mobile device platforms. Technically, we provide some strategies for device-
independent delivery of the educational materials. Target devices in our platform
include cell phones, smart phones, PDAs, tablet PCs, laptops and PCs.
Basically, we consider our platform with the following characteristics:
• Creating educational materials We believe that creating educational
materials including course materials, quizzes, and labs should be done once by
the instructor. Then the same content should be transformed into the various
target devices in the compatible formats just by clicking a button.
• Delivery of materials Our platform supports both offline and online learning.
This means our platform is able to deliver educational materials online as well
as offline. In the case of online learning, delivery of materials to mobile
devices is done in XML format which is a standard and universal data format.
On the other hand, for offline learning there are some challenges, including the
variety of cell phones. Students have different brands of cell phones with
different capabilities and constraints. Our platform is able to support this
variety. In other word, our platform provides device-independent delivery.

30
 In both cases of offline and online learning, it is necessary to use the
minimum bandwidth as well as the minimum memory usage that consequently
leads to less power consumption. Hence, in design and development of our
platform, we have considered these three factors to keep them at a minimum.
• Tracking and evaluating students' progress This part provides some
facilities to give feedback to student's works such as displaying correct
answers when a student submits his/her quiz, analyzing, grading the quiz, and
sending the result back to the instructor (optional). Also, students are able to
try a quiz "n" time depending on the trial number defined by an instructor.
• Learning Management System (LMS) LMS is a Web-based learning
management system which allows all educators to access, assemble, package
and redistribute educational materials including content, labs information,
quizzes and assignments. Also, it enables students to manage their courses,
access the content and educational materials generated by instructors.
In addition, a mobile version of LMS (mobiLMS) available for online
learning is supported in our platform. This mobiLMS which is deployed on
mobile phone devices enables learners to connect to our server and uses
educational materials online.
• Making educational materials knowledge-based It offers instructors the
opportunity to share their quizzes with other instructors. When instructors
create their learning materials like quizzes, they have a chance to make their
quizzes knowledge-based. If they take this opportunity; their quizzes will be
saved in a knowledge-based database. This capability gives other instructors
the chance to create their quizzes from this knowledge-based database
randomly with respect to their keywords and topics searched, although they
still have a chance to provide their quizzes from scratch.
• Supporting Learning Metadata All educational materials are saved as XML
files managed by the file management. Also, the platform has the capability to
keep track of courses, content materials, labs, assignments, quizzes and their
relationships for each instructor as metadata in the XML format.

31
 • Content distribution capability This capability offers instructors the ability
to create their educational materials like lecture notes, labs, quizzes once, then
generate and distribute them in different formats including XML, HTML,
XHTML, WML, J2ME Midlet for Java-enabled devices and Blackberry API
code for Blackberry devices.

4.2 Proposed Platform Architecture

We divide our proposed architecture into two levels (Mobile Learning Management
System and Electronic Learning Management System) and 4 different layers (storage
layer, management layer, logic layer and presentation layer). Figure 4.1 shows the
architecture of our proposed platform:
mobiLMS

eLMS

,_ Metadata for„ .Quizzes Contents

User Knowledge- instructor in in XML in XML
profiles based DB XMLformat format format

Figure 4.1: The architecture of the proposed platform

32
 mobiLMS is a platform to support online mobile learning. Its responsibility is to
manage courses, discover educational materials for each course, display scores,
messages, notifications, grade quizzes, and render educational materials adaptive to
mobile devices. Technically, we summarize the mobiLMS into the following units:
• Mobile Learning Content Management: it is used to manage courses.
• Mobile Learning Content Presentation: it is responsible for presenting
educational materials adaptive to mobile devices.
• Context Discovery: its responsibility is to discover the context including the
available courses, lectures and quizzes.
• Coursework Analysis: it analyzes coursework, assigns grades, displays
correct answers, and sends results to instructors.

eLMS is a Web-based Learning Management System categorized into the 4

following layers:
• Presentation Layer: this layer is responsible for converting generated
learning materials to the Web browser adaptive form. In this layer, we use
stylesheet, XSL, to convert these materials in the XML format (generated in
the logic layer) to the HTML.

• Logic Layer: this layer mostly provides and performs functions to generate,
compile, pre-verify, pack and deliver educational materials in different formats
including XML, WML, XHTML, J2ME Midlet and Blackberry API code.
Technically, the functions in this layer are available to instructors. The
folio wings describe these functions:
o Content Generator: it has the responsibility for converting lectures
provided in Microsoft Power Point into the WML, XHTML, XML,
J2ME Midlet, and Blackberry API compatible source code.
o Quiz Generator: it has the responsibility for converting quiz instructor-
specified inputs into the WML, XHTML, XML, J2ME Midlet, and
Blackberry API compatible source code.

33
 o Compile Engine: it has the responsibility for compiling Java source
codes to Java bytecode class files. Our architecture supports two kinds
of compiler; J2ME compiler and Blackberry API compiler.
o Obfuscator function: it improves the size, performance and security of
J2ME mobile applications.
o Pre-verification function: in order to enable J2ME bytecode class files
to be run on mobile phone devices, they should be valid bytecodes. For
this purpose, pre-verification function must be applied. Preverification
function converts J2ME bytecodes to valid bytecodes. This function
only acts on J2ME CLDC application bytecodes.
o Packing function: in order the generated codes be able to run on mobile
phone devices, it is necessary to pack bytecodes and the relevant
information to a special format such as jar or cod files.
o Delivery function: as mentioned earlier, it is responsible for delivering
educational materials in different formats on mobile devices.
o Semantic Search: it is a tool to query the Knowledge-Based quiz
database. It is responsible for searching and finding the relevant
quizzes which are matched with entered user's keywords.

• Management Layer: management layer is responsible for managing courses,

users, and files. This layer provides and performs functions available to both
educators and learners. The following shows a brief description of each
function:
o Authorization: it is used to authenticate and authorize users,
o Course Management: it is used to mange courses for both instructors
and learners. It enables students and teachers to add, delete and edit
courses,
o File Management: it provides facility to manage copy and delete
generated files into the user directories and the relevant subdirectories.

34
 o Learner Tracking: instructors are able to keep track of students'
progress by receiving students' coursework feedback and quiz result
through email or SMS.
• Storage Layer: our platform includes tables for user profiles, students,
instructors, courses and other relevant information. In addition, it includes the
other table for knowledge-based quizzes. There are other repository formats
other than tables like XML files and metadata as well. Metadata are used to
keep track of instructor's activity. The following explains a brief description
of each repository:
o Knowledge-bases Database explained earlier.
o User Profiles hold information about learners' profiles, instructors'
profiles, courses, instructors' courses, and students' courses,
o Metadata in an XML data format, keep track of quizzes, courses and
their relationships for each instructor. They are stored in the
instructor's directory managed by the file management,
o Educational materials and quizzes are stored in XML data formats.

4.3 XML Metadata

In the proposed platform, we have two kinds of XML data: one is quiz XML data in
which the quizzes are stored and the other is metadata holding information about
courses, and the relevant quizzes for each course. The following sections demonstrate
the schema format along with an XML example for these XML files:

• Quiz XML data: each generated quiz is stored in an XML data format.
DTD schema for the XML data is as follows:

<? Xml version="1.0" encoding="UTF-8"?>

<! ELEMENT QuizSet (CourseCode, Title, quizKey, Dates, QType, Trial,
CalcGrade, TimeLimit, FeedBack, Quiz*) >
<! ELEMENT Dates (Generated, Updated) >

35
 ELEMENT FeedBack (Type, Destination) >
ELEMENT Quiz (Question, Option*, Correct) >
ELEMENT CourseCode (#PCDATA) >
ELEMENT Title (#PCDATA) >
ELEMENT quizKey (#PCDATA) >
ELEMENT QType (#PCDATA) >
ELEMENT Trial (#PCDATA) >
ELEMENT CalcGrade (#PCDATA) >
ELEMENT TimeLimit (#PCDATA) >
ELEMENT Generated (#PCDATA) >
ELEMENT Updated (#PCDATA) >
ELEMENT Type (#PCDATA) >
ELEMENT Destination (#PCDATA) >
ELEMENT Question (#PCDATA) >
ELEMENT Option (#PCDATA) >
ELEMENT Correct (#PCDATA) >

The following is an example of the DTD schema:

<?xml version = "1.0" encoding="UTF-8" ?>

<QuizSet>
< CourseCode> CIS 6000</CourseCode>
<Title>Web Service Introduction</Title>
<quizKey></quizKey>
<Dates>
<Generated>Thu Sep 20 22:50:43 EDT2007</Generated>
<Updated>Sun Sep 23 21:16:44 EDT 2007 </Updated>
</Dates>
<QType>mChoice</QType>
<Trial>2</Trial>
< CalcGrade>yes</CalcGrade>
< Tim eL imit>15 </TimeL im it>
<FeedBack>
< Type>fb_email</Type>
<Destination>cmel@uoguelph.ca</Destination>

36
 </FeedBack>
<Quiz>
<Question>What is WSDL?</Question>
<Option>WSDL is an XML-based language for defining Web
services. </Option>
<Option>A service directory </Option>
<Option>A service discovery</Option>
<Option>none</Option>
<Correct> K/Correct>
</Quiz>
<Quiz>
<Question>What is UDDI?</Question>
<Option>UDDI can be queried for specific Web Services</Option>
<Option>It is like a yellow page for web service</Option>
<Option> Universal Description, Discovery, and Integration. </Option>
<Option>All</Option>
< Correct>4</Correct>
</Quiz>

</QuizSet>

• Metadata in XML data format keep track of the courses and the relevant
quizzes for each instructor. The schema format for XML (DTD) is as
follows:

<? Xml version=" 1.0" encoding="UTF-8"?>

<! ELEMENT instructor (course*)>
<! ELEMENT course (name, quiz*)>
<! ELEMENT quiz (title, filename, xmlfile, GenDate)>
<! ELEMENT name (#PCDATA)>
<! ELEMENT title (#PCDATA)>
<! ELEMENT filename (#PCDATA)>
<! ELEMENT xmlfile (#PCDATA)>
<! ELEMENT GenDate (#PCDATA)>

37
 The following is the xml file example of this DTD schema:

<?xml version="1.0" encoding="UTF-8" ?>

<instructor>
<course>
<name>cis6000</name>
<quiz>
<title>Chapter 1- Secure Web Servers</title>
<filename>http://localhost:8001/ws/bob/skyList.htm</filename>
<xmlfile>sws</xmlfile>
<GenDate>Thu Aug 09 14:58:02 EDT2007</GenDate>
</quiz>
<quiz>
<title>chl-pioneer•</title>
<Jilename>http://localhost:8001/ws/bob/historyList.htm</filename>
<xmlfile>history</xmlfile>
<GenDate>Thu Aug 09 14:59:40 EDT 2007</GenDate>
</quiz>
</course>
<course>

</course>
</instructor>

4.4 Platform Classification

Georgieva, et al [18] proposed a general classification for the existing mobile learning
systems. This classification takes into account the differences in the implementation
of the existing mobile learning systems as well as their common characteristics. We
model our platform specifications based on this classification (See Figure 4.2).

38
 CPI! Phonos
Smart Phones L I <)ut a: of,
Lf:arn«.'is
PDAs.
! Laptops
I Tablet PCs Supported
Supported Users H
Target Devices Off-Li ne
1 PCs
On-Line
Supported

'n
Supported Access Met hods
Cbmmunication
Tedinologies I non-line
learning, supported
both technologies
GPKS .
GW
Supported Learning! <--V
%st em types Browser-Based
llluftoiith Native
IrEJA r;Lodrmnp,

Figure 4.2: The proposed platform specification model

As seen in Figure 4.2, our platform specifications can be defined in five categories:
• Supported target devices: It includes cell phones, smart phones, PDAs,
laptops, tablet PCs and PCs.
• Supported connectivity: It includes GPRS, GSM, WiFi, Bluetooth, and IrDA.
• Supported users: It supports both learners and educators.
• Supported access methods to educational materials: Our platform supports
offline as well as online access to educational materials. In other words, we
have both offline and online learning. In online learning, we support both
browser-based and native technology. In browser-based, applications are
developed in a markup language such as HTML, and XHTML. In our
platform, we used HTML, XHTML and WML markup languages. In Native
technology, we compile applications which can be run in a runtime
environment. The example of these kinds of applications in mobile devices is
midlets or games which can be loaded on the mobile devices and run. The
reason to support both models is because of customization which will be
explained completely in Chapter 5.
• Supported e-learning as well as m-learning system.

39
Chapter 5

Implementation and Evaluation

5.1 Overview

In this chapter, we will discuss technical details about our implementation. Then we
will demonstrate our system by an operating scenario. Indeed, it will be considered
from two perspectives: the instructor and the learner. In the rest of this chapter, we
will illustrate quiz processing flow and evaluate our system from different aspects
including system capabilities, constraints and considered criteria in the
implementation.

5.2 Technical Details

The architecture of proof of concept implementation (prototype) is organized as a 3-

tier application system (See Figure 5.1). It includes:
- Mobile phone devices and PC's users as the client tier (PC's users themselves are
categorized into two groups: educators and learners)
Servlets which support LMS functions as the logic tier
- Databases, metadata and XML files as the data tier

40
 Educators
PC Platform

A
Learners
PC Platform
}
J
!

Learners >
Mtotofle Platform \
i
•i

Client Her ! Sirvlet as Lojpc Tier Matter

Figure 5.1: 3-tier system architecture

In the client tier, we have two kinds of clients including PC clients and mobile
clients. As you can see in Figure 5.1, PC clients talk to the server through HTTP
protocol. This means PC clients (end users), either learners or educators, send their
requests to the server through a HTTP request message. The server receives the
requests and sends the response back as a HTTP response message. In order to
implement this Web-based system, we used J2EE (Java 2 Enterprise Edition) and took
benefits of the evolving technologies including Web-tier components, Java servlets,
JSP pages (Java Server Pages), JavaMail APIs, and JAXP (Java API for XML
Processing).

On the other hand, for mobile clients who are learners, we implemented two
methods to access learning materials, online and offline. In online, we used XML and
HTTP protocol. This means mobile clients request for content materials in XML
format over HTTP protocol. For offline learning, we generated educational materials
for various platforms. Indeed, there are several mobile platforms. Attewell
categorized the possible mobile platform options: Pocket PC, Windows CE, Symbian,
Palm OS, J2ME, Pogo and Blackberry [26].
However, in the current version, we only generated offline learning materials for two
kinds of platforms including:

41
 • J2ME covers wide variety of mobile devices due to the following reasons:
o Today, most of the mobile devices are java-enabled. According to the
statistics, in 2005 more than 7 out of 10 handsets were shipped with
Java [27].
o Java is a device-independent platform.
• Blackberry: we extended our implementation to support Blackberry API as
well. It was because of the following reasons:
o To demonstrate the capabilities of the particular platform,
o To customize the learning objects and demonstrate a better
performance.

In developing the J2ME application, we used JSR 118 for MIDP 2.0, JSR139 for
CLDC1.1, JSR 172 for XML parsing (KXML), JSR 30 for CLDCL0 and JSR 37 for
MIDP 1.0.
In the logic tier, we implemented the servlet using J2EE. In the data tier we used
MYSQL as the back-end tier and JDBC as an interface between logic and data tier.

5.3 Operating Scenario

In this section, we demonstrate the operating scenarios from two perspectives:

learners as well as educators. Our implemented system is able to generate quizzes for
a variety of devices.
As Figure 5.2 shows, the operating scenario from the perspective of an educator
can briefly be categorized in 4 actions as follows:
• Creating educational materials including quizzes
• Generating quizzes for a variety of devices
• Tracking learners
• Storing/Retrieving quizzes to/from the knowledge-based database.

42
These actions can give a basic idea on how the system works from the instructor point
of view. If we want to elaborate on it, technically we have to break up our system into
the 3- tier applications as we already saw in Figure 5.1.

Figure 5.2: Operating scenario from the instructor's viewpoint

We start from the client tier. Because we have two kinds of clients, we have to
demonstrate the operating scenarios in different groups as follows.
• Instructors group In order to illustrate how our system works, we follow an
instructor named Bob who uses our platform to generate his quizzes. First of all,
he logs in to the system and chooses the course for which he wants to design a
quiz. If the course does not exist, he can add it to his profile. Bob has three
options to generate a quiz from scratch, the knowledge-based quiz DB, or an
existing quiz XML file (in order to update quiz). In the following, you can see a
description of each:
o Generating a quiz from scratch: In order to generate a quiz from scratch,
Bob has to set up some configurations. These configurations include:

43
- Course ID: it includes the course code.
Question Type: it determines the type of the quiz. It can be either
"multiple-choice" or "true-false". For future works, more types will be
supported including "fill-in-the-blank", "multi-answer" and "essay".
- Title: it defines the title of the quiz.
Quiz File Name: it specifies the name of the quiz file in which the
generated quiz will be saved.
- Target Devices: it determines the mobile target devices for which Bob
wants to generate quizzes. For instance, whether Bob wants to generate
quiz for Blackberry devices or J2ME supported devices or both.
Technically, with selecting these options, system is able to generate codes
to support offline learning.
- How many Trial(s): it identifies the number of trials that a student is able
to submit the quiz.
- Time Limit: it determines whether the quiz has limited time. If so, how
many minutes will it be?
- Feedback methods: it determines the way the results and students'
feedback should be sent to the instructor. It can be sent by e-mail or SMS.
- Number of Question: it specifies the number of quiz questions.
- Calculate Grade: it determines whether the result should be sent back to
the instructor or not. Sometimes instructors do not care about results. They
just want students to practice. If the instructor chooses this option and the
learner has several trials to submit the quiz, the last trial's result will be
sent to the instructor only.
- Knowledge-based Quiz: This option offers instructors an opportunity to
share their quizzes.

Figure 5.3 illustrates these configurations.

44
 <.'ourv« lt>: < . . •>" i

i
<ji.»tion type: wiA*<r t. .-« »

IW»». An I'-tt-.i i i i i r h l . i '

N i r q r t l>ev.re: "J; K i * J r « r i " » » i »

• : . T- • "i
Hum MJily Tliat(v",-
" •: •• f
•r
' • " • * Limit: : ,i- • rwmv*
" MP
. »• • . i . .

I » * # k Wflt'Olh. •• (-, I- f-*f . -Tn .,.,W>i-

• • s _ -- «-j^ • -
Munibi r of Qimtlou v 1'

• • , IL

<"iiliidii'• <ir.i»|r: .' , » ,

' if.
T i i -
i ' A . . .
Knov»tei!<|C t l i t e : """ ' * - " ' * * « ' """*" t l * mTrW-'ll

* i - : ' !*• wtfr- • • • . • * \

Figure 5.3: Quiz configuration

Since Bob configured the quiz, he is then able to enter the quiz questions
and their answers. Plus, he should select the correct answer for each
question as it is shown in Figure 5.4.

Vhmz. i s 8SM.7

1)

i * • -• -.
A
) ® WSDL is an XML-based languageforjggnmo Web services.
B) O A service diractoy
C) C Asaivics discoveiy
D) O none

45
 What i s ODDI?
2)

•• i . ..- . Tr*- •--<• ••- --.•••.- . „ •_•%-,

• . . . . . ' •' • V ' • • • • " . • » :• •" •'>" '**
" ' " *• ' •--" "• : - : V *#.••.!:•-•*.•v»^|

A ) Q ;UDDi can be queried for speoficWeb Services '

U ) Q It is like oyellow page for web service

' ^ ) O Universal Description. Discovery anrilnteqrolion

U ) © Ail

Figure 5.4: Sample quiz question

In the final step, Bob clicks on "Generate" button and generates the quiz in
various formats for different target devices (See Figure 5.5):

Target
••••• 2 . D e s c r i p t i o n ; I ' ^'.^JXCAi^on^tX^
Please download XML file, then you can always have your
XML quiz and you will be able to update or change the quize just Download XML
by click on the "Update Quiz" link at the toolbar.
CLDC-i.x. MIDP-2.0. Please download jar and jad file. In
Ci.DC-lJ< order to upload the code into the mobile device , Please read Java ME Devices-JAD
MXDP-2.0 the instruction file. Java ME D^VM-**;-1AR
Download Instruction

CLDC-1.0, MiDP-1.0. Please download jar and jad file. In

CLDC-1.0 order to upload the code into the mobile device , Please read Java ME Devices-IAD
MIDP-1.0 the instruction n.e: Java ME Devices-JAR
Download Instruction

Please downlead the code file. In order to upload the code

into the mobile device , Please read the instruction file:
BtackBerry BlackBerry Devices

WAP 1.x WAP l.x . Please download WML pages in order to enable
students to have online quiz. WML file WMI.Sr.rint

Otaktop Browser- You can also generate Desktop Browser Client Generate Browser Desktop
bused Client ( HTML/XHTML) simply by clicking the following link. Client

Figure 5.5: The quiz generated for various devices

o Generating a quiz from the KB (Knowledge-Based) quiz database: with

choosing this item, Bob has option to generate quizzes randomly from the
KB database. For this purpose, first he enters the quiz topics which he is
looking for and the number of questions. Quiz topics can be a single topic
or a multiple keywords set delimited by comma. Then, he clicks on

46
 "submit" button. Regarding the keywords he is searching, he will get a list
of quizzes questions. Now he can select the desired questions among them
and then click on "Generate" button; likewise, the quiz will be generated
for various target devices and added to the quizzes list of the relevant
course.

o Generating a quiz from an existing quiz XML file (update quiz): with this
item Bob is able to change his quiz and update it. For this purpose, first of
all he should upload the quiz XML file. Then he is able to change the quiz
configurations as well as the questions and answers. At the last, he can re-
generate the quiz again just by clicking on the "Regenerate" button. The
quiz will be generated again for all targeted devices and the updated date
and time will be added to the quiz XML file.

• Learners group Clients are categorized into 3 groups including:

- Instructors who use e-leaming platform to take advantage of teaching tools,
mange courses and students
- Learners who use e-learning platform
- Learners who use m-learning platform.
There are two platforms for learners. One is e-learning platform which provides
Web-based facility to access their educational materials, view their grades, display
the instructor's feedbacks, messages and notifications. The other is m-learning
platform. For this platform there are two kinds of materials including offline and
online educational content. The following sections describe each kind:

o Offline materials
Using offline materials has several advantages: Firstly, learners are always
able to upload offline materials into their cell phones in different ways
including infra-red, USB and Bluetooth. Secondly, when learners upload
offline materials into their devices, these materials will always be available; no
matter there is signal disruption or network disconnectivity. Lastly, using

47
offline materials is very cost effective for learners because they do not have to
use bandwidth to upload these materials.
In contrast, there are several challenges as well. One of these challenges is
memory usage. As we know, mobile devices are limited in memory and
processing power. Hence, we need some strategies to optimize mobile
applications and learning content in size and memory usage. The other
challenge is wide variety of mobile devices. Today, students have different
mobile phones with different capabilities and constraints. Hence, we need
some strategies to support customization and device-independent delivery of
content materials to these mobile devices.
We have developed our system regarding the criteria discussed above.
According to our investigation on available target devices, our system has
ability to generate offline quiz materials for the following targets:

Table 5.1: Offline quiz formats

Taij-i'ls Devices
CLDCLx Java-enabled device
MIDP 2.0
CLDC 1.0 Java-enabled device
MIDP 1.0
Blackberry API Blackberry device

This means our system covers a variety of mobile devices. We run the
educational materials generated by the system on various simulators, including
Motorola, Nokia, Sun, and Blackberry. Figure 5.6 shows some screen shots of
an offline quiz in the Blackberry 8800 device simulator.

48
 Spnri f eedback to nmeBielp@gnail.com Question: 1/15
W E B SERVICE INTRODUCTION Qwz WHATKW3DL?

D U R A T I O N : IS M I N S ' g W S D L IS AN X M L FORMAT FOR DESCRIBING

•KTWCMWK SEHVtees A S A S E T O F e o p o w n r s
DATES Z4- SEP 3 0 0 7 CMS34 | A SERVICE SPECnCATKW
• A SERVICE DISCOVERY
5 NONE

•IHHHH NEXT OUESTIOM • H H H H

Send feedback to ctnelhelp^matLcom Question: Z/B

W B B SERVICE ! M T P » « J C T » W QUIZ WHAT IS UDDt?
YOUR SCORE IS: HUDEM CAN BF QUERIED FOR SPFOFtC WEB
SERVICES
©OUTOFE3 • IT IS UKE A VI ' -R WEB SERVICE
S UNIVERSAL Dt COVERY, AND
YOU NEED MORE PRACTICE
•jAUL
^^^^^^^^^H TRY AGAIN ^^^^^^^^^B

Send freriuuek to nrruHhcJpPJkjnad.eani Question: 1/15_

WBB SERVICE J*reoouerK»i Qwz WHAT IS WSDL?

YOUR SCORE KB: _S W S D L M A H X M L FORMAT FOR DESCRIBING

N E T W O R K SERVICES A C A S E T O F EM0POWTS
70UTOFI5 G | A SERVICE SPECIFICATION
M A SERVICE DISCOVERY
YOU MEED MORE PRACTICE SNONE

CS&33I • H H H H I NEXT QUESTION ^ ^ m | ^ ^

THE CORRECT ANSWER I B 1

Figure 5.6: An offline quiz on the Blackberry 8800 device simulator

49
|55S!S^
Web Service Introduction Quiz WhatisWSDL?
Tue, 02 Oct 2007 ®WSDL is an XML-based language for defining
Web services.
Duration: 15Min(s)
O A service directory
Are you readytotetothe quiz? OA service discovery
Onone
Yesl;. Ho
[Next Question)

wmm :^mi3
ZE^ak U4- • i-^ m^m^^m?mm
Figure 5.7: An offline quiz on a Java-enabled device simulator

ifOtDHOM

QuestiDn:l/15
Questian:l/15
IWhatisWSOL?
mrnumsDii
iDt is m XMLHbanad language ftr d
8|fV5DL is an XM_Hbased fan] I \Ateb servicer.
puage ftr defining Web ^.sarvtedHedarf
Sfioes.
° A service diredarv °rwr»
° A service discovery
:
°none

Exit text J P MP*" I

Figure 5.8: An off-line quiz on Figure 5.9: An off-line quiz
Motorola L7I emulator on Motorola vl 100 emulator

As you can see in Figures 5.6 to 5.9, the quiz has been generated and customized on
different mobile devices.

o Online materials
In online quiz, we use two different technologies: Browser-based and Native.
Here is a brief description of the evolving technologies:

50
 Native: we have a light- weight mobile LMS (mobiLMS). This platform is
deployed on mobile phone devices. Like eLMS, it provides facilities for
learners to manage their courses, connect to the server, use the generated
materials and take quizzes. In addition, it is able to analyze the coursework,
calculate the scores, display the correct answers to the learner and finally send
the results back to the instructor (optional). In online quiz, mobiLMS
communicates with the eLMS over the open standards HTTP, and XML.
Browser-based: using this technology, quizzes are designed using markup
languages. We used two most popular markup languages currently used in
mobile devices. For those devices which support WAP 1.x, our server
generates quiz in WML format. For those which support WAP 2.x, it generates
XHTML. Table 5.2 shows the supported markup languages:

Table 5.2: Supported markup languages

Protocol | Supported Markup l.:iii»ua<rc
WAP 1.x WML, WML script
WAP 2.x XHTML

Figures 5.10, 5.11 show some screen shoots of an online quiz in both format
WML and XHTML on Nokia Mobile Browser 4.0:
IMDKfA
Question 1/15 Select
What is WSDL?
WSDLis. O A service directory
O A service discovery
Onone

Options Back Select Back

Figure 5.10: A browser-based quiz (WML format) on

Nokia Mobile Browser 4.0 simulator

51
 XHTML Quiz format

Web Service Introduction

Quiz
Course: CIS 6000
Please answer the following
questions:

l ) W h a t is WSDL?
® W S D L i s an XML
f o r m a t f o r describing
network services as a
set of endpoints
O A service specification 1
O A service discovery
Onone

2 ) W h a t is U D D I ?
B U D D I can be queried
Options V Biack

Figure 5.11: A browser-based quiz (XHTML format) on

Nokia Mobile Browser 4.0 simulator

5.4 Mobile Technologies in our Implementation

Technically, there are various brands of mobile devices with different capabilities and
constraints not only in hardware such as screen dimension, resolution, processing
power, memory usage, and bandwidth but also in software such as operating system
and internet browser. Hence, it is important for mobile application developers that
their applications support a wide range of mobile devices with different mobile
technology. Thus, in order to find a road mapping process, we categorized the mobile
technologies in 4 categories based on [26]. It includes content delivery methods,
connectivity methods, content development languages, and mobile platforms as you
can see in the following:
Content delivery methods: diversity in mobile phone devices is one of the
issues facing in developing applications for handheld devices. Hence, content

52
 delivery to these devices needs lots of efforts in order to overcome the
differences on mobile devices. However, in order to address this issue, our
platform supports various content delivery methods as follow: WAP 1 .x, WAP
2.x, HTTP, Email and SMS.
Content development languages: our platform is able to generate educational
materials in the following languages all at once: XML, HTML, XHTML,
WML, J2Me (Midlet) and Blackberry API (application).
- Supported Connectivity: in our platform, transport is available via GPRS,
GSM, Infra Red, Bluetooth, PC download, and WiFi.
Platforms support: J2ME, Blackberry , and Windows OS.

5.5 Considered Criteria

There are also some important criteria that we have considered during the
development of our system. These criteria can be summarized as follows:
• Cost issues We always considered it as a high propriety criterion while we were
developing the mobile applications. Because we believe that if a system wants to
be applicable especially in our scopes which mostly are students, it should be cost
effective. That's why our platform generates both offline and online materials. In
fact, in order to reduce the cost of communication, learners can use online
materials while they are in campus and offline materials while they are in off-
campus. Using offline materials also can avoid inconvenience of signal disruption.
Besides, in our implementation, we considered to keep the size of midlets,
platforms and generated educational materials at a minimum. This can save the
bandwidth and reduce the cost of communication when using the online materials.
Also, it can save the memory usage.

• Data availability It determines whether generated educational materials are

always available for learners either in on-campus or off-campus. In order to
support this criterion, our platform generates both offline and online educational
materials and quizzes. Hence, students and learners can always have access

53
 learning materials as well as quizzes. It does not matter whether they are in on-
campus or off-campus.

• Ease of use It has been tried to design and develop the application and tool very
friendly and easy to use for both learners and teachers. Indeed, teachers put
minimum efforts to generate the materials.

• Support the existing educational materials and existing systems Our system
has capability to generate the educational materials and quizzes for mobile phones
as well as Personal Computers, tablet PCs and laptops all at once. Besides, we are
extending our system in order to support the existing educational materials as
well.

5.6 The Processing Flow

The work flow diagram for learners is presented in Figure 5.12. We already
explained the operating scenario from the learners' perspective in both models online
and offline. However, what we want to highlight here is the quiz features for learners.
As it is shown in Figure 5.12, the quiz generated by our proposed system has the
following features:
• Time-Limit: it may have time limit or not depending on the quiz configuration. In
the case of time-out, the result is displayed on the mobile phone and the student
goes for the other trial (if applicable).
• Trial number: learners are able to try the quiz several times depending on the
trial number defined by the instructor. Each time student is able to view his/her
score and after the last trial, correct answers are displayed on the mobile device's
screen.
• Feedback: the result may be sent to the instructor or not. It depends on what the
instructor set up in the quiz configuration. Also, the feedback methods are defined
by the instructor.

54
 On-line quiz

Ftegjsterthe
-^
service

Select Courses

*_
Select Quiz

Offrline.quiz

Time Out
\ i

Submit quiz

Display the
score to learner

Doesthe ^ * ^ - « . Yes
resultssend Send results back to
-^back to teacher^---'' teacher by email or SMS
-'
No '£
Logout s
V ./

Figure 5.12: The work flow diagram of learners

5.7 Evaluation

In order to evaluate our system, we propose an evaluation model (See Figure 5.13). In
this model, some of the indicators including the kind of supported mobile devices,
support of e-learning, and availability of content (offline and online access) are
adopted from [18]. We add the extra indicators. They include the supported data
formats, context-aware service supports, costs and ease of creating educational

55
materials. Choosing these indicators has been according to the usability of the mobile
learning systems. In fact, we argue that the usability of mobile learning systems can
be affected without which one of these indicators. We have already explained the
reasons throughout this thesis.

Every Time
Adaptability of content Availability of content Everywhere

Supported data
Support of e-learning f Device
formats
[_ Independency

Context-aware Service Ease of creating

Support content materials

Figure 5.13: Evaluation roadmap to examine

current mobile learning systems

Hence, we evaluate our system based on this proposed model. The result is specified
in the following sections:
• Adoptability of content: in our system content are adopted in a variety of devices
including PDAs, smart phones, cell phones, Tablet PCs, laptops and PCs.
• Support of e-learning: our Web-based learning management system (eLMS)
supports e-learning.
• Availability of content: the proposed system supports offline as well as online
access to educational materials.

56
• Supported data formats: our system generates educational materials in XML,
XHML, HTML, WML, J2ME midlet and Blackberry API code which are
compatible with most mobile devices.
• Ease of creating educational materials: in our system we consider creating the
content materials from the instructor viewpoint. In fact, we provide teaching tools
that facilitate creating educational materials. One of these tools is our quiz
generator system in which quizzes are created by the instructor once through a
Web-based interface and then the same content is generated for various devices
automatically. The other tool is our knowledge-based quiz database which offers
educators to create their quizzes from this DB according to their requested topics.
• Costs: the only cost included is the cost of using online access to materials, that of
course, it can be ignored when we use offline learning. To note that the offline
version of content materials are always available on our e-learning platform
(eLMS).
• Context-aware service support: in this version of the system, we do not support
context-aware services.

5.8 Lessons Learned

We have learned a lot during the design and implementation of this project. In the
following section, we demonstrate some of these findings.

5.8.1 Customized Environment

In online-learning, we use both technologies including Browser-based and Native to

access content materials. In Browser-based technology, we generate content materials
in WML and XHTML format whereas in Native technology, we use content materials
in XML format. The reason to support both technologies has been for customization
purpose. In the following, we explain explicitly the issues which lead us to support
both technologies:

57
• WML: WML (Wireless Markup Language) is used for mobile devices which
support WAP 1.x protocol. Actually, there are two drawbacks in using this
markup language. One is limitation in WML page sizes. In fact, in WAP1 the
WAP gateways refuse WML pages that their size is beyond a very little
amount depending on the browser between 3K and 20K. The other
disadvantage is inconsistent layout and different presentation of content
materials on different mobile devices [37]. These two issues make limited
variety and quality of content that it is in contrast with what is expected in
mobile learning. We mean high quality content materials. Hence, although we
generate WML version of learning materials, we look forward for other
solution.
• XHTML: XHTML is currently known as the official Web markup standard
replacing HTML. It is also used as the markup language for mobile devices
which supports WAP 2.x protocol. In fact, it provides better content
presentation abilities than WML. In addition, it offers the opportunity to
provide the same content for mobile devices and PC environment. However,
using XHML features are restricted in mobile phones due to their constraints.
In addition, it does not support scripts on mobile phones. These two issues
made difficulty in presenting interactive materials like quizzes.
The other issue is the depth of pre-fetching. Actually, we have two
ways to present quiz questions on mobile devices. One way is presenting all
questions in one XHML Web page. Regarding the small screen size of mobile
phones, it is not a convenient way for learners to scroll Web page down and up
to view questions. We prefer learners to view the questions page by page. This
means one question per page. The other way is presenting each question in one
Web page that of course, it is not an optimize solution because it increases
data traffic across the network. Besides, if we want to pre-fetch a few Web
pages then what will be the depth of pre-fetching? How many level of Web
paged should be pre-fetched? Due to these issues, we go towards the third
solution using XML.

58
 • XML: in order to access content materials online using Native technology, we
use XML. In this method, we develop a light-ware platform which should be
deployed on the mobile devices. We named this platform mobiLMS. It talks to
our server over XML and HTTP. Figure 5.14 shows this relationship:

Figure 5.14: Online access to learning materials using

native technology

This method is very cost effective, decrease the data traffic across the network
and very customized. In addition, it has capability to do customized works
such as analysing and grading quizzes without using bandwidth. In developing
this platform, we considered size to be kept at a minimum.
To sum up, in the context of mobile learning having a user friendly and customized
environment is necessary.

5.8.2 Text Inputs

Actually, one of the main problems on mobile phones is text input. Today, most
mobile phones are equipped with a standard keypad [37]. In this keypad a mapping
of more than one character is used per button. This has made typing text on the
mobile devices cumbersome and most of people just ignore it.
In PDAs and smart phones there is similar problem. Basically, in these kinds
of devices there is variety of methods to input text. One is using keypad and another

59
one is a pointer device or touch screen. Although using pointer device is more
comfortable than using keypad, there is not convenient to input long piece of text yet.
The reason is because users still have to work with a small button on the keypad or a
small touch screen button on the small screen. As a result, in the context of mobile
learning, having a convenient environment is an essential thing.
Hence, in developing mobile learning applications, we should consider these
issues. In other words, we should design our applications in a way that needs
minimum efforts to input text.

5.8.3 Content Material Design

According to the literature, the content materials designed for mobile devices should
be different from the content materials designed for PC environment. This has been
because of the limitations on mobile devices. Although we adopt this solution in some
applications such as designing learning materials for informal and/or lifelong
learning, we argue that creating educational materials (by educators) doubled for both
PC and mobile devices environments is a tedious work. On the other hand, content
materials are an important part in the context of mobile learning. In fact, they should
be designed in some ways that encourage learners to learn and read. Perhaps, here it is
better to back to the term "right content". Actually, what learners need is right
content.

5.8.4 Wide Spread Use of M-Learning Applications

There are lots of debates on the effectiveness of m-learning to enhance the learning
process. However, we adopt that before any judgment, learning modules and systems
for mobile devices need to be developed and used in mobile environments vastly [35].
In fact, the most challenge in mobile learning is in developing innovative, usable and
affordable software [7].

60
Chapter 6

Conclusions and Future Work

6.1 Conclusions

In this thesis, we have discussed various forms of educational and training systems.
We have also seen how trends are moving towards a new form of education called
mobile learning or m-learning. We have explained the impacts of mobile learning on
the establishment of pervasive learning environments and how it can facilitate and
enhance learning. We then investigated the obstacles in mobile learning. We have
identified some of the major obstacles, including: (1) various definitions of mobile
learning; (2) no framework for mobile learning systems; (3) many technical issues in
developing mobile learning systems that have not been addressed. For instance,
disconnectivity of wireless networks, lack of location-aware services, cost of online
connections, a variety of mobile devices available with different capabilities and
constraints, and reusability of content.
In order to address these issues, we have designed and implemented a device-
independent platform for mobile learning. To this end, the contributions of this thesis
are:
1) A mobile learning model has been proposed. In this model, we contend that an
efficient learning strategy can be achieved by providing the "right content" to
learners, and this is accomplished by integrating knowledge-based data and
contextual information in a semantic way.
2) We designed a platform that utilizes the above mobile learning model. This
platform address many issues raised, including device-independent delivery of
content, support for a variety of mobile devices and reusability of learning
materials. It also offers learners online as well as offline access to the
materials from any device.

61
 3) A proof of concept implementation has been constructed. The implemented
platform supports learners as well as educators. It allows educators to create
the educational materials once and generate content for a variety of devices
including cellular phones, smart phones, PDAs, laptop computers and PCs. It
uses both native as well as browser-based technology to support offline as well
as online learning. It also provides tools to generate quizzes and use the
knowledge-based repository to share educational materials with other
instructors. In addition, it allows learners to access the generated content at
anytime from anywhere.
4) An evaluation model for mobile learning system has been developed. This
model includes some indicators that determine the usability of mobile learning
systems including availability of content, supported data formats, support of e-
learning, supported devices, context-aware service support, and ease of
creating content.

6.2 Future Work

The current implementation of the quiz generator supports text-based quizzes, but we
plan to support other mobile learning objects such as images, multimedia, audio,
video and animation.
Adobe Acrobat and Microsoft PowerPoint are the most popular software
packages used by educators to create the educational materials and lecture slides. We
will extend our platform to make it possible to transform these kinds of educational
materials to our adapted formats compatible with various target devices. This would
be convenient from the educators' viewpoint.
Finally, we will be adding support for synchronous learning. In a synchronous
system, students are able to communicate with teachers or other students in real time.

62
Bibliography

[1] Sharpies, M., Taylor, J., and Vavoula, G. (2007). A Theory of Learning for the
Mobile Age, In R. Andrews and C. Haythornthwaite (eds.) The Handbook of
Elearning Research. London: Sage, pp. 221-247.

[2] Siemens, G. (2004). Connectivism: A learning theory for the digital age.
http://www.elearnspace.org/Articles/connectivism.htm (accessed August 4, 2007).

[3] Vavoula, G., Sharpies, M., Scanlon, E., Lonsdale, P., and Jones, A. (2005),
Report on literature on mobile learning, science and collaborative activity (D33.2.2),
European Commission Project Report, Contract No. IST-507838.

[4] European Commission. (2001). Memorandum on Lifelong Learning: Making a

European Area of Lifelong Learning a Reality, Brussels, http://eur-lex.europa.eu/
LexUriServ/site/en/com/2001/com2001_0678en01.pdf (accessed June 12, 2007).

[5] Cook, J. and Smith, M. (2004). Beyond Formal Learning: Informal

Community eLearning. Computers and Education, CAL03 Special Issue, 43(1-2), pp.
35-47.

[6] Department for Education and Employment (DfEE). (1998). The Learning Age: A
Renaissance for a New Britain (Green Paper CM 3790), London: The Stationery Office.

[7] Holzinger, A., Nischelwitzer, A., Meisenberger, M. (2005). Lifelong-Learning

Support by M-learning: Example Scenarios. ACM eLearn Magazine, 2005(11).

63
[8] Reardon, M. (2007). Emerging markets fuel cell phone growth,
http://www.news.com/2100-1039_3-6159491.html (accessed September 23, 2007).

[9] Kanellos, M. (2004). A billion PC users on the way,

http://www.news.com/A+billion+PC+users+on+the+way/2100-1003_3-5290988.html
(accessed September 23, 2007).

[10] Mobile Phone Culture or Customs. (2006). http://apsidalkat.blogspot.com/

(accessed September 23, 2007).

[11] NTT DoCoMo and KDDI Develop Platform for Immediate Notification of
Emergency Earthquake Alerts. (2007). http://www.kddi.com/english/corporate/
news_release/2007/0530/index.html (accessed September 20, 2007).

[12] Wagner, E. D. (2005). Enabling mobile learning. EDUCAUSE Review, 40(3),

pp. 40-53.

[13] Oksman, V. (2006). Young People and Seniors in Finnish 'Mobile Information
Society'. Journal of Interactive Media in ZWwcaft'o«,http://www-jime.open.ac.uk/
2006/02/oksman-2006-02-04.html (accessed July 9, 2007).

[14] Wagner E. (2007). Mobile Learning: The Next Step in Technology-Mediated

Learning. http://www.clomedia.com/content/templates/clo_article.asp?articleid=1686
&zoneid=182 (accessed July 10, 2007).

[15] Hundebol, J. and Helms, N. (2006). Pervasive Learning Environments. In C.

Crawford et al. (eds.), In Proceedings of Society for Information Technology and
Teacher Education International Conference, Chesapeake, VA, 2006, pp. 2226-2231.
VA: AACE.

64
[16] Sharpies, M. (ed.) (2006) Big Issues in Mobile Learning: Report of a workshop
by the Kaleidoscope Network of Excellence Mobile Learning Initiative. Nottingham:
Learning Sciences Research Institute.

[17] Lehner, F., Nosekabel, H. (2002). The Role of Mobile Devices in E-Learning
First Experiences with a Wireless E-Learning Environment, In Proceedings of IEEE
International Workshop on Wireless and Mobile Technologies in Education, Sweden,
2002. pp. 103-106.

[18] Georgieva, E., Smrikarov, A., Georgiev. T. (2005). A General Classification of

Mobile Learning Systems, In Proceedings of the International Conference on
Computer Systems and Technologies (CompSysTech' 2005), Technical University,
Varna, Bulgaria, 2005.

[19] AvantGo, http://www.avantgo.com (accessed August 11, 2007).

[20] Joanneum, FH. (2004). The Mobile Learning Engine (MLE) - A Mobile,
Computer-aided, Multimedia-based Learning Application. In Proceedings of
Multimedia Applications in Education Confernece (MApEC), Graz, Austria, 2004.
http://drei.fh-joanneum.at/mle/docs/Matthias_Meisenberger_MApEC_Paper
_mLearning.pdf.

[21] Trifonova, A., Ronchetti, M. (2003) Design and Development Approaches for
Creating a Mobile Learning System - Mobile ELD IT. In Proceedings of the Twelfth
International World Wide Web Conference (WWW2003), Budapest, Hungary, 2003.

[22] Chan, Y. Y., Leung, C. H., Wu, A. K. W. and Chan, S. C. (2003). MobiLP: A
Mobile Learning Platform for Enhancing Lifewide Learning, In Proceedings of the
3rd IEEE International Conference on Advanced Learning Technologies, Athens,
Greece, 2003, pp. 457-457.

65
[23] Quinn, C. (2000). Mobile Learning: Mobiles, Wireless, In-Your-Pocket
Learning. LiNE Zine, http://www.linezine.eom/2.l/features/cqmmwiyp.htm (accessed
June 15, 2007).

[24] Keegan, D. (2005). Mobile Learning: The Next Generation of Learning.

Dublin: Distance Education International, http://learning.ericsson.net/mlearning2
/files/ workpackage3/corvinus_technical_summary.doc (accessed June 22, 2007).

[25] Brown, T. H. (2005). Towards a Model for m-Learning in Africa.

International Journal on ELearning, 4(3), pp. 299.

[26] Attewell, J. (2005). From Research and Development to Mobile Learning:

Tools for Education and Training Providers and their Learners. In Proceedings of the
4th World Conference on Mobile Learning (mLearn 2005), Cape Town, South
Aferica, 2005, http://www.mlearn.org.za/CD/papers/Attewell.pdf (accessed June 12,
2007).

[27] Munn, L. Design of Cool Mobile Java Application, http://developers.sun.com

/asiasouth/news/cebu2.pdf.

[28] M-Learning India. (2007). Introduction to Mobile Learning (M Learning).

http://www.mlearningindia.com/index.php?option=com_content&task=view&id=12&Ite
mid=l.

[29] Weiser, M. (1991).The Computer for the 21st Century. Scientific American,
September 1991, pp. 94-104.

[30] TASC workshops. (2002). Learning Styles Exploration,

http://www.trcc.commnet.edu/ed_resources/tasc/success%20workshops/learning_styl
es.htm (accessed October 28, 2007).

66
[31] MOBIlearn. (2002). User Trial at Uffizi Gallery, http://www.mobilearn.org/
results/trial.htm, (accessed September 15, 2007).

[32] Brugnoli, M. C , Morabito, F., Bo, G., Murelli, E. (2007) Augmented

Itineraries: Mobile Services Differentiating What Museum has to Offer. PsychNology
Journal 4 (3), pp. 311 - 335.

[33] Distance Education. (2007). http://www.educationconnection.com/global

/DistanceEducation.aspx, (accessed September 14, 2007).

[34] Tan, T., Liu, T., (2004). The MObile-Based Interactive Learning Environment
(MOBILE) and A Case Study for Assisting Elementary School English Learning, In
Proceedings of IEEE International Conference on Advanced Learning Technologies,
Joensuu, Finland, 2004, pp.530 - 534.

[35] Goh, T., Kinshuk (2003). Developing an Adaptive Mobile Learning System, In
Proceedings of the International Conference on Computers in Education (ICCE
2003), Hong Kong, 2003, pp. 1062-1065.

[36] Yang, S. J. H. (2006), Context Aware Ubiquitous Learning Environments for

Peer-to-Peer Collaborative Learning, Educational Technology & Society, 9 (1), pp. 188-
201.

[37] Wigdor, D., Balakrishnan, R. (2003). TiltText: Using Tilt for Text Input to
Mobile Phones, In Proceedings of the 16th Annual ACM Symposium on User Interface
Software and Technology, Vancouver, Canada, 2003, pp. 81 — 90.

67