WEB-BASED SUPPORT FOR RESOURCE-EFFECTIVE

E-LEARNING

M. A. Rentroia-Bonito, Frederico Figueiredo, André Martins, Vitor Fernandes, Joaquim Jorge

e-Learning-in-Organization Research Group, Departamento de Engenharia Informática, IST/INESC-ID,

Av. Alves Redol, 9, Lisbon, Portugal

Keywords: e-Learning, Learning Management System, Moodle, Webcast, e-Presence, Customization, SEMINOLE.

Abstract: This article presents work done at the Technical University of Lisbon, on web systems to support

technology-enhanced learning. Based on eighteen months experience with webcast technologies and

commonly used Learning Management Systems, this work has integrated both concepts and developed a

simple, yet effective learning approach. This approach supports both students and professors in moving

from the traditional classroom to technology-enhanced online settings. We present the evolution of our

system and the results obtained from testing and evaluating the prototype, during the last Spring semester.

One of the major features of our approach lies in combining both online and offline components of the e-

Learning experience and incorporating desirable traits from traditional and technology-mediated learning.

This way, we have identified significant issues for users in order to better manage the changes required to

adapt our system to organizational processes and context of use. Our assessment points to changes in the

teaching method, course organization, and highlights new factors of students’ motivation. Results, however

preliminary, seem to indicate both the usefulness of our approach and the feasibility of deploying e-learning

initiatives with efforts optimization to complement traditional offerings in university settings.

1 INTRODUCTION

In the past few years there have been substantial

advances in e-Learning environments spurned by

developments in Communications, Media, and

Computing Technologies. However, technology

itself is not the single, most important, driving factor

in improving e-Learning experiences. As we gain

more understanding on the dynamics of online

learning, new challenges emerge. Indeed, there is a

need for proper methodologies and tools to address

these new challenges, as students and educators

migrate from traditional classrooms to online

environments. This is because both Human Factors

and Technology Investments need to be managed in

articulation with learning strategies to explore new

possibilities in a more resource-effective manner.

For courseware developers, this translates into

requirements to shorten the development time of

courseware that are both cost-effective and

acceptable to students. This is challenging work

practices in all kind of organizations. The Instituto

Superior Técnico (IST), of the Technical University

of Lisbon, Portugal, is no exception. During the last

two years, our research group has been developing

an e-Learning system to support the transition of

traditional class-room environments to technology

enhanced-learning, whether online or offline.

This paper’s goal is to share the experimental

aspects of our approach to develop, implement, and

test an e-Learning solution. We are developing

through third-party integration and customization of

off-the-shelf components a SEaMless INtegrated

Online Learning Environment (SEMINOLE).

The remainder of this paper presents a

conceptual overview of our approach and describes

our concrete experience at IST. Preliminary

assessments and lessons learnt, jointly with potential

benefits for each of the main stakeholders, namely

instructors, students, system administrators, and

educational institutions; and future work are

presented. This work aims at setting up a

comprehensive framework to analyze online

learning and guide development efforts towards

resource-effective solutions for e-Learning

(Rentroia-Bonito, Tribolet, Jorge & Ghaoui, 2005).

281

A. Rentroia-Bonito M., Figueiredo F., Martins A., Fernandes V. and Jorge J. (2006).

WEB-BASED SUPPORT FOR RESOURCE-EFFECTIVE E-LEARNING.

In Proceedings of WEBIST 2006 - Second International Conference on Web Information Systems and Technologies - Society, e-Business and

e-Government / e-Learning, pages 281-286

DOI: 10.5220/0001256702810286

 SciTePress

2 CONCEPTUAL FRAMEWORK

Creating and sustaining e-learning settings indicate

institutional readiness to assume this new paradigm

as part of its strategy, culture, and internal practices.

The literature reveals that this readiness covers

structural and relationship issues (Dix et al, 1998;

Dougiamas & Taylor, 2002; Preece, Rogers &

Sharp, 2002; Rosson & Carroll, 2002) and previous

works (Rentroia-Bonito, Figueiredo, Martins, Jorge

& Ghaoui, 2006b). As shown in Figure 1, as

technology gets in the process of learning, the need

for structure highly increases and the need for

relationship differs.

However, effectiveness and efficacy of this

paradigm still relates to how well courseware

designers: (a) address the specificities of users’

learning needs, characteristics, goals and priorities

taking into account their immediate context and

available resources; and (b) cover not only the

production and distribution of learning content in

educational settings, but also its articulation with

internal work processes and labour market dynamics

and needs.

Technology and instructional design are main

issues when structuring an e-learning experience. At

this regard, organizational investment in technology

(e.g. network, equipment, software), and human

resources (e.g. instructors, knowledge or task expert,

speakers, helpdesk staff, teachers, and administrative

assistants) should meet users’ local and remote

conditions in order to improve e-learning

effectiveness (Rentroia-Bonito, Jorge & Ghaoui,

2006a). Technology should support and explore all

four different combinations. Our vision is that a

workable technology should address all

combinations even if deployed progressively. This

involves defining clear milestones in terms

organizational acceptance, skill development,

required accessibility and infrastructure supporting

e-learning experiences to allow for harmonious

development.

The following section presents the strategies and

progress of our research work with e-Learning at the

Departamento de Engenharia Informática (DEI) at

IST.

3 STRATEGIES AND PROGRESS

After defining an e-learning vision and assuring

required resources, the development approach for

the solution was iterative and incremental. This first

prototype was tested during the Fall semester 2003,

giving a e-lecture to 77 HCI students from three

Portuguese universities, (Rentroia-Bonito, Jorge &

Ghaoui, 2006a). Then, another test was realized

during the Spring semester of 2004. Back then, a

total of 11 PCM students participated.

This proof-of-concept allows the retrieval of

structured and unstructured user feedback that

contributes to improve the current understanding of

e-learning dynamics reflecting it on prototype

requirements and architecture. SEMINOLE was

architected during Fall 2004, tested and evaluated

during Spring semester 2005 with a class of 28 PCM

students. Next, the main development stages until

achieving current version of SEMINOLE are

presented.

3.1 Stage I: Archive and Webcast

Before this stage was completed, the course was

supported by a static Website (PCM, 2004) where

contents (announcements, goals, course program,

schedules, grading details, delivery dates, and

information about the teachers, classifications, and

bibliography) were available to all students. All

class contents were edited and published manually

through the edition of HTML pages and later

uploaded to the Website server. In this first stage, we

introduced recording and video archiving of all

classrooms, and live webcast of invited lectures.

Figure 1: Conceptual Framework.

WEBIST 2006 - E-LEARNING

282

Our first approach consisted of a software tool,

supported by FCCN (Portuguese Foundation for

Science and Research), that enabled audio and video

streaming with support for synchronized

presentation slides. These were defined to simulate

the classic classroom scenario allowing remote

students to interact with other students and the

speaker, through the use of email messages and an

IRC (Internet Relay Chat) online session. This tool,

in its simplicity, had shortcomings related to: (a)

high demands for resources (skills, time and media)

complex pre and post-production processes, and (b)

inefficacies of the integration of IRC session and

quizzes evaluation.

In an attempt to get users’ feedback, the last live

webcast session was evaluated. Based on this user

feedback, the alignment between the learning

process-system for this user group was improved by

implementing a dynamic and more efficient content

management tool tightly adapted to the internal

teaching process (Rentroia-Bonito, Jorge & Ghaoui,

2006a).

3.2 Stage II: Development of

SEMINOLE

The version of the system was designed to meet

three main requirements: learning content

management, class webcast and archive, and

evaluation methods. Its main functionalities were

identified based on defined vision, priorities;

university’s teaching process and analysis of

strengths and weaknesses of available Learning

Management System (LMS) platforms.

SEMINOLE is based on an opensource LMS

integrated with a streaming system, as shown in

Figure 2. The former allows students to access all

class of contents, participate in online fora, take

quizzes, check grades, etc. The latter allows

webcasting of all events in a course. This way,

students can attend classes remotely and view slides

which are synchronized with audio and video

streams. They can also participate in classes, via

chat-room, as well as place questions to teachers and

other colleagues.

The learning content management process was

tightly supported by a LMS called Moodle (Moodle,

2004) that embraces the social constructivism as an

educational philosophy (Dougiamas & Taylor,

2002). Moodle provides the necessary tools for

building and managing the course Website (PCM,

2005), facilitating, among others: (a) scheduled

delivery of learning content in different formats, (b)

possibilities of diverse resources and tasks

supporting different learning methods, and (c) fast

feedback to students after doing quizzes. For the

webcast feature, we chose to use ePresence

(ePresence, 2004) from KMDI Labs in the

University of Toronto, along with additional

customization, the component in SEMINOLE

currently supports: (a) lectures by audio and video

synchronized with slides; (b) integrated and

moderated live chat; (c) question submission, and

(d) semi-automated generation of structured,

navigable, searchable lecture archives.

For students to evaluate the usability of

SEMINOLE, online questionnaires were used.

Registered students participated in evaluation

sessions. Motivation-to-elearn items were captured

before students used the system and ten weeks later.

The system’s usability was evaluated in the last

session. Also students reported their system usage

frequency and average access duration time.

Confidentiality and anonymity were assured before

evaluation sessions started. Data was analyzed to

identify improvement areas based on results by: (a)

comparing motivation-to-elearn perceptions,

measured before and after using SEMINOLE; (b)

evaluating the usability of the system, and (c)

content-analysing responses to open questions.

Figure 2: Current SEMINOLE Architecture.

WEB-BASED SUPPORT FOR RESOURCE-EFFECTIVE E-LEARNING

283

4 RESULTS AND LESSONS

During this stage four main lessons were identified.

First, from technical perspective, several areas of

improvement were identified: (a) integrating user

logging to allow further analysis of students' access,

usage and communication patterns; (b) facilitating

content publication by implementing an automated

schedule; and (c) taking full advantage of the

possibilities of the platform in terms of delivering,

and reutilizing, current and archived contents in

different formats. This will address the structural

and relationship issues shown in Figure 1, improving

the alignment between PCM course and

SEMINOLE.

Second, structuring e-learning experiences

demands strong organization skills from instructors

as well as enforcing the application of standard

operating procedures. This is required to coordinate

efforts from involved staff (e.g. Helpdesk, Content

managers and producers, invited speakers, teaching

assistants, among others) in order to assure

consistent communication during the different

events (e.g. posting news, instructions, feedback).

Third, from a class dynamics perspective,

learning tasks were facilitated. In particular,

communication with local and remote students

improved due to the possibility for them to talk to

each other. Within webcast sessions, a total of 910

messages were sent, using e-Presence’s chat-room

feature. 73% of messages were related to

communication between the students and the

moderator, mainly concerning technical aspects of

the session. Fora optimized the process of class

communication by turning individual answers into

collective ones. We also verified that “Course

Program”, “Evaluation Details” and “How to

perform exams (quizzes)” were the most viewed

resources, which calls for planning when designing

e-learning experiences. For instance, self-reported

responses indicated that 58% of students that

evaluated SEMINOLE, accessed the system

frequently (three or more times per week). 91% of

the students reported to have spent less than ten

minutes each time they accessed the system, 58% of

them took less than five minutes. This raises specific

concerns regarding the usability of SEMINOLE.

Usability evaluation of the system was done at the

week of course by 86% of registered students.

Based on system data and user questionnaires, out of

79 posts, 53% were discussions initiated by students.

On average, posting was reported as an easy task to

perform in the system. Yet, almost 42% of

participating students never posted anything. Also,

few students reported not to have accessed archived

learning content and uploaded materials using

SEMINOLE. This translated into specific actions to

improve the usability of SEMINOLE’s interface.

Fourth, involving students as users into the

development cycle, proved to be useful and

resource-effective. They were asked to report what

they liked the most, the least and what technical and

physical conditions must be improved. 45% of

responses were given in the first evaluation session.

Short-term actions were taken to address some

issues (e.g. physical conditions, class rules, site

usability, communication issues, among others).

Those suggestions that could not be implemented

during this semester will be considered for a new

version of SEMINOLE (e.g. content organization

and structure, improving class dynamics, better user

manuals, etc.).

5 EXPECTED BENEFITS

Using our system brought forth several benefits.

First, two specific system administration tasks were

simplified: (a) edition / re-edition is done directly on

the platform, and (b) pre and post-production of

webcast classes (live and archive) had their time

significantly reduced from one working hour per

class hour to about 15 minutes per class.

For instructors, administrative tasks related to

quizzes and standard communication instruments

(e.g. program, course details, etc.) and events (e.g

meeting with students, orientation sessions, etc) are

also simplified. This raised other issues, such as the

effectiveness of mixing paper-and-pencil and online

quizzes, the need for a timely implementation of the

course plan and the amount of time required to

structure the learning experience once compared

with traditional ones. Nevertheless, it is

unquestionable that a home-grown and easy-to-use

e-learning system, which is also resource-effective

to build and maintain, can be a competitive tool to

deliver learning content.

For students, the dynamics of e-learning, during

SEMINOLE testing, created a “common

communication space” for those that actively

participated. In this way, online and offline

communication reinforce each other, if consistently

given. At this regard, frequency of use could be a

good indicator of adequate user engagement and

participation in online experiences. This will also

allow identifying or anticipating specific

performance issues and thus setting up the agenda

for offline meetings with students, when needed.

For system designers, structured user feedback

allowed to concentrate on critical issues negatively

affecting engagement and performance (resources

and time). Generally speaking, this will likely

WEBIST 2006 - E-LEARNING

284

improve development time, user acceptance levels

and help anticipate major changes in organizational

process and required competencies.

For process owners or managers, the results of

data analysis (self-reported and coming from system

database) will allow obtaining performance metrics.

They will be of great value to estimate the

usefulness, effectiveness, articulation of results and

strategies, return of investment of this kind of

initiatives, and support later business-driven

decisions. Within current business environment, this

will be a useful feature for all kind of organizations

to be competitive.

6 FUTURE WORK

As our knowledge of the fundamentals of e-Learning

requirements and praxis improves, we expect to

derive useful models to assist courseware developers

in developing high-quality resource-effective

learning materials and interactive courseware. Four

areas for future work were identified

First, we believe that collaboration and

personalisation are two essential requirements for

learning effectiveness. SEMINOLE lacks the ability

to: (a) allow students to specify personalization

features; (b) promote cooperation and collaboration

among students and instructors from local and

remote universities, and (c) foster collaborative work

among instructors and students In traditional

classrooms, students have their own materials

available; they chose what to take to a classroom and

in what way they want to take notes. Thus,

improvements could be added in order to: (a)

enhance the feeling of belonging and social presence

(Tu, 2002); (b) allow students to specify

personalized features, for example, each student

could be notified of the archived materials he/she

has not yet seen, and (c) enhance the publication

workflow. Currently, materials are prepared outside

the learning environment, and once they are ready

for publication they are added to the archives.

Again, as students should be able to share

documents and learning materials among

themselves, instructors could also be able to create

their materials online in a collaborative manner.

Second, from a technological standpoint,

SEMINOLE’s development will be directed towards

two main goals: (a) improving cost-efficiency, and

(b) increasing instructor and student immersion/

interactivity. The system’s interface and workflow

will be simplified by optimizing pre and post-

production processes. This way a “one man show”

scenario can be possible. Tests with low-cost and

highly available hardware have already begun. In

these tests, a camera was replaced by a webcam, and

the media-encoding server and instructor

presentation computer were replaced by a single

laptop. Additionally, multimedia content

distribution presents several challenges for

SEMINOLE: (a) webcast delay, and (b) network

bandwidth requirements. To face these challenges,

three options were analyzed. Replacing unicast

streaming with multicast streaming can reduce

bandwidth requirements (Multicast Streaming,

2005), but raises several issues when deployed in a

heterogeneous network environment since “a large

portion of the Internet is not multicast-enabled”

(Yeo, Lee & Er, 2002). Tunneling between

multicast-enabled networks (Francis, 2000)

potentially solves the bandwidth issue, but the live

media delay remains. Research in peer-to-peer

content distribution may solve both bandwidth

concerns and live content delay. This is based on the

principle that interested content receivers will be

willing to resend it (Yeo, Lee & Er, 2002;

Xu,

Xianliang, Mengshu & Chuan, 2005).

Third, planning standardized contents using

Sharable Content Object Reference Model

(SCORM) for multiple assets, such as presentations,

simulations or webcasts, is under way. Also,

building basic learning objects from previous raw

classes’ contents, through the process of

standardization, is another step. Then, with sets of

basic learning objects, we will develop SCOs

(Shareable Content Object). A SCO represents the

lowest level of granularity of learning resources that

can be tracked by a Learning Management System

(LMS) (Advanced Distributed Learning, 2004).

Each SCO will have a predetermined sequence and

can be seamlessly used for multiple course lessons.

Lessons are created from groups of SCOs that can

contain a concept, fact, procedure, process or

principle (Cisco Systems, 2003). Adding Learning

Objects Metadata (LOM) is also planned. LOM

makes SCO more reusable, since instructors can

search and find them easily. Issues resulting from

SCO usage will be analyzed, among them:

increasing or decreasing contents granularity;

choosing a set of basic learning objects to build a

SCO; selecting the right SCOs to build a lesson, and

finding an organization or navigation structure.

This approach, using standards like SCORM (or

others), brings advantages by making it possible to

use materials from different sources into the future

standard-compliant SEMINOLE. Changing the

learning environment will have no extra costs,

beside a new course deployment, because there will

be no need to change the SCO itself. This achieves

interoperability among user technologies. Standards

provide a common data model and by using a

standard-compliant SEMINOLE potentially assures

WEB-BASED SUPPORT FOR RESOURCE-EFFECTIVE E-LEARNING

285

that the course will be correctly delivered,

interactions logged and student grades saved.

Standards also provide a common packaging format

that ensures an easy storage and distribution process.

Moreover, building a repository of standardized

SCOs is also resource-effective, because reusability

increases.

Last, there is a need to rethink business models

for effective deploying ee-learning solutions. This is

required to develop cost-effective and sustainable

approaches to enhance existing curricula through

judicious use of IT. This should be the subject of

further studies.

7 CONCLUSIONS

We have presented our ongoing work towards a

web-based system for supporting resource-effective

e-Learning. The main advantages of this solution lie

in its cost-effectiveness and just-in-time approach to

content production and distribution. We achieved

this by using opensource tools and making

functionalities available based on user feedback and

system data analysis, while trying to respect

organizational culture, processes and idiosyncrasies.

Usability evaluation results showed that: (a)

SEMINOLE is easy to use for the main

stakeholders: instructors and students, and (b) e-

learning is a team-based process with a complex

nature due to its non-linear impact on expected

outcomes.

REFERENCES

Advanced Distributed Learning, (2004). Shareable content

object reference model (SCORM) 2004 overview.

Alexandria, VA: ADL Co-Laboratory.

Cisco Systems, Inc. (2003). Reusable learning object

authoring guidelines: How to build modules, lessons

and topics. San José, CA: Cisco Systems, Inc.

Dix, A., Finlay, J., Abowd, G., & Beale, R. (1998).

Human-Computer Interaction. Second edition.

Prentice Hall, England.

Dougiamas, M. y P. Taylor (2002) Interpretative analysis

of an internet-based course constructed using a new

courseware tool called Moodle, en: Goody, A., J.

Herrington y M. Northcote (eds.), Proceedings of the

2002 Annual International Conference of the Higher

Education Research and Development Society of

Australasia (HERDSA) Perth, Western Australia:

HERDSA, Inc.

ePresence Interactive Media (2004). Retrieved January 4,

2005 from http://epresence.tv/.

Francis, P (2000), Yoid: Extending the internet multicast

architecture. Retrieved November 10, 2005 from

http://www.icir.org/yoid/docs/

Moodle, Open Source CMS (2004). Retrieved October 17,

2004 from http://moodle.org/.

Multicast Streaming with Windows Media Services 4.1

(2005). Retrieved December 5, 2005, from

http://www.microsoft.com/windows/windowsmedia/se

rve/multiwp.aspx

Preece, J., Rogers, Y. & Sharp, H., (2002). Interaction

Design: Beyond Human-Computer Interaction. NY:

John Wiley & Sons. New York.

PCM - Produção de Conteúdos Multimédia (2004),

Lecture course notes (in Portuguese) IST. Retrieved

December 5, 2005, from http://immi.inesc-id.pt/pcm04

and http://immi.inesc-id.pt/pcm05.

Rentroia-Bonito, M. A., Tribolet, J.M., Jorge, J. &

Ghaoui, C. (2005). Just-in-time approach to virtual

learning: Arguing for cost-effective knowledge

dissemination in organizations. Knowledge-Based

Virtual Education: User-Centred Paradigms Pt. 1

(Studies in Fuzziness & Soft Computing S.) In Ghaoui,

C.; Jain, M.; Bannore, V.; Jain, L. (Eds.). Berlin:

Springer-Verlag and Heidelberg GmbH & Co. K.

Rentroia-Bonito, M. A., Jorge, J. & Ghaoui, C. (2006a).

Motivation to e-Learn within organizational settings:

An exploratory factor structure, International Journal

of Distance Education Technologies. In print.

Rentroia-Bonito, M. A., Figueiredo, F., Martins, A., Jorge,

J. & Ghaoui, C. (2006b). An Integrated Evaluation

Method for e-Learning: A Case Study. International

Journal of Interactive Technologies and Smart

Education, Special issue. In print.

Rosson, M. B. & Carroll., J. (2002). Usability

Engineering. Scenario-based Development of Human-

Computer Interaction. CA : Morgan Kaufmann

Publishers

Tu, C. (2002). The Measurement of Social Presence in an

Online Learning Environment. International Journal

on E-Learning. 34-45.

Yeo, C.K., Lee, B.S. & Er, M.H. (2002), A peering

architecture for ubiquitous IP multicast streaming,

ACM SIGOPS Operating Systems Review, Volume

36 Issue 3, ACM Press.

Xu, Z., Xianliang, L., Mengshu, H. & Chuan, Z. (2005) A

speed-based adaptive dynamic parallel downloading

technique, ACM SIGOPS Operating Systems Review,

Volume 39 Issue 1, ACM Press.

WEBIST 2006 - E-LEARNING

286