AN INVESTIGATION INTO THE REQUIREMENTS

for an e-Learning System

Yaen Y. Sofer, Steve B. McIntosh

School of Computer Science, Cardiff University, Queen's Buildings,

Newport Road, Cardiff, UK

Keywords: Learning environment, Distance learning, e-Learning, Web-based Technology, SSM, CPTM, Requirements,

Activities, Persona, Viewpoints, Business model.

Abstract: The learning environment where students of the same age group learn together instructed by a teacher, was

developed over the years and is known today as the traditional classroom. This traditional classroom may be

changed by using the latest Web-based technology to replace and/or support the learning process. These

new learning environments are accessible using the Internet as the main communication medium and by

other remote means such as CD-ROM, and video. Many aspects of the current use of these new

technologies reflect an approach to teaching and learning reminiscent of the “programmed learning” training

material of the 1970s. This paper uses Soft Systems Methodology (SSM) to construct a Consensus Primary

Task Model (CPTM) to analyse the requirements for a distance or e-learning system. In conducting the

analysis, we investigate the alternative methods proposed for the construction of a CPTM.

1 INTRODUCTION

Since the early days of education, people have

searched for appropriate ways to spread and share

their knowledge with their peers. In the 18

and 19

centuries, Jewish pupils used to study in small

basements instructed by the community Rabbi. This

study environment used to be called the ‘Heder’

(‘The Room’). The pupils used to study the Bible

and gained reading and writing skills.

According to Dr. Katz, Director of the School of

Education at Bar-Ilan University, The study of the

evolution of communication technology has

considerably influenced the development of distance

learning. Katz (Katz, 2001) categorises the stages of

the development of distance learning as three

generations: First generation: utilising traditional

printed material and communications via mail and

telephone. Second generation: audio recordings,

radio and television broadcasts. Both first and

second generation Distance Learning delivery

systems were designed primarily to produce and

distribute learning materials as efficiently as the

technology of the day permitted without any

attention being focused on the lack of interactive

communication between students and teachers.

Third generation: includes interactive video,

email, and world wide web (www) technologies,

learning activities via these Distance Learning

systems has been redefined to include teacher-

student interaction. Interactive video-conferencing

or interaction by way of on-line Internet-based

instructional and learning packages offer one-to-

many tuition in which teachers and students are able

to communicate synchronously thereby solving

instructional and learning problems in real time.

This may transport the student to a new cognitive

environment which motivates and stimulates the

student (Katz, 1998, 2000, 2001).

This evolution of the learning environment and

methods has arisen in response to technology. We

are unconvinced that the development has taken

place in a systematic way, through an analysis of the

requirements of the basis of the needs of various

stakeholders in the learning system. In general,

requirements elicitation is a complex and difficult

task, and there is evidence (see for example the

British Computer Society survey “IT Projects: Sink

or Swim”, January 2000) that failure to get the

requirements right is clearly associated with the

failure of IS/IT projects. In this context, failure may

include lack of user acceptance.

233

Y. Sofer Y. and B. McIntosh S. (2004).

AN INVESTIGATION INTO THE REQUIREMENTS - for an e-Learning System.

In Proceedings of the Sixth International Conference on Enterprise Information Systems, pages 233-237

DOI: 10.5220/0002604202330237

 SciTePress

A number of stakeholders, with different

viewpoints, can be identified in considering learning

systems. Their views may be surmised, and having

been identified will impact the perception of purpose

of the system to be designed. For example:

The “learner” has some motivation to learn, but

may be assumed to wish to do so at minimum cost in

time and effort.

The instructor may seek to transfer knowledge in

the most effective and efficient way.

The employer may wish to reduce costs by

delivering learning remotely, avoiding travel costs

and loss of production.

The infrastructure owner may wish to minimise

communications overhead.

The material owner may wish to protect

intellectual property rights.

Because the various stakeholders have different

views of the purpose of the system, requirements

elicitation implies the need to consider an

appropriate methodology. A methodology according

to Dr. Brian Wilson is a description of how to think

about the process of analysis prior to doing it. The

methodology can be described as a set of guidelines

which simulate the intellectual process of analysis

(Wilson, 2001).

The question of which methodology to choose is

itself problematic. It must reflect characteristics of

the problem, scope and compass of the methodology

and the skills and knowledge of the analyst, among

other considerations A valid first question to ask is,

is the situation considered as a ‘soft’ or a ‘hard’

problem? Wilson indicates that the design of a piece

of a software to meet a given specification is a

‘hard’ problem whereas the specification of

information requirements to meet business needs is a

‘soft’ problem particularly if the needs as specified

by potential users. He states that the assumption

upon which Soft Systems Methodology (SSM) is

based is that:

Whatever the nature of the organisations,

assume that the individuals within it are pursuing

purposeful activity.

Individuals may well be pursuing different

purposes, but they are not acting randomly. This

means that providing we can identify their purposes

and accommodate them, we should be able to

alleviate the problem of competing viewpoints.

2 THE ARGUMENT

According to Dr. Owston (Owston, 2000) the growth

of Web-based courses over the last several years has

been extraordinary. Despite the widespread adoption

of this new technology by educational institutions, it

seems that we know very little about Web-supported

pedagogy. Owston (Owston, 1997) cautions that

before embracing the innovation of this new

technology we need to be able to answer three

questions:

(1) Does the Web increase access to learning?

(2) Can the Web promote improved learning?

(3) Can increased access and improved learning

be attained without increasing the costs of

education?

Owston indicates that unless we have evidence

of satisfying these criteria we may be doomed to

promoting just another educational bandwagon

(Owston, 2000).

On the question of access, Owston suggests that

each of us may have a different interpretation of

what ‘access to learning’ means, although most will

agree that it means making education more

attainable for more people. Owston suggested that

this implies an increase in the availability of

educational opportunities for those unable to attend

formal classes (i.e. school, university, corporate

training etc) because of cultural, economic, or social

barriers. According to Owston, Web-based

educational methods can break down these barriers.

Owston suggests, however, that although the

Web breaks down the long-standing physical and

temporal barriers of access to education, it can create

new kinds of barriers for students. These include

shortages of computer hardware, malfunctions of

hardware, skills difficulties and bandwidth

problems.

Owston suggests that there are promising

indications that the Web is a viable means to

increase access to education. Evidence on how it can

promote improved learning is not as readily

available. In fact, there is debate in the instructional

design literature about whether there are any unique

attributes of media that can promote improved

learning (Clark, 1983, 1994; Kozma, 1991, 1994).

Owston suggests that the issue becomes further

complicated when the Web is used as a ‘tool for

learning’, as opposed to a medium for delivering

pre-determined content, which requires the users to

gain the skills needed to use the tool; this may cause

new barriers as mentioned above.

With regard to the cost of education, Owston

suggests that there are three main areas of cost for a

Web based course:

(1) Hardware and software - includes the Internet

connection itself and all necessary computer

hardware and Web related software required for

delivering as a course.

(2) Course development - includes planning the

course content and suitable pedagogy for developing

the Web resources associated with the course.

ICEIS 2004 - HUMAN-COMPUTER INTERACTION

234

(3) On-going course support – includes posting

new materials and removing dated materials,

verifying the validity of the links, improving the

layout and design, adding functionality and uses of

new technology.

Based on these assumptions, we can define a

purpose for a Web-based learning environment,

offering efficient, effective learning and reducing the

cost of education. Using an appropriate

methodology, we can derive system requirements.

3 SELECTING SSM AS THE

METHODOLOGY

The terms ‘hard’ and ‘soft’ are used frequently in

explanations of the Soft Systems approach, but first

we need to make clear the distinction between the

two. The terms are essentially comparative ones and

are used to distinguish between methods of

examination that address clearly defined problems

(techniques) and others that are used when the

problem is not clear at the outset. Here a preliminary

investigation is required to identify and select the

problem to be solved. The latter type of examination

applies to situations that are regarded as

unstructured or soft, inevitably involving people

working as individuals or groups towards some

organisational or other goal (Patching, 1990). This is

illustrated in Figure 1.

Figure 1: The Soft/Hard Division (Checkland and Scholes,

1999)

Checkland (Checkland, 1999) describes the use

of SSM in relation to ‘problem-solving’. He argues

that ‘Real-world problems’ are more of a case of

‘perceptions of problems’, this means that any

problem can be perceived differently by different

individuals or groups. Furthermore, Checkland states

that: “a fixed element in every problem situation will

be the existence of the role of ‘problem owner’,

occupied by those who perceive the problem. A

second fixed element will be the role, of the would-

be problem solver, the occupants of which wish to

tackle the perceived problem(s).”

SSM is concerned with defining what problem(s)

needs to be solved, clarifying the problems that exist

as a prerequisite for defining the options for

improvement. These raise the questions that the

designed system should give an answer to.

4 ROOT DEFINITION

According to both Checkland and Wilson, the Root

Definition (RD) is a way of trying to capture the

sense (root) of the purpose to be served. Like

differential equations, RDs do not exist in reality but

represent a precise framing of the system’s purpose,

achieved with the participation of available

stakeholders. The equivalent in real-world

terminology might be business objectives, mission

statement, specifications etc.

The next stage is to define the area of concern

more precisely (i.e. to synthesise the ‘Root

Definition’ (Checkland, 1999)). That will lead us

towards a well defined statement about the area of

concern, its activities and components. This may

represent a minimum that can be agreed in terms of

the domain of the real activity. It will offer people

who have an interest in the system the opportunity to

see what they are agreeing on and what has been left

out. As suggested by Checkland, the Root Definition

used as a statement of purpose in SSM can be

validated in terms of the mnemonic “CATWOE”

(Checkland and Scholes, 1999). This is illustrated

below.

According to the assumptions identified earlier,

we can define a Root Definition representing one

view of a learning system:

“A client owned system to provide appropriate

training to trainees by using the Internet as the main

communications medium and learning environment,

to provide an effective and efficient training solution

to allow trainees to take a course anywhere at any

time in order to reduce the training costs, while

maintaining the privacy of the system and complying

with training policy and directives, and while

learning from the operation of the total system”.

Customer: Not stated (payer of training costs)

Actor: Trainees

Transformation: Untrained trainees Æ

appropriately trained trainees

Weltanschauung: Using the Internet as the main

communication medium to deliver effective and

efficient training will reduce the training costs

AN INVESTIGATION INTO THE REQUIREMENTS FOR AN E-LEARNING SYSTEM

235

Owner: Client

Environment: Internet; training policy and

procedures

Using this RD allows us to develop a model of

the activities that the system must logically do, if it

is to be the system described in the RD. The words

in the RD allow us to make an initial assessment of

the likely subsystems in the overall model, with a

view to facilitating the development of a full model

at a relevant level of detail. The following candidate

subsystems:

• Provide effective and efficient training

• Assess training (needed to assure effectiveness)

• Comply with the client’s training policy and

procedures

• Assess the cost of training (necessary in order to

reduce costs)

• Maintain privacy

• Maintain knowledge base

• Monitor and control the system’s performance

(required by systems theory)

5 GUIDANCE AND METHODS OF

CONSTRUCTING THE

SYSTEM’S CPTM

The Consensus Primary Task Model (CPTM)

reflects an accommodation of stakeholders’

viewpoints. Each viewpoint may give rise to

different ideas on the transformation carried out by

the system, and different “Weltanschauungen”

, or

beliefs that underlay the purpose of the system.

The four methods for the construction of the

CPTM are:

A. mission-statement: this is the most defensible

method since the starting point is a ‘definition of

purpose’ arrived at by personnel in the situation

itself. The drawback of this method is that it uses

only one RD. As we saw from our root definition

earlier, even this suggests seven subsystems models

to consider. Using this method for construction will

not result in a model offering sufficient detail, which

will lead us to a CPTM model of limited utility.

B. W-decomposition: this is the most difficult

method to be used since the combination of the

resultant W-dependent models has to be based upon

a well-specified non-contentious RD and model.

Weltanschauung – German word, described by

Checkland and Wilson as world-view or viewpoints of

the system.

C. Wider-system Extraction: this method is

relatively easy to use once a wider-system model has

been derived. In our case we offer an

economic/effective online training system delivered

remotely using the Internet Technology. Our main

aim is to reduce the training costs on the one hand

and on the other hand to offer an efficient and

effective training process. To use the Wider-system

Extraction method of construction we should

examine the system Ws. In our case these are the

other methods of training (i.e. learning from books

and etc.); as this is not our main purpose we will not

use this method.

D. Enterprise Model Assembly: this is the

method most widely used by Dr Wilson in his

consulting and has proved the most acceptable from

the client ‘buy-in’ point of view. Of course, the

client must have some initial appreciation of the

status and purpose of the models being generated so

that acceptance can be pursued. The method, apart

from being based upon a very simple generic model

of any enterprise, is an appropriate method, given

widely differing Ws. With a little practice, it is a

relatively easy method, but relies on the ability to

construct a logically defensible conceptual model

from a set of RDs (Wilson, 2001).

Wilson argues that in a model of an enterprise

there will be a set of activities which represents its

core purpose, T–Core Transformation. There will be

other activities which facilitate, or support, this

process-S–Support. Since the enterprise is bound by

the limitations of its environment

, other activities

must exist which link its activities to the

environment -L–Linking, providing opportunities to

use the Internet Technology and other remote

methods. Finally in a managed enterprise there will

activities of planning, monitoring and control to

enable it to survive in a changing environment –

P,M,C–Planning, Monitoring and Controlling (in

this case, making sure that the online training

solution is effective and efficient) (Wilson, 2001).

6 CONCLUSIONS

The main purpose of modelling a system using this

approach is to derive as wide a range as possible of

the system’s functional and non-functional

In a “Human Activity System (HAS)”, the assignment of

a boundary is a subjective matter. Wherever the

boundary is considered to be, an “open” system

exchanges information with its environment.

ICEIS 2004 - HUMAN-COMPUTER INTERACTION

236

requirements, reflecting the differing views of

relevant stakeholders.

The list of activities and their logical

relationships derived from the system’s CPTM allow

us to investigate the system’s functional and non-

functional requirements. Choosing the appropriate

way of constructing the CPTM has an impact on the

utility of the resulting model, because of the

different assumptions underlying the approach.

Analysing the activities represented in the CPTM,

will give us the basis of the list of functions that the

system will be required to support, and should meet

the expectations of the stakeholders considered in

the analysis.

7 FURTHER WORK

In this research paper we have used Dr. Wilson’s

Enterprise Model in order to construct the system’s

CPTM. A number of unfinished questions were

raised during that process and require further

investigation and work. While the CPTM addresses

“what” the system has to achieve, it remains an open

question “how” to carry out each activity. This

question raises concerns about the non-functional

requirements, which are to some extent ignored in

the approach we have used.

One of our main concerns was the importance of

the system’s End-user. For consideration is what the

End-user seeks in the system. One of the proposed

solutions can be the use Cooper’s idea of the

Persona (Cooper, 1999) to describe the system’s

user. The Persona is an elastic imaginary user that

has identity (i.e. name) and will be addressed by

name (i.e. David) and not as ‘User’. As David is

determined by his set of characteristics we will be

able to know what David seeks in the system and

build a solution dedicated to David and his

colleagues. This will be approached by building a

Business Model that will indicate:

The need for the proposed solution (market)

The market sector for which the solution will be

aimed at (David and his colleagues).

The second question relates to information

analysis. The CPTM provides a list of the necessary

system activities, which we regard as functional

requirements. Analysing those requirements will

give us a list of Software based Functions that the

system needs to support, which will lead to ‘Data

Requirements’ which can be divided into:

Performance data (i.e. how do we know how

well we are doing each activity)

Operational data (i.e. what information do we

need in order to carry out the activity)

In terms of the development of an on-line

learning system, we believe that our approach will

lead us to a defensible set of functional

requirements. We propose to carry out the necessary

further analysis and implement the design arrived at.

With respect to our consideration of the candidate

approaches to the development of a CPTM, we need

to justify our confidence in the enterprise assembly

approach by constructing the CPTM using

alternatives, and determining whether there is any

substantive difference in the resulting models.

REFERENCES

Checkland, P., 1999, Systems Thinking, Systems Practice,

Includes 30-year retrospective, John Wiley and Sons,

Chichester, A5-A57,4-11,16,18,78,94,164,225-

227,314,199, 203.

Checkland, P., Scholes, J., 1999, Soft systems methodology

in action, John Wiley and Sons, Chichester 35,163.

Clark, R. E., 1983. Reconsidering research on learning

from media, Review of Educational Research, 53(4),

445-459.

Clark, R. E., 1994, Media will never influence learning,

Educational Technology Research and Development,

42(2) ,21-29.

Cooper, A., 1999, The inmates are running the Asylum,

why High-Tech Products Drive us Crazy and How to

Restore the Sanity, SAMS, 121-152.

Katz, Y. J., Yablon, Y. B., & Sagee, R., 2001, In:

Sponberg,H.,Lustigova, Z., & Zelenda, S. (Eds.),

Charles University, Prauge. 114-117.

Katz., Y. J., 2000, The comparative suitability of three

ICT distance learning methodologies for college level,

Educational Media International, 37 (1), 25-30.

Katz., Y. J., 1998, The relationship between distance

learning methods and satisfaction and achievement of

college students, In G. Davies (Ed.) Teleteaching '98:

distance learning, training and education, Austrian

Computer Society (OCG) and IFIP, 569-578.

Kozma, R. B., 1991. Learning with media, Review of

Educational Research, 61(2), 179-212.

Kozma, R. B., 1994. Will media influence learning?

Reframing the debate, Educational Technology

Research and Development, 42(2), 7-19.

Owston, R. D., 1997. The World Wide Web: A technology

to enhance teaching and learning?, Educational

Researcher, 26(2), 27-33.

Owston, R. D., 2000. Evaluating Web-based learning

environments: Strategies and insight,

CyberPsychology and Behavior, 3(1), 79-87.

Patching, D., 1990. Practical soft systems analysis,

Pitman, London, 40-53.

Wilson, B., 1990, Systems: Concepts, Methodologies and

Applications, John Wiley and Sons, Chichester.

Wilson, B., 2001. Soft System Methodology: Conceptual

Model Building and its Contribution, John Wiley and

Sons, Chichester, 73-75, 111-122, 133.

AN INVESTIGATION INTO THE REQUIREMENTS FOR AN E-LEARNING SYSTEM

237