Notesale: Turn your study into money

Document Preview

Extracts from the notes are below, to see the PDF you'll receive please use the links above

---

See discussions, stats, and author profiles for this publication at: https://www
...
net/publication/228906207

An overview of systems design and development methodologies with regard
to the involvement of users and other stakeholders
Article · January 2003
CITATIONS

READS

23

2,518

2 authors:
Shawren Singh

Paula Kotzé

University of South Africa

University of Pretoria

21 PUBLICATIONS   94 CITATIONS   

131 PUBLICATIONS   1,106 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Human-Computer Interaction View project

Designing with and for children View project

All content following this page was uploaded by Paula Kotzé on 21 May 2014
...


SEE PROFILE

An Overview of Systems Design and Development Methodologies with
Regard to the Involvement of Users and Other Stakeholders
SHAWREN SINGH AND PAULA KOTZÉ
University of South Africa
________________________________________________________________________________________________
Systems development methodologies have been developed with the purpose of designing and developing effective information systems
...
One general critical flaw we have identified in these processes is that the interface design is not effectively
conducted or implemented, but neither are the many other stakeholder issues addressed
...

Categories and Subject Descriptors: D
...
2 [Information Interfaces and
Presentation]: User Interfaces - Theory and methods
General Terms: Design, Standardization
Additional Key Words and Phrases: Human-computer interaction, Object-Orientation, Systems Development Life Cycle

________________________________________________________________________________________________
1
...
For this reason issues
and methods from the field of human-computer interaction (HCI) affect the overall process of software engineering
(SE) tremendously
...
There are also major gaps of communication between the
HCI and SE fields – the methods and vocabulary being used in each community are often foreign to the other
community
...

The goal of the development of any information system (IS) is to make human endeavour easier and safer
...
[2001](structured systems development life cycle authors) define an IS as an arrangement of people, data, process,
information presentation, and information technology that interact to support and improve day-to-day operations in a
business as well as support the problem-solving and decision-making needs of management and users
...

[2002] (object-oriented development authors) have a much narrower definition of an IS: an IS is a collection of
interrelated components that collect, process, store, and provide as output the information needed to complete a business
task
...
There is, however, an inherent defect in both approaches in that both the
structured systems development life cycle (SDLC) and object-oriented (OO) approaches tend to ignore the human
aspects during system development
...
In fact the critical flaw, of the Satzinger et al
...

This paper addresses this issue and compares various systems development methodologies
...
In Section 3 we discuss information systems development
...
Section 5 assesses the methodologies
given the issues identified in Section 2
...


________________________________________________________________________________________________
Author Addresses:
S
...
ac
...

P
...
ac
...

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that the copies are
not made or distributed for profit or commercial advantage, that the copies bear this notice and the full citation on the first page
...
Abstracting with credit is permitted
...

© 2003 SAICSIT
Proceedings of SAICSIT 2003, Pages 37 – 47

38

•

S Singh and P Kotzé

Economic forces
Political forces

Domestic- environment (uncontr olable)
Competitive forces

(Controlables)
Price
Cultureal forces

Product
IS

Business
Channels of
Promotion
distribution

Level of
technology
Economic climate
Geography

Structure of distribution

Foreign-enviroment (unc ontr ollables)

Figure 1
...


THE BUSINESS ENVIRONMENT

The typical environment in which contemporary businesses operate, as illustrated in Figure 1, is complex and fragile
...
Outside those spheres are uncontrollable factors that affect the business
organisation
...
For example, in the outer most
sphere, the following aspects are beyond the direct control of the business, but severely impacts on its operation:
— Cultural and language forces: This can be illustrated with an example
...
The name
Polo [Paroz 1988] sounds inoffensive in English, but in Sotho (a local African language) the word Polo means
penis
...
If a young African female owned a Polo Playa, it could add a rather negative
connotation to her image
...

— Geography: The world has become smaller with the development of the Internet
...

— Levels of technology: The average South African does not have access to the Internet
...

All these aspects have some human component or connotation
...
With the development of the correct IS
that keep the human element in mind, an organisation will have a competitive advantage over other business
...
This system
is rather inflexible and largely closed to the wider company or world
...
Traditional Approach [Shelly 2003]

•

39

Figure 3
...
We can also see from the
Figure 2 that the user interface was the domain of the IT department, be it command driven or graphical user interface,
the IT staff would somehow get your information to you
...
In Figure 3 the system
and all stakeholders interface with each other
...
The domain for the development of the IS and ultimately the user interface is, however,
still that of the IT department
...


COMPARING INFORMATION SYSTEMS DEVELOPMENT TECHNIQUES

There are various development methodologies that are used in developing ISs, some more conventional than others
...
The
proponents of HCI and interaction design propose lifecycle models with a stronger user focus than the conventional
approaches
...

Before we look at these approaches we need to argue about the method of comparing and assessing the various
methodologies
...
Therefore
before we compare the two approaches, we would refer to these possible problems and how we tried to overcome them
in our approach
...
The same term often
has different meanings in different approaches
...
We did
not compare terminology, but rather how these approaches addresses the elements identified in Figure 3
...
Any comparison would necessarily be
restrictive in the information it reviews
...

— Using an inappropriate framework with which to conduct the evaluation
...
e
...
We did not quantify issues, but merely identified if a particular issue was addressed or not
...
Although we looked at certain terms, we did not limit our assessment to that terms, but rather
analysed the entire methodology in how it addressed the human issues
...
The definition of ‘methodology’ should
be clearly stated within the introduction to any methodology comparison
...
We did
not look at definitions per se, but rather to whether the human elements were addressed in generic terms
...
This can lead to studies showing a number of methods as being relatively equal in support from
a numerical standpoint (the number indicating the method ‘addresses’ certain topics, when in reality this is not the
case)
...
We
did not add a summative value to these issues
...
Since few (if
any) of the comparisons cite page references for where a particular methodology comparison item (e
...
a term,
concept, or example) is found in the methodology under review, it is difficult, if not impossible, to verify the
Proceedings of SAICSIT 2003

40

•

S Singh and P Kotzé

accuracy of these methodology comparisons
...
We have to acknowledge that methodologies are always in a
state of flux
...

4
...
[2001] mentions people as part of their definition
...
Thus if the key components such as humans and
the IS do not communicate effectively with each other, this business organisation is bound to fail
...
The aim of
all these methodologies is to design effective and efficient ISs
...
1
Traditional systems development approaches
Traditional (structured) development approaches include methodologies such as Structured Analysis and Design
Techniques (SADT) [Ross 1985], The Yourdon Systems Method (YSM) [Yourdan 1993], Specification and Description
Language (SDL) [Belina and Hogrefe 1989], Information Engineering and Jackson System Development (JSD)
[Jackson 1983], the Dennis and Wixom Approach [Dennis and Wixom, 2000], et cetera
...

Most of these development approaches follow the waterfall approach or is an iterative variation to the waterfall
approach
...

4
...
1 The Dennis and Wixom Approach
The most contemporary of the structured development approaches, namely the Dennis and Wixom Approach [2000],
consists of the following phases:
— Planning (why build the system): Identifying business value, analyse feasibility, develop work plan, staffing the
project, and control and direct project
...

— Design (how will the system work): Physical design, architecture design, interface design, database and file design
and program design
...

— Dennis and Wixom [2000] describe their user interface design aspect as consisting of:
— Develop use scenarios (generic user)
...

— Design interface standards
...

— Design user interface
...

Although included in the Dennis and Wixom approach, these steps are conducted much too late in the design phase
...
The structured
development approach therefore relegates the design of the user interface (and the human related issues) to the design
phase of the development life cycle – to quite late in the development process
...

We assessed the Dennis and Wixom approach against each of the components depicted in Figure 3
...
1
...

4
...
Many object-oriented designers and developers, for
example, seem to focus almost entirely on programming language issues
...
For example, to fully leverage IBM's
design approach, IBM assumes the target language to be Smalltalk
...

Another sector of the object-oriented community is interested in formality and rigour
...
They view object-oriented software engineering as
primarily an engineering process with well-defined deliverables
...


Proceedings of SAICSIT 2003

An Overview of Systems Design and Development Methodologies

•

41

There are various OO methodologies such as Objectory [Jacobson 1994], Unified Modelling Language [Bahrami
1999], Coad and Yourdon Method [1991], Booch Method [1987], Object Modelling Technique [Rumbaugh et al
...

Although diverse in approach most object-oriented development methodologies follow a defined system
development life cycle, and the various phases are intrinsically equivalent for all the approaches, typically proceeding
as [Schach 2002]: requirements phase; OO analysis phase (determining what the product is to do) and extracting the
objects; OO (detailed) design phase; OO programming phase (implementing in appropriate OO programming
language); integration phase; maintenance phase; and finally retirement
...

We analysed three OO methodologies: The Rumbaugh, Coad and Yourdan, and IBM approaches and their
relationship to the aspects illustrated in Figure 3
...

4
...
1
The Object Modelling Technique
The Rumbaugh et al
...
The model is expressed
in terms of objects and relationships, dynamic control flow, and functional transformations
...
The analysis phase
has the following sub-phases: write or obtain an initial description of the problem (problem statement); build an
object model; develop a dynamic model; construct a functional model; and verify, iterate, and refine the three
models
...
There are several canonical
architectures that can serve as a suitable starting point
...
[1991] the OO paradigm
introduces no special insight into system design
...

Decisions are made that are necessary to realise the system without descending into the particular details of an
individual language or database system
...

We assessed the Rumbaugh approach against each of the components depicted in Figure 3
...
2
...

4
...
2
The Coad and Yourdan Model
The Coad and Yourdon [1991a, 1991b] model consists of two phases, the OO analysis and the OO design phases:
— The OO analysis phase identifies and defines classes and objects which directly reflect the problem domain and the
system’s responsibilities within it
...

— The OO design phase identifies and defines additional classes and objects, and reflects on the implementation of
the requirements
...

In the Coad and Yourdon [1991b] model there is therefore an active component for the design of the interaction
between human and machine
...

We assessed the Coad and Yourdan approach against each of the components depicted in Figure 3
...
2
...

4
...
3
The IBM model
The IBM model [1990a, 1990b, 1990c, 1999] consists of two phases, the OO design phase and the design the
business model phase:
— The OO design phase has the following sub-phases: define business transactions; capture the user's model; specify
the objects; build the view; and code and test the view
...

Proceedings of SAICSIT 2003

42

•

S Singh and P Kotzé

Figure 4
...
1987]

The model therefore includes two specific user-oriented phases, i
...
the capture the user’s model phase and build the
view phase
...

— Design direct manipulation actions; decompose the model objects; define instance variables; and design the
windows
...
Table 1 (section 4
...
3)
summarises our findings and will be placed in context in Section 5
...
3

HCI focused life cycle approach
The HCI proponents aim to focus more on the human and end-user aspects
...
With the
widespread introduction of information and communication technology in our everyday lives, most computer users
today have limited computer experience, but are expected to use such systems
...
One
broad dimension of usability is how easy to learn the user interface is for novice and casual users [Mayhew 1999]
...
Often, novice users expect to walk up and use an application as easily as
they use the telephone or a car [Turban 1995]
...
e
...
Another usability dimension is how easy to use (efficiency, flexibility, powerfulness,
etc
...
Expert users are therefore usually most concerned with ease of use [Dennis and Wixom 2000]
...

Williges et al
...
In their model interface design drives the whole
process
...
[2002] suggest a simple lifecycle model, called the Interaction Design Model, consisting of
identifying needs/establish requirements; evaluate; build an interactive version; and (re)design
...
Other lifecycles models that focus on HCI aspects include the Star
Model of Hartson and Hix [1989], the Usability Engineering Lifecycle of Mayhew [1999], and Hackos and Redish’s
model [1998]
...

Although varying greatly as far as individual phases are concerned, they are all based on an iterative system
development approach and essentially contain more explicit main phases than the structured and OO systems
development approaches
...
Only time
will tell whether or not this is a cost-effective strategy
...
The Williges et al
...


Proceedings of SAICSIT 2003

An Overview of Systems Design and Development Methodologies

•

43

Project
Start

Iterate

Corporate objectives

long-range goals

decisions on markets

feasibility/cost decisions

User and task analysis

users

users’ goals and tasks

users’ environments

Iterate

Technology decisions

platforms supported

architecture

Task model

users profiles

task analyses

environment profiles

Iterate

Iterate
Systems analysis

dataflows

objects

User and task analysis

Use model

Data model
Iterate

Iterate

Paper prototype
Iterate

Iterate
Usability testing

Prototype with dataflow
and interface
Iterate

Iterate
Usability testing

Implementation of design
Iterate

Iterate
Function testing

Usability testing

Release

Figure 5
...
3
...
[1987]
...


Proceedings of SAICSIT 2003

44

•

S Singh and P Kotzé

no

yes

Design

yes

yes

Implementation

yes

yes

attempts

attempts

System design

no

Object design

no

not
involved
not
involved

The Coad and Yourdan Model
Analysis

no

attempts

Design

yes

attempts

The IBM model
OO design phase

yes

attempts

The OO methodologies
4
...
1
The Rumbaugh model
Analysis phase

4
...
2

4
...
3

Government

Analysis

IT Department

yes

Suppliers

Traditional systems development approaches
4
...
1
The Dennis and Wixom Approach
Planning
no

Customers

Internal Users

Approach

UI Component

Section No
...


not
involved
not
involved
not
involved
not
involved

not
involved
not
involved
not
involved
not
involved

actively
part of
actively
part of
actively
part of
actively
part of

not
involved
not
involved
not
involved
not
involved

not part
of
not part
of
not part
of

not part
of
not part
of
not part
of

actively
part of
actively
part of
actively
part of

not part
of
not part
of
not part
of

not part
of
not part
of

not part
of
not part
of

actively
part of
actively
part of

not part
of
not part
of

not part
of
not part
of

actively
part of
not part
of

not part
of
not part
of

actively
part of
actively
part of
actively
part of

attempts

actively
part of
actively
part of
actively
part of
actively
part of

attempts

The design the business model
phase
HCI focused life cycle approach
4
...
1
Williges et al
Initial Design

no

not part
of

not part
of
not part
of

yes

yes

attempts

attempts

Formative Evaluation

yes

yes

attempts

attempts

Summative Evaluation

yes

yes

attempts

attempts

Hackos and Redish Approach
Systems development

no

attempts

attempts

attempts

Interface design

yes

yes

attempts

attempts

Design and implementation

yes

yes

attempts

no

Testing phase

yes

yes

attempts

no

4
...
2

attempts
attempts

attempts
no
no

Table
...
Methodology Matrix

— The summative evaluation stage consists of the following phases: operational software interface; benchmarking;
formal experimentation; and a feed-forward results phase
...
Table 1 (section 4
...
1)
summarises our findings and will be placed in context in Section 5
...
3
...
Figure 5 shows how the usability analysis fits into both the design of the interface and the design
of the underlying system
...
The systems development phase and
interface design phase are conducted in parallel:
— The systems development phase is concerned with the overall system and interaction with outside stakeholders
...

— The interface design phase is an in-depth analysis of the design of the interface for the proposed system
...

— The design and implementation phase is a marriage of the above two phases to produce the final system
...

The testing phase tests functionality and usability of system
...

We assessed the Hackos and Redish approach against each of the components depicted in Figure 3
...
3
...

5
...
User interface concerns are ‘mixed in’ with wider
development activities
...
Either HCI is ignored, or it is relegated to the later
stages of design as an afterthought
...
If HCI is ignored then there is a
good chance that problems will occur in the testing and maintenance stages
...
In either case the cost of introducing ‘usability’ issues will rise,
the later one postpones it in the development cycle
...
The
components of Figure 3 only partially maps on to this approach, with no reference to the customers, suppliers, the
IT department specifically, or the governmental issues
...
model [1991] there is no special consideration given for the design of the user interface or
any of the other components reflected in Figure 3
...
This model is a partial fit onto Figure 3
...

— The IBM model considers the users in the development of the system, however Figure 3 is still only a partial fit
onto this model
...

It is clear from the above that there are several missing components in all these SDLCs
...
The same applies
for the IBM approach, but the process is much shorter
...
Although Rumbaugh actively employs use case scenarios,
get users involvement in systems design, it does not map directly into the system user interface design
...
Both approaches also ignore the greater business environment as shown in Figure 1
...

Our findings are a confirmation of the work of Monarchi and Puhr [1992]
...
But there is, as yet, little
consensus or standardization among these new development techniques
...

— Interface objects are associated with the user interface
...

— User interfaces are a prevalent and integral part of systems development today and should be a part of any software
analysis and design methodology
...
Iivari [1991]
recognizes the absence of interface objects in other analysis methods and includes them in his framework for
identifying objects
...

Eleven years on this still applies
...
The HCI focused models attempt to do this, but they all also still suffer major shortcomings as
illustrated in Table 1
...

When we consider Table 1 with regards to the HCI focused development models we find that:
— Williges et al
...

— The Hackos and Redish model [1998] as depicted in Figure 5, seems to be a most comprehensive model we
assessed
...
Hackos and Redish are however silent on this issue and does not elaborate
what they mean by corporate objectives
...
In fact if this is the case the model maps onto Figure 3 and considers
most of the aspects addressed in Figure 1
...

The shortcomings of all the methodologies are therefore related to the complexity of the wider environment
introduced by the issues highlighted in Figures 1 and 3, and how these aspects should inform the systems development
process
...
Both the traditional SDLC and OO approaches fall short on the issue of human aspects and
stakeholder involvement
...
Which one
faired the best? Although the Williges et al [1987] and Hackos and Redish [1998] approach focussing on the user goes a
long way in achieving this, several shortcomings can still be identified
...

6
...

The authors of this paper recently attempted to submit a paper to an international conference
...
The instructions were downloaded from
the conference web site
...
Upon arrival of the closing date for submission of papers, the authors in true fashion (like most of the other
authors) left it to the last minute
...
The authors followed all the instructions religiously, managed to create a user account, password
and then it was time to upload the paper
...
The instructions clearly stated that papers that did not meet all the criteria, including the two versions of
the paper will be disqualified
...
It was too late when the organisers realized what had
happened and had to e-mail all corresponding authors with apologies, and change the business rules midway
...
Some way in the development processes something went wrong
...
The supplier provided an offthe-shelf system and did not customise the data model
...
The question on all developers’ lips is: which development methodology did they
follow? Nobody is saying at this stage, and given our investigation it is hard to tell which would have been the most
beneficial, and solved the problem before implementation
...


•

47

CONCLUSION AND PROPOSAL

In order for IS development to stay relevant and deliver systems fitting the demand of the current business environment,
our research indicates that there is a dire need for the establishment of a unified process to the development of ISs,
including all of the components identified in Figure 3
...

We believe that many of the shortcomings of the development models could be catered for by making the end-user
of the system a primary element in the entire process, and include explicit guidelines for the inclusion of other external
issues such as laws and regulations, human rights issues (including accessibility), the abilities and skills of the human
resource complement of the IT department, the supplier chain and availability of technology, et cetera
...
g
...
; the user
context (geographical, cultural, socio-economic, educational, etc
...

We further argue that most of the stakeholders identified in Figure 3, should ideally be involved in the formative and
summative evaluation of the proposed and delivered system
...
Extensive further research would be required to establish processes and procedure to achieve this
...


REFERENCES

BAHRAMI, A
...
Object-Oriented Systems Development: Using the Unified Modeling Language
...

BELINA, F AND HOGREFE, D
...
The CCITT-specification and description language SDL
...

BOOCH G
...
Software Engineering with Ada, 2nd ed
...

COAD P
...
1991a Object-Oriented Analysis
...

COAD P
...
1991b
...
Englewood Cliffs, NJ: Yourdon Press Computer Series
...
and WIXOM, H
...
2000
...

VANDERDONCKT J and HARNING M
...
2003
...
Interact 2003
Workshop
...
interact2003
...
html
...
J
...
J
...
User and Task Analysis for Interface Design
...

HARTSON, H
...
1989
...
ACM Computing
Surveys 21(1): 5-92
...
1990a
...
Raleigh, North Carolina: IBM International Technical Support
Center
...
1990b
...
Raleigh, North Carolina: IBM International
Technical Support Center
...
1990c
...
Boca Raton, Florida: IBM International Technical
Support Center
...
1999
...
Raleigh, North Carolina: IBM International Technical
Support Center
...
1991
...
IEEE Trans
...

JACOBSON, I
...
Object-Oriented Software Engineering: A use Case Driven Approach
...

JACKSON, M
...
Systems Development
...

MANDEL, T
...
The Elements of user interface design, Wiley, NY
...
J
...
The Usability Engineering Lifecycle: a practitioner's handbook for user interface design, Morgan Kaufmann, San Francisco
...
G
...
A Research Typology for Object-Oriented Analysis and Design
...
35, No
...
35 - 47
...
R
...
1988
...
Morija Sesuto Book Depot
...
Lesotho
...
, ROGERS, Y
...
2002
...

RICKS, A
...
1993
...

ROSS, D
...
Application and extension of SADT
...

RUMBAUGH, J
...
, PREMERLANI, W
...
, AND LORENSEN W
...
Englewood
Cliffs, NJ: Prentice-Hall
...
W
...
B
...
D
...
Systems Analysis and Design in a Changing World, Thomson Learning, Canada
...
W
...
Object-Oriented and Classical Software Engineering, McGraw Hill, Boston
...
J and ROSENBLATTH
...
2002
...
Thomson Learning, Canada
...
1993
...
http://www
...
com/pub/mcr
...

TURBAN, E
...
Decision Support and Expert Systems: Management Support Systems
...

WHITTEN, L
...
, BENTLEY, D
...
and DITTMAN, C
...
2001
...

WILLIGES, R
...
, WILLIGES, B
...
and EIKERTON, J
...
In Handbook of Human Factors(Ed, Salvendy, G
...
1414-1449
...
1993
...
Prentice-Hall, Englewood Cliff, NJ

---