CRYSTALLIZATION OF AGILITY
Back to Basics
Asif Qumer
Facutly of Information Technology, University of Technology, Broadway, Sydney, Australia
Brian Henderson-Sellers
Facutly of Information Technology, University of Technology, Broadway, Sydney, Australia
Keywords: Agility, Traditional software development, Agile software development.
Abstract: There are a number of agile and traditional methodologies for
software development. Agilists provide agile
principles and agile values to characterize the agile methods but there is no clear and inclusive definition of
agile methods; subsequently it is not feasible to draw a clear distinction between traditional and agile
software development methods in practice. The purpose of this paper is to explain the concept of agility in
detail; and then to suggest a definition of agile methods that would help in the ranking or differentiation of
agile methods from other available methods.
1 INTRODUCTION
Traditional plan-based software development
methods work well if the project requirements are
fixed; but are often considered slow and insensitive
when the project requirements are being changed
frequently (Paetsch et al., 2003). The concept of
agile software development evolved when people
argued that the traditional software development
approach often fails to produce valuable software
products in certain situations; consequently, it was
argued, new software development methods were
needed (Beck, 2000). Furthermore Nandhakumar
and Avison (1999) point out that traditional software
development methods are too mechanistic to be used
in detail (Abrahamsson et al., 2002). Such concerns
have encouraged software engineering practitioners
to develop new approaches to software development,
including lightweight and agile methods.
Cockburn (2002) defines the core of agile
m
ethods as “the use of light-but-sufficient rules of
project behaviour and the use of human-and
communication-oriented rules” (Abrahamsson et al.,
2002). The Agile Manifesto (2005) provides agile
principles and agile values that qualitatively
characterize the agile methods, but there is no
precise and comprehensive definition of agile
methods, per se. Conboy and Fitzgerald (2004) point
out that the agile methods are significantly
dependent on the principles embodied in the Agile
Manifesto (2005) and in agile values; and there is no
inclusive definition of agile methods.
The objective of this paper is to describe in detail
th
e concepts of agility (basic elements of agility)
that have been identified in our current research; and
then to propose a more formal definition of agile
software development methods in terms of the
underlying concepts of agility, existing agile
software development methods and agile principles
(AgileManifesto, 2005). This paper has three
sections. Firstly, it describes the concept of agility
and its application. Secondly, it gives an overview of
agile software development methods: Extreme
Programming (XP) (Beck, 2000), Feature Driven
Development (FDD) (Palmer & Felsing, 2002),
Adaptive Software Development (ASD) (Highsmith,
2000), Dynamic Software Development Method
(DSDM) (DSDM, 2003) and Scrum (Schwaber &
Beedle, 2002) and building on both the definitions of
agility and their manifestations in current agile
methods, it proposes a definition of what should be
the essence of an agile software development
method. Finally, the paper concludes with a
discussion of options for future research.
121
Qumer A. and Henderson-Sellers B. (2006).
CRYSTALLIZATION OF AGILITY - Back to Basics.
In Proceedings of the First International Conference on Software and Data Technologies, pages 121-126
Copyright
c
SciTePress
2 WHAT IS AGILITY?
The concept of “agility” conceals many facets such
as nimbleness, suppleness, alertness, responsiveness,
swiftness and activeness – yet it is difficult to give a
rigorous or complete definition of agility. Agility
may be taken as the demonstrable ability of anything
that is capable of adapting to changes quickly and
allowing anything to occur whenever it is required –
and to do so with flexibility. According to Dove
(1997), agility is a very seductive word, evidencing
confusion for many with immediate and personal
definitions. Hence, a clear, technical understanding
of this concept needs to be crystallized.
2.1 Attributes of Agility
The concept of agility is not new; to understand the
concept of agility, we first need to study the
underlying concepts of flexibility, speed, leanness,
learning and responsiveness. Flexibility is the ability
to respond to the expected change whereas leanness
accentuates lower cost, reduced timeframe and
quality production (Dove, 1997). According to
dictionary definitions, flexibility is the ability to
adapt to change; speed characterizes rapid and quick
behaviour; leanness refers to compactness and
tidiness; responsiveness refers to life, reaction and
sensitivity; and learning refers to knowledge and
improvement.
2.2 Existing Definitions of Agility
Wong and Whitman (1999) argue that agility refers
to the effective response to rapid and unexpected
change with flexibility, which is a characteristic of
agility (Table 1). This implies adaptability and
versatility in the domain in order to respond to
unexpected changes. Conboy and Fitzgerald (2004)
propose the definition of agility as: “the continual
readiness of an entity to rapidly or inherently,
proactively or reactively, embrace change, through
high quality, simplistic, economical components and
relationship with its environment”. These two
definitions seem to overlook two important factors
of agility: the learning factor and factors external to
the domain and the environment. The learning factor
demonstrates the capability of an agile entity that
improves over a period of time as it gains in
experience and acquires knowledge from its internal
and external environment (Henderson-Sellers &
Serour, 2005). Boehm and Turner (2004a; 2004b)
assert that “agility applies memory and history to
adjust to new environments, react and adapt, take
advantage of unexpected opportunities, and update
the experience base for future”. This definition
incorporates the learning factor but seems to have
only a vague concept of environment since it
overlooks the external environment factor. An agile
entity should consider such external factors that may
affect its working.
Table 1: Attributes of agility.
Features Wong &
Whitman
(1999)
Conboy &
Fitzgerald
(2004)
Boehm &
Turner
(2004a;
2004b)
Flexibility X X X
Speed X X
Leanness X
Learning X
Responsiveness X X X
2.3 Comprehensive Definition of
Agility
We suggest a definition of agility in the light of
agility concepts and existing definitions proposed by
different researchers (Section 2.1). Indeed, we have
already applied and tested this definition (which is a
basic requirement of a science) to measure the
degree of agility of two well-known agile methods
(Qumer & Henderson-Sellers, 2006b and Section
4.3). This definition may be used to measure the
degree of agility of any method or technique, not
only so-called agile methods. For example, Software
Development Life Cycle (SDLC) (a.k.a. “waterfall”:
Royce, 1970) or Rapid Application Development
(RAD) may be assessed for agility since we believe
it is possible for any method to encompass some
degree of agility, which may be ranked from weak to
strong. This is an independent definition of agility
(yet in the light of above definitions) that defines the
concept of agility in terms of flexibility, speed,
leanness, learning and responsiveness; and covers
the inadequacy of existing definitions.
Our proposed definition is as follows: “Agility is
a persistent behaviour or ability of a sensitive entity
that exhibits flexibility to accommodate expected or
unexpected changes rapidly, follows the shortest
time span, uses economical, simple and quality
instruments in a dynamic environment and applies
updated prior knowledge and experience to learn
from the internal and external environment.”
Here, we justify and expand upon this
definitional statement, which has five facets. In
future we will evaluate this definition more
extensively as we proceed further in our research.
ICSOFT 2006 - INTERNATIONAL CONFERENCE ON SOFTWARE AND DATA TECHNOLOGIES
122
Flexibility (FY)
Flexibility is the ability or behaviour (flexible) of an
entity that allows adapting to changes whenever it is
required. A method or phase in a method may
demonstrate flexibility by accommodating expected
or unexpected changes.
Speed (SD)
Speed of an entity characterizes rapid and quick
behaviour to get to the desired destination or to
achieve goals. A speedy method may help to show
the results quickly by following a specific approach.
Leanness (LS)
Leanness refers to compactness and tidiness. A lean
method gives the desired quality output,
economically, in the shortest possible time frame by
applying simple and quality means of development.
Learning (LG)
Learning refers to knowledge and improvement and
is an indispensable ability of an entity, which is
achieved primarily by using up-to-date knowledge
and experience, gained from previous practices. A
learning method shows continuous improvement
over the period of time.
Responsiveness (RS)
Responsiveness refers to life, reaction and
sensitivity. A responsive method is method that does
not remain silent when response is required in
different situations
3 ANALYSIS OF PROPOSED
DEFINITION OF AGILITY
This section discusses and analyzes the proposed
definition of agility in the context of a software
development method. We will discuss to see which
elements may affect the five attributes of agility and
help to classify a software development method as
flexible, speedy, lean, learner and responsive. A
method will be classified as flexible if we can
change, delete or add new practices (software
development) in the method during software
development; speedy, if it produces workable code
in the form of small increments; lean, if it takes
minimal possible timeframe and resources to
produce such increments; learning, if it improves
primarily by using up-to-date knowledge and
experience gained from previous practices and
feedback mechanisms (feedback loops in an iterative
development) in a dynamic environment (where
things are not fixed and are handled as we progress
towards our targets); responsive, if it responds to the
questions (asked) by responding to its internal and
external entities
. A question may be asked of a
method: “When and how to code the design?” The
method should have a phase in which different
practices may be available to produce the code from
design. The following sections present the analysis
of the agility in more detail.
3.1 Agility Priority Patterns (APP)
All the five attributes of agility are equally important
but we may add weights to show the priority among
the agility attributes according to the specific
situation (project). There are five attributes;
therefore we can use the priority weights value from
1 (minimum priority) to 5 (maximum priority). The
‘null’ value weight will be used if the specific agile
attribute is not present in the agile entity and, as a
result, there is no point of assigning weight to that
attribute. Two or more agility attributes may have
the same priority in some specific pattern. Table 2
presents the different agility priority patterns (APP)
as an example. We will discuss APP in detail in a
future paper.
Table 2: Agility Priority Patterns.
Attributes FY SD LS LG RS
Priority Pattern 1 4 3 3 null 5
Priority Pattern 2 1 3 5 2 4
Priority Pattern 3 1 5 5 4 2
3.2 Application of Agility (Testing
the Definition)
Agility demonstrates attributes that may be applied
to any object (e.g. Organizations, Systems, Methods,
Processes, Software and Documents) to make them
agile. Agility, measured in terms of the five
variables described above (flexibility, rapidness,
leanness, responsiveness and learning) may exist to
varying degrees in an object at some specific level or
lifecycle phase. For example, a software
development method may encompass agility at the
design phase level, planning phase level or at the
requirements engineering phase level (labeled LA) –
but not necessarily all three. We characterize the
degree of agility (DA) for each of these
phases/levels as the fraction of the five agility
variables that are encompassed and supported. If all
five variables are supported, then we categorize the
level and the object has possessing full agility (FA).
CRYSTALLIZATION OF AGILITY - Back to Basics
123
If 1-4 variables are supported, we label it partial
agility (PA). We have applied attributes of agility to
measure the degree of agility of both XP and Scrum
at process level and practices level (see Qumer and
Henderson-Sellers, 2006b and results for the degree
of agility of both XP and Scrum summarized here in
Section 4.3 for reference). In future, we will apply
definition of agility to measure the degree of agility
of other available agile and non-agile methods.
The following equations may describe the
application of agility to different entities.
Object (OB) = {Organization, System,
Method, Process, Software,
Documentation, Activities, Techniques,
Metamodel, Method Engineering,
Development…….}
Agility (AG) = {flexibility, rapidness,
leanness, responsiveness, learning}.
We may apply agility (AG) to any of the objects
(OB) and that object may have some degree of
agility at a specific level. For example, we could
write the fact that for OB = Method, LA =
Requirements Engineering, the value of DA was
classed as partial (PA), i.e.
DA (Object=Method, LA=Requirements
Engineering) = PA
Of especial interest is the degree of agility at the
phase and the practice level, i.e.
DA (Object, Phase or Practices) =
{PA,FA}
4 AGILE SOFTWARE
DEVELOPMENT
According to the concepts that have been outlined in
different agile methods, agile manifesto values and
agile principles (2005), agile software development
is characterized by: incremental development,
cooperative development, a simple and adaptive
development (Stapleton, 1997l Abrahamsson
et al.,
2002, 2003). We can say in a broad spectrum that
agile software development methods mainly develop
the software product iteratively and in small
releases; where project stakeholders cooperate and
collaborate (people focused and communication
oriented) to follow simple development steps in an
adaptive manner.
This section first explains the concept of
software development method and methodology;
and then proposes a definition of an agile software
development method in the light of tested and
applied definition of pure agility (Qumer and
Henderson-Sellers, 2006a), existing agile software
development methods and agile principles
(AgileManifesto, 2003).
4.1 What is a Software Development
Method and Methodology?
Software Engineering is the practice of using
processes, tools, techniques and guidelines to
produce high quality and defect-free software. A
method (a.k.a. methodology: Jayaratna, 1994) in
software engineering guides the process of
developing a software product. Brinkkemper (1996)
describes a software development method as being a
systematic approach that encompasses directions,
rules and specific way of thinking in order to
perform development activities with corresponding
development products. According to Brinkkemper
(1996), a software development methodology for
information system is the systematic description and
evaluation of all aspects of methodical information
systems development. There is another definition
that describes the software development method.
“The documented collection of policies, processes
and procedures used by a development team or
organization to practice [sic] software engineering is
called its software development methodology”
(Chapman, 1997).
4.2 What is an Agile Software
Development Method?
An agile software development method may be
described by the attributes of agility: flexibility,
speed, leanness, responsiveness and learning. We
may apply agility to any method to make that
method an agile method. According to the above
agility definition, any method (entity) that expresses
agility is called an agile method. An agile method
may be partially or fully agile according to the level
and degree of agility encompassed in that method.
We may propose a definition of an agile software
development method in the light of above study that
will help us to differentiate between agile and non-
agile software development methods.
ICSOFT 2006 - INTERNATIONAL CONFERENCE ON SOFTWARE AND DATA TECHNOLOGIES
124
Figure 1: The elements of an agile method.
Our proposed definition for an agile method
(Figure 1), paralleling the definition of agility given
in Section 2.3, is as follows: “A software
development method is said to be an agile software
development method when a method is mainly
people focused and communication-oriented,
flexible (ready to adapt to expected or unexpected
change at any time), speedy (encourages rapid and
iterative development of the product in small
releases), lean (focuses on shortening timeframe and
cost and on improved quality), responsive (reacts
appropriately to expected and unexpected changes),
and learning (focuses on improvement during and
after product development)”.
4.3 Calculated Degree of Agility in
XP and Scrum
As two examples of the application of our definition
of agility to software development methods, Table 3
and Figure 2 summarize the results for the degree of
agility (DA) for both XP and Scrum (at practices and
phases level) (see Appendix for a brief description
of XP and Scrum), measured in terms of the five
variables (features) in the agility definition:
flexibility (FY), speed (SD), leanness (LS), learning
(LG) and responsiveness (RS) (Qumer and
Henderson-Sellers, 2006b). The degree of agility
(DA) and related agility priority pattern (APP) will
be used to select a particular agile method for a
specific project in-hand. Generally, we may say that
the methods with a greater degree of agility are
suitable for small and medium size projects; but for
large and complex projects, methods with a less
degree of agility (more formal) would be better.
Table 3: Degree of agility (Qumer and Henderson-Sellers,
2006b).
Agile Method
Flexible
People Focused
Focused
Learning
Responsive
Lean
Speedy
Communication-Oriented
Phases & Practices XP Scrum
Phases 21/30
= 0.70
9/15
= 0.60
Rank 1 2
Practices 44/60
=0.73
28/35
=0.80
Rank 2 1
0,70
0,60
0,73
0,80
0,00
0,20
0,40
0,60
0,80
1,00
XP Scrum
Degree of agility
Phases
Pr a c t ic e s
Figure 2: Degree of agility (Qumer and Henderson-Sellers,
2006b).
5 CONCLUSION
We have explained here the concept of agility in
detail and, by applying this concept to software
development methods, proposed a definition of agile
methods. We intend to ratify and, if necessary,
extend the definition of agility and agile methods as
we proceed further in our research. The definition of
agility and agile methods will help us to measure the
degree of agility of any method and then to rank
software development methods from weakly agile
method to strongly agile methods. Such ranking of
methods will aid in the differentiation of agile
methods from other traditional methods or partially
agile methods. In the future, we will also apply these
updated concepts of agility and agile methods to
develop an agility measurement model.
ACKNOWLEDGEMENTS
We wish to thank the Australian Research Council
and Eagle Datamation International for financial
support under the Linkage Grants Scheme. This is
contribution number 06/04 of the Centre for Object
Technology Applications and Research.
CRYSTALLIZATION OF AGILITY - Back to Basics
125
REFERENCES
Abrahamsson, P., Wasta, J., Siponen, M.T. & Ronkainen,
J., 2003. New Direction on agile Methods: a
Comparative Analysis. 25th International Conference
on Software Engineering. IEEE Computer Society,
Portland Oregon.
Abrahamsson, P., Salo, O., Ronkainen, J. & Warsta, J.,
2002.
Agile software development methods. Review
and analysis Report. VTT Publications 478.
AgileManifesto., 2005. Manifesto for Agile Software
Development.
Beck, K., 2000. Extreme Programming Explained,
Addison-Wesley, Pearson Education.
Boehm, B. & Turner, R., 2004a. Balancing Agility and
Discipline: A Guide for the Perplexed, Pearson
Education, Inc. Boston.
Boehm, B. & Turner, R., 2004b. Balancing Agility and
Discipline: Evaluating and Integrating Agile and Plan-
Driven Methods. Proceedings of the 26th
International Conference on Software Engineering.
IEEE Computer Society, Washington DC USA.
Brinkkemper, S., 1996. Method engineering: engineering
of information systems development methods and
tools. Information and Software Technology, 38.
Chapman, J.R., 1997. Software Development
Methodology.
http://www.hyperthot.com/pm_sdm.htm.
Cockburn, A., 2002. Agile Software Development,
Addison-Wesley. Boston.
Conboy, K. & Fitzgerald, B
., 2004. Toward a conceptual
framework of agile methods: a study of agility in
different disciplines. Proceedings of the 2004 ACM
workshop on Interdisciplinary software engineering
research. ACM Press, New York USA.
Dove, R., 1997. The Meaning of Life and the Meaning of
Agility. Paradigm Shift International
www.parshift.com/library.htm.
DSDM., 2003. DSDM Consortium. Dynamic Systems
Development Method Ltd.,
http://www.dsdm.org..
Henderson-Sellers, B. and Serour, M.K., 2005. Creating a
dual agility method - the value of method engineering.
J. Database Management, 16.
Highsmith, J.A.I., 2000. Adaptive Software Development:
A Collaborative Approach to Managing Complex
Systems, Dorset House Publishing. New York.
Jayaratna, N., 1994. Understanding and Evaluating
Methodologies, NIMSAD: A Systemic Approach,
McGraw-Hill.
Nandhakumar, J. & Avison, D.E., 1999. The fiction of
methodological development: a field study of
information systems development. Information
Technology & People, 12. IEEE Computer Society,
Washington DC USA.
Paetsch, F., Eberlein, A. & Maurer, F., 2003.
Requirements Engineering and Agile Software
Development. Proceedings of the IEEE International
Workshops on Enabling Technologies: Infrastructure
for Collaborative Enterprises. Linz, Austria.
Palmer, S.R. & Felsing, J.M., 2002. A Practical Guide to
Feature-Driven Development, Prentice-Hall Inc,
Upper Saddle River.
Qumer, A. and Henderson-Sellers, B., 2006a. Measuring
agility and adoptability of agile methods: A 4-
Dimensional Analytical Tool. Procs. IADIS
International Conference Applied Computing 2006
(eds. N. Guimarães, P. Isaias and A. Goikoetxea),
IADIS Press, 503-507
Qumer, A. & Henderson-Sellers, B., 2006b. Comparative
evaluation of XP and Scrum using the 4D Analytical
Tool (4-DAT), Proceedings of the European and
Mediterranean Conference on Information Systems
2006 (EMCIS2006) (eds. Z. Irani, O.D. Sarikas, J.
Llopis, R. Gonzalez and J. Gasco), CD, Brunel
University, West London
Royce, W.W., 1970. Managing the development of large
software systems. Procs. IEEE WESCON.
Schwaber, K. & Beedle, M., 2002. Agile Software
Development with SCRUM, Prentice Hall.
Stapleton, J., 1997. DSDM: The Method in Practice,
Reading, MA, Addison-Wesley.
Wolak, R.G., 2001. System Development: Research Paper
1, SDLC on a Diet.
http://www.itstudyguide.com/
papers/rwDISS725researchpaper1.htm
.
Wong, S.-P. & Whitman, L., 1999. Attaining Agility At
The Enterprise Level. Proceedings of the 4th Annual
International Conference on Industrial Engineering
Theory, Applications and Practice. San Antonio.
APPENDIX – OVERVIEW OF XP
AND SCRUM
Extreme Programming (XP)
The XP software development process focuses on
iterative and rapid development. XP is characterized
by six phases: Exploration, Planning, Iterations to
first release, Productionizing, Maintenance and
Death (Beck, 2000; Wolak, 2001). XP stresses
communication and coordination among the team
members all the time; and requires cooperation
between the customer, management and
development team to form the supportive business
culture for the successful implementation of XP.
Scrum
Schwaber and Beedle (Schwaber & Beedle, 2002)
report that Scrum is a flexible, adaptable, empirical,
productive and iterative method that uses the ideas
of industrial process control theory for the
development of software systems. According to
Schwaber and Beedle (2002), Scrum has threes
phases: Pre-Game, Development and Post-Game.
The Pre-Game phase has a further two sub-phases:
planning and high level design (architecture).
ICSOFT 2006 - INTERNATIONAL CONFERENCE ON SOFTWARE AND DATA TECHNOLOGIES
126
