Agile Systems Development

Neil McBride
 Agenda

What’s the problem?

What’s the metaphor? – Building Construction versus Jazz
Improvisation
Agile Methods
Examples
Agile Manifesto
XP as an example of agile method
Four Values
Basic Principles
Planning games, pair programming and metaphor.
Critique

Neil McBride Centre for IT Servic

e Management Research
 The Problem

Failure of software project is endemic.

Schedules slip,
Projects are cancelled,
Business needs never met,
Changes made faster than system can be developed,
Testing is inadequate,
Quality not managed,
Staff turnover,
Systems become too large to manage.

Neil McBride Centre for IT Servic

e Management Research
Computer systems cannot cope with change.

Difficult to change,
Cannot adapt to organisational change,
Require certainty.

But organisations’ and businesses’ problems change

New customers,
Transactions changes,
New products,
Demands of Internet-based systems

Neil McBride Centre for IT Servic

e Management Research
 How do we deal with change?

Large scale project management approaches try to:

Reduce change
Increase certainty
Eliminate risk by planning, documentation and contracts.

We do what it says on the contract

We freeze requirements and we formalise such freezing though stage
meetings.
Changes become rewrites.

Our metaphor is the construction industry and our project management

practices are drawn from the industry.
What if the metaphor is wrong?
Neil McBride Centre for IT Servic
e Management Research
 The Construction Analogy

Assumption: sequential development process, predictable and stable

environment, goals of achieving efficiency and reducing uncertainty

Structure
IS stable artifacts? Requirements ‘fixed in concrete’ IS as social
artifacts

Process
Linear? Formal join points? SSADM-like?

Roles
Cultural ghettos, divergent, conflict and formalisation. Blinkered view.
Neil McBride Centre for IT Servic
e Management Research
 Film Production

Script based on literary sources

Revision during production

Interpretation by different creative artists

Pre-production, Filming, Post-Production

Neil McBride Centre for IT Servic

e Management Research
Structure
More pliable product
Builds on body of work
But little active role of users.

Process
Dynamic development
Greater scene of teamwork, changing roles, continuous involvement of
many creative staff.

Roles
Variety of roles.
Wide-skill set.
Some role changing.
Neil McBride Centre for IT Servic
e Management Research
 Jazz Improvisation

Use of minimalist musical structures including harmonies, melodies and

rhythm.
Small team elaborating on simple structures in complex ways
Musicians operate with a set of social norms, with changing roles and intense
interaction.

Minimalist Structures
Constrain the turbulence of the jazz process by specifying particular ways of
inventing and co-ordinating musical ideas’

Neil McBride Centre for IT Servic

e Management Research
Social Structure
Behavioural Norms
Communicative Codes
Partnering in autonomous
ensemble
Trust within wide zones of
manoeuvre and constructive
controversy
Alternate soloing and comping.
For leadership and personal
development
Risk-taking and continuous
learning
Technical Structure
Keys, Chord progressions,
repertoire
Templates of songs, choruses etc
on which to improvise
Wide stock of talent and
performative competence
Knowledge of music technology
and instrumentation
Experimenting with new
instruments, styles and textures
of sound
Refashioning performance in
response to colleagues and
audience

Neil McBride Centre for IT Servic

e Management Research
Structure
Dynamic systems, subject to change in response to organisational change
Minimal componentised structures
Patterns

Process
Variation with socially determined process structured
Iterative development and continuous delivery
Theme development

Role
Role rotation.
Importance of mentors
Continuous communication and listening.
Neil McBride Centre for IT Servic
e Management Research
Construction Analogy
Sequential progression
Product structure defined and
fixed at start
Roles fixed in skill base and
temporal involvement with
project
Changes difficult to achieve.
User input limited temporally
and functionally
Technology fixed at start of
project
Focus on technology and
rigorous adherence to pre-
defined design.
Neil McBride Centre for IT Servic
e Management Research
Film Production/ Jazz
Improvisation Analogy
Iterative dynamic progression
Product structure evolves in
response to new ideas and
environmental change
Roles rotate and project
involvement is continuous
throughout the project's lifetime
Changing structure and new
structure is the norm
User input critical throughout
project.
Technology adaptation occurs
as new tools and methods are
explored.
Focus on creativity and
development of new ideas
within a minimal design
framework
 Agile Software Development Approaches

Ancestry in Rapid Application Development

Held up as antithesis of traditional software development.
Focus on:
Early delivery priority business requirements.
Dealing with partial knowledge
Reduced documentation
Small groups of programmers
Iteration
Continuous testing
Integral customer involvement

Neil McBride Centre for IT Servic

e Management Research
 The Agile Alliance

Snowbird 2001 : Meeting of representatives of agile methods

Purpose to get all the leaders of lightweight methods into one room

‘Define a developer community freed from the baggage of Dilbertesque

corporations.’
[Respond to] ‘the failure of the standard "fixed" process mindset that so
frequently plagues our industry.’
‘The Agile movement is not anti-methodology, in fact, many of us want to
restore credibility to the word methodology. We want to restore a balance.
We embrace modeling, but not in order to file some diagram in a dusty
corporate repository. We embrace documentation, but not hundreds of
pages of never-maintained and rarely-used tomes. We plan, but
recognize the limits of planning in a turbulent environment.’

Neil McBride Centre for IT Servic

e Management Research
 Agile Methods

Adaptive Software Development

Feature Driven development
Crystal Clear Method
Dynamic Systems Development Method
Rapid Application Development (James Martin)
Scrum
Pragmatic Programming
Extreme Programming

Neil McBride Centre for IT Servic

e Management Research
 Agile Alliance Values

Individuals and interactions over processes and tools

Working software over comprehensive documentation

Customer collaboration over contract negotiation

Responding to change over following a plan

Neil McBride Centre for IT Servic

e Management Research
 Principle 1

Our highest priority is to satisfy the customer

through early and continuous delivery
of valuable software.

Software as the business output.

Continuous delivery provide regular feedback.
Enables customer evaluation

Neil McBride Centre for IT Servic

e Management Research
 Principle 2

Welcome changing requirements, even late in

development. Agile processes harness change for
the customer's competitive advantage.

Promoting software ‘evolvability’

But still avoiding radical changes at the end of the development project.

Neil McBride Centre for IT Servic

e Management Research
 Principle 3

Deliver working software frequently, from a

couple of weeks to a couple of months, with a
preference to the shorter timescale.

Rapid feedback.
Daily build
Requirements can be tested and altered quickly.
Key importance of iterative and incremental development
From fixed pricing to adaptive pricing?
Defining releases to match business deadlines and user ability to
absorb changes.

Neil McBride Centre for IT Servic

e Management Research
 Principle 4

Business people and developers must work

together daily throughout the project.

Role of developers changing towards business orientation.

Development managers with both business and technical understanding and
still writing code

Neil McBride Centre for IT Servic

e Management Research
 Principle 5

Build projects around motivated individuals.

Give them the environment and support they need,
and trust them to get the job done.

Agile approaches put emphasis on people factors – sociability, talent, skill

communication.
‘Personnel attributes and human relation activities provide by far the largest
source of opportunity for improving software productivity’ Barry Boehm

XP very demanding of people skills.

We’re not all Kent Becks or Ward Cunninghams!
Need for extraordinary skills, strong tacit knowledge and discipline

Neil McBride Centre for IT Servic

e Management Research
 Principle 6

The most efficient and effective method of

conveying information to and within a development
team is face-to-face conversation.

Stand-up meetings.
Informal communication
Tacit knowledge transfer
Role of written communication.
Problem of unrecognised shortfalls of tacit knowledge

Neil McBride Centre for IT Servic

e Management Research
 Principle 7

Working software is the primary measure of progress.

Early delivery
Measurement. .. Not lines of code
Measurements of what? Functions? Passed tests?

Neil McBride Centre for IT Servic

e Management Research
 Principle 8

Agile processes promote sustainable development.

The sponsors, developers, and users should be able
to maintain a constant pace indefinitely.

Workaholism.
Ineffectiveness of work-all-hours mentality.
Creativity needs recreation
Sustained overtime is a bad sign

Neil McBride Centre for IT Servic

e Management Research
 Principle 9

Continuous attention to technical excellence

and good design enhances agility.

Refactoring and personal quality requirements

Expertise and attitude of the programmer

Neil McBride Centre for IT Servic

e Management Research
 Principle 10

Simplicity--the art of maximizing the amount

of work not done--is essential.

Avoiding bells and whistle.

Simple solutions are easy to understand
Who decides what is simple?

Neil McBride Centre for IT Servic

e Management Research
 Principle 11

The best architectures, requirements, and designs

emerge from self-organizing teams.

Self-governing teams
Emergent behaviour
The ‘jelled’ team

Complexity Theory and Post-modernism

Neil McBride Centre for IT Servic

e Management Research
 Principle 12

At regular intervals, the team reflects on how

to become more effective, then tunes and adjusts
its behavior accordingly.

Value of organisational learning.

Time cost of review

Neil McBride Centre for IT Servic

e Management Research
 eXtreme Programming

Now let’s see the Agile Manifesto Principles in Action.

Comes from a group of American programmers : –

Kent Beck
Ward Cunningham
Ron Jefferies.

Assumes an object-oriented approach.

Closely connected with the patterns movement.

Neil McBride Centre for IT Servic

e Management Research
 Extreme Programming

Little documentation. Source code is only documentation

No software specification
No separate design and testing phase
Design for change prohibited
No formal reviews or inspections

Neil McBride Centre for IT Servic

e Management Research
 Extreme Programming

A view of what development is.

Four values.

Some basic principles.

Twelve practices.

An approach to development .. The development episode.

Neil McBride Centre for IT Servic

e Management Research
What’s important in software development management?

Cost
Time
Quality
Scope

What’s important in the process?

Coding!
Testing
Listening
Designing.

Neil McBride Centre for IT Servic

e Management Research
 Extreme Programming: Four Values

Communication Making communication essential. Face-to-face.

Environment. Pair programming.

Simplicity What is the simplest thing that can possibly work?

Removing complexity (by refactoring)
Challenging complexity (by continuous integration,
pair programming)
Feedback Coaching. Testing. Starting with test case. Daily
integration. Quick releases.

Courage. Make a decision. Team support. Ready to start again.

Neil McBride Centre for IT Servic

e Management Research
 Extreme Programming. Fundamental principles.

Rapid Feedback. Learn immediately. Stimulus/response. Days not weeks/

months.
Assume simplicity. Sort today’s problems today. No belts and braces. Using
tests. No design for reuse.
Incremental Change. Series of small changes. Evolve solutions. Plan, design,
team, change a little at a time. XP adopted a little at a time.
Embrace Change. No frozen specs.
Quality work. How. Constant review (pair programming) refactoring. Removing
redundancy. Pride in craftsmanship. Art..
Others (see for yourself)
Teaching learning, Small initial investment, play to win (not (not to loose)),
Concrete experiments (leave no abstract decisions untested), open, honest
communication, Work with people instincts, accepted responsibility, local
adaptation, travel light, honest measurement.
Neil McBride Centre for IT Servic
e Management Research
 Extreme Programming : Basic Practices

The Planning Game - Developers and customers scope and plan using
story cards
Small releases - Release as small as sensibly possible.
Metaphor – Identify overarching metaphor e.g. contracts and contract
management, tracking robot.
Simple design Design as simple as possible at any given moment. Only
leave whatever is really needed. Eliminate complexity
Testing – Write the test first. These must run flawlessly for development to
continue.
Refactoring – restructuring system without changing its behaviour.
Pair Programming – All production code is written with two programmers at
one machine.

Neil McBride Centre for IT Servic

e Management Research
 Extreme Programming : Basic Practices

Collective Ownership – Anyone can change code anywhere in the system

at anytime. Opposite of Software Configuration Management. Works if
there’s continuous integration, simple writing, complete visibility.
Continuous integration. Integrate and build system many times a day,
every time a task is complete.
40-hour week. Never two weeks in a row overtime.
On-site customer – real live customer (not line manager) sitting with
team all the time. More value out of system with more business
contribution.
Coding standards – programmers write all the code in accordance with
rule emphasizing communication though the code.

Neil McBride Centre for IT Servic

e Management Research
 Extreme Programming : Roles

Programmer – adds actual value, at heart of development

Customer – importance of customer engagement
Tester – helps customer write functional tests
Coach – responsible for process as a whole – XP expert. Provides toys
and food.
Tracker – conscience of XP team. ‘Is the feedback loop being closed?
Consultant – ‘wizard’ supplies deep technical knowledge.
Big Boss – Champion of development team.

Neil McBride Centre for IT Servic

e Management Research
 The Planning Game

Business writes stories.

Estimate time for each story. Break down stories if necessarily. Clarify with
customer.

Sort stories and decide what’s in the first iteration. Sort on value and risk.

Turn stories into tasks.

Neil McBride Centre for IT Servic

e Management Research
 Programmer

Accepts tasks and estimates.

Finds a partner.
Writes test cases (which will all fail at the start because there’s nothing
there).
Gets all tests working.

Get something, probably with many small classes, up and running as fast
as possible. ‘Go like the clappers’

Minimalist design.
Neil McBride Centre for IT Servic
e Management Research
 Critique

Origins – OO, Patterns movement, key group of ‘gurus’

Refactoring – programmer psychology. Programmer versus manager.
Business / IT gap approach to crossing it?
Scalability Debate. Size does matter.
Reaction to outsourcing and ‘big thump’ delivery of outsourcing.
Philosophy: Chaos and Complexity, Emergence, Post-modernist, Eastern
Philosophy.
Culture . ‘Biggest barrier to XP is culture’ Beck.
Precedence of code as the lingua franca.
A programming high priesthood?

Neil McBride Centre for IT Servic

e Management Research
 Bibliography and Sources

Beck, K. (1999) Embracing change with extreme programming. IEEE

Computer, October 1999, 70-77.
Beck, K. (2000) Extreme Programming Explained. Addison Wesley
Agile Methods List (CERN)
Agile Software Development Ecosystems
Build better software
Agile Alliance
Questioning Extreme Programming
Agile Software Development in Theory and Practice (MSc Thesis,
Finland)
Martin Fowler articles on XP and Agile Methods
Boehm, B and Basili, V.
Gaining Intellectual Control Over Software Development

Neil McBride Centre for IT Servic

e Management Research
 Bibliography and Sources

Jim Highsmith Links and resources

Xtreme programming.com (Ron Jefferies)
Extreme Programming Roadmap (Ward Cunningham)
Scrum
Crystal Clear
Extreme Programming Case Study - Teaching Extreme Programming in
a university
DaimlerChrysler C3 Case Study
On the Productivity of Agile Software Practices: An Industrial Case
study
Hi-Tech Workplace no better than factories (BBC News)
Avison, D and Wilson, D. (2001) A viewpoint on software engineering
and information systems: what we can learn from the construction
industry. Information and Software Technology 43, 795 - 799.
Kamoche,K and Pina e Cunha, M. (2001) Minimal Structures: From
Jazz Improvisation to Product Innovation. Organisational Studies 22(5)
733 - 764.
Neil McBride Centre for IT Servic
e Management Research