M.Sc Computing Science Software Engineering Lecture 1

1
M.Sc Computing Science Software Engineering
Lecture 1

• Niki Trigoni
• Department of Computer Science
• and Information Systems
• Birkbeck College, University of London
• Email niki_at_dcs.bbk.ac.uk
• Web Page http//www.dcs.bbk.ac.uk/niki

2
The S/E course at a glance

• Main objectives
• Become adept at understanding and defining a
  problem
• Learn how to design a solid solution
• Approach to achieving these objectives
• Follow a specific methodology (e.g. the Unified
  Process)
• Build robust and maintainable systems by using
  object-oriented analysis and design (OOA/D)
• Use UML to illustrate analysis and design models
• Apply design patterns to improve solution

3
Resources

• Textbooks
• Applying UML and Patterns An Introduction to
  Object-Oriented Analysis and Design and the
  Unified Process,
• Craig Larman, 2nd edition, Prentice-Hall 2002
• Design Patterns Explained A New Perspective on
  Object-Oriented Design, Alan Shalloway and James
  R. Trott, Addison Wesley 2002
• Lecture Notes
• http//www.dcs.bbk.ac.uk/niki/teaching.htm
• Coursework
• Announced 19 February
• Due 26 March

4
Roadmap
Lecture 1 Overview of OOA/D and the UP (Unified
Process) Lecture 2 Use-Case model write
requirements Lecture 3 Domain model identify
conceptual classes Lecture 4 Design model
design objects and their interactions Lecture 5
Design model (cont) the GRASP patterns Lecture
6 Design examples and Implementation model
5
Roadmap (cont.)
Lecture 7 Demo of Borland Together Lecture 8
More GRASP and GoF patterns Lecture 9 Design
examples using GRASP and GoF patterns Lecture
10 Architectural analysis Lecture 11 Revision
Lecture
6
Overview of lecture 1

• Initial observations on software development
• Object-oriented analysis and design
• An intro to the Unified Process (UP)

7
What makes good software?

• Relevance
• Usability
• Robustness
• Fault-tolerance
• Maintainability
• Efficiency
• The emphasis on these metrics varies across
  projects.

8
Examples of software products

• Real time air traffic control
• Embedded systems digital camera, GPS
• Data processing telephone billing, pensions
• Information systems web sites, digital
  libraries
• Sensors weather data
• System software operating systems, compilers
• Communications routers, mobile telephones
• Offices word processing, video-conferences
• Scientific simulations, weather forecasting
• Graphical film making, design
• etc.

9
Putting software production in perspective

• Biggest program that you have written?
• Biggest program that you have worked on?
• Biggest project team that you have been part of?
• Longest project that you have worked on?
• Most people who have used your work?
• Longest that your project has been in production?
• The coursework for S/E will be about 0.01
  personyears.
• A big project may be 100 to 1,000 personyears.
• Software production requires team work.
• Large software projects are built on older ones.
• Software is expensive.

10
Approach to building large systems

• Analysis and design
• Object-oriented technology
• Follow a process that prescribes analysis and
  design in a well-defined manner (e.g. the Unified
  Process)
• Clear documentation and sharing of information
  during software development (e.g. using UML)
• Human factor work in teams of people
• proficient in S/E and
• with good communication skills

11
Analysis and design

• Analysis do the right thing
• investigation of a problem
• requirements
• e.g. requirements analysis or object analysis
• Design do the thing right
• conceptual solution that fulfills the
  requirements
• can lead to implementation
• not the implementation
• e.g. object design or database design
• Object-oriented Analysis and Design (OOA/D)
• think (analyze and design) in objects

12
Object-oriented analysis

• OO analysis create a description of the domain
  from the perspective of classification by
  objects.
• identify main concepts (objects)
• add attributes
• associate objects with each other

13
Object-oriented design

• OO design define software objects and their
  collaborations
• static view design class diagrams
• dynamic view interaction diagrams

14
Object-oriented design

• OO design define software objects and their
  collaborations
• static view design class diagrams
• dynamic view interaction diagrams

15
Software engineering is not just coding

• Knowing an OO language, like Java or C, is
  necessary, but not sufficient in order to create
  object software.
• In between a nice idea and a working software,
  there is much more than programming.

16
OOA/D in the Unified Process (UP)

• In which context do we apply OOA/D ?
• We need a process
• with well defined steps
• that uses OOA/D
• in order to achieve a software development goal.
• The Unified Process is an example process
  particularly suited for building OO systems.
• Why use the Unified Process to develop software?

17
Central ideas in the UP

• Use of OO technologies
• Iterative lifecycle
• Continuous feedback and adaptation
• Managed complexity
• Early visible progress
• Time-boxed iterations
• Risk-driven development
• Early mitigation of high risks
• Cohesive and well-documented process
• Widely used process
• UML a well-defined modeling language for
  documenting and communicating ideas and designs

18
The UP iterative development

• Feedback from iteration N leads to refinement and
  adaptation of the requirements and design in
  iteration N1
• Iterations are fixed in length

Requirements
Requirements
Analysis Design
Analysis Design
Time
Implementation
Implementation
Testing Delivery
Testing Delivery
Iteration N
Iteration N1
19
UP phases

• Inception
• Feasibility phase and approximate vision
• Elaboration
• Core architecture implementation, high risk
  resolution
• Construction
• Implementation of remaining elements
• Transition
• Beta tests, deployment

20
UP artifacts and disciplines

• The UP describes work activities, which result in
  work products called artifacts. Examples of
  artifacts
• Code
• Web graphics
• Database schema
• Use cases (text documents explaining business
  scenarios)
• Diagrams, models and so on
• A discipline (or workflow) is a set of work
  activities. Examples
• Requirements
• Business modeling
• Design

21
UP disciplines and iterations

• An iteration includes work in most disciplines.

Sample UP Disciplines
Business modeling
Requirements
Design
Implementation
Test
Deployment
Configuration and Change Mngmnt
Project Mngmnt
Environment
22
An overview of the Unified Process
23
The waterfall model

• How does it differ from the UP?

24
Case study Banking application

• Inception envision the product scope, vision and
  business case.
• What is the vision for this project?
• Is it feasible?
• Who will do it?
• In-house development
• Buy external packages and customize them
• Sub-contract the project to a software company
• How much will it cost ?
• This depends on who will do the work.
• Will it require training of the personnel of the
  bank, or hiring new people?
• Think of the main risks and constraints involved.
• Define the most important functional requirements
  (opening an account, deposit, withdrawal)?
• Is it worth investigating in serious
  investigation?

25
Case study Banking application

• Elaboration build the core architecture and
  resolve the high risk elements, define most
  requirements and estimate the overall schedule
  and resources.
• Work in short time-boxed risk-driven iterations
• How do we describe the functional requirements of
  the banking application?
• Which are the main domain concepts involved in
  the business of a bank (account, customer,
  transaction)?
• How do domain concepts translate to design
  objects?
• Design the database schema used to represent data
• Build the User Interface
• Which are the key architectural issues?
• How will we test the system? What will be tested?
• Implement a good part of the system

26
Questionnaire

• InceptionRequirements, ElaborationDesign,
  ConstructionImplementation (true? or false?)
• Requirements must be fully defined before
  starting analysis and design (true? or false?)
• A system must be perfectly designed before being
  implemented (true? or false?)
• A suitable iteration is four week long, as
  opposed to four month long (true? or false?)
• An accurate estimate of the project development
  cycle is possible at the end of the Inception
  Phase (true? or false?)
• The most critical functional requirements are not
  defined during the Inception Phase (true? or
  false?)
• The code written during the first iterations of
  Elaboration is a throw-away prototype (true? or
  false?)