System Development Approaches Chapter 10

Chapter 10: Approaches to System Development

• This chapter details system development approaches.
• Two main approaches are predictive and adaptive SDLCs.
• Predictive SDLC (waterfall model) is used when project planning is possible upfront.
• Adaptive SDLC uses iteration when requirements are unclear and flexibility is needed.
• All new system development projects use a methodology.
• A methodology includes:
  • SDLC
  • Tools
  • Techniques
  • Models
• Agile development is a current trend in system development.
• Unified Process (UP) is a formal iterative approach using UML models with agile philosophies and UP disciplines.
• Extreme Programming (XP) is an iterative approach taking industry practices to the extreme.
• Scrum is an iterative approach using a Scrum Sprint.

The System Development Life Cycle (SDLC)

• Two general approaches to the SDLC:
  • Predictive: Waterfall model, assumes project planning in advance and requirements are well-understood.
  • Adaptive: Iterative development, assumes flexibility and adaptation to changing needs.
• Most projects fall on a continuum between predictive and adaptive.

Traditional Predictive SDLC

• Based on sequential phases:
  • Initiation
  • Planning
  • Analysis
  • Design
  • Implementation
  • Deployment
  • Support
• Phases are related groups of development activities.
• Waterfall model assumes phases are completed sequentially without overlap or iteration.
• Once a phase is complete, you move to the next phase, with no going back.

Modified Waterfall with Overlapping Phases

• Provides more flexibility than traditional waterfall.
• Still assumes predictive planning and sequential phases to some extent but allows for some overlapping of phases.
• Allows for additional activities in each phase.

Newer Adaptive SDLC

• Emerged in response to complex and uncertain technological environments.
• Iterations are common, where design and implementation happen from the beginning.
• Often claimed as the only way to develop information systems.

Adaptive Approaches

• Incremental development: Completing portions of the system in small increments as the project progresses.
• Walking Skeleton: Building the complete system structure initially with basic functionality, making it available sooner which allows business to begin.

A Generic Adaptive Approach

• Six core processes repeated across iterations:
  • Identify the problem and obtain approval
  • Plan and monitor the project
  • Discover and understand details
  • Design system components
  • Build, test, integrate components
  • Complete system tests and deploy
• Multiple iterations are required.

Methodologies, Models, Tools, and Techniques

• Methodologies provide guidelines for system development.
• Models abstract real-world aspects; crucial for communication.
• Tools assist in creating models or other project components.
• IDEs provide comprehensive development environments.
• Visual modeling tools create graphical models.
• Techniques are collections of guidelines to complete activities and tasks.

Agile Development

• A guiding philosophy and set of guidelines for developing information systems in rapidly changing environments.
• Chaordic -- a term for adaptive projects, meaning they are both chaotic and ordered.
• Agile values: Responding to change, individuals and interactions over processes, working software over comprehensive documentation, and customer collaboration over contract negotiation.
• Agile Modeling (AM) philosophy emphasizes building necessary models at the correct level of detail.

Agile Modeling Principles

• Build only necessary models.
• Models should be useful and detailed appropriately.
• Focus on model content rather than representation.
• Use informal models when appropriate.
• Emphasize stakeholder needs.

Agile Principles

• Focus on developing software as the primary goal.
• Prioritize enabling the next effort as secondary.
• Be mindful of versioned deliverables.
• Build only the necessary models or documents.
• Embrace change and incremental changes.
• Use small, reversible steps to maintain progress and flexibility.
• Build models with a purpose for understanding, communication, and problem-solving.

The Unified Process (UP)

• Was an early leader in adaptive approaches.
• UP and UML (Unified Modeling Language) were developed together.

UP Phases

• Inception: Getting the project started.
• Elaboration: Understanding system requirements.
• Construction: Building the system.
• Transition: Preparing for deployment of the system.

UP Disciplines

• Two types of disciplines:
  • Development: Business modeling, requirements, design, implementation, testing, and deployment.
  • Management: Planning, control, configuration and change management, and project management.

Extreme Programming (XP)

• Takes best software development practices to the extreme.
• Focus on proven industry practices with unique ways to get better results.
• Core values: Communication, simplicity, feedback, courage.

XP Practices

• Planning based on user stories, thorough testing, pair programming, simple designs, refactoring, collective code ownership, continuous integration, on-site customer, a system metaphor, small releases, forty-hour work week, and coding standards.

XP Project Approach

• Three-level approach (three rings): Outside ring (user stories and acceptance tests), middle ring (testing and overall planning), inside ring (iterations of coding and testing).

Scrum

• Combines principles of Rugby and Agile.
• Defines an intense effort involving the entire team for a defined period of time.
• Involves a prioritized list of user requirements (product backlog).
• Defines product owner.
• Defines Scrum master.
• Scrum Sprint is a time-controlled mini-project.

Scrum Practices

• The scope of each sprint is often frozen to provide clarity and limit unnecessary expansion.
• Consistent time periods are important to maintain a predictable rhythm.
• It includes daily scrum meetings.
• Scrum is very iterative.

Explore the different approaches to system development in Chapter 10. This chapter compares predictive and adaptive SDLCs, detailing methodologies like Agile, Unified Process, Extreme Programming, and Scrum. Understand how each methodology suits various project requirements.