Software Engineering: Agile, Scrum, and Processes

Software Engineering

• Agile Software Engineering includes agile methods, Scrum, and Extreme Programming.

Course Learning Outcome

• Course Learning Outcome 1 (CLO1) focuses on recognizing software engineering principles, life cycle phases, processes, and activities.

The Software Process

• A software process is a structured set of activities needed to develop a software system.
• Different software processes exist, but generally include:
  • Specification: Defining what the system should do.
  • Design and implementation: Defining the organization and implementing the system.
  • Validation: Checking that the system meets customer needs.
  • Evolution: Changing the system based on evolving customer needs.
• A software process model is an abstract representation of a process that describes it from a particular perspective.

Software Process Descriptions

• Process descriptions include activities or actions like specifying a data model or designing a user interface, including the order of these activities.
• The descriptions also include:
  • Products: The outcomes of a process activity.
  • Roles: The responsibilities of people involved in the process.
  • Pre- and post-conditions: Statements true before and after a process activity.

Plan-Driven and Agile Processes

• Plan-driven (Prescriptive) processes plan all activities in advance, measuring progress against this plan.
• Agile processes plan incrementally, making process changes easier to meet changing customer requirements.
• Most practical processes utilize elements from both plan-driven and agile approaches.
• There is no single right or wrong software process.

Software Process Models

• The Waterfall Process Model is a plan-driven model with separate phases for specification and development.
• The Incremental Process Model is a plan-driven model with separate phases of specification and development in increments.
• The Evolutionary/Iterative Process Model is an iterative model that builds increasingly complete software versions, emphasizing risk management.

Waterfall Model Phases

• Requirements analysis and definition.
• System and software design.
• Implementation and unit testing.
• Integration and system testing.
• Operation and maintenance.
• Each phase is marked by milestones and deliverables, also known as artifacts.

When to Use Waterfall Model

• Works well for well-understood problems with minimal or no changes needed in the requirements.
• Easy to explain to customers.

Waterfall Model Drawback

• A key drawback is the difficulty to accommodate changes after starting the process.
• Each phase must be complete before moving to the next.
• It may be challenging for customers to explicitly state all requirements, and responding to changes is difficult due to the inflexible partitioning into distinct stages.
• It is suitable when requirements are well-understood and changes will be fairly limited.

Modified Waterfall

• This allows going back to previous phases and this opened the door for other iterative models

Incremental Model

• Instead of delivering a system as a single delivery, it is broken down into increments, each adding part of the functionality.
• Combines elements of linear and parallel process flow.
• Focuses on delivering an operational product with each increment.
• User requirements are prioritized and included in early increments. Once development starts on an increment, its requirements are frozen, but requirements for later increments can evolve.

Incremental Model Activities

• Choose features to include in an increment from the planned product features.
• Add details to feature descriptions for team understanding to enable implementation.
• Implement and test the chosen feature with automated tests to validate system behavior.
• Integrate the developed feature with the existing system, testing it to ensure proper functioning with other features.
• Deliver the system increment to the product manager for review; a version may be released for customer use if there are enough implemented features.

Incremental Model Benefits

• Reduced costs for accommodating changes to customer requirements.
• Easier to get customer feedback on development work.
• Rapid delivery and deployment of useful software.
• Higher priority system services receive more system testing.
• Reduced risk of overall project failure, compared to the waterfall model.

Incremental Model Problems

• System structure can degrade as new increments are added.
• Regular change corrupts its structure unless time and money are spent on refactoring.
• In the classical incremental model, the list of features is defined from the beginning.

Evolutionary Model

• Evolutionary models are iterative and develop increasingly complete software versions in increments.
• It is used when core or system requirements are well understood, but additional system requirements are not yet defined.
• Two common types of evolutionary process models are the Prototyping Model and Spiral Model.

Evolutionary Models - Prototyping

• Customers define general objectives without specific details for functions and features Or developers are unsure of the efficiency of an algorithm.
• The prototype can be "the first system,” but is evolutionary by it slowly evolves into the actual system.
• Users can feel the system and developers build something immediately.

Evolutionary Models - Prototyping Paradigm

• Begins with communication, where you will meet with stakeholders to outline overall software objectives, known requirements, and further definitions.
• Quick iteration and planning occurs.
• Quick design occurs through modeling.
• A prototype construction takes place once the quick design is done.
• The prototype is used by stakeholders who provide feedback for improvements.
• Iteration occurs to satisfy the differing stakeholders needs while understanding what needs to be done.

Evolutionary Models - Prototyping - Benefits

• Both Stakeholders and developers immediately improve the design before systems deployment.

Evolutionary Models - Prototyping - Drawbacks

• Stakeholders see working versions while the software quality is poor.
• The stakeholders did not consider overall software quality or long-term maintainability.
• Use of inappropriate OS, programming language or poor algorithm.

Evolutionary Models - Spiral

• Mixes prototyping iterative nature with the systematic parts of the waterfall model.
• It supports increasing software versions development.
• It is a risk-driven model, with 2 features:
  • Cyclic nature for growing system's implementation while decreasing risks.
  • It is a set of milestones for supporting commitment to solution.

Evolutionary Models - Spiral

• The first circuit iteration can create a product specification.
• Subsequent versions may create a prototype with more sophisticated software.
• All circuits can adjust to the project plan.
• All costs can be adjusted through customer feedback.
• Spiral models use prototyping for minimizing risk.

Specialized Process Model

• Component-Based Development
• Unified Process

Component-Based Development

• Involves software re-use by integrating existing components or application systems (Commercial-off-the-shelf or COTs).
• Reused elements can be customized for needs functionality.
• Re-use creates a standard for all types of business

Component Based Development

• The proper process to apply is when the use is a development process.
• Use already made components or use reused components
• Reusability helps developers reduce the development time to create software.

Types of reusable software

• Stand-alone application systems configured for a particular environment.
• Object Collections developed as a package integrated with component frameworks such as .NET or J2EE.
• Web services developed using standards available with remote invocation.

Reusability Advantages and Disadvantages

• Advantages:
  • Reduced costs are created as less software is created from scratch.
  • System deliveries are completed faster.
• Disadvantages:
  • Requirements compromises are inevitable so the software may fail to meet some of the proper needs
  • Losses of control of elements of reused systems.

The Unified Process (UP)

• The unified process is an iterative and incremental development process.
• Its Elaboration, construction and transition phases are timeboxes.
• Each iteration creates a new increment which is a new release of the current system

The Unified Process (UP) Phases

• Inception: Identify all business use-cases as requirements. Propose rough system architecture.
• Elaboration: Refine and expand use-cases. Expand the architecture into 5 views: -Use-case model, analysis model, design model, implementation model & deployment model
  • Construction: Develop and obtain initial software components for release.
  • Transition: Give users beta to conduct necessary changes. Make support manual guides for troubleshooting.
  • Production: Monitor all on-going uses and evaluate report defects.

Agile software engineering

• Software products must be brought to market quickly via rapid software development and delivery.
• The majority of software products are developed using agile approaches.
• Agile approaches deliver quickly, adapt based on specifications and minimize overheads.

Values of the Agile Manifesto

1 - individuals and interactions over processes and tools 2 - working software over documentation 3 - customer collaboration over contract negotiation 4 - responding for change

Agile Manifesto and Mindset

• Agile relies on a mindset of four values with 12 guidelines that have different approaches.
• All agile managers select based on the need.

Agile principles

• Satisfying customer through early and continuous delivery of valuable software
• Welcome changing requirements late in development.
• Deliver working software frequently (weeks to months with higher timeframe rate)
• Daily worker cohesion to developer and stakeholders.
• High environment with motivated teams
• Transparent in-person discussions.
• Working code is valuable
• Agile supports development and constant pace
• Continual attention and design
• Simplicity to maximize amount of work
• Team-led teams
• Regular team tuning

Agile Definition

• Agile is term for "Lean" with multiple varieties of Kanban.

Key points

• Software processes are producing systems.
• Software models are producing the systems.
• Traditional models describe the organizations.
  • EX: waterfall
• Requirement engineering creates a software specification

Key points

• Transform an specifications to executable system by design and implementation.
• Software Validation inspects conformity by user needs.
  • Software evolution can always change and remain useful.
• Implementation processes support prototyping and delivery for easier coping of changes.

Key points

• Processes build for iterative system delivery with changes disrupting the system from whole.