Software Development Lifecycle

This lesson covers the Software Development Lifecycle (SDLC) for the OCR A-Level Computer Science (H446) specification. You need to understand the stages of software development, common development methodologies, documentation practices, and version control.

---

What is the SDLC?

The Software Development Lifecycle is a structured process that guides the development of software from initial concept to deployment and maintenance. It ensures that software is developed systematically, meets user needs, and is delivered on time and within budget.

---

Stages of the SDLC

Stage	Description	Key Activities
Requirements	Gather and document what the system must do.	Interviews, questionnaires, observation, document analysis.
Design	Plan the system architecture and components.	Data structures, algorithms, UI design, database schema.
Implementation	Write the code according to the design.	Programming, unit testing, code reviews.
Testing	Verify the system works correctly.	Unit, integration, system, acceptance testing.
Deployment	Release the system to users.	Installation, data migration, user training.
Maintenance	Support and improve the system after release.	Bug fixes, updates, enhancements.

---

Requirements Gathering

Types of Requirements

Type	Description	Example
Functional	What the system must DO.	"The system must allow users to register with an email and password."
Non-functional	HOW the system should perform.	"The system must load pages within 2 seconds."
User requirements	Written from the user's perspective.	"As a student, I want to view my grades online."
System requirements	Technical specifications for developers.	"The database must use PostgreSQL 14 or later."

Requirements Gathering Techniques

Technique	Advantages	Disadvantages
Interviews	Detailed, can ask follow-up questions	Time-consuming, subjective
Questionnaires	Reach many people quickly	Limited depth, low response rates
Observation	See real workflows	Time-consuming, may alter behaviour
Document analysis	Understand existing systems	Documents may be outdated
Prototyping	Users can see and interact with a model	Can create unrealistic expectations

---

Design

Key Design Activities

Activity	Description
System architecture	Overall structure: client-server, microservices, etc.
Data design	Database tables, relationships, normalisation.
Algorithm design	Pseudocode, flowcharts for key processes.
Interface design	User interface mockups and wireframes.
Module design	Classes, functions, and their interactions.

Design Notations

Notation	Purpose
Pseudocode	Describe algorithms in structured English.
Flowcharts	Visual representation of program flow.
UML class diagrams	Show classes, attributes, methods, and relationships.
Entity-relationship diagrams	Show database entities and their relationships.
Data flow diagrams	Show how data moves through the system.

---

Implementation

Implementation is the stage where the design is translated into working code.

Best Practices

• Follow coding standards for consistent style (naming, indentation, commenting).
• Write modular code -- break the program into small, focused functions/classes.
• Use meaningful variable names that describe their purpose.
• Add comments to explain complex logic (not obvious code).
• Apply DRY (Don't Repeat Yourself) -- avoid duplicating code.

Code Review

Aspect	What to Check
Correctness	Does the code do what it should?
Readability	Is the code clear and well-structured?
Efficiency	Are there unnecessary operations or redundant code?
Security	Are there vulnerabilities (e.g., SQL injection, unsanitised input)?
Standards	Does the code follow agreed conventions?

---

Testing

Testing has been covered in detail in the previous lesson. The key testing activities in the SDLC are:

1. Unit testing during implementation.
2. Integration testing as modules are combined.
3. System testing when the complete system is ready.
4. Acceptance testing with end users before deployment.

---

Deployment

Deployment Strategies

Strategy	Description	Risk
Direct changeover	Old system switched off; new system switched on immediately.	High -- no fallback if it fails.
Parallel running	Old and new systems run simultaneously until confidence is built.	Low -- but expensive (running two systems).
Phased implementation	New system introduced in stages (e.g., one department at a time).	Medium -- issues found early in each phase.
Pilot implementation	New system tested by a small group before full rollout.	Low -- limited impact if problems occur.

Exam Tip: OCR may ask you to recommend a deployment strategy for a scenario. Always justify your choice: direct changeover is fast but risky; parallel running is safe but costly.

---

Maintenance

After deployment, software requires ongoing maintenance: