Software Development Lifecycle (SDLC) models

Take aways from other methods

The Agile manifest

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

Or: “Agile = characterized by the division of tasks into short phases of work and frequent re-assessment and adaptation of plans.”

Most of the following models are Agile…

Dynamic system development method (DSDM)

  • Timeboxing

  • MoSCoW prioritization:

    • must-have

    • should-have

    • could-have

    • and won’t-have, or will not have right now.

      • Some companies also use the “W” in MoSCoW to mean “wish.”

  • Iterative and incremental approach

Feature-driven development (FDD)

  • Feature teams

  • Parallel development

  • Tracking completion status (reporting, milestones, percentage completion)

Crystal methods

  • People-centric

  • Frequent delivery

  • No “one size fits all”

  • Modern principles: automated tests, frequent integration

Scrum (For very large projects)

  • Business value-driven comprehensive framework

  • Scrum roles

    • Owner

    • development team:

    • testers

    • documenters

    • DB admins

    • scrum master: agile coach, servant leader, removes barriers, coaches and convinces; does not manage

  • scrum workflow

    • sprint planning->sprint->review->retrospective

  • Potentially shippable product increment each sprint

  • Suitable for very large projects

Lean

  • Minimize waste, visualize production, look for bottlenecks, inefficiencies

  • Last responsible moment: wait ‘til we have enough facts

    • not make decisions based on preset timelines.

  • Deliver as fast as possible: small iterations are easier to manage

Kanban

  • Visualize your work, limit work in progress

Extreme Programming

Branching and merging is not listed as a core XP practice

  • is generally not a favorable practice in Agile approaches.

  • Branching has the potential for creating technical debt if changes are not merged frequently.

  • The risks an be somewhat mitigated by merging frequently.

  • In any case, branching and merging .

Pair programming

  • Collaborate, continuous code inspection

  • real-time development technique to increase algorithm implementation

Test-driven development

  • Write test

  • Write function to fail test

  • Write code to pass test

Spotify

  • Culture

    • Innovation: hackathons

    • Learning from each other

    • Failure-friendly culture

    • (internal) code repositories, visible and open for other teams

    • Minimize handoffs (help from other teams)

    • Bureaucracy minimized

DevOps (Development Operations)

  • Developers and testers

  • Ops (operations): people working with release, servers, middleware, network, storage configs, monitoring techniques

  • DevOps aims to extend fast and frequent software feature development approach to build an efficient delivery pipeline.

  • Phases: development –> test –> stage –> production

    • development: unit testing (code)

    • test: integration testing (User-interface, web server, app server, back-end)

    • stage: user acceptance

    • production

  • Continuous…

    • Integration: Central repository

      • development + test

    • Development: Ensuring stable product after every release

      • development + test + stage

    • Deployment: automating updates to production

      • development + test + stage + production

And more…