OIM3640 - Problem Solving and Software Design

Functions: Decomposition & Abstraction

How do we write code?

• So far, we've learned:
  • Basic programming concepts
  • Writing small programs in separate files
  • Code is made up of step-by-step instructions
• The challenge:
  • This works for small tasks, but quickly becomes messy for bigger projects
  • Hard to keep track of details
  • Not always clear which code should go where

Writing Better Programs

• Writing more code doesn't mean writing better code
• A good programmer is judged by what the program can actually do
• Functions help us:
  • Decomposition → break big problems into smaller, manageable parts
  • Abstraction → hide the messy details, focus on the inputs and outputs

Example - Drone

• Think of a drone as a black box:

  • We don’t know how it works inside
  • We only see the interface: input (commands) → output (actions)
  • We control it with a remote or an app
  • It flies and takes videos based on our commands

• ABSTRACTION: To use a drone, you don’t need to understand its inner mechanics

Example – Drone Light Show

• In a drone light show, thousands of drones work together:

  • Each drone gets its own task
  • Each one follows a programmed path
  • Together, they form amazing patterns in the sky

• DECOMPOSITION: A big show becomes possible by splitting it into many smaller drone tasks

From Drones to Code

• DECOMPOSITION

  • Break a big problem into smaller, self-contained pieces
  • Each part can work on its own

• ABSTRACTION

  • Hide the inner details of how something works
  • Focus on the inputs, outputs, and the result

Decomposition

• With drones, we split the work among many units

• In programming, we split code into modules or functions:

  • Each part is self-contained
  • Easier to manage and reuse
  • Keeps the program organized and clear

• Today: we’ll use functions for decomposition

• Later: we’ll see how classes (OOP) extend this idea

Abstraction

• With the drone, you don’t need to know how it’s built

• In programming, a function is like a black box:

  • Inner details are hidden
  • We only care about input → output → result

• Abstraction saves us from messy details

• We describe it with function specs or docstrings

Functions

• A function is a reusable block of code
  • It just sits there until you call (use) it
  • Think of it as a tool you can pick up when needed
• Key parts of a function:
  • Name → how we call it
  • Parameters → inputs it can take
  • Docstring (optional but strongly recommended) → explanation of what it does
  • Body → the code it runs
  • Return value → the result it gives back

Civilization advances by extending the number of operations we can perform without thinking about them.

- Alfred North Whitehead (British mathematician and philosopher, 1861–1947)