2.3.2 Characteristics of Good Algorithms

When we create an algorithm, we want it to be a good one. Just like a good recipe helps you bake a delicious cake every time, a good algorithm helps a computer solve a problem efficiently and correctly. Here are the main things that make an algorithm good:

1. Correctness: The most important thing is that an algorithm must be correct. This means it always gives the right answer for every possible input it's designed to handle. If you tell a computer to add two numbers, a correct algorithm will always give you the exact sum, not something close or completely wrong. It should also handle special or unusual situations properly, like if someone tries to add zero or a very large number.

Source: Codecademy. "What Are Algorithms?" https://www.codecademy.com/resources/docs/general/algorithm

2. Efficiency (Time and Space): A good algorithm is also efficient. This means it gets the job done without wasting too much time or using too much computer memory (space).

Time Efficiency: How quickly does the algorithm run? Imagine you have two different ways to sort a pile of homework. One way might take you five minutes, and another might take an hour. The five-minute way is more time-efficient. For computers, this means less waiting for results, especially with huge amounts of data.
Space Efficiency: How much memory does the algorithm need? Computers have a limited amount of memory, like the space on your desk. An algorithm that needs less memory is better, especially for smaller devices or when running many programs at once. A good algorithm finds a balance between speed and memory use. Sometimes, an algorithm might be a little slower but uses much less memory, which could be a good trade-off.
Source: GeeksforGeeks. "Analysis of Algorithms | Set 1 (Introduction)." https://www.geeksforgeeks.org/analysis-of-algorithms-set-1-introduction/

3. Clarity/Simplicity: A good algorithm should be clear and simple to understand. Even though computers follow instructions, humans are the ones who write and read algorithms. If an algorithm is too complicated or messy, it's easy for people to make mistakes when writing it or trying to fix it later. A simple algorithm is easier to explain, debug (find and fix errors), and improve.

Source: TutorialsPoint. "Algorithm - Characteristics." https://www.tutorialspoint.com/data_structures_algorithms/algorithms_basics.htm

4. Finiteness: As we learned, an algorithm must always stop after a certain number of steps. It can't run forever. This characteristic is called finiteness. If an algorithm never stops, it's not useful because it will never give you a final answer.

Source: Wikipedia. "Algorithm." Last modified June 28, 2025. https://en.wikipedia.org/wiki/Algorithm

5. Robustness: A good algorithm should be robust. This means it can handle unexpected or unusual inputs without crashing or giving completely wrong results. For example, if an algorithm is supposed to find the average of a list of numbers, a robust algorithm wouldn't crash if someone accidentally gave it an empty list or a list with text instead of numbers. It might give an error message, but it wouldn't break down completely.

Source: Algocademy.com. "A Comprehensive Guide to Designing Robust Algorithms". https://algocademy.com/blog/a-comprehensive-guide-to-designing-robust-algorithms/

6. Generality: Finally, a good algorithm is often general. This means it can solve a whole class of problems, not just one specific instance. For example, an algorithm to sort numbers should be able to sort any list of numbers, whether it has 10 numbers or a million, and not just one specific list you give it. This makes the algorithm much more useful.

Source: Wikipedia. "Algorithm." Last modified June 28, 2025. https://en.wikipedia.org/wiki/Algorithm