13.2. Characteristics of an Algorithm

An algorithm must terminate after a finite number of steps.

Finiteness is a crucial property of an algorithm. This means that every algorithm must come to an end after a specific number of steps. It should not run indefinitely; otherwise, it may result in an infinite loop. For instance, if you are developing a recipe for cooking, you must specify that the cooking process should last for a certain time. If you don't, the cooking might never stop, resulting in a burnt dish.

Imagine you are putting together a jigsaw puzzle. The instruction to complete the puzzle illustrates finiteness, showing that you should stop once the final piece is placed, ensuring the process concludes successfully instead of continuing forever.

Each step must be precisely defined.

Definiteness refers to the clarity in each step of the algorithm. Each instruction must be clear and unambiguous, allowing anyone to understand what needs to be done. If the steps are vague or open to interpretation, the algorithm may yield incorrect results or confuse users. This ensures that each action can be followed without any doubt about what is expected at each stage.

Think of a GPS navigation instruction. If it says 'drive this street,' it leaves room for confusion. However, if it specifies 'turn right at the next traffic light,' it gives clear directions that lead to the right destination.

The algorithm must accept zero or more inputs.

Input refers to the data that an algorithm needs to process to produce an output. An algorithm can work with no input parameters (like a default action), or it can require several inputs to function correctly. This flexibility allows algorithms to tackle various problems by processing the data they receive.

Consider a vending machine. It requires an input (the money you drop in) to dispense an item. But imagine a scenario where a vending machine just gives out a free token periodically without requiring any input—that's an algorithm accepting zero inputs.

The algorithm must produce at least one output.

Output refers to the results generated by an algorithm after processing the inputs. At least one output must be produced, whether it’s a final result, a report, or an action taken. Without producing an output, the algorithm wouldn’t have any tangible effect and could be considered ineffective.

Picture a calculator: when you input numbers and mathematical operations, the display provides an output (the result). Without an output, the calculator would simply be a black box with no usefulness.

The steps of the algorithm must be basic enough to be carried out, in principle, by a person with pencil and paper.

Effectiveness indicates that an algorithm's steps should be simple and achievable. Each instruction must be feasible and manageable enough that a person could carry it out without needing advanced tools—just the basic ability to perform operations theoretically with simple manual methods.

Think of writing instructions for a friend to bake a cake. If your recipe calls for mixing ingredients in a way that requires specialized machinery, it fails the effectiveness test. However, if you describe how to mix flour, sugar, and eggs using just a bowl and a whisk, it passes, because anyone can do it with basic knowledge.

--