deskcheck

Deskcheck is a manual method used in software development to verify the correctness of an algorithm or piece of code by simulating its execution without running it on a computer. Practitioners read the code, select representative inputs, and trace how data and control flow change as each statement executes. The goal is to catch logical errors, boundary cases, and misunderstandings of the algorithm early in the development process.

Process: create or copy the code onto paper or a whiteboard; annotate initial values for variables; step through one statement at a time, updating the state (variable values, loop counters, stack if applicable); record the expected outputs and side effects; note any deviations from the intended behavior. Common focus areas include loop termination, conditional branches, indexing, and function calls.

Relation to other activities: desk checks are informal, non-automated checks performed before or alongside debugging and unit testing. They are especially useful for clarifying understanding during code reviews or when introducing a new algorithm. They are not a substitute for automated tests and can be time-consuming for large or complex systems.

Limitations: human error can be introduced, and desk checks may not scale well to large codebases, concurrent or I/O-heavy programs, or stateful components with many interactions. They rely on careful notation and discipline but are valuable as a low-cost early verification step.

Examples: a simple function that sums an array can be walked with a sample input, tracking total and index. Edge cases such as empty inputs or negative values can be tested this way.