Pseudo code

Pseudo code is an informal, language-agnostic way to describe an algorithm using plain structured statements instead of the syntax of any one programming language.

Why it matters

It lets you reason about what an algorithm does before committing to how a specific language expresses it. It’s the lingua franca of textbooks, interviews, and design discussions: a Python programmer and a Rust programmer can read the same pseudo code and agree on the logic. Writing pseudo code first also separates two hard things — getting the algorithm right, and getting the syntax right — so you tackle them one at a time.

How it works

There is no single standard, but good pseudo code shares a few conventions:

• Sequence — one statement per line, top to bottom.
• Assignment — total ← 0 (or total = 0).
• Selection — if … then … else … end if.
• Iteration — for i ← 0 to n-1 / while condition.
• Subroutines — function name(args) … return value.
• Indentation marks blocks, like Python.

Example — find the maximum of a list:

function max(A):
    best ← A[0]
    for i ← 1 to length(A) - 1:
        if A[i] > best:
            best ← A[i]
    return best

The goal is to be precise enough to translate mechanically into code, but free of language noise (no semicolons, imports, or memory management).

Example

Most lessons in this track open with pseudo code before any concrete implementation — searching states binary search as an invariant-driven loop in pseudo code first, then links a runnable Rust version in playgrounds/rust/binary-search.

Pitfalls

• Too vague — “sort the list” hides the very thing you’re trying to describe. Spell out the steps.
• Too concrete — if it has full Java syntax, it’s just Java. Drop the ceremony.
• Ambiguous indexing — be explicit about 0- vs 1-based and inclusive vs exclusive bounds; this is where off-by-one bugs are born.

See also

• programming-paradigms