A topological space in which every open cover has a finite subcover: for every collection of open sets whose union contains the space, finitely many of them already cover the space.

Principle

Compactness captures the idea that global coverage can be reduced to finitely many local pieces; it converts potentially infinite local data into finite combinatorial certificates.

Demonstration

The closed interval [0,1] in R with the usual topology is compact (Heine-Borel in R), while R itself is not; in any compact metric space every sequence has a convergent subsequence (sequential compactness coincides with compactness in metric spaces).

Misapplication

Assuming sequential compactness equals compactness in arbitrary topological spaces is false; treating compactness as mere boundedness or finiteness is also wrong—compact spaces can be infinite yet tightly controlled by their topology.

Consequence

Compactness yields many useful results: continuous images of compact spaces are compact, every net has a convergent subnet, continuous real-valued functions attain maxima and minima, and many theorems reduce global problems to finite checks.

Reversal

The opposite condition—requiring every finite subcover to be extendable to an open cover—does not capture compactness and instead trivializes covering arguments; dually, local compactness focuses on neighborhood-level compactness rather than global finiteness-of-cover properties.

Boundary

Compactness is a topological property independent of metrics; in R^n compactness is equivalent to closed and bounded (Heine-Borel), but in general topological spaces closed and bounded need not imply compactness; compactness excludes properties like local compactness or Lindelöfness which are distinct.

Semantic Tension

Compact vs sequentially compact: in metric spaces the notions coincide, but in general spaces sequential compactness is strictly weaker; compact vs complete: compactness implies completeness in metric spaces but completeness alone does not imply compactness.

Synthesis

A compact space is one where every open cover admits a finite subcover, a global finiteness condition that ensures powerful continuity and extremal properties and that in metric contexts equates to several other finiteness-like compactness notions.