A real-valued function on a vector space that assigns a nonnegative size to each element and satisfies positive definiteness (zero only at the zero vector), absolute homogeneity (scaling by scalars), and the triangle inequality (subadditivity).

Principle

A norm measures magnitude in a way that is compatible with vector space operations and induces a metric d(x,y)=||x-y||; norms can be induced by inner products or defined more generally (p-norms, sup-norm, operator norms), and they determine topology and notions of convergence.

Demonstration

Common examples include the Euclidean norm ||x||_2 = (sum x_i^2)^{1/2} on R^n, the p-norm ||f||_p = (∫|f|^p)^{1/p} on function spaces, the sup-norm ||f||_∞ = sup |f|, and operator norms defined by suprema over unit vectors.

Misapplication

Using a seminorm (which may vanish on nonzero vectors) or a non-subadditive function as if it were a norm, or assuming equivalence of norms in infinite-dimensional spaces without justification, are typical misuses; also assuming induced inner products exist for every norm is incorrect.

Consequence

A norm induces a metric, a topology, and notions of continuity, boundedness and completeness; in finite-dimensional spaces all norms are equivalent (same topology), while in infinite-dimensional settings choice of norm affects compactness, convergence and functional-analytic properties.

Reversal

A reversal would be a pseudometric or a seminorm that relaxes one of the defining axioms (e.g., permitting nontrivial kernel or violating triangle inequality), which changes the induced topology and algebraic consequences dramatically.

Boundary

Norms are defined on vector spaces over R or C and on modules in functional-analytic contexts; they are not general set functions and require linear structure. Many function-space norms are only seminorms until equivalence classes (modulo null sets) are taken into account.

Semantic Tension

There is tension between 'norm' and 'metric' because every norm induces a metric but not every metric arises from a norm; there is also tension between 'norm' and 'seminorm' in contexts where degeneracy matters.

Synthesis

A norm is a scale-compatible, subadditive size function on a vector space that induces metric and topological structure; it underpins concepts of length, distance and convergence, with finite-dimensional equivalence and infinite-dimensional sensitivity to the chosen norm.