← all systems · see it in The System
Metric space
'A metric space is a set together with a notion of distance between its points. The distance is measured by a function called a metric or distance function.' Formally an ordered pair (M, d) with d: M x M -> R such that d(x, x) = 0, d(x, y) > 0 for x != y, d(x, y) = d(y, x), and d(x, z) <= d(x, y) + d(y, z).
Properties published scienceour proposal
Sourced fields, and our own readings.
| boundary | Inside: a set of points and one distance function obeying four axioms (zero self-distance, positivity, symmetry, the triangle inequality), with everything the distance defines: balls, open sets and topology, convergence, completeness, boundedness, compactness, and the maps that respect distance (isometries, Lipschitz maps, contractions, quasi-isometries). Outside: extra structure some metric spaces carry (norm, inner product, measure, smooth structure, group action) and the relaxed notions (pseudometrics, quasimetrics, semimetrics) that drop an axiom. |
| nested level | Field scale, rung 1, below the first fold. In the hierarchy 'Inner product spaces are a subset of normed vector spaces, which are a subset of metric spaces, which in turn are a subset of topological spaces'. Riemannian manifolds, normed spaces and graphs are kinds of it. |
| scale | field |
Measures (7) published science
| axioms of a metric | 4 (identity, positivity, symmetry, triangle inequality) count |
| comparison of plane metrics | d_inf <= d_2 <= d_1 <= 2 d_inf inequality |
| taxicab length of the shortest paths in the illustration / Euclidean length of the diagonal | 12 / 6 sqrt 2 (about 8.49) length |
| discrete metric distance between distinct points | 1 length |
| Frechet's abstract space with metric properties | 1906 year |
| Hausdorff coins 'metric space' | 1914 year |
| characteristic clock: none of its own; in the seismology reading distance tracks seismic travel time | roughly proportional (straight-line metric) time |
Partners and parts published science
Built from / builds (4)
- Gromov hyperbolic space
- Distance function (metric)
- Set (of points)
- Metric tree (real tree, R-tree; walked as an idea)
Sits inside / holds (5)
- Riemannian manifold (positive-definite geometry; space at one time)
- Topological space
- Sub-Riemannian manifold
- Euclidean distance
- Metric separation (distance-like measures)
Trades with (valence) (11)
- Gromov hyperbolic space via sys:spacetime · Metric space even
- Earth via sys:gravitational-field · Metric space even
- Internet via sys:electromagnetic-field · Metric space even
- Riemannian manifold (positive-definite geometry; space at one time) via sys:spacetime · Metric space gain
- Euclidean space via sys:spacetime · Metric space gain
- Hyperbolic plane via sys:spacetime · Metric space even
- Sub-Riemannian manifold via sys:spacetime · Metric space gain
- Euclidean distance via sys:spacetime · Metric space gain
- Metric separation (distance-like measures) via sys:spacetime · Metric space gain
- Metric tree (real tree, R-tree; walked as an idea) via sys:spacetime · Metric space even
- Limit (mathematics): the value a sequence or function approaches, and the practice of approaching, bounding and proving it via sys:sensory-environment · Metric space gain
floor filled by (18)
- Euclidean metric (dot product)
- Gromov hyperbolic space
- Earth
- Riemannian manifold (positive-definite geometry; space at one time)
- Euclidean space
- Seismic waves
- Topological space
- Distance function (metric)
- Set (of points)
- Normed vector space (Banach space when complete)
- Length space / geodesic metric space
- Graph with path distance (finite or discrete metric space)
- Metric measure space
- Sub-Riemannian manifold
- Euclidean distance
- Metric separation (distance-like measures)
- Metric tree (real tree, R-tree; walked as an idea)
- Limit (mathematics): the value a sequence or function approaches, and the practice of approaching, bounding and proving it
represents (1)
Kinds (9)
- Riemannian manifold (positive-definite geometry; space at one time)
- Euclidean space
- Gromov hyperbolic space
- Topological space
- Normed vector space (Banach space when complete)
- Length space / geodesic metric space
- Graph with path distance (finite or discrete metric space)
- Ultrametric space
- Metric measure space
wired (2)
Notes published science
- The Earth has two honest distances: the great circle for ships and planes, and the straight chord through the planet for seismic waves, whose travel time roughly tracks it.
- A Riemannian manifold can be rebuilt from distances alone: 'all information about a Riemannian manifold can be recovered from its distance function'.
- Walking times between mountain villages are not a metric but a quasimetric, because uphill takes longer than downhill.
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
distances on the Earth (great-circle and straight-line metrics)
- L1: The surface of the Earth as a set of points with two metrics: the great-circle distance 'as the crow flies' and the straight-line distance through the interior
- L2: Uses: the great-circle metric 'is particularly useful for shipping and aviation'; the straight-line metric 'is natural in seismology, since it roughly corresponds to the length of time it takes for seismic waves to travel between those two points'
metric space (own: a set with a distance function)
- L1: Points: the elements of a set M; 'Given a metric space (M, d) and a subset A, we can consider A to be a metric space by measuring distances the same way'
- L2: The distance function with four axioms: d(x, x) = 0; d(x, y) > 0 for x != y; d(x, y) = d(y, x); d(x, z) <= d(x, y) + d(y, z) ('you can arrive at z from x by taking a detour through y, but this will not make your journey any shorter')
- L3: Topology from balls: open balls B_r(x) = {y : d(x, y) < r} form a base; convergence, Cauchy sequences, completeness (every metric space has a unique completion), boundedness, total boundedness, compactness ('complete and totally bounded')
- L4: Maps that respect distance: isometries ('essentially identical'), K-Lipschitz maps, contractions (Banach fixed-point theorem: on a complete space every contraction 'admits a unique fixed point'), quasi-isometries preserving 'large-scale structure' (R^2 and Z^2 are quasi-isometric)
- L5: Extra structure: normed vector spaces (Banach when complete), length and geodesic spaces, Riemannian manifolds ('distance infinitesimally'), metric measure spaces (CAT(k), Alexandrov and RCD spaces generalise curvature bounds), graphs
- L6: Distances between objects: Hausdorff distance between subsets; the Gromov-Hausdorff distance between compact metric spaces themselves; Wasserstein metrics ('the cost of transporting one to the other'); graph edit distance; Hamming distance between strings
space at one time as a Riemannian metric space (distance as infimum of path lengths)
- L1: Distance from path length on a connected Riemannian manifold: d_g(p, q) = inf {L(gamma)}; '(M, d_g) is a metric space, and the metric topology on (M, d_g) coincides with the topology on M'
- L2: Recovering the geometry: 'The Riemannian metric is uniquely determined by the distance function; this means that in principle, all information about a Riemannian manifold can be recovered from its distance function'
Who fills each layer
- L1: Metric separation (distance-like measures)
- L2: Euclidean distance, Gromov hyperbolic space, Metric separation (distance-like measures)
- L3: Metric separation (distance-like measures)
- L4: Metric separation (distance-like measures)
- L5: Metric tree (real tree, R-tree; walked as an idea)
- L6: Metric tree (real tree, R-tree; walked as an idea)
Sources (10) published science
Hyperbolic metric space (Wikipedia, via Wikimedia Enterprise cache) (behind its links)
Real tree (Wikipedia), via the Wikimedia Enterprise snapshot 2026-11-01 (behind its links)
Euclidean space (Wikipedia, rev 1380354378, via Wikimedia Enterprise cache) (behind its links)