← all systems · see it in The System
Affine space
In mathematics, an affine space is a geometric structure that generalizes some of the properties of Euclidean spaces in such a way that these are independent of the concepts of distance and measure of angles, keeping only the properties related to parallelism and ratio of lengths for parallel line segments. ... Unlike for vectors in a vector space, in an affine space there is no distinguished point that serves as an origin.
Properties published scienceour proposal
Sourced fields, and our own readings.
| boundary | Inside: a set of points, an associated vector space of translations acting freely and transitively, the subtraction of points (Weyl's axioms), affine combinations (weights summing to 1), flats and parallelism, ratios of lengths of parallel segments, affine maps. Outside: an origin (choosing one turns it into a vector space), distance and angle (an inner product makes it Euclidean; an indefinite form makes it pseudo-Euclidean), points at infinity (adding them gives projective space). |
| nested level | Field scale, rung 1, below the first fold. The flat skeleton under Euclidean and pseudo-Euclidean (Minkowski) space; contained in projective space; the model of physical space, of time and of flat spacetime with no preferred origin. |
| scale | field |
Measures (7) published science
| dimension | the dimension of the associated vector space count |
| sum of coefficients in an affine combination | 1 dimensionless |
| Weyl's axioms | 2 count |
| axioms of an affine plane (Cameron 1991) | 3 count |
| points in an affine basis of an n-dimensional Euclidean (affine) space | n + 1, not contained in a hyperplane count |
| physical models | E(3) (non-relativistic space), E(1,3) (relativistic spacetime), 1-D (time) dimension |
| characteristic clock: none of its own; time itself is one of its models (durations are displacements) | none n/a |
Cross-scale patterns seen here (2) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Affine space. Each links to every other system that shows it.
- templating (structural)
One flat skeleton recurs under every flat geometry: Euclidean spaces have 'affine properties' (lines, subspaces, parallelism), pseudo-Euclidean ones keep 'the affine structure ... unchanged', and Minkowski spacetime is 'an affine space when the origin is dropped'. - same shape, opposite sign (surface)
Two completions of the same skeleton that pull opposite ways: adding an origin makes it a vector space (one point singled out), adding points at infinity makes it projective (parallels meet); affine space sits between, with neither.
Valence: what it trades, and with whom (7) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Projective space via sys:spacetime · Affine space even
- Quantum vacuum (vacuum state) via sys:quantum-vacuum · Affine space loss
- Euclidean space via sys:spacetime · Affine space even
- Minkowski spacetime via sys:spacetime · Affine space even
- Pseudo-Euclidean space via sys:spacetime · Affine space even
- Affine connection via sys:spacetime · Affine space even
- Time via sys:spacetime · Affine space even
Partners and parts published science
Built from / builds (3)
- Projective space
- Spacetime event
- Vector space: sets whose elements can be added and scaled, as mathematics describes it, with bases and dimension, linear maps, inner products, and their use as the state spaces of physics and data
Sits inside / holds (2)
expressed by (1)
floor filled by (12)
- Projective space
- Quantum vacuum (vacuum state)
- Spacetime event
- Minkowski spacetime
- Observer (frame of reference, physics sense)
- Vector space: sets whose elements can be added and scaled, as mathematics describes it, with bases and dimension, linear maps, inner products, and their use as the state spaces of physics and data
- Affine plane
- Affine group
- Time
- Three-dimensional Euclidean space
- Pseudo-Euclidean space
- Affine connection
represents (1)
Kinds (8)
- Euclidean space
- Projective space
- Shapes (geometry): the practice and the structure
- Pseudo-Euclidean space
- Vector space: sets whose elements can be added and scaled, as mathematics describes it, with bases and dimension, linear maps, inner products, and their use as the state spaces of physics and data
- Affine line (number line)
- Affine plane
- Complex affine space
wired (1)
Notes published science
- 'An affine space is nothing more than a vector space whose origin we try to forget about' (Marcel Berger): physics does this every time it says only energy differences matter, until gravity, which feels the vacuum energy itself, hands the origin back.
- Alice and Bob, holding different origins, disagree on almost every sum of points, except the ones whose weights add up to 1: that is exactly the geometry no observer can argue about.
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.
affine space (own: points acted on by translations, no origin)
- L1: Points and translations: a set A with a vector space acting on it freely and transitively; 'there is no distinguished point that serves as an origin'; 'There is no predefined concept of adding or multiplying points together'
- L2: Subtraction and combination: b - a is 'the unique vector' taking a to b (Weyl's axioms; the parallelogram property); affine combinations, 'weighted sums with numerical coefficients summing to 1', define barycentres independent of the origin
- L3: Flats and parallelism: an affine subspace a + V has direction V; 'two subspaces that share the same direction are said to be parallel'; generalized Playfair: 'for any point a of A there is one and only one affine subspace of direction V, which passes through a'; ratios of lengths of parallel segments kept
- L4: Affine maps: translations and linear maps centred at a point, L(a) = b + M(a - b); 'After making a choice of origin b, any affine map may be written uniquely as a combination of a translation and a linear map centred at b' (the affine group)
- L5: Closure at infinity: 'Affine spaces are contained in projective spaces'; adding a line at infinity whose points are classes of parallel lines gives a projective plane; 'any affine linear transformation extends uniquely to a projective linear transformation'
energy as an affine quantity (quantum vacuum)
- L1: Energies as an affine space for R: 'it is often not meaningful to talk about absolute energy, but it is meaningful to talk about energy differences'
- L2: 'The vacuum energy when it is defined picks out a canonical origin'
physical quantities modelled as affine spaces (time and position)
- L1: Time: 'Specific points in time (such as a date on the calendar) are points in the affine space, while durations (such as a number of days) are displacements'
- L2: Position: 'Physical space is often modelled as an affine space for R^3 in non-relativistic settings and R^(1,3) in the relativistic setting', called E(3) and E(1,3); Minkowski spacetime is 'an affine space when the origin is dropped'
- L3: Alice and Bob: two observers who take different points as origin get different linear combinations, but 'if the sum of the coefficients in a linear combination is 1' they agree
Who fills each layer
- L1: Time
- L5: Projective space