nLab span in simplicial type theory

Context

(,1)(\infty,1)-Category theory

Internal (,1)(\infty,1)-Categories

Directed homotopy type theory

Contents

Idea

The concept of a span in simplicial type theory. Note that in the context of simplicial type theory, spans make sense for any type, not just the Segal types.

Definition

Using morphisms

Let AA be a type in simplicial type theory, and let x:Ax:A and y:Ay:A be two elements in AA. A span of x:Ax:A and y:Ay:A is a record consisting of

  • an element z:Az:A,

  • a morphism p x:hom A(z,x)p_x:\mathrm{hom}_A(z, x),

  • and a morphism p y:hom A(z,y)p_y:\mathrm{hom}_A(z, y),

The type of spans span A(x,y)\mathrm{span}_A(x, y) between x:Ax:A and y:Ay:A is then the respective record type.

Using functions from shapes

Let Λ 2 2\Lambda^2_2 denote the (2,2)(2,2)-horn type.

Let AA be a type in simplicial type theory, and let x:Ax:A and y:Ay:A be two elements in AA. A span of x:Ax:A and y:Ay:A is a record consisting of

  • a function f:Λ 2 2Af:\Lambda^2_2 \to A,

  • an identification p x:f(0,1)=xp_x:f(0, 1) = x,

  • and an identification p y:f(1,1)=yp_y:f(1, 1) = y,

The type of spans span A(x,y)\mathrm{span}_A(x, y) between x:Ax:A and y:Ay:A is then the respective record type.

Left quotients

A left quotient of a span f:hom A(x,y)f:\mathrm{hom}_A(x, y) and h:hom A(x,z)h:\mathrm{hom}_A(x, z) is a morphism g:hom A(y,z)g:\mathrm{hom}_A(y, z) such that hh is the unique composite of ff and gg:

leftQuotient(f,h) g:hom A(y,z)isUniqueComposite(f,g,h)\mathrm{leftQuotient}(f, h) \coloneqq \sum_{g:\mathrm{hom}_A(y, z)} \mathrm{isUniqueComposite}(f, g, h)

We say that the span (f,h)(f, h) is left divisible if it has a left quotient.

If AA is a Segal type, this is equivalent to the usual definition of a left quotient

leftQuotient(f,h) g:hom A(y,z)gf=h\mathrm{leftQuotient}(f, h) \coloneqq \sum_{g:\mathrm{hom}_A(y, z)} g \circ f = h

Given a morphism f:hom A(x,y)f:\mathrm{hom}_A(x, y), the span (id x,f)(\mathrm{id}_x, f) is always left divisible with left quotient ff. A retraction is a left quotient of the span (f,id x)(f, \mathrm{id}_x).

Morphisms of spans

Using morphisms

Given elements x:Ax:A and y:Ay:A and spans f:span A(x,y)f:\mathrm{span}_A(x, y) and g:span A(x,y)g:\mathrm{span}_A(x, y), a morphism of spans between ff and gg is a record consisting of

  • a morphism h:hom A(pr 1(f),pr 1(y))h:\mathrm{hom}_A(\mathrm{pr}_1(f), \mathrm{pr}_1(y))

  • a witness that pr 2(g)\mathrm{pr}_2(g) is the unique composite of pr 2(f)\mathrm{pr}_2(f) and hh

  • a witness that pr 3(g)\mathrm{pr}_3(g) is the unique composite of pr 3(f)\mathrm{pr}_3(f) and hh.

The type of morphisms of spans homspan A(x,y,f,g)\mathrm{homspan}_A(x, y, f, g) is then the respective record type.

If AA is a Segal type, then the record consists of fields

  • a morphism h:hom A(pr 1(f),pr 1(y))h:\mathrm{hom}_A(\mathrm{pr}_1(f), \mathrm{pr}_1(y))

  • an identification pr 2(g)=pr 2(f)h\mathrm{pr}_2(g) = \mathrm{pr}_2(f) \circ h

  • an identification pr 3(g)=pr 3(f)h\mathrm{pr}_3(g) = \mathrm{pr}_3(f) \circ h.

Using functions from shapes

Given elements x:Ax:A and y:Ay:A and spans f:span A(x,y)f:\mathrm{span}_A(x, y) and g:span A(x,y)g:\mathrm{span}_A(x, y), a morphism of spans between ff and gg is a record consisting of

  • a function h:𝕀Ah:\mathbb{I} \to A

  • an identification p x:h(0)=pr 1(f)p_x:h(0) = \mathrm{pr}_1(f),

  • an identification p y:h(1)=pr 1(g)p_y:h(1) = \mathrm{pr}_1(g),

  • a function j:Δ 2Aj:\Delta^2 \to A,

  • a witness that jj is the unique element of the fiber of the canonical function (Δ 2A)(Λ 2 2A)(\Delta^2 \to A) \to (\Lambda^2_2 \to A) at ff

  • a witness that jj is the unique element of the fiber of the canonical function (Δ 2A)(Λ 2 2A)(\Delta^2 \to A) \to (\Lambda^2_2 \to A) at gg

The type of morphisms of spans homspan A(x,y,f,g)\mathrm{homspan}_A(x, y, f, g) is then the respective record type.

Last revised on April 10, 2025 at 17:24:20. See the history of this page for a list of all contributions to it.