Spectral/homotopy/susp.hlean
Floris van Doorn c3650048f0 fixes and additions
add some properties about pointed maps and groups
2018-09-10 18:04:28 +02:00

192 lines
8.4 KiB
Text
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

import .pushout types.pointed2 ..move_to_lib
open susp eq pointed function is_equiv lift equiv is_trunc nat
namespace susp
variables {X X' Y Y' Z : Type*}
definition iterate_susp_iterate_susp_rev (n m : ) (A : Type*) :
iterate_susp n (iterate_susp m A) ≃* iterate_susp (m + n) A :=
begin
induction n with n e,
{ reflexivity },
{ exact susp_pequiv e }
end
definition iterate_susp_pequiv [constructor] (n : ) {X Y : Type*} (f : X ≃* Y) :
iterate_susp n X ≃* iterate_susp n Y :=
begin
induction n with n e,
{ exact f },
{ exact susp_pequiv e }
end
open algebra nat
definition iterate_susp_iterate_susp (n m : ) (A : Type*) :
iterate_susp n (iterate_susp m A) ≃* iterate_susp (n + m) A :=
iterate_susp_iterate_susp_rev n m A ⬝e* pequiv_of_eq (ap (λk, iterate_susp k A) (add.comm m n))
definition plift_susp.{u v} : Π(A : Type*), plift.{u v} (susp A) ≃* susp (plift.{u v} A) :=
begin
intro A,
calc
plift.{u v} (susp A) ≃* susp A : by exact (pequiv_plift (susp A))⁻¹ᵉ*
... ≃* susp (plift.{u v} A) : by exact susp_pequiv (pequiv_plift.{u v} A)
end
definition is_contr_susp [instance] (A : Type) [H : is_contr A] : is_contr (susp A) :=
begin
apply is_contr.mk north,
intro x, induction x,
reflexivity,
exact merid !center,
apply eq_pathover_constant_left_id_right, apply square_of_eq,
exact whisker_left idp (ap merid !eq_of_is_contr)
end
definition loop_susp_pintro_phomotopy {X Y : Type*} {f g : ⅀ X →* Y} (p : f ~* g) :
loop_susp_pintro X Y f ~* loop_susp_pintro X Y g :=
pwhisker_right (loop_susp_unit X) (Ω⇒ p)
variables {A₀₀ A₂₀ A₀₂ A₂₂ : Type*}
{f₁₀ : A₀₀ →* A₂₀} {f₁₂ : A₀₂ →* A₂₂}
{f₀₁ : A₀₀ →* A₀₂} {f₂₁ : A₂₀ →* A₂₂}
definition susp_functor_psquare (p : psquare f₁₀ f₁₂ f₀₁ f₂₁) :
psquare (⅀→ f₁₀) (⅀→ f₁₂) (⅀→ f₀₁) (⅀→ f₂₁) :=
!susp_functor_pcompose⁻¹* ⬝* susp_functor_phomotopy p ⬝* !susp_functor_pcompose
definition susp_to_loop_psquare (f₁₀ : A₀₀ →* A₂₀) (f₁₂ : A₀₂ →* A₂₂)
(f₀₁ : susp A₀₀ →* A₀₂) (f₂₁ : susp A₂₀ →* A₂₂) : psquare (⅀→ f₁₀) f₁₂ f₀₁ f₂₁ →
psquare f₁₀ (Ω→ f₁₂) (loop_susp_pintro A₀₀ A₀₂ f₀₁) (loop_susp_pintro A₂₀ A₂₂ f₂₁) :=
begin
intro p,
refine pvconcat _ (ap1_psquare p),
exact (loop_susp_unit_natural f₁₀)⁻¹*
end
definition loop_to_susp_square (f₁₀ : A₀₀ →* A₂₀) (f₁₂ : A₀₂ →* A₂₂)
(f₀₁ : A₀₀ →* Ω A₀₂) (f₂₁ : A₂₀ →* Ω A₂₂) : psquare f₁₀ (Ω→ f₁₂) f₀₁ f₂₁ →
psquare (⅀→ f₁₀) f₁₂ (susp_pelim A₀₀ A₀₂ f₀₁) (susp_pelim A₂₀ A₂₂ f₂₁) :=
begin
intro p,
refine susp_functor_psquare p ⬝v* _,
exact psquare_transpose (loop_susp_counit_natural f₁₂)
end
open pushout unit prod sigma sigma.ops
section
parameters {A : Type*} {n : } [HA : is_conn n A]
-- we end up not using this, because to prove that the
-- composition with the first projection is loop_susp_counit A
-- is hideous without HIT computations on path constructors
parameter (A)
definition pullback_diagonal_prod_of_wedge : susp (Ω A)
≃ Σ (a : A) (w : wedge A A), prod_of_wedge w = (a, a) :=
begin
refine equiv.trans _
(comm_equiv_unc (λ z, prod_of_wedge (prod.pr1 z) = (prod.pr2 z, prod.pr2 z))),
apply equiv.symm,
apply equiv.trans (sigma_equiv_sigma_right
(λ w, sigma_equiv_sigma_right
(λ a, prod_eq_equiv (prod_of_wedge w) (a, a)))),
apply equiv.trans !pushout.flattening', esimp,
fapply pushout.equiv
(λ z, ⟨pt, z.2⟩) (λ z, ⟨pt, glue z.1 ▸ z.2⟩) (λ p, star) (λ p, star),
{ apply equiv.trans !sigma_unit_left, fapply equiv.MK,
{ intro z, induction z with a w, induction w with p q, exact p ⬝ q⁻¹ },
{ intro p, exact ⟨pt, (p, idp)⟩ },
{ intro p, reflexivity },
{ intro z, induction z with a w, induction w with p q, induction q,
reflexivity } },
{ fapply equiv.MK,
{ intro z, exact star },
{ intro u, exact ⟨pt, ⟨pt, (idp, idp)⟩ ⟩ },
{ intro u, induction u, reflexivity },
{ intro z, induction z with a w, induction w with b z,
induction z with p q, induction p, esimp at q, induction q,
reflexivity } },
{ fapply equiv.MK,
{ intro z, exact star },
{ intro u, exact ⟨pt, ⟨pt, (idp, idp)⟩ ⟩ },
{ intro u, induction u, reflexivity },
{ intro z, induction z with a w, induction w with b z,
induction z with p q, induction q, esimp at p, induction p,
reflexivity } },
{ intro z, induction z with u w, induction u, induction w with a z,
induction z with p q, reflexivity },
{ intro z, induction z with u w, induction u, induction w with a z,
induction z with p q, reflexivity }
end
parameter {A}
-- instead we directly compare the fibers, using flattening twice
definition fiber_loop_susp_counit_equiv (a : A)
: fiber (loop_susp_counit A) a ≃ fiber prod_of_wedge (a, a) :=
begin
apply equiv.trans !fiber.sigma_char, apply equiv.trans !pushout.flattening',
apply equiv.symm, apply equiv.trans !fiber.sigma_char,
apply equiv.trans (sigma_equiv_sigma_right
(λ w, prod_eq_equiv (prod_of_wedge w) (a, a))), esimp,
apply equiv.trans !pushout.flattening',
esimp,
fapply pushout.equiv (λ z, ⟨pt, z.2⟩) (λ z, ⟨pt, glue z.1 ▸ z.2⟩)
(λ z, ⟨star, z.2⟩) (λ z, ⟨star, glue z.1 ▸ z.2⟩),
{ fapply equiv.MK,
{ intro w, induction w with u z, induction z with p q,
exact ⟨q ⬝ p⁻¹, q⟩ },
{ intro z, induction z with p q, apply dpair star,
exact (p⁻¹ ⬝ q, q) },
{ intro z, induction z with p q, esimp, induction q, esimp,
rewrite [idp_con,inv_inv] },
{ intro w, induction w with u z, induction u, induction z with p q,
esimp, induction q, rewrite [idp_con,inv_inv] } },
{ fapply equiv.MK,
{ intro w, induction w with b z, induction z with p q, exact ⟨star, q⟩ },
{ intro z, induction z with u p, induction u, esimp at p, esimp,
apply dpair a, esimp, exact (idp, p) },
{ intro z, induction z with u p, induction u, reflexivity },
{ intro w, induction w with b z, induction z with p q, esimp,
induction p, reflexivity } },
{ fapply equiv.MK,
{ intro w, induction w with b z, induction z with p q, exact ⟨star, p⟩ },
{ intro z, induction z with u p, induction u, esimp at p, esimp,
apply dpair a, esimp, exact (p, idp) },
{ intro z, induction z with u p, induction u, reflexivity },
{ intro w, induction w with b z, induction z with p q, esimp,
induction q, reflexivity } },
{ intro w, induction w with u z, induction u, induction z with p q,
reflexivity },
{ intro w, induction w with u z, induction u, induction z with p q,
esimp, induction q, esimp, krewrite prod_transport, fapply sigma_eq,
{ exact idp },
{ esimp, rewrite eq_transport_Fl, rewrite eq_transport_Fl,
krewrite elim_glue, krewrite [-ap_compose' pr1 prod_of_wedge (glue star)],
krewrite elim_glue, esimp, apply eq_pathover, rewrite idp_con, esimp,
apply square_of_eq, rewrite [idp_con,idp_con,inv_inv] } }
end
include HA
open is_conn trunc_index
parameter (A)
-- connectivity of loop_susp_counit
definition is_conn_fun_loop_susp_counit {k : } (H : k ≤ 2 * n)
: is_conn_fun k (loop_susp_counit A) :=
begin
intro a, apply is_conn.is_conn_equiv_closed_rev k (fiber_loop_susp_counit_equiv a),
fapply @is_conn.is_conn_of_le (fiber prod_of_wedge (a, a)) k (2 * n)
(of_nat_le_of_nat H),
assert H : of_nat (2 * n) = of_nat n + of_nat n,
{ rewrite (of_nat_add_of_nat n n), apply ap of_nat,
apply trans (nat.mul_comm 2 n),
apply ap (λ k, k + n), exact nat.zero_add n },
rewrite H,
exact is_conn_fun_prod_of_wedge n n A A (a, a)
end
end
end susp