Update backlog and changelog
This commit is contained in:
parent
ac01b786a7
commit
4ae898dfe0
32
BACKLOG.md
32
BACKLOG.md
|
@ -4,17 +4,37 @@ Backlog
|
|||
Prove postulates in `Cat.Wishlist`:
|
||||
* `ntypeCommulative : n ≤ m → HasLevel ⟨ n ⟩₋₂ A → HasLevel ⟨ m ⟩₋₂ A`
|
||||
|
||||
Prove that these two formulations of univalence are equivalent:
|
||||
|
||||
∀ A B → isEquiv (A ≡ B) (A ≅ B) (id-to-iso A B)
|
||||
∀ A → isContr (Σ[ X ∈ Object ] A ≅ X)
|
||||
|
||||
Prove univalence for the category of
|
||||
* the opposite category
|
||||
* sets
|
||||
This does not follow trivially from `Cubical.Univalence.univalence` because
|
||||
objects are not `Set` but `hSet`
|
||||
* functors and natural transformations
|
||||
|
||||
Prove:
|
||||
* `isProp (Product ...)`
|
||||
* `isProp (HasProducts ...)`
|
||||
|
||||
Ideas for future work
|
||||
---------------------
|
||||
|
||||
It would be nice if my formulation of monads is not so "stand-alone" as it is at
|
||||
the moment.
|
||||
|
||||
We can built up the notion of monads and related concept in multiple ways as
|
||||
demonstrated in the two equivalent formulations of monads (kleisli/monoidal):
|
||||
There seems to be a category-theoretic approach and an approach more in the
|
||||
style of functional programming as e.g. the related typeclasses in the
|
||||
standard library of Haskell.
|
||||
|
||||
It would be nice to build up this hierarchy in two ways: The
|
||||
"category-theoretic" way and the "functional programming" way.
|
||||
|
||||
Here is an overview of some of the concepts that need to be developed to acheive
|
||||
this:
|
||||
|
||||
* Functor ✓
|
||||
* Applicative Functor ✗
|
||||
* Lax monoidal functor ✗
|
||||
|
@ -26,9 +46,7 @@ Prove:
|
|||
* Monoidal monad ✓
|
||||
* Kleisli monad ✓
|
||||
* Kleisli ≃ Monoidal ✓
|
||||
* Problem 2.3 in [voe]
|
||||
* Problem 2.3 in [voe] ✓
|
||||
* 1st contruction ~ monoidal ✓
|
||||
* 2nd contruction ~ klesli ✓
|
||||
* 1st ≃ 2nd ✗
|
||||
I've managed to set this up so I should be able to reuse the proof that
|
||||
Kleisli ≃ Monoidal, but I don't know why it doesn't typecheck.
|
||||
* 1st ≃ 2nd ✓
|
||||
|
|
19
CHANGELOG.md
19
CHANGELOG.md
|
@ -1,6 +1,25 @@
|
|||
Changelog
|
||||
=========
|
||||
|
||||
Version 1.4.1
|
||||
-------------
|
||||
Defines a module to work with equivalence providing a way to go between
|
||||
equivalences and quasi-inverses (in the parlance of HoTT).
|
||||
|
||||
Finishes the proof that the category of homotopy-sets are univalent.
|
||||
|
||||
Defines a custom "prelude" module that wraps the `cubical` library and provides
|
||||
a few utilities.
|
||||
|
||||
Reorders Category.isIdentity such that the left projection is left identity.
|
||||
|
||||
Include some text for the half-time report.
|
||||
|
||||
Renames IsProduct.isProduct to IsProduct.ump to avoid ambiguity in some
|
||||
circumstances.
|
||||
|
||||
[WIP]: Adds some stuff about propositionality for products.
|
||||
|
||||
Version 1.4.0
|
||||
-------------
|
||||
Adds documentation to a number of modules.
|
||||
|
|
Loading…
Reference in a new issue