Zulip Chat Archive

Stream: mathlib4

Topic: Homeomorph.curry


Vasilii Nesterov (Nov 06 2025 at 16:23):

In #26149 I proved

def Homeomorph.curry {X Y Z : Type*} [TopologicalSpace Z] : (X × Y  Z) ≃ₜ (X  Y  Z)

but found that we already have

def Homeomorph.curry {X Y Z : Type*} [TopologicalSpace X] [TopologicalSpace Y] [TopologicalSpace Z]
    [LocallyCompactSpace X] [LocallyCompactSpace Y] :
    C(X × Y, Z) ≃ₜ C(X, C(Y, Z))

Which one should be called curry?

Patrick Massot (Nov 06 2025 at 19:36):

Your statement looks too specialized to get this name, so I would search another name for it, clearly mentioning the product topology.

Aaron Liu (Nov 06 2025 at 19:59):

Patrick Massot said:

Your statement looks too specialized to get this name, so I would search another name for it, clearly mentioning the product topology.

What do you mean by this?

Jireh Loreaux (Nov 06 2025 at 22:37):

Patrick's point is that the topology on X × Y → Z is the product topology (or, in Lean parlance, docs#Pi.topologicalSpace), and likewise for X → Y → Z.

Aaron Liu (Nov 06 2025 at 22:54):

oh yeah that's yeah ok

Aaron Liu (Nov 06 2025 at 22:54):

I thought Patrick was referring to the binary product


Last updated: Dec 20 2025 at 21:32 UTC