Zulip Chat Archive

Stream: mathlib4

Topic: Iso VS equality in cartesian product of categories


Nicolas Rolland (Jul 20 2024 at 09:01):

We have a lemma for the cartesian product of categories, CategoryTheory.prod.etaIso, proving the isomorphism (X.1, X.2) X

I don't find any corresponding lemma for the equality.
Is there any deep reason for this ?

This product is not any universal product, characterized up to iso.
It is the product that we chose and built up, so we have

def prod.etaEq (X : C × D) : (X.1, X.2) = X := rfl

Kevin Buzzard (Jul 21 2024 at 07:27):

Equality of objects is not a well-behaved concept in category theory, is presumably why we don't have it

Kevin Buzzard (Jul 21 2024 at 07:28):

Like groups are "only defined up to isomorphism", categories are "only defined up to equivalence", and equality of objects is not preserved by equivalence. I can't imagine anyone objecting to you adding it though.

Kim Morrison (Jul 21 2024 at 13:01):

I suspect rather that here that statement, which is just about products and nothing about categories, already exists.


Last updated: May 02 2025 at 03:31 UTC