Zulip Chat Archive
Stream: maths
Topic: topology rational addition is continuous
RustyYato (Jan 17 2025 at 22:48):
I'm trying to prove that addition over the rationals is continuous in both arguments.
But I'm now sure how to proceed. I think there is something I am fundamentally misunderstanding about how the product topology works.
Here's a MWE of what I'm trying to prove. I know that the line this proof went down is hopeless, but I'm not sure what else I can do.
(NOTE: I would like to avoid using rat_add_continuous_lemma, since I'm trying to learn to do this myself).
import Mathlib.Topology.Constructions
import Mathlib.Topology.Instances.Rat
open TopologicalSpace
instance : Continuous (fun x: ℚ × ℚ => x.fst + x.snd) where
isOpen_preimage := by
intro S S_open
apply GenerateOpen.basic
dsimp
left
dsimp
refine ⟨S, S_open, ?_⟩
sorry -- what to do now?
RustyYato (Jan 17 2025 at 23:00):
I was able to prove that it's continuous in each argument separately (i.e. fun x: ℚ => x + a
with some fixed a
). But not when generalizing to above.
Jireh Loreaux (Jan 17 2025 at 23:19):
The answer is that there are 15 different ways to prove this result. The key question is: what is the proof on paper that you want to formalize? The answer to that will influence the advice we give.
(e.g., do you want to use a standard ε
and δ
proof? Is there a reason you chose ℚ
specifically, and not ℝ
?)
RustyYato (Jan 17 2025 at 23:25):
Yes, a standard ε-δ
proof will work. I've chosen ℚ
because, in reality, I'm not directly using mathlib. I'm just using it as a reference, and am writing out the proofs myself, sometimes making slightly different decisions from mathlib.
At this point, I don't have ℝ
setup enough to be viable, and as part of that I'm building out my theories on ℚ
.
Last updated: May 02 2025 at 03:31 UTC