Difference between revisions of "Exercises:Mond - Topology - 1/Question 7"
From Maths
(Saving work) |
m (→Solution body: Added note) |
||
Line 31: | Line 31: | ||
*** Suppose {{M|1=\pi(x)=\pi(y)}}, we must show that in this case {{M|1=f(x)=y(y)}}. | *** Suppose {{M|1=\pi(x)=\pi(y)}}, we must show that in this case {{M|1=f(x)=y(y)}}. | ||
**** Suppose {{M|x\in D^2-\partial D^2}} (meaning {{M|x\in D^2}} but {{M|x\notin \partial D^2}}, ie {{M|-}} denotes [[relative complement]]) | **** Suppose {{M|x\in D^2-\partial D^2}} (meaning {{M|x\in D^2}} but {{M|x\notin \partial D^2}}, ie {{M|-}} denotes [[relative complement]]) | ||
− | ***** In this case we must have {{M|1=x=y}}, as otherwise we'd not have {{M|1=\pi(x)=\pi(y)}} {{ | + | ***** In this case we must have {{M|1=x=y}}, as otherwise we'd not have {{M|1=\pi(x)=\pi(y)}} (for {{M|x\in D^2-\partial D^2}} we have {{M|1=\pi(x)=[x]=\{x\} }}, that is that the {{plural|equivalence class|es}} are singletons. So if {{M|1=\pi(x)=\pi(y)}} we must have {{M|1=\pi(y)=[y]=\{x\}=[x]=\pi(x) }}; so {{M|y}} can only be {{M|x}}) |
***** If {{M|1=x=y}} then by the nature of {{M|f}} being a [[function]] we must have {{M|1=f(x)=f(y)}}, we're done in this case | ***** If {{M|1=x=y}} then by the nature of {{M|f}} being a [[function]] we must have {{M|1=f(x)=f(y)}}, we're done in this case | ||
**** Suppose {{M|x\in \partial D^2}} (the only case not covered) and {{M|1=\pi(x)=\pi(y)}}, we must show {{M|1=f(x)=f(y)}} | **** Suppose {{M|x\in \partial D^2}} (the only case not covered) and {{M|1=\pi(x)=\pi(y)}}, we must show {{M|1=f(x)=f(y)}} |
Revision as of 22:08, 11 October 2016
Contents
[hide]Section B
Question 7
Let D2 denote the closed unit disk in R2 and define an equivalence relation on D2 by setting x1∼x2 if ∥x1∥=∥x2∥=1 ("collapsing the boundary to a single point"). Show that D2∼ is homeomorphic to S2 - the sphere.
- Hint: first define a surjection (:D2→S2) mapping all of ∂D2 to the north pole. This may be defined using a good picture or a formula.
Solution
Definitions:
- H denotes the hemisphere in my picture.
- E:D2→H is the composition of maps in my diagram that take D2, double its radius, then embed it in R3 then "pop it out" into a hemisphere. We take it as obvious that it is a homeomorphism
- f′:H→S2, this is the map in the top picture. It takes the hemisphere and pulls the boundary/rim in (along the blue lines) to the north pole of the red sphere. f′(∂H)=(0,0,1)∈R3, it should be clear that for all x∈H−∂H that f′(x) is intended to be a point on the red sphere and that f′|H−∂H is injective. It is also taken as clear that f′ is surjective
- Note: Click the pictures for a larger version
- D2∼ and D2/∼ denote the quotient space, with this definition we get a canonical projection, π:D2→D2/∼ given by π:x↦[x] where [x] denotes the equivalence class of x
- Lastly, we define f:D2→S2 to be the composition of E and f′, that is: f:=f′∘E, meaning f:x↦f′(E(x))
The situation is shown diagramatically below:
Outline of the solution:
- We then want apply the passing to the quotient theorem to yield a commutative diagram:
- Lastly, we will show that ˉf is a homeomorphism using the compact-to-Hausdorff theorem
Solution body
First we must show the requirements for applying passing to the quotient are satisfied.
- We know already the maps involved are continuous and that π is a quotient map. We only need to show:
- f is constant on the fibres of π, which is equivalent to:
- ∀x,y∈D2[π(x)=π(y)⟹f(x)=f(y)]
- f is constant on the fibres of π, which is equivalent to:
- Let us show this remaining condition:
- Let x,y∈D2 be given.
- Suppose π(x)≠π(y), then by the nature of logical implication the implication is true regardless of f(x) and f(y)'s equality. We're done in this case.
- Suppose π(x)=π(y), we must show that in this case f(x)=y(y).
- Suppose x∈D2−∂D2 (meaning x∈D2 but x∉∂D2, ie − denotes relative complement)
- In this case we must have x=y, as otherwise we'd not have π(x)=π(y) (for x∈D2−∂D2 we have π(x)=[x]={x}, that is that the equivalence classes are singletons. So if π(x)=π(y) we must have π(y)=[y]={x}=[x]=π(x); so y can only be x)
- If x=y then by the nature of f being a function we must have f(x)=f(y), we're done in this case
- Suppose x∈∂D2 (the only case not covered) and π(x)=π(y), we must show f(x)=f(y)
- Clearly if x∈∂D2 and π(x)=π(y) we must have y∈∂D2.
- E(x) is mapped to the boundary/rim of H, as is E(y) and f′(any point on the rim of H)=(0,0,1)∈R3
- Thus f′(E(x))=f′(E(y)), but f′(E(x)) is the very definition of f(x), so clearly:
- f(x)=f(y) as required.
- Clearly if x∈∂D2 and π(x)=π(y) we must have y∈∂D2.
- Suppose x∈D2−∂D2 (meaning x∈D2 but x∉∂D2, ie − denotes relative complement)
- Let x,y∈D2 be given.
We may now apply the passing to the quotient theorem. This yields:
- A continuous map, ˉf:D2/∼→S2 where f=ˉf∘π
In order to apply the compact-to-Hausdorff theorem and show ˉf is a homeomorphism we must show it is continuous and bijective. We already have continuity (as a result of the passing-to-the-quotient theorem), we must show it is bijective.
Notes
- Jump up ↑ Technically a diagram is said to commute if all paths through it yield equal compositions, this means that we also require f=f′∘E, which we already have by definition of f!
References