- #1
Petrarch
- 2
- 0
Completion of Metric Space Proof from "Intro. to Func. Analysis w/ Applications"
I have started studying Functional Analysis following "Introduction to Functional Analysis with Applications". In chapter 1-6 there is the following proof
For any metric space [itex]X[/itex], there is a complete metric space [itex]\hat{X}[/itex] which has a subspace [itex]W[/itex] that is isometric with [itex]X[/itex] and is dense in [itex]\hat{X}[/itex]
(Page 1 & 2) http://i.imgur.com/CRXjh.png
(Page 3 & 4) http://i.imgur.com/PogqC.png
I think I understand parts (a) and (b). At the top of page 3, section (c) where it is proving [itex]\hat{X}[/itex] is complete it states:
Let [itex](\hat{x_{n}})[/itex] be any Cauchy Sequence in [itex]\hat{X}[/itex]. Since [itex]W[/itex] is dense in [itex]\hat{X}[/itex], for every [itex]\hat{x_{n}}[/itex], there is a [itex]\hat{z_{n}}\varepsilon W[/itex] such that [itex]\hat{d}(\hat{x_{n}},\hat{z_{n}}) < \frac{1}{n}[/itex]
I do not understand why we choose [itex] \frac{1}{n}[/itex], would some ε > 0, for each n, not suffice? I assume it must not, but I don't see why, so I must not understand this proof.
Any help would be greatly appreciated, i am pretty dumb and this has puzzled me for a couple days.
Homework Statement
I have started studying Functional Analysis following "Introduction to Functional Analysis with Applications". In chapter 1-6 there is the following proof
For any metric space [itex]X[/itex], there is a complete metric space [itex]\hat{X}[/itex] which has a subspace [itex]W[/itex] that is isometric with [itex]X[/itex] and is dense in [itex]\hat{X}[/itex]
(Page 1 & 2) http://i.imgur.com/CRXjh.png
(Page 3 & 4) http://i.imgur.com/PogqC.png
I think I understand parts (a) and (b). At the top of page 3, section (c) where it is proving [itex]\hat{X}[/itex] is complete it states:
Let [itex](\hat{x_{n}})[/itex] be any Cauchy Sequence in [itex]\hat{X}[/itex]. Since [itex]W[/itex] is dense in [itex]\hat{X}[/itex], for every [itex]\hat{x_{n}}[/itex], there is a [itex]\hat{z_{n}}\varepsilon W[/itex] such that [itex]\hat{d}(\hat{x_{n}},\hat{z_{n}}) < \frac{1}{n}[/itex]
I do not understand why we choose [itex] \frac{1}{n}[/itex], would some ε > 0, for each n, not suffice? I assume it must not, but I don't see why, so I must not understand this proof.
Any help would be greatly appreciated, i am pretty dumb and this has puzzled me for a couple days.