## Quick question on basic topology

For the discussion of math. Duh.

Moderators: gmalivuk, Moderators General, Prelates

Harg
Posts: 130
Joined: Thu Jul 10, 2008 8:24 pm UTC

### Quick question on basic topology

Could someone oblige me and give a hint for a homeomorphism between [imath]\mathbb{Q}[/imath] and [imath]\mathbb{Q}\cap(0,\infty)[/imath]. Thank you.

On a side note, does anyone know if AC is required to prove that every second-countable space is separable or can this be done without it?

jestingrabbit
Factoids are just Datas that haven't grown up yet
Posts: 5967
Joined: Tue Nov 28, 2006 9:50 pm UTC
Location: Sydney

### Re: Quick question on basic topology

Harg wrote:Could someone oblige me and give a hint for a homeomorphism between [imath]\mathbb{Q}[/imath] and [imath]\mathbb{Q}\cap(0,\infty)[/imath].

Think rational functions with small powers of x, and maybe a little bit of piecewise too.
ameretrifle wrote:Magic space feudalism is therefore a viable idea.

Harg
Posts: 130
Joined: Thu Jul 10, 2008 8:24 pm UTC

### Re: Quick question on basic topology

I've been thinking about it like that. What I've tried to do is to contract one ray of the rationals into an interval and then extend it on the other side with a linear function or something. The problem I keep running into is that I loose bijectivity on the nonlinear side.

jestingrabbit
Factoids are just Datas that haven't grown up yet
Posts: 5967
Joined: Tue Nov 28, 2006 9:50 pm UTC
Location: Sydney

### Re: Quick question on basic topology

You might have more luck trying to construct it in the other direction and then taking the inverse.
ameretrifle wrote:Magic space feudalism is therefore a viable idea.

skeptical scientist
closed-minded spiritualist
Posts: 6142
Joined: Tue Nov 28, 2006 6:09 am UTC
Location: San Francisco

### Re: Quick question on basic topology

There's probably some really clever way of doing it with a nice formula, but it's pretty easy to construct a piecewise linear bijection.
I'm looking forward to the day when the SNES emulator on my computer works by emulating the elementary particles in an actual, physical box with Nintendo stamped on the side.

"With math, all things are possible." —Rebecca Watson

Yakk
Poster with most posts but no title.
Posts: 11129
Joined: Sat Jan 27, 2007 7:27 pm UTC
Location: E pur si muove

### Re: Quick question on basic topology

What about the isomorphism that takes the group (R,+,0) to the group (R+,*,1)? That generates a topological isomorphism as well, I think, admittedly on a slightly different space.

And you can use it as inspiration to solve the problem for Q (admittedly, it won't be a group isomorphism I don't think!)
One of the painful things about our time is that those who feel certainty are stupid, and those with any imagination and understanding are filled with doubt and indecision - BR

Last edited by JHVH on Fri Oct 23, 4004 BCE 6:17 pm, edited 6 times in total.

ihope127
Posts: 66
Joined: Mon Jul 30, 2007 3:59 am UTC
Location: 127.0.0.1
Contact:

### Re: Quick question on basic topology

skeptical scientist wrote:There's probably some really clever way of doing it with a nice formula

How about = 1/(y - x)?
There is a significant chance that an artificial intelligence created within the next few decades will not value humanity and therefore will treat us as we treat animals. It would be awesome if xkcd mentioned this.
----
Find me on freenode as uorygl.

skeptical scientist
closed-minded spiritualist
Posts: 6142
Joined: Tue Nov 28, 2006 6:09 am UTC
Location: San Francisco

### Re: Quick question on basic topology

Yakk wrote:What about the isomorphism that takes the group (R,+,0) to the group (R+,*,1)? That generates a topological isomorphism as well, I think, admittedly on a slightly different space.

And you can use it as inspiration to solve the problem for Q (admittedly, it won't be a group isomorphism I don't think!)

The nice isomorphisms from (R,+,0) to the group (R+,*,1) are of the form [imath]x \mapsto a^x[/imath] for some positive a different from 1. Clearly that doesn't induce a homeomorphism from Q to Q+. Moreover, (Q,+) and (Q+,*) are not isomorphic as groups, so the same trick wouldn't work. I'm not quite sure what you mean by using it as inspiration here.
I'm looking forward to the day when the SNES emulator on my computer works by emulating the elementary particles in an actual, physical box with Nintendo stamped on the side.

"With math, all things are possible." —Rebecca Watson

ihope127
Posts: 66
Joined: Mon Jul 30, 2007 3:59 am UTC
Location: 127.0.0.1
Contact:

### Re: Quick question on basic topology

ihope127 wrote:
skeptical scientist wrote:There's probably some really clever way of doing it with a nice formula

How about = 1/(y - x)?

...well, if that isn't the most unusual equation I've ever seen. I meant y = 1/(y - x). I see now, though, that that isn't much of a bijection:

1/y = y - x
y - 1/y = x
y^2 - xy - 1 = 0
y = (x + sqrt(x^2 + 4))/2

I'll have to try again.
There is a significant chance that an artificial intelligence created within the next few decades will not value humanity and therefore will treat us as we treat animals. It would be awesome if xkcd mentioned this.
----
Find me on freenode as uorygl.

Yakk
Poster with most posts but no title.
Posts: 11129
Joined: Sat Jan 27, 2007 7:27 pm UTC
Location: E pur si muove

### Re: Quick question on basic topology

skeptical scientist wrote:
Yakk wrote:What about the isomorphism that takes the group (R,+,0) to the group (R+,*,1)? That generates a topological isomorphism as well, I think, admittedly on a slightly different space.

And you can use it as inspiration to solve the problem for Q (admittedly, it won't be a group isomorphism I don't think!)

The nice isomorphisms from (R,+,0) to the group (R+,*,1) are of the form [imath]x \mapsto a^x[/imath] for some positive a different from 1. Clearly that doesn't induce a homeomorphism from Q to Q+. Moreover, (Q,+) and (Q+,*) are not isomorphic as groups, so the same trick wouldn't work. I'm not quite sure what you mean by using it as inspiration here.

Under the group isomorphism mentioned, R- maps to (0,1), {0} maps to {1}, and R+ maps to (1, infinity). That's about the only inspiration.

It isn't that hard to change those Rs to Qs:

Now, a: Q-->(0,1) in Q := x -> 1/(-x+1) maps Q- to (0,1) in Q topologically nicely.
And b: Q+->(1, infinity) in Q := x -> x+1 maps Q+ to (1, infinity) topologically nicely.

They do not, however, behave topologically nicely near 1. And that won't do.

The first step is obvious: fill in the gap.
And c: {0} -> {1} defined in the only way.

The next step is to make the stitch work better. We need small negative numbers to map to a number near 1, instead of near 0. That's easy:
Take d: Q- -> (0,1) defined by x -> 1-a(x).

Then each of b, c, and d map a distinct portion of Q to a distinct portion of Q+, and the limits along the points that they stitch together behave topologically nicely.

Edit: I missed a +1 and a -.
Last edited by Yakk on Sat Nov 15, 2008 4:40 pm UTC, edited 1 time in total.
One of the painful things about our time is that those who feel certainty are stupid, and those with any imagination and understanding are filled with doubt and indecision - BR

Last edited by JHVH on Fri Oct 23, 4004 BCE 6:17 pm, edited 6 times in total.

Harg
Posts: 130
Joined: Thu Jul 10, 2008 8:24 pm UTC

### Re: Quick question on basic topology

Thanks to everyone. I had to go for a drive across half the country after jestingrabbit's second reply, so I had a lot of time to come up with $f(x)=\begin{cases}\frac{x-1}{x};& 0<x\leq1\\x-1;&x>1\end{cases}$This seems to work and it looks so obvious now, but then it always does .
So, any ideas about the other thing?

Token
Posts: 1481
Joined: Fri Dec 01, 2006 5:07 pm UTC
Location: London

### Re: Quick question on basic topology

Well, countable choice suffices to prove that every second-countable space is separable. The converse is, as far as I can tell, not easily attacked. However, it's possible to prove [ref] that if every subspace of the reals (each such is certainly second-countable) is separable, then countable choice holds for subsets of the reals. So you're clearly going to need some form of choice.
All posts are works in progress. If I posted something within the last hour, chances are I'm still editing it.

Harg
Posts: 130
Joined: Thu Jul 10, 2008 8:24 pm UTC

### Re: Quick question on basic topology

Ooh, strange paper. Shiny! *drools slightly*

Harg
Posts: 130
Joined: Thu Jul 10, 2008 8:24 pm UTC

### Re: Quick question on basic topology

I would like to confirm this. Yes or no question:
is the topology generated by the subbasis $\mathcal{P}=\{[a,\infty)\subseteq\mathbb{R}\mid a\in\mathbb{R}\}\cup\{(-\infty,b]\subseteq\mathbb{R}\mid b\in\mathbb{R}\}$ in fact the discrete topology?

Yakk
Poster with most posts but no title.
Posts: 11129
Joined: Sat Jan 27, 2007 7:27 pm UTC
Location: E pur si muove

### Re: Quick question on basic topology

Let x be an arbitrary element of R. Is {x} an open set?
One of the painful things about our time is that those who feel certainty are stupid, and those with any imagination and understanding are filled with doubt and indecision - BR

Last edited by JHVH on Fri Oct 23, 4004 BCE 6:17 pm, edited 6 times in total.

Harg
Posts: 130
Joined: Thu Jul 10, 2008 8:24 pm UTC

### Re: Quick question on basic topology

Ah, very nice. I didn't notice this characterization of the discrete topology before. I kinda just saw that this was strictly stronger than the Sörgenfrey topology on R and thought: "Well, Sörgenfrey is pretty big, so I suppose this could be discrete." Yes, thank you.
It is interesting, though, that adding a bit of points to the subbasis elements should give something so different from the euclidean topology.

Suffusion of Yellow
Posts: 42
Joined: Thu Jan 01, 2009 6:15 pm UTC

### Re: Quick question on basic topology

Even quicker, even more basic: the closure of the set of rationals is the set of reals, and the irrationals are boundary points of the set of rationals. True?

Token
Posts: 1481
Joined: Fri Dec 01, 2006 5:07 pm UTC
Location: London

### Re: Quick question on basic topology

Suffusion of Yellow wrote:The closure of the set of rationals is the set of reals

Yes: Q is dense in R.
Suffusion of Yellow wrote:The irrationals are boundary points of the set of rationals

Yes - in fact, the boundary of Q is the whole of R.
All posts are works in progress. If I posted something within the last hour, chances are I'm still editing it.

Suffusion of Yellow
Posts: 42
Joined: Thu Jan 01, 2009 6:15 pm UTC

### Re: Quick question on basic topology

Token wrote:
Suffusion of Yellow wrote:The closure of the set of rationals is the set of reals

Yes: Q is dense in R.
Suffusion of Yellow wrote:The irrationals are boundary points of the set of rationals

Yes - in fact, the boundary of Q is the whole of R.

Cool, thanks -- my school offers "Topology is a prerequisite" courses but curiously not a Topology course.