Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

compact space

0 views
Skip to first unread message

stochastician

unread,
Jan 13, 2008, 11:10:58 PM1/13/08
to
Is it possible to define a topology on N (the set of natual numbers) s.t. to make it a _compact_ space? Thanks! Obviously the trivial topology does not work.

quasi

unread,
Jan 13, 2008, 11:30:01 PM1/13/08
to
On Sun, 13 Jan 2008 23:10:58 EST, stochastician <wuy...@gmail.com>
wrote:

>Is it possible to define a topology on N (the set of natual numbers) s.t. to make it a _compact_ space? Thanks! Obviously the trivial topology does not work.

Actually, the _trivial_ topology, where the open sets are just the
empty set and the full space, does work.

Perhaps you meant that the _discrete_ topology doesn't work.

quasi

quasi

unread,
Jan 13, 2008, 11:42:10 PM1/13/08
to

Some ideas to try ...

(1) What if you take a topology on N which has only finitely many open
sets? Wouldn't it automatically be compact?

(2) How about cofinite sets? Consider covers of N using such sets.

quasi

stochastician

unread,
Jan 14, 2008, 12:03:29 AM1/14/08
to
Thank you for ur reply! I did mean discrete topo in my prev posting. I thought about cofinite subsets of N, but it seems it's not easy to be closed under intersection? Let me try more.

William Elliot

unread,
Jan 14, 2008, 4:25:51 AM1/14/08
to

Is it possible to give N a compact Hausdorff topology?

Exercise.
Give an example of a countable, compact Hausdorff space.

José Carlos Santos

unread,
Jan 14, 2008, 5:03:57 AM1/14/08
to
On 14-01-2008 4:10, stochastician wrote:

> Is it possible to define a topology on N (the set of natual numbers)
> s.t. to make it a _compact_ space?

Take the topology for which the open sets are:

1) All finite subsets of N of which 1 is not a member.

2) All subsets of N which contain a set of the form

{1, n, n + , n + 2, n + 3, ...}

for some natural _n_.

It is compact and metrisable.

Best regards,

Jose Carlos Santos

Philippe Gaucher

unread,
Jan 14, 2008, 5:35:04 AM1/14/08
to
William Elliot <ma...@hevanet.remove.com> writes:

> On Sun, 13 Jan 2008, stochastician wrote:
>
>> Is it possible to define a topology on N (the set of natual numbers)
>> s.t. to make it a _compact_ space? Thanks! Obviously the trivial
>> topology does not work.
>>
> Is it possible to give N a compact Hausdorff topology?

Consider {0,1} with the discrete topology: it is compact
Hausdorff. Consider the product {0,1}x{0,1}x{0,1}x... (countable
product) equipped with the product topology. It is compact Hausdorff
by Tychonoff's theorem. Any bijection between
{0,1}x{0,1}x{0,1}x... and N yields a compact Hausdorff topology on N.

pg.

José Carlos Santos

unread,
Jan 14, 2008, 5:40:37 AM1/14/08
to
On 14-01-2008 10:35, Philippe Gaucher wrote:

>>> Is it possible to define a topology on N (the set of natual numbers)
>>> s.t. to make it a _compact_ space? Thanks! Obviously the trivial
>>> topology does not work.
>>>
>> Is it possible to give N a compact Hausdorff topology?
>
> Consider {0,1} with the discrete topology: it is compact
> Hausdorff. Consider the product {0,1}x{0,1}x{0,1}x... (countable
> product) equipped with the product topology. It is compact Hausdorff
> by Tychonoff's theorem. Any bijection between
> {0,1}x{0,1}x{0,1}x... and N yields a compact Hausdorff topology on N.

Well, yes. But is is more natural to take, for instance

{0} U { 1/n | n natural }

with its natural topology. It is a compact Hausdorff space. No need to
use Tychonoff's theorem here.

brieucs

unread,
Jan 14, 2008, 5:56:10 AM1/14/08
to

>> Is it possible to give N a compact Hausdorff topology?

> Consider {0,1} with the discrete topology: it is compact
> Hausdorff. Consider the product {0,1}x{0,1}x{0,1}x... (countable
> product) equipped with the product topology. It is compact Hausdorff
> by Tychonoff's theorem. Any bijection between
> {0,1}x{0,1}x{0,1}x... and N yields a compact Hausdorff topology on N.

do you mean that the product {0,1}x{0,1}x{0,1}x...
(countable product) is countable itself ?

quasi

unread,
Jan 14, 2008, 5:54:11 AM1/14/08
to

Good point.

In fact, it's clearly _uncountable_.

quasi

Philippe Gaucher

unread,
Jan 14, 2008, 5:55:36 AM1/14/08
to
brieucs <brieuc@orange> writes:

> do you mean that the product {0,1}x{0,1}x{0,1}x...
> (countable product) is countable itself ?

Oopps sorry :-). I found a compact Hausdorff topology on R, not on N.
And {0,1}+{0,1}+{0,1}+... (disjoint sum) does not work either. It is
countable but not compact anymore.

pg.

Philippe Gaucher

unread,
Jan 14, 2008, 6:10:58 AM1/14/08
to
José Carlos Santos <jcsa...@fc.up.pt> writes:

> On 14-01-2008 10:35, Philippe Gaucher wrote:
>
>>>> Is it possible to define a topology on N (the set of natual numbers)
>>>> s.t. to make it a _compact_ space? Thanks! Obviously the trivial
>>>> topology does not work.
>>>>
>>> Is it possible to give N a compact Hausdorff topology?
>>
>> Consider {0,1} with the discrete topology: it is compact
>> Hausdorff. Consider the product {0,1}x{0,1}x{0,1}x... (countable
>> product) equipped with the product topology. It is compact Hausdorff
>> by Tychonoff's theorem. Any bijection between
>> {0,1}x{0,1}x{0,1}x... and N yields a compact Hausdorff topology on N.
>
> Well, yes.

Not well :-/. Here is another example which should work (I wanted to
give a conceptual example). Consider N with the discrete topology. It
is locally compact. I am a little bit confused between French and
English terminology: I mean Hausdorff and any point admits a compact
Hausdorff neighbourhood. Then consider the Alexandroff one-point
compactification. The space N u {oo} is still countable (no mistake
this time). Then using a bijection between N u {oo} and N, one obtains


a compact Hausdorff topology on N.

pg.

quasi

unread,
Jan 14, 2008, 6:25:44 AM1/14/08
to
On Mon, 14 Jan 2008 12:10:58 +0100, Philippe Gaucher
<p...@crepe.flambee> wrote:

>José Carlos Santos <jcsa...@fc.up.pt> writes:
>
>> On 14-01-2008 10:35, Philippe Gaucher wrote:
>>
>>>>> Is it possible to define a topology on N (the set of natual numbers)
>>>>> s.t. to make it a _compact_ space? Thanks! Obviously the trivial
>>>>> topology does not work.
>>>>>
>>>> Is it possible to give N a compact Hausdorff topology?
>

>Then consider the Alexandroff one-point compactification. The
>space N u {oo} is still countable (no mistake this time). Then
>using a bijection between N u {oo} and N, one obtains a
>compact Hausdorff topology on N.

That works perfectly,

quasi

G. A. Edgar

unread,
Jan 14, 2008, 8:47:26 AM1/14/08
to
Is there a countable subset of the real line that is compact?
What is the relevance of that to your problem?

--
G. A. Edgar http://www.math.ohio-state.edu/~edgar/

quasi

unread,
Jan 14, 2008, 8:59:47 AM1/14/08
to
On Mon, 14 Jan 2008 08:47:26 -0500, "G. A. Edgar"
<ed...@math.ohio-state.edu.invalid> wrote:

>Is there a countable subset of the real line that is compact?
>What is the relevance of that to your problem?

Nice hint.

quasi

Zdislav V. Kovarik

unread,
Jan 18, 2008, 4:52:48 PM1/18/08
to

Easier yet: the range of a convergent sequence with distinct points,
together with its limit, is a compact countable subset of the real line.
Make up your own bijection.

Cheers, ZVK(Slavek).

0 new messages