4
$\begingroup$

For some the question might seem trivial but the concept is new to me and I have been wondering why there are no constant functions on $\mathbb{R}$ with compact support?

Following wiki:

Def.1) Functions with compact support on a topological space $X$ are those whose support is a compact subset of $X$.

Def.2) $\operatorname{supp}f=:\overline{\{x\in X\;,f(x)\neq 0\}}$

What if we take such function $f:\mathbb{R}\to\mathbb{R}$ defined as $f(x)=0\;,\forall x\in\mathbb{R}$ then $\operatorname{supp}f=\overline{\{x\in\mathbb{R}\;,f(x)\neq 0\}}=\overline{\emptyset}=\emptyset$ but $\emptyset$ is a compact subset of $\mathbb{R}$ I think....

Could someone enlighten me? Thank you.

$\endgroup$
  • $\begingroup$ Where were you told that no constant functions on $\mathbb{R}$ have compact support? $\endgroup$ – Hayden May 31 '14 at 18:28
  • $\begingroup$ well, anyway that's the only one there is, $f \equiv 0$ $\endgroup$ – mm-aops May 31 '14 at 18:30
  • $\begingroup$ @Hayden: here for example www-personal.umich.edu/~wangzuoq/437W13/Notes/Lec%2030.pdf (12th line) $\endgroup$ – user124471 May 31 '14 at 18:32
  • $\begingroup$ Well, I'd beg to differ with the claim, for precisely the example you gave. But as mm-aops pointed out, it's the only example. $\endgroup$ – Hayden May 31 '14 at 18:35
  • $\begingroup$ the statement is very clear : ''there are NO functions'', (such functions don't exists). The real question here is if my counter-example is really correct (i.e that I followed all definitions correctly and so on). $\endgroup$ – user124471 May 31 '14 at 18:37
6
$\begingroup$

The author in the article you link to (www-personal.umich.edu/~wangzuoq/437W13/Notes/Lec%2030.pdf) is just being sloppy. He really should have said there are no non-zero constant functions with compact support on $\mathbb R$ (and thus $H_c^0(\mathbb R)$.

So yes, your example is correct. The empty set is (tautologically) compact.

$\endgroup$
  • 2
    $\begingroup$ Hi Fredrik. Thank you. Unfortunately it is not the first time I have seen it....Exactly the same statement I found in Bott, Tu, ''Differential forms in algebraic topology''. Cheers $\endgroup$ – user124471 May 31 '14 at 18:44
  • 1
    $\begingroup$ @user124471 Yep, I've seen it as well. In fact, I was just reading in the book by Bott & Tu a week ago, and had to think about the exact same question as you asked. $\endgroup$ – Fredrik Meyer May 31 '14 at 18:47
  • $\begingroup$ @user124471 I was wondering the same thing! Now I know. Can you include Bott Tu in your question? $\endgroup$ – Selene Auckland May 29 at 11:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.