Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I can't find a metric $\delta$ in $\mathbb{R}^2\setminus\{0\}$ such that be equivalent to euclidean metric, be equal to euclidean metric in the unitary circle and for all $r>0$ the set $\{(x,y)\in\mathbb{R}^2\setminus\{0\}; 0<x^2+y^2<r\}$ be $\delta$-unbounded and the set $\{(x,y)\in\mathbb{R}^2\setminus\{0\}; x^2+y^2>r\}$ be $\delta$-bounded.

I tried defining $\delta$ by cases, but is really dificult to obtain a metric in this form.

share|cite|improve this question
I don't understand the question. What does $\delta$-bounded mean? – Olivier Bégassat May 22 '12 at 22:22
@GastónBurrull $1/x$ is exactly what you need, but remember that you want to apply it to the points, not the metric. – Phira May 22 '12 at 22:37
@GastónBurrull: That is not an inversion. See my answer. – Beni Bogosel May 22 '12 at 22:42
@GastónBurrull Yes, I am sure. But you should check it, as it is quite easy to check. Also, $1/z$ is an involution that differs from the inversion $1/\overline z$ just by a distance-preserving reflection. For this kind of calculation, it is preferable because it is simpler. – Phira May 22 '12 at 22:46
@BeniBogosel Since, I didn't say that 1/x is an inversion, it is hardly an argument against my metric that it isn't. – Phira May 22 '12 at 22:51
up vote 2 down vote accepted

Consider the inversion $I : \Bbb{R}^2\setminus \{0\} \to \Bbb{R}^2 \setminus \{0\},\ I(x)=\frac{x}{\|x\|^2}$.

The define $d(x,y)=d_E(I(x),I(y))$, where $d_E$ is the Euclidean distance.

share|cite|improve this answer
Nice pullback. Ty – Gastón Burrull May 22 '12 at 22:49

Do you remember how I used the function $f$ to define the metric $\delta$ in this answer? You can use the same idea here with the function $f(x)=\dfrac{x}{\|x\|^2}$: let $\delta(x,y)=d(f(x),f(y))$, where $d$ is the Euclidean metric.

share|cite|improve this answer
Yes is a good pullback =). Ty. – Gastón Burrull May 22 '12 at 22:49
$f$ is homeomorphism, then metric are equivalent right? – Gastón Burrull May 22 '12 at 23:04
The fact that $f$ is a homeomorphism doesn’t guarantee that the metrics are equivalent $-$ we saw that in the earlier problem $-$ but in this case they are equivalent, because unlike the function in that example, $f$ is a homeomorphism with the same topology in domain and range, not just homeomorphic topologies. – Brian M. Scott May 22 '12 at 23:10
Yes, youre right, I must say if $f$ is homeomorphism the induced metric (or pullback) is equivalent with domain metric. – Gastón Burrull May 22 '12 at 23:14
Your comment was very clarificator. Is important the distinction between homeo topologies and the same topologies (equivalent metrics iff same topologies not homeo). For example in the set $\{a,b,c\}$ the topology given by $\{c\}$ and whole and empty set may be homeomorphic to topology given by $\{b\}$ but if both topologies were metrizable (probably not) metrics cant be equivalent right? – Gastón Burrull May 22 '12 at 23:39

The simplest choice is $$d(x,y)= \left|\frac 1x - \frac 1y\right|.$$

This metric is identical to the metric found by using the inversion, because the difference between the involution $\frac 1 z$ and the inversion $\frac 1 {\overline z}$ is just complex conjugation which is a distance-preserving reflection about the $x$-axis.

share|cite|improve this answer
I've proved that is a metric. Is equivalent to euclidean metric, and must be the same metric that given by the pullback of Benis post – Gastón Burrull May 22 '12 at 22:55
Ty for clarification – Gastón Burrull May 22 '12 at 22:57

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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