Consider whether a certain type $\mathcal C$ of convergence of filters, nets or sequences in a set $X$ is topologizable. If I understand correctly, "topologizable" means there exists a topology, s.t. the given convergence is the same as the convergence wrt the topology.

If $\mathcal C$ is topologizable, is the topology induced by $\mathcal C$ on $X$ the smallest or biggest topology over $X$ s.t. $\mathcal C$ is the same as the convergence wrt the topology?

To answer the above question, I would like to consider the following one. If $\mathcal C$ is the same as the convergence wrt a topology $\tau$ over $X$, and $\tau_1$ and $\tau_2$ are two other topolgoies over $X$ s.t. $\tau_1 \subseteq \tau \subseteq \tau_2$, is $\mathcal C$ the same as the convergence wrt the topology $\tau_1$ or $\tau_2$? I think $\mathcal C$ is the same as the convergence wrt the topology $\tau_1$ but not necessarily wrt $\tau_2$, right?

Thanks and regards!


Definitely not.

Look at a familiar case: let $\tau$ be the usual topology on $\Bbb R$, $\tau_1$ the indiscrete topology, and $\tau_2$ the discrete topology. In $\langle\Bbb R,\tau_1\rangle$ every sequence converges to every point; in $\langle\Bbb R,\tau_2\rangle$ the only convergent sequences are the eventually constant sequences, each of which converges to the point at which it is eventually constant. And in $\langle\Bbb R,\tau\rangle$ the sequence $\langle 2^{-n}:n\in\Bbb N\rangle$ is not constant converges only to $0$. The three topologies have three distinct families of convergent sequences.

| cite | improve this answer | |
  • $\begingroup$ Thanks! Can a given family of convergent filters, nets or sequences only be topologizable by at most one topology? Or it may or may not? $\endgroup$ – Tim Feb 24 '13 at 3:41
  • $\begingroup$ @Tim: Two different topologies on the same set can have the same convergent sequences (e.g., the discrete and co-countable topologies on an uncountable set). However, if two topologies on $X$ have the same convergent nets or filters, then they’re the same topology. $\endgroup$ – Brian M. Scott Feb 24 '13 at 4:02
  • $\begingroup$ So topologizing a type of convergence has nothing to do with fineness of topologies, does it? $\endgroup$ – Tim Feb 24 '13 at 4:09
  • $\begingroup$ @Tim: I’d not go quite that far: the finer the topology, the fewer the convergent nets/filters. But if your set of convergent nets/filters is the set of all convergent nets/filters in some topological space, then it determines a unique topology. With sequences this need not be the case. $\endgroup$ – Brian M. Scott Feb 24 '13 at 4:19

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.