It is well-known that each uncountable compact metrizable space $X$ contains a homeomorphic copy of the Cantor cube $\{0,1\}^\omega$. What about copies of Cantor cubes of larger weight?

Problem. Does every uncountable Dugundji compact space $X$ contain a topological copy of the Cantor cube $\{0,1\}^{\kappa}$ of weight $\kappa=w(X)$ ?

A compact Hausdorff space $X$ is Dugundji if and only if $X$ is an AE(0)-space, which means that each continuous map $f:B\to X$ defined on a closed subspace $B$ of a zero-dimensional compact Hausdorff space $A$ admits a continuous extension $\bar f:A\to X$.


It was proved independently by Efimov and Gerlits that the answer is yes if $\kappa$ has uncountable cofinality. In fact they proved this for any dyadic $X$ (it is well known that Dugundji compacta are dyadic). Their theorem (see "Mappings and imbeddings of dyadic spaces" by Efimov or "On subspaces of dyadic compacta" by Gerlits) actually says:

For a dyadic $X$ with $w(X)=\kappa$ the following are equivalent:

1) $X$ contains a copy of $2^\kappa$.

2) $X$ maps continuously onto $[0,1]^\kappa$.

3) $X$ is not a countable union of closed subspaces of weight less than $\kappa$.

I don't know if this can be improved at all using the stronger hypothesis of $X$ being Dugundgi.

| cite | improve this answer | |

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.