2
$\begingroup$

Let us consider a finite CW complex $X=\cup X_j$ with a given metric, compatible with the topology (maybe a reasonable one coming from some embedding into some $\mathbb{R}^n$). The characteristic maps are the maps $\phi_j^k:\mathbb{S}^{j-1}\to X_{j-1}$ for $k=1,\ldots, N_j$ which prescribe how to glue the boundaries of the $j$-dimensional cells $e_j^1,\ldots,e_j^{N_j}$ to the $(j-1)$-dimensional skeleton of $X$.

Do there always exist maps $\psi_j^k:\mathbb{S}^{j-1}\to X_{j-1}$ which are homotopic to $\phi_j^k$s and Lipschitz? If the answer is negative, in general, under which hypothesis on $X$ could this be true?

NOTES

  1. If $X$ is actually a simplicial complex, by simplicial approximation this should be true.
  2. If $X$ is a CAT(k) space, this is probably true by baricentric subdivision.
$\endgroup$
2
$\begingroup$

Let $X$ be the mapping cone of a space-filling curve $S^1\to D^2$ (here $D^2$ is the 2-dimenionsal disk). Then $X$ is homotopy equivalent to $S^2$, but I doubt that you can homotope the map $S^2\to X$ to make it Lipschitz.

$\endgroup$
  • $\begingroup$ Which metric do you chooose on $X$ ? $\endgroup$ – HenrikRüping Sep 11 '13 at 14:10
  • $\begingroup$ Do You doubt it can be done or do you know it cannot? $\endgroup$ – Samuele Sep 11 '13 at 15:34
  • $\begingroup$ I was thinking of the metric coming from some embedding in $\mathbb R^n$ (as suggested in the question). $\endgroup$ – André Henriques Sep 11 '13 at 15:41

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.