Consider the set of continuous maps $C^0([0,1],[0,1])$ equipped with the compact-open topology. It is metrisable, and therefore sequential. It is also a k-space: see http://neil-strickland.staff.shef.ac.uk/courses/homotopy/cgwh.pdf Proposition 1.6. The proof relies on the facts that every k-closed subset is in particular $\overline{\mathbb{N}}$-closed where $\overline{\mathbb{N}}$ is the one-point compactification of $\mathbb{N}$, that every $\overline{\mathbb{N}}$-closed subset is sequentially closed, and therefore the kelleyfication functor adds no open subsets in the topology. Since $\overline{\mathbb{N}}$ is not $\Delta$-generated (its $\Delta$-kelleyfication is the discrete space $\overline{\mathbb{N}}^\delta$), the preceding proof does not work for $\Delta$-generated spaces.

I am (almost) sure that $C^0([0,1],[0,1])$ is not $\Delta$-generated and I would appreciate to see a proof.

Motivation: This question is important for me because I am trying to understand specific things about the topology of the space of execution paths of a cellular multipointed $d$-space having a finite number of cells (in the sense of https://arxiv.org/abs/1904.04159). And the space above appears everywhere.

  • $\begingroup$ I have a (maybe incorrect) note that say that $C(I,I)$ (compact-open topology) is uniformly locally contractible. In particular it is locally path-connected, so should be $\Delta$-generated since it is first-countable. Feel free to correct me on this. $\endgroup$ – Tyrone Jan 8 at 13:00
  • $\begingroup$ @Tyrone I don't understand your argument. $\endgroup$ – Philippe Gaucher Jan 8 at 13:12
  • $\begingroup$ Every locally path-connected first-countable space is $\Delta$-generated. I learned this from one of Dan Christensen's papers. $\endgroup$ – Tyrone Jan 8 at 13:19
  • $\begingroup$ @Tyrone Do you have a reference please ? I will accept that as an answer by the way. $\endgroup$ – Philippe Gaucher Jan 8 at 13:20
  • $\begingroup$ It's Proposition 3.11 in The D-topology for Diffeological spaces. I don't have a proof that $C(I,I)$ is locally path-connected to hand, but that doesn't seem to hard to sort out. If you're happy I'll post an answer. $\endgroup$ – Tyrone Jan 8 at 13:23

The mapping space $C([0,1],[0,1])$ in the compact-open topology is in fact $\Delta$-generated.

The reason for this is that every locally path-connected first-countable space is $\Delta$-generated. This was proved by Christensen, Sinnamon, and Wu in Proposition 3.11 of their paper The D-Topology for Diffeological Spaces, Pacific J. Math., 272, (2014). As has already been noted, $C([0,1],[0,1])$ is metrisable, and hence first-countable. In addition it's not difficult to see that it is also locally path-connected (in fact it is locally contractible in a strong sense).


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.