MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

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

Hello !

It's well know that any sublocale of regular locale is the intersection of a familly of open sublocale. Hence if $X$ is a regular locale, the map which to a sublocal $Y \subset X$ associate $ \lbrace o \in \mathcal{O}(X), y \subset o \rbrace $ is injective. my question is : do we know it's image, at least when $X$ is compact and regular ?

If I'm not mistaken, a subset $I$ of $\mathcal{O}(X)$ correspond to a sublocale of $X$ if and only if :

  • $u\in I, v \geqslant u \Rightarrow v \in I $

  • if $\forall i, u_i \in I$ and $u = \bigcap u_i $ as sublocale, then $ u \in I$.

So I'm wondering if there is a way to make the last condition more explicit... (or equivalently to detect if $\cap u_i = \emptyset$ as sublocale ).

Thank you !

share|cite|improve this question
Surely the word is "locale"? – Yemon Choi Sep 15 '12 at 17:58
Well, "locale" is locally "local" . . . – Noah Schweber Sep 15 '12 at 21:07
Sorry, that was a translation problem... – Simon Henry Sep 15 '12 at 22:34
up vote 1 down vote accepted

Here is some sort of characterization; it is not really satisfactory but may be useful for further improvements.

Let us talk in terms of nuclei on frames. For a frame $A$ let $\mathbf NA$ be the frame of its nuclei. Then the question is how to characterize those subsets of $A$ which are equal to $\varphi_*j:=\{\ a\in A\ |\ ja=1\ \}$ for some $j\in\mathbf NA$.

Each such $\varphi_*j$ is obviously a filter on $A$. Let further $\mathbf FA$ be the frame of all filters on $A$. Then it is straightforward to check that the map $\varphi_*:\mathbf NA\to\mathbf FA$ preserves all meets, so it has a left adjoint $\varphi^*:\mathbf FA\to\mathbf NA$.

To describe this $\varphi^*$ more explicitly, some shorthand terminology seems to be useful.

Let us say that a filter $\mathscr F$ on $A$ is $\neg\neg\textit{trivial}$ if all its elements are dense, that is, $\forall\ f\in\mathscr F\ \neg f=0$. And let us say that $\mathscr F$ is $\neg\neg\textit{trivial above}$ $a\in A$ if $\mathscr F\cap[a,1]$ is a $\neg\neg$trivial filter on the frame $[a,1]$. Explicitly this means$$ \forall\ f\in\mathscr F\ \ \ f\to a=a. $$

Now in these terms we may describe, for a filter $\mathscr F$, the nucleus $\varphi^*(\mathscr F)$ by naming the set of its fixed points, which is$$ \mathrm{Fix}(\varphi^*(\mathscr F))=\{\ a\in A\ |\ \mathscr F\textrm{ is $\neg\neg$trivial above }a\ \}. $$ This then gives the following "characterization" of filters in the image of $\varphi_*$:

A subset $\mathscr F$ of $A$ has form $\{ a\in A\ |\ ja=1\ \}$ for some nucleus $j$ if and only if it is a filter and satisfies$$ \forall\ a\notin\mathscr F\ \exists\ a\leqslant a'\notin\mathscr F\ \textrm{such that $\mathscr F$ is $\neg\neg$trivial above }a'. $$

share|cite|improve this answer

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.